What is a belief? The view from economics.

Following the efforts of James Farrell as to the many different things meant by lay folk and professionals by the word ‘belief’, I wanted to try to tackle the question from an economics points of view. Given that the methods and mindsets of economists are an amalgam of other scientists, we firstly need to review how different stereotypical scientists from various disciplines would answer this question. Before getting to the perspective of economists, this medium-sized essay will therefore first present the view of a mathematician, a statistician, and a modern cognitive psychologist (or at least how I think of them).

For a mathematician, a belief is not a well-defined concept: in mathematics land, things are either assumed, taken as proven, or yet to be (dis)proved. There are no ‘shades between’ and hence nearly all common-sense uses of the word ‘belief’ would be uninterpretable within the language of mathematics. Yet, various mathematical concepts themselves come close to what the layman thinks of as a belief, ranging from an axiom to a conjecture.

On the far certainty end of the spectrum there is the axiom, which is the undoubted premise that something is true and can be presumed to be true forever and for everything one wants to use the axiom for. To say you believe an axiom to hold means talking about an ‘inner belief’ in the sense that it is not possible to verify or refute an axiom by any outside measure: it is an article of faith stemming from revealed internal knowledge. The notion of an axiom is almost religious in content in that it is dependent of some inner revelation of truth immune to all observations. A pure mathematician does not even worry about whether an axiom holds ‘in reality’ because a mathematician thinks of an axiom as an unquestioned assumption: there doesn’t have to be any outside reality in which it holds. There is then still the mystery of why one axiom is interesting to a mathematician and another is not, but one can go through life as a mathematician without ever worrying about that. Yet, as soon as one is to take the results of mathematical inquiry as useful in any outside context, the validity of ‘translated’ axioms do matter.

The other end of the spectrum of mathematical language that comes close to the word ‘belief’ is the ‘conjecture’, such as the conjecture that Fermat’s last theorem is true (something now considered proven subject to the axioms of calculus, but long considered ‘probably true but unproven’). A conjecture is a conditional statement that is either a tautology with its assumptions (and hence true) or that is not a tautology with its assumptions and hence untrue. The interesting thing about the conjecture is that its truth only depends on assumptions already made, but it is not trivial to establish whether the assumptions encapsulate the conjecture or not. This creates philosophical distinctions between various types of knowledge. One needs a certain degree of fudge though to have any interpretation of what it would have meant to ‘believe’ that Fermats theorem was true. One would almost have to envisage the possibility of hundreds of such theorems of which some high proportion would eventually turn out to be true. Thinking of such an ex ante universe of such theories requires fudge and the statement that one believes an unproven theorem to be true is not itself interpretable within mathematics.

For a statistician, the notion of a belief has a meaning in the context of a measuring process: before the measurement of any phenomenon, one can meaningfully say that one believes the outcome to be within a certain range with a certain frequency of observations. A belief is then a kind of prediction. A properly-stated belief would be a statement of the form ‘I believe that if we measure occurrences of X, the observed values will fall between the values Y and Z at least M percent of the time if the number of measured occurrences goes to infinity’.

There are a couple of important points here. The first is that the question of what X ‘actually’ is, is a metaphysical one, i.e. not of real concern to the statistician. What matters to the statistician is how X is measured because its measurement defines X as an empirical concept. Whether god really exists is hence not a proper question to a statistician. What really matters is how god is measured. In this context, the labelling of X as god is actually arbitrary to the statistician: a statistician deals purely in the relation between measured objects without necessarily allowing himself an opinion as to the relation between measured objects and unmeasured abstractions. Another interesting point is that to the statistician, a belief is a prediction that either turns out to be confirmed by the data or not with a certain probability, i.e. in finite data the initial prediction can only be refuted or confirmed with certainty if the stated prediction was in terms of ‘all or nothing’. Any stated belief that is fractional (such as that half the number of dice thrown will show a number above 4) can only be dismissed with a certain probability, never for certain. Hence to a statistician, there is no such thing as true or false in most cases, there is only likely and unlikely.

To a statistician hence, lay-mans’ beliefs can sometimes be sensible and sometimes not. Someone who says he believes it more likely that Gillard will win the next election would be expressing a fairly well-defined belief for a statistician on which he would be able to give an ex-post confidence interval, because it will be fairly clear to the statistician what the observed future event is and how to interpret observations on it. Yet, someone who says he believes he is a better than average car-driver is not making any sense to the statistician unless he gives an empirical operationalisation of the notion of ‘better’. Someone who says he believes the circumference of a circle is always two times pi times its radius would only be making sense to a statistician if they gave the statistician the heuristic via which to measure abstract instances of this statement. Someone who says they believe the world just started 5 minutes ago is making a statement that is unverifiable in the absence of time-travel and hence the statistician would not be able to assign it any probability unless more structure was put on the statement. As a singular statement it would hence be non-sensical.

A final thing to say about statistician’s view of beliefs is that knowledge within statistics is, like mathematics, ultimately internally revealed: a statistical statement about probabilities depends on prior information. Put crudely, a probability of a certain outcome is the relative frequency with which that outcome would be observed given all the instances in which one would have the exact same starting information. The validity, presence, and interpretation of that prior information derives on unquestioned truth, i.e. assumptions. Hence the statistical world, even in the Bayesian subsection, is full of internal truths, such as ‘priors’, knowledge of the ‘sampling universe’, ‘distributional assumptions’, etc. Only within these truths (such as that an actual dice can only have 6 outcomes and is a fair dice) can one meaningfully speak of information and therefore probabilities. At the end of the day hence, the statistician’s view of beliefs is much like that of the mathematician in that the only real truth is the knowledge of assumptions and axioms that have been revealed internally, and all statements of beliefs about reality are statements conditional upon that internal knowledge. Data only exists within prior theory. An important issue that makes any real world application of statistics subject to leaps of pure faith is that the assumptions must truly be unquestionable: the dice must truly be fair and have only six outcomes (or at least the deviation must be known with certainty to be within particular bounds). The situation in which someone knows absolutely nothing with complete certainty (which is how you would often want to think of as a belief) does not allow for any statistical inferences to be made: somewhere along the line the application of statistics has to use the fudge of ‘if we assume this….’.

To a modern cognitive psychologist, the meaning of the word belief is itself a purely empirical issue, i.e. a question of what goes on inside the minds of the users of the word ‘belief’. The question of what a belief then is then subdivides into the question of how a human brain actually processes information and language, and how best to typify this process so as to relate the utterances of the words ‘belief’ with stylised representations of what might lie behind it. One goes on inside the mind of the subject and the other goes on in the mind of the cognitive psychologist explaining what goes on. By necessity, we can only talk about the latter whereas we will pretend to talk about the former, trusting to competition between cognitive psychologists to yield a useful representation. These two subdivisions are both subject to exceptionally tricky philosophical questions, such as what the nature of uttered, remembered, associated, and other forms of language is, and what an outside simplified model of internal cognitive processes ‘really’ means. Libraries have been written on both and I don’t feel I know enough about them to say much more than that the answer is essentially one of practicality.
Using the classic fudge that we lack true internal information via which to judge the findings of psychologists but pretending they are all above board anyway because we hope they will return that favour when it comes to us, there are several items of note that cognitive psychologists have come up with regarding how our minds work that are important for an understanding of beliefs:

1. Almost no normal human subject is capable of thinking in terms of probabilities. It requires an exceptionally trained mind to think of the world in entirely probabilistic terms. Hence in people’s minds, the world is either flat or it isn’t, not flat with probability p and unflat with probability (1-p). Therefore the statistical view of uttered beliefs is one that only trained individuals can relate to and they have to make a great efforts in each instance where they are asked things like ‘with what probability do you think X will occur?’. Answers to questions like ‘do you think X will occur’ roll off most people’s tongue in an instant. Yet, to the statistician the latter question is completely nonsensical and its interpretation requires a god-like knowledge as to how much probability a ‘yes’ refers to (this is a serious problem in my line of research where I ask people how happy they think they will be in 5 years time). To foreshadow the general importance of this issue for economics, you only need to reflect on the fact that many business confidence indicators simply add up the number of people who say ‘yes’ to a question like ‘do you think you will have more orders the next 3 months?’. To a pure statistician and mathematician (like Manski who has gone on about this in several papers fuming at the sloppiness of such questions), the answers are meaningless. Most economists using such data though, even if they know about this conceptual problem, simply ignore the fact that they don’t really know what they are measuring, clinging to usage value with lines like ‘I don’t care what it measures as long as it predicts something I do care about’.

2. What people believe to be true in one area can be inconsistent with what they believe in another area, without giving the least bit of bother to the believer. Hence someone who says they are a devout Catholic and take the bible literally, is nevertheless quite happy to ignore passages of the bible he doesn’t like (such as stoning unmarried couples, having foreign slaves, and the various other absurd dictates in Leviticus and other parts). When pressed, people simply refer to other passages that allow them a general cop-out on the consistency of interpretation. Such inconsistencies are entirely normal everywhere though. The economic theorist who one day writes down a model of the whole economy that presumes perfect markets and on the other day writes down a micro-model of a particular sub-market with strong market imperfections is in principle also guilty of double-dipping in that he writes down assumptions incompatible with previous assumptions and will only have the vaguest internal ‘gut feeling’ that this is somehow alright in some non-formalised way. To the true mathematician though, any application of one economic model to reality invalidates the application of any other non-nested model, meaning that to the true mathematician at least 99.99% of applied economic theory is false. If we’d adhere to that kind of rigour in reality, we might as well stop as economists. Inconsistent beliefs are thus a part of any applied science as well as normal life. In normal life inconsistent beliefs allow us to have a pleasant conversation with a person at one point in time, smoothed over by the tacit application of a truly temporarily belief in each other’s goodness, and nevertheless be ready to switch to a different belief in an instant when that person requests a loan.

3. People have mental models about the real world, ranging from what a tree is to how the economy functions, to the world of mathematics, to the motivations within our families. These mental models come complete with automated emotional responses, activated memory-areas, plans-of-actions, network heuristics that activate particular mental models in particular situations, etc. Humans make up mental models all the time and it appears to be a basic survival strategy for us to do this. Essentially mathematics and statistics are just examples of mental models. These mental models are situational though, with some mental models activated much more often than others. Hence, some beliefs are more integrated in our various mental response patterns than others, meaning we can act as-if we believe the basic rules of calculus 90% of the time, whilst only acting as-if we believe that we can’t trust our daughters with young men 30% of the time. Cues that allow us to switch between models can be subtle and occur many times per second regarding different objects. We for instance automatically scan our visual inputs for danger using mental models of what is dangerous and what is not, quite apart from the mental model we might simultaneously be using when speaking to someone. From a practical point of view, this gives a neat notion of what a belief is, i.e. a relation within a mental model. It makes it clear that a belief is only relevant in a context of mental models and decision situations. Outside of those contexts the belief does not need to have meaning.

4. From an empirical point of view, the word ‘belief’ can denote different aspects of our mental models. The statement ‘I believe Ireland is in Europe’ is hence really a statement about the mental model we have of Ireland in our memory. Note that to the statistician this is a very difficult belief to interpret because the notion that an individual only assigns a certain probability to whether Ireland is in Europe begs the question what the observation space is in which it is possible that Ireland is not in Europe (nevertheless, a person with this belief would, if trained, be able to make sensible statistical statements). There are also statements of belief that denote a sense of identity and hence that tell you individuals have incorporated a whole mental model of in which the belief is but one relation. Such statements include the belief that Don Bradman is the greatest cricketer ever or that Spain was favoured by the referee in the latest World Cup final. Such statements are not probabilistic or ‘true’ statements in any sense of the word but are rather statements signifying the adoption of a particular mental model by the utterer. In a similar vein, one can have statements rationalising former actions in order to uphold a particular internal mental model of something, such as when someone who beats up his wife says he ‘believed her to be egging him on’. Such a statement is again one of the adoption of a whole mental model, but this time complete with active distortions of own memory and appealing to outside mental models of appropriate blame. Finally, one has statements like ‘I believe the traffic light is red’ which is a statement of a perception and as such ‘true’ if the statement is made to the self. It neither means the traffic light is truly red nor that the perceiver takes it as possible that it is anything but red, but it is highly predictable of further action (i.e. the person believing it will stop) and we frequently trust our lives to such beliefs, i.e. it appears to be a very reliable form of mental modeling.

5. An interesting finding of neuroscience is that we are naturally prone to ‘believe’ what we see and hear, i.e. to doubt our own senses is something we have to learn and is perceived as abnormal (making magic tricks and cognitive fallacies a source of amusement). In turn, this means that beliefs do not really follow from an objective appraisal of information or even the objective building-up of an internal predictive mental model (as an economic theorist often would like to think and as a properly trained scientist should do). Rather, it needs an internal apparatus of doubt and self-checking mental models to prevent outside stimuli from automatically becoming beliefs. Given that anything experienced often gains credence, all kinds of things can become situational beliefs that are in violent disagreement with our own interests or already existing mental models.

Amongst economists, all the views above have their place.
Beliefs as a form of tautology dependent on unquestioned axioms have a good example in the ‘revealed preference’ literature where belief is entirely circular, i.e. ‘I believe X is the best choice for a person because that person chose it. If the person would not have chosen X, it would not have been the best’. The statement is circular because there is no outside definition of ‘the best’ and hence the statement is self-confirmatory, just like any proper mathematical statement. Truth in this context is one of the divine internal revelation of the appropriate assumptions and as such is not scientific at all. Economic theory is full of tautologies like this, such as that every situation is by definition in equilibrium. Within the internal world of models, beliefs like this are tautologies, but as soon as someone uses the associations between labels used in such models and the same words used in reality as a basis for statements about reality, one essentially treats the prior tautology as a form of divine internal revelation true everywhere (often in defiance of actual data, i.e. inconsistent with other mental models applied at the same time).

Statistical views of beliefs are perhaps best exemplified by the Bayesian community within economics that come up with Bayesian models of decision making and interpretation of data. For instance, economists who, on the basis of the data, think it more likely that higher minimum wages cost jobs than create jobs, are quite explicitly using a fairly Bayesian view of beliefs. Note that if one follows any interpretation to data to its logical conclusion, one once again is forced to rely on unquestioned prior knowledge about how to interpret observations, set sample spaces, define what a ‘job’ is and what ‘create’ means, etc.
Identity-type beliefs in economics are common when it comes to sub-tribes that organise themselves as insiders versus outsiders using certain beliefs, such as experimenters who pretend they believe that lying to students seriously contaminates the future pool of subjects and that hence papers by labs that do this should all be refused, or macro-economists who pretend they believe real agents are aware of the true model of the economy and refuse to publish papers with other assumptions. Such beliefs are invariable contradictory, but since the real driver is not the thirst for knowledge but the thirst for a successful career, this is glossed over. For instance, if economic agents would be truly equipped with the correct view of how the economy works, why bother doing economic research at all and not simply step outside and ask the person at the bus station what GDP is going to be next year? The very activity of economic research is hence inconsistent with the belief that agents in the economy act as-if they know what is going on, but because we find it too hard to come up with the model of everything, we muddle on ignoring such inconsistencies but still use particular beliefs to keep others away from our table.

Ex-post rationalising beliefs are common when it concerns historical events, such as the belief that the Great Depression was made worse by the gold standard, or that the 2010 introduction of the mining tax in Australia was not presented in is most positive light.

Observational beliefs are prevalent in applied economics, such as when economists believe they see a tragedy of the commons when it concerns fishing in international waters.

Memory beliefs include the belief that the economy of the Roman Empire was dependent on an increasing set of territories to supply new slaves or that Australia’s economy grew faster during the GFC than the OECD average.

The potential criticisms and observations made on these types of beliefs for other sciences carry through in the case that they are held by economists: the beliefs are situational (non-transitive), and if you follow them up invariably rely on the internally revealed structure of our mental models formed over lifetimes. In the absence of the mental model of everything, it is not clear what any of those beliefs are worth ‘in reality’ and indeed, notions of falsification, probability, and verification are not really applicable to them in any clean sense. All we have to go on are heuristics that have been seen to work in particular areas, but of which it is not clear that they are all that useful in economics. For instance, the practice of challenging any assumption popular enough to get published on the basis that they fail to perfectly predict what happens in a lab might be useful in many areas, but in economics it is not clear it buys us anything. Perhaps most interesting is that the adoption and evolution of economic ideas is linked to the adoption and evolution of mental models. The evolutionary drivers of that race are not just inter-subjective agreed upon notions of verification and falsification defining ‘scientists’ but also include whether they are inherently appealing to students in the market for additional mental models (i.e. internal success), how easily they fit onto existing mental models of anyone who hears them (whether useful or not) and of course whether they help the adaptor survive and procreate in both a literal and career sense. Theories of the evolution of scientific thought are precisely about the different directions these driving forces go into.

After this quick tour of various sciences, what is then the ‘best’ answer to the question of what a belief is? My best answer, which is a mental model in itself, is that a belief is a relation within an internal mental model. That makes it by definition situational and only unquestionably true within the world of that internal model. The various forms and limitations of beliefs come from the various forms and limitations of mental models. On the whole, they do not relate to truth or probability as a one-to-one mapping. Only for highly trained individuals can they become related to truth or probability. Also, beliefs never stand alone and are mere parts of the mental model they are an aspect of, complete with action-plans, memories, associations, etc. To treat beliefs as separate entities is like looking at windscreen wipers without thinking about cars. The problem in trying to look at cars is that everyone has a different car.

This entry was posted in Uncategorized. Bookmark the permalink.

64 Responses to What is a belief? The view from economics.

  1. conrad says:

    If you want to make your question even more convoluted and complex, then try and explain how false-beliefs fall into the belief spectrum (it isn’t just the opposite of a belief, since that of course would assume beliefs are statistical or have definable truth values). If you’re interested in that question (and it is an important question if you’re interested in things like what a delusion is and how one might classify them), then you can add another area in another field which cognitive psychology has taken from — attribution theory in social psychology.

  2. Nicholas Gruen says:

    Thx for the post Paul. Makes me think of how we all have to be foxes rather than hedgehogs when it comes to understanding our world. Even a statistician has to understand a lot, have a lot of informal knowledge and judgement to be a good statistician. I guess that may not be true of a mathematician, but it’s true of an incredible range of things, from intellectual things like ALL the social sciences and many of the sciences and living life itself.

    It’s things like that, that mean that my hat’s in the ring with the pragmatists as far as a philosophy of life or of knowledge is concerned. It’s also why I describe myself politically as influenced by an amalgam of political traditions – conservatism, liberalism and social democracy.

    It also explains my passionate weakness of character. As Robert Solow says “Sometimes I think it is only my weakness of character that keeps me from making obvious errors”.

  3. SJ says:

    After this quick tour of various sciences, what is then the ‘best’ answer to the question of what a belief is? My best answer, which is a mental model in itself, is that a belief is a relation within an internal mental model. That makes it by definition situational and only unquestionably true within the world of that internal model. The various forms and limitations of beliefs come from the various forms and limitations of mental models. On the whole, they do not relate to truth or probability as a one-to-one mapping. Only for highly trained individuals can they become related to truth or probability. Also, beliefs never stand alone and are mere parts of the mental model they are an aspect of, complete with action-plans, memories, associations, etc. To treat beliefs as separate entities is like looking at windscreen wipers without thinking about cars. The problem in trying to look at cars is that everyone has a different car.


    Some beliefs are testable and require evidence, and some aren’t and can’t.

    Your analysis necessarily puts belief in say, the roundness or lack thereof of the earth (which is easily measured), in the same category with belief in magical sky pixies (which don’t exist, as far as we can tell).

  4. Rafe says:

    “Some beliefs are testable and require evidence, and some aren’t and can’t”.

    Right on! What about shelving talk about beliefs and focus on theories and policies? This calls for a major shift in the focus of epistemology as it has been traditionally practiced.

    “This [traditional approach] has led students of epistemology into irrelevancies: while intending to study scientific knowledge, they studied in fact something which is of no relevance to scientific knowledge. For scientific knowledge simply is not knowledge in the sense of the ordinary usage of the words ‘I know’. While knowledge in the senses of ‘I know’ belongs to what I call the ‘second world’, the world of subjects, scientific knowledge belongs to the third world, to the world of objective theories, objective problems and objective arguments…Thus my first thesis is that the traditional epistemology, of Locke, Berkeley, Hume, and even of Russell, is irrelevant, in a pretty strict sense of the word. It is a corollary of this thesis that a large part of contemporary epistemology is irrelevant also.” (Popper, 1972, p.108).

  5. SJ says:

    I really can’t understand why Popper is a favorite of creationists and libertarians.

  6. Nicholas Gruen says:

    Libertarians I can understand why Popper’s a fave. Creationists? What Creationists are Popperians?

  7. Nicholas Gruen says:


    I think I agree with the Popper quote you have cited. Yet though Popper’s own work might have advanced some philosophical understanding of science (though I don’t know how it takes us beyond the pragmatism of C.S Peirce – perhaps you can enlighten me?), Popper’s work doesn’t help scientists practice science either.

    The science Popper had the greatest impact on was economics and there Popper seems to have enjoyed his fame and esteem in the eyes of economists without it seems letting on to anyone that the falsificationism that was being preached in his name was a crude anti-metaphysical positivism, which Popper claimed later to abjure (and did abjure at least formally in his response to the Vienna circle as a young man).

    The effect on economics of what I presume was Popper’s encouragement of his disciples in economics was to assist the disastrous slide of economics into obsessive compulsive formalism. Strange in a way with Hayek sitting alongside. I wonder what he was making of it all in the 1950s as Popper’s star rose with the likes of Samuelson and Friedman while he was republishing his essays from the late 40s in his polemic against scientism and positivism The Counter-Revolution of Science.

    A very interesting story don’t you think?

  8. conrad says:

    “What about shelving talk about beliefs and focus on theories and policies? ”

    Because if you want understand why and how people do things and think about things, then you need to understand what a belief is, and for most things, this is an entirely different question to saying, in SJ’s words, “Some beliefs are testable and require evidence, and some aren’t and can”. SJ is confusing scientific testing of theories with the day-to-day mental models people use for almost everything.

    For example, lets say you want to put a cover on a high-level building to stop people dropping things and endangering others. Now, let’s also say most people think only heavy things drop quickly, as they did for most of history (they probably still do in some places on Earth). Your problem now is not the fact that we know both heavy and light things drop at the same speed, it’s convincing other people to change their beliefs so you can install your cover, which they now think is a waste of money. Now convincing them of some scientific fact like this may be easy (because you have great evidence, and you don’t need to know why or how they think in this case), but it will be made more difficult for things that are much more fuzzy, as they are in the social sciences, because your model of your problem may be relatively ill-defined (and hence less convincing), so just telling people your model is the best won’t work. If you don’t believe this, try telling a compulsive gambler something only slightly less complicated than the dropping-example, that they will lose at the casino. Now it’s statistically very likely they will, and you can prove this. But this doesn’t matter — what’s important is understanding what they believe and why, since you need to change those beliefs to stop them gambling, not the fact that to you it’s obvious they’ll lose.

  9. Rafe says:

    Nicholas, those are good questions and I will address them over the weekend.

    Conrad, the question is whether all the talk about beliefs and the justification of beliefs by philosophers contributes to solving the problems that you pose. I think the answer is no.

    Insttead we need to critically discuss alternative theories about learning (or how to enforce worksafe codes) and how to encourage people to be more critical and analytical about the ideas they have picked up.

  10. conrad says:


    beliefs contribute massively in many areas — the gambling example is a classic, which is why I gave it (since the beliefs are irrational if the goal of gambling is to make money). That being said, the psychology idea of beliefs is different to the philosophy one (although certainly not orthogonal). However, if you want a good example where philosophers have helped our understanding of beliefs in the late 20th century, then there are a string of papers looking at the extent that you can infer meaning (and hence beliefs) by statistical inference from the sub-components of the beliefs. As it happens, you can’t in many situations (indeed, there are cases where more information should make it harder to get to the correct meaning based on decompositional analysis, but it doesn’t), and that’s an important observation, since it means that trying to pursue decompositional frameworks is going to be pointless in many situations.

  11. SJ says:

    Creationists? What Creationists are Popperians?

    Try googling “popper evolution”. The creationists seem to think that the theory of evolution is unfalsifiable, and thus groundless, unscientific conjecture. (Not that this is really relevant to the present discussion, I’m just answering your query).

  12. Rafe says:

    Popper has some appeal to libertarians on account of his “non-authoritarian” theory of knowledge but they are very upset by the social democratic elements of his social philosophy.

    The creationists make a big deal out of Popper’s theory of conjectural knowledge, meaning that even our most impressive theories are liable to be challenged by new evidence, new arguments or a newer, more expansive theory (like the challenge of Einstein to Newton). So they say, theories are JUST theories, and our theory is another theory just like Darwinian evolution, so it should be taken equally seriously. However Popper also pointed out that even though you cannot justify a belief in the certainty of any particular theory, YOU CAN JUSTIFY A CRITICAL PREFERENCE for one theory over another in a head to head contest if one predicts and explains more, stands up to tests, generates a productive research program, integrates different fields etc.

    Moving on to Popperism in economics. This needs to be treated at book length, so I had better stick to the executive summary here.

    The main point is that the Popperism that was discussed ad nauseum for two or three decades in the academic litature by the likes of Blaug, Hands, Hausman, Maki, Backhouse, Caldwell, de Marchi, Hoover, Latsis, McClosky, Mirowski, Weintraub, was actually the degenerative turn taken by Lakatos in his attempt to combine Popper and Kuhn. I think it is fair to say that virtually nothing has emerged from that literature than is helpful to working economists.

    The more helpful commentators are Larry Boland, Jack Binner, Boland’s student Stanlay Wong, and Bruce Caldwell after he recovered from sticking his head into too many positivist scrums. However they have been ignored and swamped in the literature by other voices. This is because Popperism is either ignored or wildly misrepresented in the US universities and in dozens of introductory philosophy books.

    Strangely, both Friedman and Samuelson are supposed to have taken on board the Popperian idea of testability but I can’t see it, and the best critique of Samuelson was provided by Wong, as explained by Mirowski.

  13. Nicholas Gruen says:


    I agree with you that the Popperianism explored in so much economic methodology from the 70s on is degenerative. But it was harmless enough. The real nonsense began with the Pigmy Popperianism of Friedman and Samuelson.

    My point is that Popper gives every sign of being thrilled by all the attention. Can you show me one place where Popper says in some serious way that causes people to think carefully about what they were doing, that they are not really following his philosophy?

  14. Rafe says:

    I don’t know how people got the idea that Friedman and Samuelson were influenced by Popper. I have seen it suggested in the case of Friedman but not Samuelson and his influence is more obviously the logical empiricists who dominated philosophy in the US before Kuhn and the Continentals offered more exciting fare.

    I don’t think he took any notice of the philosophy of economics literature because physics was his first love. He turned to the social sciences as a kind of community service during the 1930s to do something about the ideas which he thought underpinned totalitarian ideologies. Later biology became more prominent and the 1970s and early 1980s were taken up with “Objective Knowledge: An evolutionary approach” (1972), the intellectual autobiography for the volume in the Library of Living (but almost dead) Philosophers (1974) and hundreds of pages of replies to critics in that volume, then “The Self and its Brain” with Eccles (1976) and then the final dash to get “The Postscript to the Logic of Scientific Discovery” into print after standing in galleys since the 1950s. This appeared in three volumes in 1982 and 1983. He had a heap of ms for a book to be called “Philosophy and Physics” but I think he gave up on that book to get The Postscript finished.

    Because he regarded the ideas of Lakatos as a travesty of his own he would have regarded most of the literature on MSRP as a waste of space, similarly the depictions of his ideas by the likes of Hands, Hausman, Maki, Backhouse, de Marchi, Hoover, McClosky, Mirowski are a caricature of his position.

  15. Edward Mariyani-Squire says:

    [1] Paul on Belief

    a relation within an internal mental model

    A prima facie problem with this arises out of the question: the relation between what and what? This is not to deny that believing is not a relation; I can readily understand believing to be a relation between a person and some object of belief, but if it is a relation that is always and only internal to the person then that would seem to give us subjective idealism.

    [2] Paul on Mathematics and Statistics

    You seem to be relying heavily on constructivism for your conception of mathematics and conventionalism-operationalism for your conception of statistics. There is a realist school which is opposed to these conceptions.

    [3] Rafe on Unimportance of belief

    One can agree (with Popper and yourself) that belief is, in the long-run, an unnecessary condition for the existence of knowledge. But that doesn’t negate its important role in the short-run for the generation of knowledge.
    For example, beliefs can serve as general motivators of action – ‘I believe that the acquisition of knowledge will help humankind’ motivates people to seek knowledge. It also motivates policy recommendations – ‘I believe that this policy will reduce unemployment … if I didn’t believe that, I wouldn’t advocate it.’

    Also, some ‘presuppositional’ beliefs are necessary for making conjectures in the first place. For example, if one didn’t believe there was some unknown cause of X, would one bother of put forward a testable conjecture about what is causing X? And the same goes for testing: if one didn’t believe that thermometers accuracy measure temperature, would one test a conjecture about the temperature of X using a thermometer?

    [4] Rafe on Popper and Friedman

    Friedman met Popper at the Mont Perlin Society. According to Friedman, they discussed methodology. It is possible that some lines in Friedman’s “Methodology of Positive Economics” are due to that conversation, but this is sheer speculation. It is not clear that Friedman actually read anything by Popper. In an interview with Daniel Hammond, he was able to name “Conjectures and Refutations”, but on the face of it, seems to have been woefully ignorant of its content.

    [5] Nicolas on Samuelson and Popper

    There’s no evidence that Popper had an influence on Samuelson in the 1940s. Samuelson’s methodological comments from that period are closest to E. Mach’s radical empiricism. This comes out fairly clearly in Samuelson’s response to Friedman’s “Methodology of Positive Economics” and in his subsequent rejoinders to critics of his position.

  16. Edward Mariyani-Squire says:


    he [Popper] regarded the ideas of Lakatos as a travesty of his own

    Do you have a few quotable quotes from Popper where he explicitly expresses this opinion? Ta.

  17. Rafe says:

    Thanks Edward, I forgot about the MPS meeting.

    I will get back on the quotes, the most likely source is Popper’s reply to Kuhn’s criticism in the Library of Living Philosophers volume.

  18. Paul Frijters says:


    the question of how the internal mental model relates to the outside world is something not talked about above, apart from the sentence about mental models being a survival strategy and hence, ultimately, a tool. However, it was quite explicit above that mental models need have no outside relevance or descriptive powers. Hence a belief need be no more than a relation in a pure stand-alone construct. I would definitely categorise much of mathematics in that.

    As to realism, as soon as one takes some form of measurement-come-theory as more less unquestioned (like our senses), you can of course start to weave other things from that. Somewhere you need that initial leap of faith though.

    Edward, Rafe, etc. (on philosophy),

    I have a question for you: why study the words of philosophers when it comes to beliefs when such things are now part of the scientific enquiry of thousands of cognitive psychologists and neuroscientists? Whilst they are happily charging on in many exciting directions, the mutterings of the Poppers of this world seem to me no more than amateurish in comparison. Why bother? For my money, the study of beliefs has followed the study of many things once considered part of philosophy: it has come to fruition as a question in applied science.

  19. conrad says:

    “why study the words of philosophers when it comes to beliefs when such things are now part of the scientific enquiry of thousands of cognitive psychologists and neuroscientists?”

    I’m not into older philosophy nearly as much as Rafe so he might be able to defend those guys better than me, but I might be able to legitimately profess some knowledge of modern day cognitive psychology and neuroscience (indeed my first journal article was on semantics, although it’s too hard for me — let alone trying to explain how to get to meaning from there — so I don’t do much in the area anymore). At least for modern day philosophy, I don’t think what’s happening is orthogonal to what’s happening in other areas — indeed, you get the occasional philospher publishing in cognition journals now and then and vice-versa. Generally it’s just all considered cognitive science these days, although you still get alphabet soup papers from philosophers at one end of the specturm, and there are a lot of purely experimental papers that don’t even consider the philosophy of it all at the other (think of all of those papers about judgement bias, many of which now seem to get classified into the behavioral economics category). The reason modern day philosophy is important is that I think it reasnoable to say that if you a wanted to create a really good theory of beliefs that wasn’t just a taxonomy of behavior (what are they? where do they come from? How do they predict actions in different contexts? How do they go wrong? How does they relate to other human cognitive functions? etc.), you would at least have to consider some of the more philosophical arguments.

  20. Mike Pepperday says:


    I’d like to hear Rafe’s take on “C S Pearce” and Popper but you mean C S Peirce (e before i and pronounced “purse”), do you not?

    I’ve never connected him with Popper. In a context of belief, Peirce once said

    “It rather annoys me to be told that there is anything novel in my three categories; for if they have not, however confusedly, been recognized by men since men began to think, that condemns them at once.”

    Essentially, his three categories were your “conservatism, liberalism and social democracy.”

  21. Nicholas Gruen says:

    Thanks Mike,

    I realised my error in the shower. Embarrassing. And corrected it this morning – now I find you had outed me half an hour before removing the evidence – or so I thought. I did know it was spelled funny – but still had to look it up when I fixed it. Didn’t know he answered to the name ‘purse’ so thanks.

    I don’t really know what people like Popper added to Pragmatism’s understanding of how science worked, but there is certainly a lot of ink spilled over it.

    My take on Popper FWIW is that he’s ‘naturalising’ non foundationalism for people all of whose instincts are foundational. For pragmatists and various other post-Hegelian philosophies non-foundationalism was not such a big deal, and they moved on to other topics. Popper’s agenda was basically authoritarian – to provide a way of demarking science from nonsense. Not so different to the Vienna Circle’s ambitions, but he at least realised that the Vienna Circle’s logical positivism was (as I like to say ;) like the Titanic – the unsinkable that sank on its maiden voyage.

    So he laboriously works his way back to roughly where the Pragmatists were at the end of the nineteenth century. That’s the way I see it anyway.

  22. Rafe says:

    Popper went to great pains to explain that he was not concerned with demarcating science from nonsense. He wanted a handy indicator as to whether evidence was going to be used as one of the forms of criticism of theories, along with internal consistency, problem-solving capacity etc.

    There is huge affinity between Peirce, Popper and also Dewey, and a debate between them would have been a joy but Popper lived after the other two and did not have the chance to talk to them as he pressed on with some of the technical problems in quantum theory and also probability theory. He also collaborated with John Eccles on a book called “The Self and its Brain”.

    Something strange has happened to Peirce and Dewey, judging from the email discussion groups dedicated to their thoughts. I gave up on the Peirce group after it was taken over by followers of people like Deluze (a French POMO) under the guise of taking up Peirce’s early work on semiotics.

    The Dewey group is almost defunct, with about five messages a month, but over the last five or six years the main topics have been things like the extent of influence of Christianity on Dewey’s thought. I would have hoped that admirers of Popper, Peirce and Dewey could form a common front against the logical empiricism that is still dominant in the US, challenged by various brands of POMO. This book seems to give some of the flavour.


    This anthology contains over 80 extracts from the literature of science and philosophy from Democritus (circa 400 BC) onwards. Part I covers the ancient and medieval periods, the scientific revolution, the modern philosophers (Bacon to Kant), then methodology and revolution (Lavoisier to Einstein). Part II contains the received (positivist or logical empiricist) view, mostly Carnap and Hempel, confirmation and observation (more Hempel and some others), then the revisionists in methodology (Popper, Kuhn and Lakatos), explanation (mostly Salmon) and the realism debate (Boyd, van Frassen, Laudan and Fine).

    From my perspective it is scandalous that logical empiricism is still the mainstream!

  23. Edward Mariyani-Squire says:


    In response to:

    [1] Hence a belief need be no more than a relation in a pure stand-alone construct.

    [2] why study the words of philosophers when it comes to beliefs when such things are now part of the scientific enquiry of thousands of cognitive psychologists and neuroscientists?


    When one says ‘relation’, one must be presupposing at least two entities – those two entities being related to each other. So, again my question is: if a belief is a relation, what two or more entities is it a relation between?

    The notion of belief you are outlining is, as far as I can tell, a good example of this. It seems to draw upon theorisation by a large ‘cognitive science’ school in current psychology. Basically this school hypothesises the existence of ‘mental models’ in the brain that are supposed to be internal representations of external phenomena. Your statements suggest a variation on this – the big difference being that the internal representations are not even related to external phenomena: a belief-statement that is apparently referring to something external (‘Ireland is in Europe’) is really referring to a mental model (the idea ‘Ireland’ that has a set-relation to the idea ‘Europe’). That mental model is not a representation of something external (geo-political areas on the globe), however. It is an apparently free-floating construct, “a walking shadow, a poor player that struts and frets his hour upon the stage … full of sound and fury, signifying nothing.” The plain, and I think reasonable question, is this: if ‘believing’ is not a relation between a person and some entity external to the person, then what possible grounds are there for thinking that there is anything external at all. How does this avoid collapsing into solipsism? On this view, how would – how could – anyone believe, in a non-delusory manner that there is anything at all except one’s own mental model?


    Why worry about philosophical issues? Because if one does not think carefully about such matters as philosophers of the past and present have done (do), one may make basic conceptual mistakes that have already been pointed out, and a hundred thousand unreflective empirical studies will be produced that are either deeply confused or a waste of resources.

    For example, if one were interested in the arguments of dead philosophers, one would know the above problem of mental models has been played out already (in the 18th century). John Locke claimed the mind consisted of mental representations of external phenomena in the world, George Berkeley objected that if one is only acquainted one’s mental representations, then one couldn’t possibly establish there was anything ‘beyond’ them which they allegedly represented. But if that is the case, then how does one know there is anything other than one’s own mental representations? Thus, the absurdity – and bad faith – of solipsism (unless, say, with Berkeley, one somehow apprehends that there are entities beyond how own mental representations – namely, all those things ‘observed’ in the mind of God).

    For a more elaborate answer to why philosophical reflection is important to psychology, Nigel Mackay and Agnes Petocz in their “Realism And The State Of Theory In Psychology” (Draft) give a good extended answer. To give a flavour of the essay:

    What should be evident from even this brief survey is the extent of psychology’s theoretical and metatheoretical disarray. Although a certain amount and kind of variety is healthy and would not be a problem, the theoretical disarray in psychology is accompanied by inconsistencies and conceptual confusions which undermine psychology’s efforts to advance. These have not gone unnoticed and, in addition to the long standing debates over the disunity of psychology (e.g., Sternberg, 2005), each group has been criticised for its theoretical limitations and for its inability to provide a coherent metatheoretical framework for psychology2. Wittgenstein (1953) claimed that psychology’s “confusion and barrenness” were to be attributed not to its status as a “young” science, but to its odd mixture of “experimental methods and conceptual confusion” (p. 232, emphasis in original). The mainstream groups, including cognitive science, have been attacked for their misconceptions of science, their pseudoscientific methodological practices, their misunderstanding and misuse of the various data-analytic techniques at their disposal, and their implicit adherence to aspects of the Cartesian dualism which they explicitly reject (Bennett & Hacker, 2003; Bickhard, 1992; Haack, 2003). The nonmainstream alternatives, in turn, have been accused of sharing the mainstream’s misconceptions of science, and of offering inconsistent foundational alternatives (Greenwood, 1992; Hibberd, 2005a; Michell, 2004).

    It will be argued that realist critiques have lessons for the casual, anti-metatheoretical complacency of mainstream empirical psychology, which covers so much of its conceptual confusion with sheer empirical effort. It will also be argued that the information-processing and the cognitive neuroscience establishment operates with a number of deficient concepts and consequently proceeds with hamstrung theories and a misdirected research program. The deficiencies arise because the treatment of a number of key concepts in psychology (e.g., cognition, motivation) fails the standards of science-despite the appearance of scientific soundness, because they in turn rest on incoherent epistemological and ontological theses. If the underlying conceptual theses are flawed and inconsistent—and it will be argued that a number are—then the theories and research that depend on them will be similarly flawed. The essays also contain discussion of the ideas involved in several alternative nonmainstream programs (constructivism, constructionism, situated cognition and so on) that in perhaps worthy attempts to rewrite psychology in noncomputational terms, to recognise the embodied nature of cognition, or to incorporate meaning, do so hampered by strains of a similar antirealism. And behind this (and in spite of recent explicit attempts to free psychology from Cartesianism) is the constant and powerful pull of the Cartesian concept of mind, a pull that for centuries, from Descartes himself, Locke and the empiricists, through Reid and Kant to modern cognitivism seems to have defeated all attempts to escape its grip.

  24. Mike Pepperday says:

    Thank you Edward. I was looking at Paul’s

    “why study the words of philosophers when it comes to beliefs when such things are now part of the scientific enquiry of thousands of cognitive psychologists and neuroscientists?”

    and thinking that it was pollyanna-ish. As far as I can tell, after a century of furious activity psychology has achieved nothing and knows nothing. You have spelt it out. “covers so much of its conceptual confusion with sheer empirical effort” Spot on! Thousands of PhDs, thousands of articles, thousands of academic careers, millions of factor analyses ? and not a useful fact in sight. So philosophers’ words still count ? if you don’t attend to them you are in danger of reinventing their (square) wheels.

    fMRI testing is becoming less expensive and this may be different but it is a question as to whether it is really psychology. It might be a bit like trying to understand water by investigating hydrogen and oxygen atoms. Still, certain areas of the brain have been isolated as concerning cooperation and certain others light up when subjects engage in competition. That seems like a start. I don’t know if they have figured out a way to ethically test for coercion but the last time I looked they were thinking about it.

    What I can’t fathom is why the social sciences don’t adopt the natural sciences approach of theorising through idealisation. Galileo had a perfect sphere touching a perfectly flat and frictionless plane in a single point. He was frank about the fact that this does not exist in nature. Nature gives us avalanches but it is the theory constructed from the ideal which allows us to comprehend avalanches. The one social science that does take this approach is economics with its rationality and market clearing and so on. Given the success of the hard sciences (where theorising using ideal-types seems universal) and the relative success of economics, it is baffling to me why idealisation is not the standard approach in all social science.

    There are no units of measure in social science so the only idealisation is all or nothing ? rationality and market clearing are there or they are not ? and the fMRI testing has taken cooperation and competition without qualification which seems right. But what theory are they following? None that I know of. Why doesn’t psychology theorise perfect cooperation and total competition and absolute coercion (and corresponding extremes of a few dozen other concepts)? It seems the obvious thing to do.

  25. Paul Frijters says:


    thx for the reply (and sorry for not looking at the name close enough). On [1] I will merely reply that I see room for mental models with or without an outside reality which it tries to comprehend. Hence I dont really see that we disagree in any serious way on that one.

    As to the use of philosophy, we are now talking about a broad field and I have no wish to poopoo all of philosophy. Moral philosophy for instance is almost impossible to replace by something else and the advantage of the crowd of people who use words like ‘epistemological’ and ‘metatheoretical’ is that it keeps the rest honest as to the limits of what they can know and provides a monestary-lik atmosphere for those who seek it out.

    However, on the topic of beliefs [2], the passage you quote is basically one of a smug psychology-bashing. That’s a bit disappointing and there’s not much concrete to put your teeth into. Psychology also is a big church housing plenty of people who have no idea why they are there, but being there pays the bills. Economics too gets that kind of pedantic ‘we know what you should be doing whilst we are using words most of you have never heard about’ essay thrown at it all the time. I think I prefer Mike Pepperday’s kind of critique: he asks why a field doesnt go in for a certain technique (to which the answer is that at the top, this idealisation-plus-implementation is of course what people try to do but so far with limited success).

    Hence I ask again: in what sense is the divorce between cognitive science and its philosophical roots (under apparent jeers of the philosophers) any different from previous divorces that lead to economics, biology, medicine, and physics? What, apart from some following an impossible prescript, would these cognitive scientists actually do differently if they spent a few years reading about the Poppers of this world? And dont get me wrong, I read more than my fair share of philosophy, time I now mostly consider a form of conspicuous leisure.

  26. Mike Pepperday says:


    “this idealisation-plus-implementation is of course what people try to do but so far with limited success”

    Who does this? Economics specifies homo economicus and is interminably criticised for being unrealistic. Which it is. Which it has to be. It’s an ideal-type and useful theorising needs unrealistic extreme ideal-types.

    What unrealistic concepts does psychology use? Who are the angry critics who object to psychology (because they hate its conclusions)? None; nobody. Psychology does not adopt extreme concepts and has no conclusions to excite criticism. Compare with the airplay “evolutionary psychology” gets. It gets it because it actually has something to say.

    I see no difference between the divorce of psychology from philosophy and the divorces of other sciences except that psychology has failed. What would psychologists do different if they read the philosophers? Well, I don’t know if Popper explained theorising with ideal-types but Weber certainly did. And Mach, Kaufmann and Hempel.

    The fMRI experimenters have located cooperation and competition in the brain. Wherever did they get the idea of looking for these things? Is there such a thing as “cooperation”? Which school of psychology sets it out? What can these neuroscientists be thinking of, looking for psychological phenomena that academic psychology is unaware of?

    So there is no “of course”. And nor is it “idealisation-plus-implementation”; it’s idealisation-plus-deduction. And then it’s empirically testing the deduced results. And then if you’re as persuasive as the famous economists, there’s implementation.

  27. Paul Frijters says:


    I beg to differ. A number of cognitive theories, I think, have been very successful and do have an ideal-type flavour to them. Categorisation theory is a good example. If you’re looking for a clear success in neuroscience, then the implants to stop epilepsy or restore hearing come to mind. Hard to see how they could have been achieved without mountains of empirical work.

    The number of theories that include ideal types and havent made it big are very numerous. Freud’s theory of the id, theories of imprinting, theories of roles in groups, the notion of cognitive development stages, the list goes on and on. It is not for lack of thinking in terms of ideal types that have lead to only partial success in psychology. The problem is that you cant isolate traits within individuals: they are all interconnected.

  28. James Farrell says:


    The term ‘thoughtful post’ has unfortunately been overused and debased to the point that it’s no longer adequate for praising a post like that one. You should offer it as an entry in some encyclopedia somewhere.

    I came to the issse originally because I was struck by the difference between two kinds of things we call beliefs, in the context of religion:

    1. Those that resemble my belief that I’m not adopted — based on strong but not watertight evidence, and having earth-shaking potential consequences if falsified. Later this week I’m attending the funeral of an ex-Adventist friend who suffered a profound crisis on learning that Ellen Whote’s divinely inspired writings were plagiarised.

    2. Publically held positions on propositions over which achieving certainty is pretty much ruled out, either due to the limits of science, or because they can’t be falsified even in principle.

    I think your essay will help, but I’ll need more time to digest it.

    The question whether ‘normal’ people form probabilty estimates is intriguing. I don’t qualify as a ‘highly-trained individual’, but I’ve only recently become aware that I have a mental habit of forming an explicit probability estimate of almost everything. If we’re expecting an important letter, I’ll say to my wife: ‘My subjective probabilty estimate that it will arive today is about 35%. What’s yours?” She” reply that it wouldn’t even occur to her to quantify it.

    I wonder what be the consequences of a world where everbody formed complex probability estimates vis. a vis. one where no-one did.

  29. James Farrell says:

    Ellen White’s divinely inspired writings

  30. conrad says:


    if you want a good example of where philosophers are helping cognitive science people, and you’re interested in categorization and meaning, then this is the type of thing philosphers are up to these days. There are some useful points in there — if one wanted to construct a mathematical model of categorization, for example, some of those points are well worth thinking about.

    Mike, you are far too positive about fMRI testing. I’ve published papers using fMRI, and my opinion is that it’s told us suprisingly little despite all the money that is poured into it — and the reason is obvious — it’s limited in the temporal domain (amongst other things), but much of human information processing occurs extremely quickly. For example, if I was interested in how you combine two words “brown” and “cow” to form brown-cow, I’d know that much of the processing occurs within < 100 ms. fMRI is simply not sensitive to things at this speed (MEG is better). The other problem is that if you don't have an apriori theory, then you're just looking at activation differencs in different anatomical areas. That's nice, but it's like telling you that there are things called supermarkets that sell things, and ending there, versus telling you how they work etc. ., and with current technology, there's no way to get to the "supermarkets work like X, Y, & Z" answer (or the cognitive brain-problem equivalent), due to fundamental limitations of the technology. Pretty much the best you can hope for is to learn about some processing relationship between two grossly different anatomical areas.

    On that note: here's a good question the guy who I did my post-doc with used to ask the fMRI people: "Apart from anatomical localization, can you tell us something that fMRI has told us about language and language function that we didn't already know?". Despite the fact we know an enormous amount of stuff about language, you'd surprised by how few things people can come up with. Many can come up with none — this is because most of the work with fMRI looking at higher level cognitive functioning gets a well known cognitive phenomena, and then simply looks for anatomical correlates, which, in the end, often isn't very interesting, at least compared to the cognitive models which are around, which can predict result down to the item level in some domains (i.e., if I give you item X, you will respond like Y).

  31. Nicholas Gruen says:

    “I read more than my fair share of philosophy, time I now mostly consider a form of conspicuous leisure”

    Inconspicuous surely

    Do you read with a sign up? “Silence please, philosopher at work” ;)

  32. Paul Frijters says:


    “Apart from anatomical localization, can you tell us something that fMRI has told us about language and language function that we didn’t already know?”

    I love a challenge like that. Off the top of my head, I can name two things I learned from the cognitive scientists that might have come from the fMRI (though you are going to have tell me whether they really did: I now and then wade through the textbook, not all the background papers)

    1. All language has emotional content, in that the very interpretation of words (and sounds, apparently using separately connections) includes an emotional colouring. We feel emotionally warmer about sentences that include words with which we have positive emotions, like mother and father, than sentences using negative words like cold and pain. People cant turn off this emotional colouring even if they wanted to. That’s exactly the kind of thing one might ex ante suspect to be true, but an fMRI confirms it. it has heaps of non-trivial implications for how to write and talk to different audiences.

    2. Words are associated with roles and remembered as such: if you are reminded of your mathematical prowess and the tests you do as a child, you will much more quickly be able to recall what the meaning of the word ‘fractional integration’ is than if you are reminded of the cakes you used to eat. The coat-hanger organization of our word-memory banks also has quite a few non-trivial implications, such as that you need to get ‘into your role’ if you want to speak well on a topic and it pays to introduce topics slowly so that your audience is already accessing the right parts of their memory when reading what you say. Again, this is the sort of thing you might have suspected beforehand, but an fMRI would clinch it. There is a whole raft of papers coming out in top journals as we speak that do economic experiments where subjects’ identities are ‘primed’ and where subsequent competencies and market behaviour are measured. Who would have been willing to say with certainty beforehand that the same person would behave differently in a market when primed to think of herself as an economist versus as a female?

  33. conrad says:

    These are good example where people remember the fMRI data, because it’s, well, fMRI data, even though it didn’t tell us anything new.

    1. It’s well known that words activate meaning. The classic experiment is Stroop (1935), who showed that trying to read color words that were in a different color to how they are spelled is hard than trying to read words in the same color they are spelled. As it turns out, this works in some conditions and not others. However, people then investigated emotional words and found that fear related words activate their meaning in almost any circumstance (as do taboo words) and that positive words are more mixed, but still have some priority in processing compared to words of neutral valence. There is also neural evidence that fear related things get special processing status (basically, there’s some hardware in your brain that gives them a special status), but this was initially found with animal models, people with brain injuries, and with EEGs (the little caps that measure the electrical responses of your brain). The fact that the fMRI people now tell us that the amygdala gets activated by fear related words (and hence it is assumed they are getting processed at a different level to normal meaning), is really no surprise! So it’s nothing new they’ve found, but, yes, it is more evidence.

    2. This is really a few questions, but:

    1) There’s a million and one experiments looking at words in context and priming. The simplest way this has been looked at is with semantic priming, of which some of the studies/theories were reviewed by Neely in 1977 (i.e., before people got to play with fMRI in any meaningful way). Basically, if you see two related words, you are quicker at retrieving the second if it is related to the first. Lots of different types of relationships have been looked at (semantically related, superordinates (cat->animal) etc.). It’s also been looked at in sentence contexts, emotional contexts, idiomatic meaning, … everything you can think of and more probably!

    2) There are lots of experiments looking at people’s initial states (e.g., anxiety, confidence, effect of observers.. etc.) and then later performance. I’m not sure why the fMRI data is any better for this sort of thing than any other type of data you would look at. The real problem here is that for the fMRI data to be good, you need some theory of what different patterns of brain activation mean, which you don’t have (at least not ones that are specific enough to work as really strong marker effects), apart from for things like fear. For example, what pattern of brain activation do I need to find that suggests you are even imagining yourself as male versus female? I don’t know, and I don’t think anyone else does either.

    3) I know about some of the neuro-economics literature (they love it where I work).

    However, I wouldn’t be too thrilled by it, since, at least in my opinion, a lot of the neuro-marketing/behavioral economics stuff is just other areas of psychology rehashed — that’s not to say it is interesting being applied to this domain (it is).

    For example:

    “Who would have been willing to say with certainty beforehand that the same person would behave differently in a market when primed to think of herself as an economist versus as a female?”

    Probably thousands upon thousands of social psychologists. This type of experiment has going on for donkeys years.

    For example, plain old applied social psychology has always looked at questions like yours (e.g., “who would have thought that under the influence of an observer, people poor at sport perform worse but people good at sport perform better” or “who would have thought that if someone believed they were African, they would act like this compared to someone who believed they were White”)

    Plain old learning theory, for which there are even good mathematical models (from the 70s!) has been used to explain gambling and other various types of irrational behavior (e.g., “why does person X gamble even though they lose”). It’s funny now watching people telling us that people in “market simulations” behave irrationally.

    Plain old decision making theory has been used to look at the type of biases people make in simple judgments for years also (Kahneman & Tversky were initial psychologists).

  34. Mike Pepperday says:

    Paul ? Categorisation theory? Never heard of it and I am VERY interested in categorisation. I googled. Ah! Self-categorisation theory ? Turner at ANU. Shoulda known. It says almost nothing, predicts nothing, and it contradicts other theories without disproving them (of course). Its adherents perennially puff up the magnificent strides they have made and they ignore ? studiously ignore ? the main rival theory. Not all of the adherents ignore but most of them do. Oh yes it’s successful, as psychology in general is successful. I shouldn’t be surprised if psychology is the most successful academic discipline of all.

    As for id, imprinting, theory of a role in a group, the notion of cognitive development ? they are nothing to do with ideal types. If you don’t want to accept my explanation of what an ideal-type means, read Weber (or the other philosophers I mentioned). Possibly except for imprinting, these concepts effectively do not exist. They belong in the basket containing the limeniferous ether and phlogiston.

    Conrad, I don’t disagree with you about fMRI (I know SFA about MRI) and I certainly agree about the need for a priori theorising. But you can only theorise if you have some concepts. In the hard sciences the concepts ? the objects of study ? tend to be self-evidently to hand. Social science has a severe problem finding agreed objects. (Which partly explains psychology’s success: it has an endless supply of non-contradictory concepts to blather about.)

    Those fMRI researchers located distinct brain areas for concepts they called competition and cooperation. Your temporal problem wouldn’t have bothered them since competition and cooperation take place over an appreciable time. What is interesting is that it lends these two concepts some objective reality. I don’t know of anything comparable. Everyone (except psychologists during office hours) knows what competition and cooperation mean but without those brain locations how could we say that they really existed? And they are important concepts: as I said before, it only remains to locate a coercion centre and the three dimensions of social interaction would have direct neural analogues. Being so plain, and so weighty, they are good candidates for use as ideal-types. To do that it is necessary to theorise in terms of total cooperation and total competition, just as economics theorises with total rationality and perfect information.

    What is fascinating is that when those fMRI researchers posited (presumed, invented, dreamt up, fabricated) these two concepts they ignored a century of psychological concept invention and elaboration. A century of diligent academic work was, for these fMRI researchers, as relevant as astrology. There is some official American dictionary of psych terms. It probably has thousands of entries; I bet none of them will ever be located in the brain.

    I append a couple of quotes…

    “In this fMRI study individuals played a specially designed computer game, according to a set of predefined rules, either in cooperation with or in competit¬ion against another person. The hemodynamic response during these conditions was contrasted to that of the same subjects playing the game independently… …distinct regions were found to be selectively associated with cooperation and competition, notably the orbitofrontal cortex in the former and the inferior parietal and medial prefrontal cortices in the latter. This pattern reflects the different mental frameworks implicated in being cooperative versus competitive with another person.”
    (Decety, Jean, Philip L. Jackson, Jessica A. Sommerville, Thierry Chaminade, and Andrew N. Meltzoff. 2004. “The neural bases of cooperation and competition: an fMRI investigation.” NeuroImage 23:744-751.)

    “These studies use paradigms such as the ultimatum game, the prisoner’s dilemma, and the trust game in order to examine the neural responses associated with cooperation, competition, fairness, and trust. Across these studies, cooperation, trust, and fair play typically activate VMPFC, MPFC, and MPAC, whereas unfair and untrustworthy responses activate insula, caudate in the basal ganglia, or DMPFC. The finding that cooperation, relative to competition, promotes MPFC rather than DMPFC activity is consistent with previously described work…”
    (Lieberman, MD. 2007. “Social cognitive neuroscience: a review of core processes.” Annual Review of Psychology 58:259-289. Citations omitted)

  35. Edward Mariyani-Squire says:


    1. …[a belief] … based on strong but not watertight evidence, and having earth-shaking potential consequences if falsified. Later this week I’m attending the funeral of an ex-Adventist friend who suffered a profound crisis on learning that Ellen White’s divinely inspired writings were plagiarised.

    On the slightly morbid topic of death and beliefs, the former UNE philosopher that I was visiting, died recently.

    We intermittently discussed the topic of death. What he found surprisingly and disturbing – in your words, ‘earth-shaking’ – was that he was going to die.

    On the face of it, this seems absurd; obviously we all believe that we will die. Why the ‘surprise’? He accounted for this in terms of his own worked-out philosophical position. He said that we ordinarily go about our days believing/knowing we are will die, but in what he called a ‘referential’ and ‘ostensive’ manner – we can point to examples of it, evidence of it, construct arguments to demonstrate its factual inevitability. When faced with its immanence however, there is a ‘phenomenological shift’. We come to know death by what he called its ‘meaning’ – a direct experiential realisation of ‘the terrifying infinite’, the coming of nothingness – by living the event of dying. That this is disturbing is clear enough (at least to me). It is surprising because we cannot really know or understand the ‘meaning’ of death outside this phenomenological shift, and we cannot make his shift except in the act of dying.

    This was not the ‘falsifying’ of a deeply held belief. The ‘knowing-by-reference’ of the coming-death was not negated. Nor was it that a meta-belief – that this belief was all there is to know – was negated, for his entire philosophical system was built on the notion that there was more to knowledge than just the referential kind. Rather, it was experience of living the belief about dying that was shocking, ‘earth-shaking.

    As a life-long atheist, there was no funeral or ceremony of any kind for his body.

  36. conrad says:

    “Social science has a severe problem finding agreed objects. ”

    I guess the alternative way to look at this is to see what has been learnt. If you had a stroke, for example, and couldn’t speak properly, your remediation would be far better off now because of what we have learnt, despite competing models of speech production that work on quite different principles. The same is true of reading. If you were learning to read today or lost your ability due to a stroke, you would be far better off because of what we know based on theoretical concepts where people don’t agree (there are 3 main models of reading — and within those models there are parts that people think are good, bad and indifferent).

    There are even more wishy-washy areas than these cognitive ones, where you can at least start at something real (e.g., “how do people hear or see the words””), like, for example, types of parenting. Despite this, you’ll find we really know a lot about them, and can apply the theories in useful ways, and since they’re important in daily life, it would be rather silly to give up on them because people can’t agree on them. So whilst people might not agree on either high level concepts, or the low level parts that they are supposed to emerge from, it hasn’t stopped them learning something productive about human behavior.

    As for your reference, you need worry about two things: Why should certain behaviors always correlate with some particular anatomical area? The idea that things will is basically new-age phrenology, and easy to disprove. Many things won’t, and so people have had to look further afield for things like coherence amongst different areas to try and work out what’s going on. The second thing is that your orbito-frontal cortex and your inferior parietal cortex are not exactly specific regions. They’re huge, and knowing that the task correlates with them is not a predictive model (which in the end is what you want), it’s just more data, pointing to the fact that cooperation and competition require different sets of processes, which I guess is no surprise at all. The more interesting thing is what the sets of processes are. Once you ask that question, you are back in cognitive land with the concepts that people don’t agree on.

  37. conrad says:


    if you’re interested, there is a good philosophical debate over how useful fMRI can and ever will be based on ideas like falsification and so on (a debate Rafe would like!). It started in the 90s, but here is rejoinder that basically has many of the references as they go back over time.

  38. Mike Pepperday says:


    Thanks for the reference to the falsifiability, etc of fMRI; I will check it out. In social science there is not half enough emphasis on falsifiability.

    The various abilities and achievements of clinical neurology and even clinical psychology are not really the issue. The issue is the lack of scientific understanding (in psychology not neurology). For centuries soaring cathedrals supported by flying buttresses were built without any science of the forces they were coping with. At the time we would have been thankful for it. An example today in psychology might be learning theory which works with children and animals. That’s good, but really there isn’t any theory, just an empirically supported story or guidelines. Your example of reading might also be like those medieval buttresses ? or it might not if the models are in direct conflict or don’t work except in the eyes of their proponents.

    Those two brain areas may be huge but then, competition and cooperation are socially huge. The interesting thing is that the locations are distinct from one another. If they weren’t distinct, if experiments with so-called “competition” and so-called “cooperation” showed that the brain lit up all over the place or the two were mixed together, there’d be nothing to report and said competition and cooperation wouldn’t exist. Are there any other concepts which have been comparably located? If not, and if people have gone looking, then it’s extraordinary that two ordinary expressions of social interaction do have separate locations in the brain.

    “Why should certain behaviors always correlate with some particular anatomical area? The idea that things will is basically new-age phrenology, and easy to disprove.”

    I have no idea what should or shouldn’t show anatomically. I thought that these two did always show in their particular regions. Has someone conducted experiments showing that sometimes competition and cooperation show elsewhere, or that they are not distinct from each other? If so it would seem to be a clear falsification. If you know any relevant references I would very much like to have them.

    “…it’s just more data, pointing to the fact that cooperation and competition require different sets of processes, which I guess is no surprise at all. The more interesting thing is what the sets of processes are. Once you ask that question, you are back in cognitive land with the concepts that people don’t agree on.”

    Well, different processes may be no surprise but different locations impresses me. Still, I agree: it is your last sentence that matters. The fMRI stuff is a something of a sidelight. Agreement on terms is needed. Economics is in cognitive land and it does have some efficacy. My point is that the words competition and cooperation are fairly plain and are not generally to be confused with each other, so are therefore good candidates to do what the economics and the natural sciences do with concepts, namely to make them ideal-types. When you do that meaning is no longer in dispute ? at least to the extent that, like perfect market and imperfect, they not to be confused.

    As far as I know all successful, scientific theory depends on ideal-types. If, nowadays, you want to build a cathedral nave you draw a diagram and calculate vectors to analyse where the forces are. You assign weights and directions all over the place and all your numbers are ideal. That is, nothing is real. And yet no meanings are in doubt. In social science there are no ideal numbers for there are no units of measure. The only ideals are the extremes: presence or absence. (Weber pretty much said it a hundred years ago. Every sociologist and political scientist genuflects in his direction but no one takes any notice of him!)

    If this is so ? if ideal-types are universally required for understanding ? then there is no option: the social sciences must do it too. Not just competition and cooperation but all sorts of things: justice, freedom, equality, nature, human nature, trust, mistrust, self-reliance, interdependence, identity, punishment, revenge, carrot, stick ? you name it, anything social. Take a look what the philosophers have said and then theorise the extremes. Maybe it’s difficult but as Margaret said, there is no alternative. It ought to be the usual approach. It is not as if there is no social science model for it is precisely what economics does. No one attempts to theorise degrees of rationality, or parts per million of free market, or gallons of market clearing.

    No guarantees. Just because competition and cooperation are fairly clear is no guarantee of results if they are theorised as ideal-types. But they do sound like a good start and it is encouraging that the fMRI is supportive. So why not adopt the stock-standard approach of successful science? After a hundred barren years, why continue to buck the system? Progress might be slow (How long did economics take to produce some useful results?) but surely theorising with ideal-types ought to be social science’s default approach.

  39. Nicholas Gruen says:

    Mike writes “What I can’t fathom is why the social sciences don’t adopt the natural sciences approach of theorising through idealisation.”

    It’s interesting that there’s a kind of assumption that in all these disciplines there’s some growing ‘body of theory’. That further the ‘best’ minds in a field will head for the theory – that way they can light the way for the lesser lights.

    Economics is pre-eminent amongst the social sciences in this. So much so that it’s ‘theory’ is completely formal. ‘Theory’ in virtually all other disciplines is more ‘philosophical’, though there’s a range, and I take Mike to be arguing that other social sciences should be more like economics.

    Certainly a bunch of theoretical endeavours in economics have paid off very handsomely, I guess starting with Malthus on population (if you call that economics) and Ricardo. (Smith’s theory was more discursive, whereas the architecture behind Malthus’ theory of population and Ricardo’s theories of comparative advantage and rent were formal whether or not expressed in symbols rather than words.

    And of course there’s been lots of progress since then. There’s also been some very important regress of late in the sense that economists are stuffed full of technical tricks and are not too understanding of their limitations.

    Anyway, the thing is that even in economics, as I’ve argued there are severe limitations as to how useful this is. I’m not saying that it shouldn’t go on, but I’m not sure it should be regarded as the commanding heights of the discipline, the place where the best minds should go. And then there are other disciplines where Mike wills us to greater progress in theory – like psychology. If it can be made, well and good.

    But I think also of history, and Hempel’s name jumped out at me, because he was pushing the idea of History being rebuilt in the mould of the natural sciences with people going hunting for laws. It didn’t work out too well. Today history is an interesting discipline. In some sense it’s just telling stories. But really good history is a kind of continuous meditation a crab-walk between the simple building blocks of history – simple facts, mediation on the knowability and contestability of those facts on the simple ingredients of stories, perhaps on ‘theories’ of history, on the explanation of human events and so on and on.

    This holds out virtually no hope of being pieced together as some coherent body of theory. But at least in history I think the search for something more is certainly hubristic and probably bound to fail.

    Even today economics is kind of collapsing under the weight of its formalism and has become a kind of wilderness where people are full of sound and fury and signify nothing. When policy makers want to think about policy problems there are a few fields that are of some use – industrial organisation for competition policy problems, Adam Smith and his formal alter ego, perfect competition as a heuristic for thinking about trade and micro-economic reform and Keynes and Hicks’ IS-LM) if you’re panicking in a global downturn and the alternative framework on offer was built by people who think the Great Depression was caused by perfectly far-sighted workers reading Scientific American and realising that new technology made it optimal for them to take a decade off. Can people add to this list?

    Apart from that, you have all the largely theoretically unconditioned factoids being churned out by Freakonomists everywhere – and very useful they are too – they tell us that reducing class sizes is an inefficient way of raising marks compared with better teacher selection (and pay if necessary). I’m suggesting here that even in the most propitious field – economics – the preeminent importance given to theory or to some idea of expanding the (context independent) ‘body’ of the discipline is well into diminishing returns – which is not to say that we can’t get any gains out of it, but is to question the preeminence given to it in the profession.

  40. Nicholas Gruen says:


    I don’t know why developmental stages are not ideal types. What might such an area look like if it was constituted according to your ideal types.

  41. Julia says:

    What a fascinating discussion!
    My two cents worth is not to provide an argument, although that is very tempting, but to provide some readings.

    On the sort of syllogistic thinking analytical philosophers do, empirically tested (on those usual lab rats – students – by cognitive scientists). Syllogistic logic, it seems is not self evident.
    Stenning, Keith, and Michiel van Lambalgen. Human Reasoning and Cognitive Science. Cambridge Massachusetts: MIT Press, 2008.

    On Belief in Mathematics interpreted as coherence of metaphor
    Lakoff, George, and Rafael Nunez. Where Mathematics Comes from: How the Embodied Mind Brings Mathematics Into Being. New York: Basic Books, 2001.

    Nunez, Rafael, and George Lakoff. “The Cognitive Foundations of Mathematics.” In Handbook of Mathematical Cognition, edited by Jamie I. D. Campbell, 508. Psychology Press, 2005.

    On the general slipperiness of naming things and thinking they are objects Ian Hacking’s wonderful book on Mad Travellers
    Hacking, Ian. Mad Travellers: Reflections on the Reality of Transient Mental Illnesses. University of Virginia Press, 1998. and on the ‘reality’ of social construction
    Hacking, Ian. The Social Construction of What. Cambridge Massachusetts: Harvard University Press, 2000.

    On the difference between decision making and what he calls “sensemaking” – pretty much a claim that “biases” are in fact normal thinking – a straight down the line intellectual inheritor of American Pragmatism without the POMO and advocate of the idea that “Believing is seeing” ie you can’t ‘see’ anything empirical without a pre-existing belief that allows you to, and the (literally) disastrous consequences of not.
    Weick, Karl E. Sensemaking in Organizations. Thousand Oaks: Sage Publications, 1995.

    Extending this idea into the fMRI universe. A very non Women’s Weekly/ New Idea version of left brain/ right brain – the idea of two different kinds of thinking (and reasoning)
    McGilchrist, Iain. The Master and His Emissary: The Divided Brain and the Making of the Western World. Yale University Press, 2009.

  42. conrad says:

    “The interesting thing is that the locations are distinct from one another.”

    I think you are confusing (possibly due to me being sloppy with my writing) different locations with different peak activations. There’s no doubt that if you looked at the activations in the two tasks, you would find activation all over the place in both — but what they are talking about is the difference between the two tasks (i.e., one has more activation in an area than another, not one has this and not that — the activation differences are based on a subtraction of the two). All this tells you is that there is some difference in function between cooperation and competition, which no-one knowledgable would disagree with, as they two quite separate things, even from an evolutionary perspective — cooperation, for example, requires that you understand what someone else wants and how that fits in with your goals much more than competition, and the only reason you can do this is because humans evolved in groups. If we evolved as an animal that walked around by itself, then we couldn’t cooperate, but no doubt we could still compete.

    “Are there any other concepts which have been comparably located? If not, and if people have gone looking, then it’s extraordinary that two ordinary expressions of social interaction do have separate locations in the brain.”

    Yes, any cognitive neuroscience textbook will be able to give you a description about the type of tasks that the parietal and orbito-frontal regions are correlated with. My favourite, for example, is that the OFC is associated with confabulators — people who make up stories and can’t tell whether they are true or not — and it’s believed this is because the OFC plays a role in the selection and inhibition of episodic memories (or if you prefer you can think of inhibition as “competition” between memories). As for parietal cortex — it’s used for any number of things, especially those to do with numbers and spatial tasks.

    “Has someone conducted experiments showing that sometimes competition and cooperation show elsewhere, or that they are not distinct from each other?”

    You need to get away from thinking about things in terms of competition and cooperation, because what you are talking about is complex task A, which involves any number of different things, and complex task B, which also involves a whole lot of things, and you call it competition. It’s like walking into a supermarket and saying there are deliveries and sales, without worrying about the cash register, the staff, the trucks, and everything else that goes on.

    If you are interested in competition and cooperation tightly defined, then yes, there are a million and one experiments which implicitly look at it. A very simple mathematical model of how these types of dynamics could be formalized by Stephen Grossberg appears in Psychological Review in 1978 “How the brain builds a cognitive code”. Jay McClelland took these ideas, rewrote them so people could understand them in 1981, and showed how a simple model of lexical (word) retrieval could be constructed. In that model, competition dynamics are used between individual words to show how memories of other words interfere with the retrieval of the one you want and what I think you mean by cooperation is used to show how the letters of words help to activate the word form.

  43. Mike Pepperday says:


    An ideal developmental stage? A perfect, or extreme developmental stage? I think it has no meaning. Developmental stages are just empirical, chronological descriptions, not components of any theory. A scientific theory is relational: ideal relations between idealised objects.

    I, too, have the feeling economists had run out of theoretical steam and what prompted me to so think was actually all the freakonomics. That activity is, as you say, theory-free but useful. Empirical research is fine if you know what you are researching. Class size and exam results are pretty plain concepts, which means there is no need of explicit theory as it is implicit. Administrators since the dawn of civilisation have wanted statistics. If you want to know how much vaccine to order or what colour will best sell your product you go out and do some counting. The result will be a snapshot of the people interrogated and (depending on mathematical theory) you might extrapolate the result to a broader population and it may be extremely useful.

    But you will know nothing about why. The information may permit competent administration but you don’t get understanding. If you want to know why this particular colour is favoured, or why a certain age group is prone to disease, you need a theory.

    I expect that the reason economics is bumping against a theory glass ceiling is because it wants to encroach on psychology and it can find no psychology theory. I suppose it always has been nudging psychology. The freakonomics activity may be indicating that economists have run out of the easy stuff, that everything has been said about straightforward economic concepts like trade and comparative advantage. Psychological concepts are now needed ? and there are no reliable ones.

    Economics long since recognised that it lacks a theory of tastes. Its attitude has been to claim that peoples’ preferences are a given and not the business of economics. This has a sanctimonious, Pontius Pilate odour. Economics specifies rational and says rational means consistent. That’s the talk; the walk reveals the true colours. In practice ? in public choice and economics experimentation ? rational means selfish and it means selfish about money. Was there ever a game experiment that rewarded with comradeship, harmony, the opportunity to be useful, or environmental protection? The reward is dollars, doled out by the man in the lab-coat and horn-rimmed glasses. Has any experiment offered incentives of promotion, pride, dignity, or satisfaction of duty done? The people with a taste for these things are expected to be satisfied with lucre.

    Prisoner’s dilemma and ultimatum experiments show time and again that there are some subjects who do not respond selfishly and that overall women are less selfish than men. Generally, it seems that the ones who don’t respond to monetary incentive are written off as a peculiar, contrarian minority. I looked into this a couple of years ago and found some economists who do puzzle over the variation. Some have tried to apply personality “theory” to explain tastes. I concluded it was completely useless. (“Theories” of personality are mere vague descriptions. What is it for? What does it do? Personality is a total mystery.) One or two people thought evolutionary psychology might the key. EP would help explain male-female difference but looks hopeless for the variation that is observed.

    A theory of tastes would not just say tastes vary ? that is a triviality ? but would show they vary systematically. It would set out a limited range of ideal-tastes of ideal-type people. One type of person will be money-oriented; other kinds will have other priorities. (Yes, Julia, I also reckon “biases” are normal thinking.)

    Julia, thanks for those references. Do you happen to know of anyone in the last fifty years who has written on ideal-types?

  44. Paul Frijters says:


    I sympathize with 70% of what you say, so let me start with the 30% before getting to the inherent problem you and I face when thinking about the 70%.

    Developmental stages to me are ideal types with ideal relations: later developmental stages can only come after earlier ones. It is a very simplistic relation, but it is a relation. Also, developmental stages are themselves caused by other processes that can be idealised, such as hormone cycles. That is not just another ideal type with idealised relations, it is a theory that is truly causal in that it allows real-world interventions: by giving more or less of the right hormones at the right time, you can prevent people from getting too tall or you can help them get a bit taller. In a similar way, you can intervene at particular moments to stimulate mental developments of various kinds. If that’s not a useful ‘theory’ complete with notions of ideal types (‘normal development stages’) and their relations, I do not know what is. It is true that it is mostly ex-post theorising after a lot of trial-and-error, but that is not relevant for this discussion. You are hence being less kind to the cognitive sciences than you need to be. They do have theories and they have made progress.

    Then, on the 70%, i.e. the sense in which economics has run out of theoretical territory to explore. I agree that the classical theory of markets has more or less run its course and that the boundaries have been well and truly reached in various other fields quite a while ago (including game theory). Yet, there is at this very moment an explosion of theories in many different directions. They will not be visible to you because they are snowed under by the existing curriculum and they have to fight for journal space with the established ideas, but it is happening. Theories of religions, reciprocity, how hunter-gatherer societies worked, networks, money, etc. If you are willing to see it, these are very exciting times for theorists and for economics in general.

    Where the real problems for ‘truly new theories’ are, is that they are handicapped by the fact that they will only ever explain a portion of reality and that anyone with half a brain-cell will immediately be able to spot their deficiencies. Hence they are easy prey for any ‘defenders of orthodoxy’. That same orthodoxy is just as full of holes, but its adherents have the advantage of being the incumbents, i.e. they dont really need to prove their theories are useful. Hence the new theories are four-times handicapped:

    1. Even an outsider will see its imperfections, making them at best lukewarm proponents of new theory;
    2. The core data the new theories would want to rely on is not actually collected in great quantities because statistical agencies attempt to measure that which is seen as an arrived theoretical concept;
    3. Entrenched insiders have all the incentives to shoot it down and deny it journal space;
    4. Any new theory has to compete with the thousands of cranks out there who think they are misunderstood geniuses and have the theory of everything at their disposal.

    If you doubt the difficulty of getting outsiders to help you, simply ask yourself this: if you saw a theory that did do a reasonable job at idealisations in order to explain hitherto unexplained outcomes, but of which you too could spot the imperfections, would you fight for it and say that it is precisely what you are calling for, or would you do the childish thing and say ‘this is not perfect. It ignores X, Y, and Z’? I continuously see examples of supposedly open-minded men and women who fail at this test. They want perfection to replace current theory, thereby in fact setting an impossible barrier for any new theory and thus implicitly supporting the existing orthodoxy.


    I agree that much of behavioural economics is essentially discovering what’s useful about 1st year psychology and anthropology. I despair at the lack of education amongst economists that lies behind it, and the cynicism of the old men who control the journals that supports it, but what can I do?
    Of course socialisation has been around as an idea much longer, but the notion that people can switch between identities within minutes and that they will behave materially differently when it comes to money is newish, at least to me.


    :-) thanks for the references, but I would much prefer to hear what you think the killer arguments are that can be made regarding beliefs that are in these references so that I dont have to read them to get their main point. The whole point of debate (at least to me) is not to have to read everything another person has read, but rather what their distilled wisdom is from their readings, their experiences, and their thoughts. Much more efficient.

  45. conrad says:

    “I looked into this a couple of years ago and found some economists who do puzzle over the variation. ”

    There’s a very recent paper that looks at altuism that summarizes the different theoretical positions using the literature on grandparents as an example, including the economic one — and they come to the same conclusion as you, which is basically that the economic theory predicts a whole lot of things that the data doesn’t support. There is also peer commentary. Look here.

    This is a “hot” area of psychology at the moment incidentally — people have become interested in whether certain social processes are innate. There is lots of recent stuff in the developmental literature also on this, since people use the argument that if really young infants display some sort of social behavior like altruism, then it is like to be innate, or at least the precursors to it.

  46. Julia says:

    Mike, I am trying to follow your argument for ideal types from your earlier posts. I’ll do you the disservice of replicating what I think you are talking about in summary.
    1. Ideal types are useful “pure” standards against which to create theory. This method works well in natural sciences

    The standard social science response to this is that there is a profound difference between the phenomena of the natural sciences and that of the social sciences. It is possible to replicate a natural science experiment because your experimental variable pretty much stays the same in the natural world, but in the social world it is protean. Come back a second time and your social group has changed its behaviours and its meaning about those behaviours in idiosyncratic and unpredictable ways.
    However, there is a second more interesting historically based explanation of the “fixed versus protean” idea in Stephen Gaukroger’s book
    Gaukroger, Stephen. The Emergence of a Scientific Culture. Oxford: Oxford University Press, 2006.
    Roughly speaking, he argues that Aristotle regarded things that change as being lesser than things that remained fixed and it was the job of philosophers to render things fixed. So its possible to understand the history of science as a history of finding fixed reference points – principles, laws generalisations and the like. The countervailing trend (put it down to “globalisation” or similar changes if you like) is the rise of “process” thinking the phenomenologists, the pragmatists and eventual postmodernism and similar.
    Another writer on this historical development of “process” is Arran Gare in the now out of print book
    Gare, Arran. Nihilism Incorporated. Melbourne, Victoria: Eco-Logical Press, 1994.

    The sort of syllogisms that logicians use are a reflection of this search for fixed or ideal types – all swans are white..etc being a classic case. The ‘fixed’ here, being used not simply as a scientific reference point but as the basis of a particular kind of (valorised) argument style. (This is why I included the Stenning and Lambalgen book – they show how this type of thinking breaks down on examination of how people do it – and they are logicians themselves, not psychologists)
    So I guess that leaves me open to accusations of PoMo cultural relativism, but I do regard these things as culturally conditioned.

    2. Ideal types are useful for theorising as psychology in particular has taken refuge in description and not analysis. (Economics has hit a theoretical dead end and needs a kick start and this may provide it)

    I would say the retreat from theory could be levelled at a number of disciplines. In fact I would go further and say that the low risk approach of description substituting for analysis was a direct outcome of the sort of view of ourselves and our thinking that the normative picture of human nature economics fosters. It is possibly to blame for extrapolating this state of affairs. We become what we are relentlessly told we are. Economic calculation is not about “the vision thing”. That’s me going out on a limb!

    3. “The only ideals are the extremes: presence or absence. (Weber pretty much said it a hundred years ago. Every sociologist and political scientist genuflects in his direction but no one takes any notice of him!)
    If this is so ? if ideal-types are universally required for understanding ? then there is no option: the social sciences must do it too. Not just competition and cooperation but all sorts of things: justice, freedom, equality, nature, human nature, trust, mistrust, self-reliance, interdependence, identity, punishment, revenge, carrot, stick ? you name it, anything social. Take a look what the philosophers have said and then theorise the extremes.”

    My reason for including Hacking – a philosopher – in my list is his approach to words like “competition, cooperation, justice, freedom, equality, nature, human nature, trust, mistrust, self-reliance, interdependence, identity, punishment, revenge, carrot, stick”. He uses mental illness characterised by real pain, real distress and real actions resulting from this to show the transience of the category pertaining to it – in his case “fugue” but RSI serves as a more recent example. He gives (I can’t clearly remember) six or seven social conditions under which such categorisations emerge. The same could be said about for instance the binary of competition – cooperation. It is sincerely felt, and it has real consequences, but it is an arrangement of meanings which in terms of their centrality to policy or indeed to a discipline like economics may be transitory. What indeed is “presence or absence”?

    4. “Everyone (except psychologists during office hours) knows what competition and cooperation mean but without those brain locations how could we say that they really existed?”

    See above, but this sentence also presupposes that the “problem of consciousness” has been resolved and that we know how we get from an arrangement of neurons and chemical and electrical messages to subjective meaning and self consciousness.

    5. On tastes, preferences and “rational means selfish and it means selfish about money”

    Over in Anthropology, broadly defined, there is a truckload of literature on “the gift” and what it means for status and self esteem and the nuances that cannot be captured in how people know what is an “equal” transaction. Marcel Mauss and Bourdieu spring to mind. However I thought (though I am not an economist and have not exactly read deeply) that Eleanor Ostrom had ventured here.

  47. Mike Pepperday says:

    Thanks Conrad. I will pursue the altruism reference. To your earlier post (#42)…

    Qualify it as you will, Decety et al say in the above quote: “distinct regions were found to be selectively associated with cooperation and competition”. Distinct regions means locations, not activity. Generally, computers don’t compute economics problems in one part of its chip and engineering problems in another. I think specialised computers are built for rare special purposes and for mother nature to use specialised hardware for different problems says something. Anyway it is all peripheral; it’s just that the fMRI is supportive, not contradicting.

    “You need to get away from thinking about things in terms of competition and cooperation, because what you are talking about is complex task A, which involves any number of different things, and complex task B, which also involves a whole lot of things, and you call it competition. It’s like walking into a supermarket and saying there are deliveries and sales, without worrying about the cash register, the staff, the trucks, and everything else that goes on.”

    That’s got to be back to front. Cooperation, competition and coercion are pretty plain. Perhaps they are complex but everyone knows what they are. If you wanted to explain supermarket operations you would indeed start by mentioning the main parts such as deliveries and sales. Having grasped that you can mention staff and cash registers and when that is mastered you can go into details of product codes. You would not try to understand a motor car by examining the combustion of refined hydrocarbons under various conditions of temperature and pressure. To begin you have say here is the motor which drives these wheels and here is a throttle to make it go and there is a steering wheel to direct it. It follows that to explain society you have to start by explaining its main features and how they interrelate. That’s the first task and after a hundred years of effort there is nothing to show. Now, you investigate the psychology of individual words. Well, I say you have no context, that for all you know, you are trying to understand a motor car by investigating the window winders.

    “If you are interested in competition and cooperation tightly defined…”

    No way. Definitions will get us nowhere. Science does not define. For example, Newton’s second law says force equals mass multiplied by acceleration (F=ma). This equation helps put a predictable mass into a predictable orbit. There is no agreed definition of mass (apparently attempts lead to contradictions). Acceleration depends on time (a=s/t/t, where s is distance and t is time) and nobody has the foggiest idea how to define time. If physicists can’t define mass we have no chance of defining cooperation. Definitions are am endless source of disagreement and they are not the way of science. All the work of psychology to define things is phantom-chasing.

    Science consists of idealised relations between ideal-types. That way theory becomes theorist-free and anyone can confirm or disprove the relation. If the relationship works out, everyone will agree on the ideal-types (but won’t attempt to define them).

  48. Mike Pepperday says:


    “Developmental stages to me are ideal types with ideal relations: later developmental stages can only come after earlier ones.”

    You want to use “ideal” in your own sense. I see nothing extreme, pure, exaggerated and unrealistic about them.

    At the end of your second para you say I am not kind to cognitive sciences. Maybe so, but that para does not bear on this. It is not about cognition; it’s about biology, namely hormones, which is natural science.

    Consider hormones. A hormone will be described with a chemical formula (which is an ideal relationship between atoms). This formula is the ideal hormone. Nature does not produce, and we cannot produce, this perfect, pure material. The hormone will work by having molecules locking into the membranes of cells or some such story. That story is idealised. In practice all sorts of other things will be happening at the same time, disturbing the ideal picture. A textbook would contain sketches of the molecules. Such sketches are ideal-types. It is not just at atomic scale you have sketches. A modern textbook on pathology or geology or carpentry may contain high quality photos but it will usually contain many more sketches. Sketches idealise by emphasising the salient features. They are unrealistic; things do not actually happen as the sketches show but it is through idealisation that we understand things. Idealisation helps to communicate. An ideal is a sort of caricature and you could even say a cartoon sketch of a politician is an idealisation ? for it is through some idealisation (big ears, hooded eyes) that the artist made the polly recognisable. The politician becomes essentialised as someone with those ears and eyelids. In those everyday things, as in natural science, the idealisation slips under our guard; we take it for granted.

    I would actually love to see a theory that used ideal types. My whole complaint is that no one in social science does it, except economics. I have seen jurisprudence suggested as another exception and I can kind of see that. Jurisprudence is, like economics, a successful endeavour.

    “…what’s useful about 1st year psychology and anthropology.” You give them too much credit. Here’s a big name:

    “For economists to rest a large part of their theory of choice on differences in tastes is disturbing since they admittedly have no useful theory of the formation of tastes, nor can they rely on a well-developed theory of tastes from any other discipline in the social sciences, since none exists.
    Becker, Gary S. 1976. The economic approach to human behavior Chicago: University of Chicago.

    None exists, he says. Does Becker lack education? Has anything changed since 1976?

  49. conrad says:

    “Qualify it as you will, Decety et al say in the above quote: “distinct regions were found to be selectively associated with cooperation and competition”. ”

    Yes, that’s because they’re sloppy. What they mean for people that don’t read fMRI papers all the time is this:

    “If you get the activation from cooperation and subtract it from the activation from competition, then what you find is that there is a different pattern of activation left over in different regions of the brain”….”This suggests that either one or more aspects of cooperation is done differently to competition or that one or more shared processes involved in cooperation and competition are used to a different extent”

  50. Paul Frijters says:


    of course I agree that idealisation is an important part of science. I am merely disputing your interpretation as to what counts as an idealisation and what not. Phycisists talk about time which is of course ultimately based on observation. Psychologists talk about developmental stages. Both concepts are hard to define, but in both cases is it true that what matters in the implementation is how they are measured and both physicists and psychologists spend an awful lot of time on ‘accurately’ measuring time and development. I fail to see how you can accuse the whole of psychology of being without ideal types. I can go along with idea that doing theory the economist way is not practised a lot amongst psychologists, but to dismiss them out of hand is too much.

    The quote from Gary Becker is a classic case of willful ignorance. Economics has seen plenty of theories of tastes, such as the theory of conspicuous consumption and conspicuous leisure by Veblen, written up at the end of the 19th century and which I think is a very useful theory (which would undoubtedly have been known to Becker). What is true is that economics doesnt have a universally accepted theory of taste-formation, either then or now, although I would say that reference point theory is by now an accepted theory of how norms change.

    Your conception of the brain as one of large functional areas is not realistic. You should think of the brain more as being made up of tens of thousands of small units doing very focussed things (like recognising the height of a pitch or calculating the movement of a lip in order to pronounce a vowel). Something like cooperation is a much higher-order phenomenon, involving thousands of sub-processes which will each be used for all kinds of other things. There simply is no such thing as a cooperation centre.

  51. Nicholas Gruen says:

    What a lot of the comments lack it seems to me is some instinctive sense of how useful various moves are likely to be. Of course you can’t know for sure in advance how useful things are likely to be, but you can have some idea. The (fading?) obsession with ‘micro-foundations’ in economics is an illustration of the problem I’m talking about.

    I guess the search for micro-foundations is a worthy search. And just as Paul has said that the micro-foundations of the brain are going to be found in the interplay of thousands of small units doing focused things, so micro-foundations might be useful in economics if they can somehow reflect the chaotic multiplicity that we know is there in fact (and which we ought to be able to intuit from what we do know, goes on in the brain.) They were the micro-foundations that are in Keynes, not formalised, because they were so indeterminate, but driven by the cultural and structural factors of the market interacting with human psychology as generally understood – animal spirits, and prone to periods of widespread and self reinforcing bouts of gloom and optimism.

    A potentially useful micro-foundations project then would, I would have thought, focus in on a range of ideal types, and then investigate their relative prevalence and try to build realistic models of the economy with those micro-foundations. I don’t fancy anyone’s chances, but that’s what the project ought to look like shouldn’t it?

    But what we see if one unitary micro-foundation to another. Rational expectations, monopolistic competition, menu costs etc etc. If that were regarded as some very early stage research, then well and good (and I’d have my doubts as to when it would overtake Keynes discursive microfoundations in useful analysis, but then who am I?) In fact what happens is that each of these ideas has a few years in the sun and gets seen as the latest key, the latest lens through which to see the economy. But it’s ridiculously partial and doesn’t take us anywhere immediately useful.

    Paul Krugman helped build micro-foundations for trade theory, development economics and economic geography. In each case there is virtually no discussion in the literature as the project was underway as to how useful such a project might be. (Like I said, you can’t know for sure in advance but if you want to use your talents productively it makes sense to consider and discuss the issue.) Just as Krugman and co had started settling in to base camp for their climb of mount microfoundations of intra-industry trade (1988), Peter Gray wrote this.

    To be able to embrace all of the various forces which act on IIT [intra-industry trade] in a single model which yields a precise solution, is impossible because the variables are too numerous and their effects can be inconsistent across cases and through time. Even if it were possible to produce a definitive treatment of IIT, the model would be so general and would not be operational in the sense of either being subject to empirical test or useful for policy prescription. . . . Such is the complexity of IIT that our understanding of this phenomenon will be better served by . . . a general ‘loose and untidy’ model than by a selectively precise or rigorous one.

    Explaining his Nobel Prize Krugman says this:

    The broad pattern of what countries produce is determined by things like resources and climate, but there’s a lot of additional specialization due to economies of scale, and there’s much more trade, especially between similar countries, than you would expect from a purely resource-based theory.

    You may think all this is obvious, and it is – now. But it was totally not obvious before 1980 or so – except for some prescient quotes from Paul Samuelson, you really can’t find anyone describing trade this way until after the theory had been laid out in mathematical models. The plain English version came later.

    And you should bear in mind that economists have been thinking and writing about international trade for a couple of centuries; to come along and say, “Hey, we’ve been missing half the story” was a pretty big thing.

    This is an extraordinary claim. How about Linder, Vernon and his theory of the product cycle. Gruber’s and Australia’s Peter Lloyd’s articles on Intra Industry Trade in 1971 (I think) and Lloyd’s book on intra-industry trade in 1975. Indeed in 1976 Peter Gray tried to sketch out the kind of ‘heuristic’ model he was speaking about above in a book. But it wasn’t a new determinative model – so it was ignored. A $100 if anyone can find any serious discussion of Gray’s point above in the corpus of Strategic Trade Theory. Yet Krugman now concedes that strategic trade theory generated next to nothing in useful policy insights, something that Gray predicted at the outset from broad, commonsensical methodological reflection – though to my knowledge his arguments were never even seriously considered.

    I was thinking of this also when I made the points I did about history above. Perhaps because it hasn’t been crowded out by the demands of formal technique, the highest academic status history is almost invariably a dialectic between the practical doing of history – digging around in archives, telling stories – along with a highly refined methodological awareness.

    And this is precisely the kind of considered reflection on the methodological implications of what one is doing as one is doing it that is a central feature of almost all the great economists I can think of up to and including J. R. Hicks. Have a read of Value and Capital and of Hicks’ choice of perfect competition to model general equilibrium or Marshall or Pigou on the criteria of welfare for an illustration of what I mean. There’s virtually no place for this kind of consideration to be built into economics as it is done any more.

    So it seems to me that to be helpful, and to minimise the extent to which it leads to destructive hubris, all social scientists should understand their craft (or science) as operating in a constant tension between an awareness of the likely strengths and weaknesses of formal reasoning (including using ideal types) and the strengths and weaknesses of less formal and structured – more commonsensical and story telling – approaches.

    Finally a challenge for Mike. Are we being ‘scientific’ in our discussion here? If so how come since we’re not using ideal types. If not, what are we being, and what is the role of the kind of methodological reflection in this thread in social science?

  52. Julia says:

    Finally, I think I understand what you, Mike, are getting at.
    Your words, “A modern textbook on pathology or geology or carpentry may contain high quality photos but it will usually contain many more sketches. Sketches idealise by emphasising the salient features. They are unrealistic; things do not actually happen as the sketches show but it is through idealisation that we understand things” made the penny drop.
    There is such a thing in social science. It is the case study. Before you all instantly stop reading here, because you have bad memories of the “unscientific” nature of cases, or have ploughed through Yin, that apologist for scientific method who has written so extensively and influentially about how a case study is so defectively a sample of one and turned so many off case study with all the rules he says therefore follow, bear with me for a moment.
    Mike is writing about a “pointing” function. Ideal types he says (and I argue, cases and stories and examples) perform a function of attentional pointing. As Mike says, they “emphasise the salient features”. This is why we teach with example – it contextualises, it makes concrete and it highlights the knowledge or relationship on which we want students to focus.
    However case studies can do other useful things that might also be thought of as properties of ideal types.
    Bent Flyvbjerg my favourite writer on cases, (http://poli.haifa.ac.il/~levi/res/fivemisunder.pdf) points out that a well selected case study can falsify. You only need one instance (case)of a black swan. Case studies can also provide boundaries for your phenomena in much the same way as ideal types. Whereas an ideal type uses an unrealistic idealisation as the “outlier”, a case can use actual outliers. The way for instance “wild children” have been used in sociological literature to circumscribe the boundaries of the question “what is the social?” is a demonstration. Similarly, there is a book which I found quite jaw dropping in its implications for the boundaries of what little it takes of what we consider common human attributes to operate perfectly functionally as a happy and well adjusted society- such as an idea of object persistence after that object has vanished from sight – or an inability to count or draw as a cultural condition. (See Everett, Daniel L. Don’t Sleep, There Are Snakes. New York: Pantheon Books, 2008.) This is a “case” of a single South American tribe, but it speaks volumes about us all in refining the delimitations of basic human capacities.
    Flyvbjerg identifies four uses of cases, all of which have some of the functions of ideal types.
    1 Extreme/deviant cases – To obtain information on unusual cases, which can be especially problematic or especially good in a more closely defined sense.
    2 Maximum variation cases – To obtain information about the significance of various circumstances for case process and outcome, e.g. three to four cases that are very different on one dimension: size, form of organization, location, budget, etc.
    3 Critical cases – To achieve information that permits logical deductions of the type, ‘if this is (not) valid for this case, then it applies to all (no) cases’.
    4 Paradigmatic cases – To develop a metaphor or establish a school for the domain that the case concerns.

    However to get back to the original thread topic – belief.
    Cases and ideal types are part of the formal processes of how we get to pay attention to salient features. Beliefs do this internally (as part of our “Bias” system of separating and acquiring useful information from the flow of experience.)
    Beliefs have a “pointing” function. They prime you for salience and allow you to make comparisons between expectations generated by beliefs and things thrown up by actual experience of events. But their second function is compaction. Given a decent set of beliefs – eg there is such a thing as “market forces” it is not a conceptual metaphor drawing on a larger mental frame of “mechanism”, for instance – one can then happily regard those aspects of the flow that comfortably fit with this as dealt with. They do not need to be revisited. This frees up that valuable resource, short term memory, to deal with “salience” factors that fit though the filter as they hove into view.
    A formal tool (ideal types / cases) that more or less matches the attributes of a “bias” you are already using is bound to be a comfortable tool indeed.

  53. Mike Pepperday says:

    There’s a lot going on in this thread.

    “Are we being ‘scientific’ in our discussion here?”

    I shouldn’t think so; we are not developing a scientific theory.

    “If so how come since we’re not using ideal types. If not, what are we being, and what is the role of the kind of methodological reflection in this thread in social science?”

    Is this a bit meta-analytic? Like asking about the form of forms, the ideal-type of the ideal-type? I guess we’re philosophising. The barrow I’m pushing is science in social science but there are other things in the universe. We have words, we convey things even if imperfectly. We can learn without being scientific. (I have learnt that the ideal-type concept is not easy to put across.) The point of a discussion like this is to set a thought free and see how it flies. I might find out where the tangents are that I better not overlook if I wish to be understood. (See? I am conducting an investigation into the vagaries of others’ misperceptions.)

    Take Veblen. Paul says his theory was useful. I am not saying that all social science is empty of content. Veblen was entertaining which alone is useful but he offered a different view, an angle which disturbed or countered conventional wisdom. It enabled people to see that forces were operating that they hadn’t been aware of. That’s got to be useful. But his theory makes no pretence at being eternal truth. Science (and economics) seeks universal laws.

    If swans were in social science then if some disturber of established wisdom alleged there was a black swan, he or she would be ignored. If the rebel notion got legs, it would be shown that the black thing is not a swan, or that this black is the same as white. There would be no way to settle the question with the meanings of “swan” and “white” varying from theorist to theorist. Schools might then develop publishing on shades of swans. In social science the terms depend on the people using them. There is probably not much confidence in soc sci that there exists any truth that is independent of theorists. Someone mentioned Eleanor Ostrom. She referred to John Searle’s ideas and said:

    “The basic difference between the social world and the biophysical world is that the biophysical world exists whether or not humans reflect on it, but the social world is constituted by human thought, language, and action. Given the importance of language, a more serious threat to the future of our discipline than the lack of universal laws is our lack of common definitions for key terms we use including power, norms, and institutions.”
    Ostrom, Elinor. 2006. “The 2005 James Madison Award lecture: converting threats into opportunities.” Political Science & Politics 39(1):3-12.

    Her diagnosis that the social world is constituted of human reflection is obviously right and this means there is a problem of agreement on terms. And I concur: there can be no universal theory without agreement on terms. Her proposed remedy of agreed definitions can never work as I have pointed out. On the contrary, agreement on terms for concepts has to be FREE of human definition. If someone has to define it, it depends on the definer so the definition can be changed and universal laws become impossible. She is exactly wrong way up: first you need universal laws and THEN you will have agreement on what the terms mean. Ostrom overlooks that biophysical theories ? not the biophysical world but the biophysical theories ? are also the product of human reflection. It follows that the process of natural science theorising and social science theorising should be the same.

    I met her once. I think I was the only one in the large, day-long seminar that had done the reading. That was a book of hers, hard to follow, probably because of jargon.

    Bent Flyvbjerg is nearly as entertaining as Veblen. He thinks that the quest for a science of society is simply hopeless (If you think I condemn social science, I am sweetness and light compared with Flyvbjerg) and recommends we give up the quest and adopt some hazy Aristotelian notion which I have forgotten and the Wikipedia entry doesn’t remind me.

    Science is Galileo, not Aristotle. Aristotle looked at what things are; Galileo looked at how things relate. Aristotle is concerned for being; Galileo for doing. A is a philosopher; G is a scientist. To write F=ma is to express what its parts do. What they actually are, no one can say and it doesn’t matter. (Everyone in the game knows what they are but no one can say what they are.)

    Maybe that last para consists of a string of ideal-types.

  54. conrad says:


    you are too hard on the social sciences for defining things, since almost all sciences start somewhere that isn’t the bottom or the top level of analysis that they have.

    For example, many chemists don’t care too much about things that go on below the atom (e.g., quarks), but that’s not because there aren’t interesting things going on below that level — lots of things do. It’s just that many problems have solutions that can be found by worrying about atoms and nothing below them, despite the loss of information that this entails.

    I also think you’re wrong on this: “Science (and economics) seeks universal laws.”. In most areas of science, people are not seeking universal laws — they seek laws or models that have very good predictive capability. For example, in the area where people do sometimes have the pretense of looking for universal laws, the laws are not universal anyway. Take Newtonian physics — these are laws that apply in many many situations, but they are not laws of the universe — they’re just very good approximations of what goes on. If you want to predict where a planet is a few light years away, for example, you will have take into account much more than what Newton taught us. Thus, the best you can hope for is that the laws which you use that are at the bottom level of your analysis don’t have too much error in them such that the things you infer from them are not useful.

  55. Julia says:

    Sorry Mike, I am giggling at your last post as I write. Not for any mocking reasons but because it was entertaining. The “hazy Aristotelian notion” you refer to is Phronesis, or practical ethics, according to Aristotle (and Flyvbjerg), another form of knowledge that we don’t use much. But that argument is in his book “Making social science matter”, and not part of the “case” argument I was making.
    I too think you are wrong on “science seeks universal laws” but I tend to be more comfortable with a “cultural practice” idea of science (eg as Actor Network theorists think) or of the sort of historical situatedness of scientific ideas that Gaukroger (cited earlier) presents. In practice, scientists don’t seek universal laws (but I think that makes me a Pragmatist)

  56. Mike Pepperday says:


    “you are too hard on the social sciences for defining things…”

    I’m neither hard nor soft and this is not relevant. I put an argument.

    I wouldn’t care about definitions if those using them knew they are a dead end as far as science is concerned. But they don’t; on the contrary, they ? Nobel winner Ostrom is Exhibit A ? think definitions are the answer. Definitions are useful in their place but they are not required for natural science theory; they won’t yield fundamental understanding and they get in the way.

    The rest of your first sentence I cannot understand. And the point of the next para I also can’t figure out.

    A scientific law is without exceptions. That is what universal means. We had this discussion before. Exceptions to Newton’s laws crop up all the time; they are called paranormal and are complete crap. You also allege that Newton’s laws are approximations. They are ideal-types. (Did I forget to mention that?) They are not approximate; they are imaginary and perfect. If you had said they were not correct, you’d have been right for they have been found to be incorrect. No scientific law can claim to be correct.

    If you want to call making predictions “science” you can but that is not the natural sciences I have been talking about it ? which is the endeavour of discovering nature’s (universal) laws. No doubt there are plenty of urgers who would like to dignify their activity as “science” but anyone who is not in the game of seeking (refining, testing) her universal laws is not doing anything relevant to this conversation ? what I am calling science.

    To a related matter mentioned by Julia in an earlier post:

    Scientists do not perceive any cultural conditioning. They presume a truth is out there to be discovered. This attitude has been successful and is supported by the practical evidence. The equation F=ma is unaffected by culture; it works the same in every culture, and once written it was independent of Isaac Newton and every theorist. Newton had no choice: he could not write an equation containing heat or happiness or the price of fish. God or nature decreed that F, m and a are the ideal-types that work here; the only ones. As ideal-types, each is a product of human reflection, none actually exists in nature. Yet nature insists on these particular ideal-types; anything else will be rubbish (unless you are Einstein). It is a strange thing: ideal-types don’t exist in nature and yet they have to be discovered. It is the way science is.

    Any genuine social science will have to be like this too. Ideal-types must be discovered. A handicap to this in the social sciences are widespread normative attitudes. Science is not an “ought”; it is an “is” ? description, not prescription. Machiavelli’s attitude is needed. M preferred a republic but when he advised the dictator he told politics as he saw it. He has never been forgiven. Economics calls it as it sees it and is also vilified for it.

  57. Paul Frijters says:

    James #28,

    thanks for leg-up and advice how to get more mileage out of these things. I am lazy about these things.


    I agree with most others who said they have a different view of science than you do, but in your defense I agree that the search for universal laws (and looking for ideal types) is a very helpful way at reaching whatever level of knowledge is attainable. Hence, though I think you are essentially wrong about what science is in practice (even physics. Just think of the various models of the atom that fit different needs. Physics, like all other sciences is quite happy using mutually inconsistent models side-by-side as the situation dictates), you are right about the desire needed to get as far as possible.

  58. Julia says:

    Mike said
    “Scientists do not perceive any cultural conditioning. They presume a truth is out there to be discovered. This attitude has been successful and is supported by the practical evidence. The equation F=ma is unaffected by culture; it works the same in every culture, and once written it was independent of Isaac Newton and every theorist. Newton had no choice: he could not write an equation containing heat or happiness or the price of fish. God or nature decreed that F, m and a are the ideal-types that work here; the only ones. As ideal-types, each is a product of human reflection, none actually exists in nature. Yet nature insists on these particular ideal-types; anything else will be rubbish (unless you are Einstein). It is a strange thing: ideal-types don’t exist in nature and yet they have to be discovered. It is the way science is. ”

    If you are interested in an alternative explanation of how mathematics got to look like this but one which has culture and cognition at its heart, you could look at one of the works I mentioned earlier.
    Lakoff, George, and Rafael Nunez. Where Mathematics Comes from: How the Embodied Mind Brings Mathematics Into Being. New York: Basic Books, 2001.
    Nunez, Rafael, and George Lakoff. “The Cognitive Foundations of Mathematics.” In Handbook of Mathematical Cognition, edited by Jamie I. D. Campbell, 508. Psychology Press, 2005.

    I suggest you take it on in the spirit of “willing suspension of disbelief” at least until you have read the whole argument. Nunez and Lakoff write a novel interpretation of how mathematics gets to look universal. They suggest that what is “discovered” is a new conceptual metaphor, one which can be extended to a greater realm of operation than previous metaphors. They don’t discuss the mechanism of conceptual metaphor in depth in this book, they have elsewhere.
    However one could argue that their cognitive approach to mathematics depends on that “bias” in human thinking towards a distinct preference for internal coherence in a belief.

    Indeed the test of most logic is a self – consistency test. We tend to see cognitive dissonance as a bias but cognitive consistency is the other side of the same coin. Its the stuff we mentally keep trying to make. We are carving it out of a sometimes intractable or at least indifferent “reality” however. I’m not at all sure the universe presupposes it has to be logical or consistent.

    These limits are imposed by our framing which has never really left off reading events and experiences of the world as if they were reflections of a guiding personality somewhat like ours. There is an irresistible urge to attribute agency to nature or the universe. This used to look like God deciding for each individual whether to punish or reward and personally intervening to do so. These days in a more clockwork, deistic conception, our attribution to nature is that things will turn out to be good. Meaning at base we have our fingers crossed that things are logical, consistent or balanced (like market forces).

    Why should we believe in a universe that is biased in our favour and looks and awful lot like our own (slightly optimistic – another demonstrable bias) thinking? If you are being truly scientific and truly self consistently logical you cannot admit of a universe that is anything but utterly neutral, and that means no bias in favour of comprehensibility like logic and consistency or favourable outcomes like “balance”.

  59. Nicholas Gruen says:

    Well put Julia, and your last paragraph is of course an appeal to consistency ;)

  60. Julia says:

    Well yes and no as academics are famously wont to say. One of the charges levelled against PoMo is that the statement “everything is bound by context” is of course appealing to universalism in the “everything” part of the sentence. I think things break down in a welter of irreconcilables at the other end of the epistemological spectrum as well. You can’t appeal to a preternaturally ordered universe unless you take on board the problem of where the order comes from.

    I think it was the mathematician Godel who pointed to the problem of sets which have themselves as a member. I think science metaphors behave the same way. Both sets and metaphors are “containers” which separate like from not like and which explode if you try to do too much with them. This is complicated by my own belief that sets and metaphors are “in here” and not “out there” and are to do with our own attempts to impose order on general messiness by separating like from not like – using our own definitions of what such a pattern might be.

  61. Nicholas Gruen says:

    Yes, and why some of it ends up in humour – our ultimate response to things which are and are not. I have a grand narrative of humour in which it comes from play. Reptiles don’t play (I don’t think), but birds and certainly mammals do. Especially young mammals. A lion cub will bite another cub and then let go. He’s biting it, but he’s not biting it. Add a few million years and you get irony.

    Anyway, apologies for that bit of heaviosity, it’s just something I’ve been pondering. I’m always surprised that philosophers don’t take a bit more interest in humour – though it’s hard to do so without killing it. The other thing about humour is that it seems to be a better way of leaving off of these subjects. You end up just musing on something, and somehow getting a kick out of not getting anywhere. In philosophy you don’t get far, but the form seems to suggest that you will.

    This is just me musing away, not pretending to any deep relevance to your last comment, but perhaps it is quite relevant. And for some reason I’m thinking of Woody Allen commenting on Annie Hall’s parking skills by saying “No that’s OK. We can walk to the curb from here”.

  62. Mike Pepperday says:

    Julia (#58), I’m baffled. You quote me at length and then do not respond to the quote. I say F=ma and machine guns and Toyota utes work the same for Muslims, Christians and pagans. It is independent of any person. Can you show me I am wrong? I reject the implication that I have not read the books you list. If you have an argument, make it. And do please illustrate with specific examples or the reader may suspect there are none and it is all empty abstractions.

    Yes Paul (#57), physics has controversies and conflicting models. It was ever thus. Many have been resolved and more will be. This disagreement over the atom you mention ? how old is it? A few years? A few decades? In social science the sages of two and half thousand years ago are quoted along with those of a hundred years or five years ? indeed, the older the more the respect. What is resolved? New theories do not replace older ones and this, along with the fact that economics can’t pick up anything from which it could even begin to build a theory of tastes, shows the problem. In terms of knowing something useful the situation seems hopeless. Physicists differ over models at the electron and quark level but social science has not reached the Galileo stage and it shows no sign of reaching it.

    This is not to say there is no knowledge at all. There are scattered bits and pieces ? factoids. In political science I think quite a lot is known. What’s solid in non-clinical psych? Learning theory, helplessness theory, possibly one or two other things. Freud and his offshoots are long since passé, behaviourism is dead, buried and cremated, and sociobiology has made a bigger, clearer contribution to understanding human behaviour than a century of psych. The parlous state of psychology is well illustrated by the following example.

    The Psychological Bulletin, vol. 128, No. 1, 2002 was devoted to the so-called IND-COL distinction. The keynote paper by Oyserman et al cites about 300 references and carries appendices tabulating a similar number of post-1980 studies using IND-COL. They would entail many thousands of questionnaires or interviews. This is a massive review paper in the most prestigious psych journal and the authors are, of course, sympathetic to IND-COL research.

    They find, inter alia, that “European Americans were not… less collectivistic than Japanese or Koreans”. Blind Freddie knows this is wrong. This distinction, which goes back to Maine and Tönnies (whom the paper cites), and which has been noted by every relevant thinker since, is invisible to academic research. That alone should be a devastating indictment. Oyserman et al incriminate themselves further:

    “…the greatest strength of the IND-COL framework is its theoretical parsimony… However the value of this approach is contingent on the concepts of IND and COL being defined clearly, allowing them to be operationalized, assessed, and manipulated. Only under this condition will social scientists be able to evaluate the usefulness of the IND-COL framework.”

    A veritable army of researchers spends vast resources to find nothing and then they say they don’t know if it is useful! Are they taken out and shot? Are they sent to the salt mines? Are they sacked for incompetence? They are not even embarrassed. It’s ho-hum, par for the course: give the perpetrators more grant money. Non-clinical psychology must be the greatest lurk since astrology and alchemy. They are unable to evaluate the usefulness of IND-COL but its great strength is that it is parsimonious. I have a theory that the moon is made of cheese; the great strength of my theory is that it is parsimonious.

    Wait ? all is not lost for they know where the solution lies: individualism and collectivism must be defined clearly. But if there was any prospect of such definitions wouldn’t they have been formulated by now? That alone shows the fruitlessness of this proposal. And if physics can’t define its most basic terms, to wit mass and time, why are they even thinking of definitions? Science theory does not depend on definitions. A definition is an attempt to tell nature what she is. She scorns the impertinence. Science theory is discovery ? not diktat, but discovery. Science theory describes what nature does ? not is, but does. And it is what she does in principle ? not in practice, but in principle.

    Following, say, Tönnies, they invent IND-COL and then try to confirm the division exists. Yet we know ? surely we all know by now ? that science research is not about confirmation; it’s about falsification. When researchers publish biology or geophysics findings there are dozens of people in the particular sub-sub branch ready to falsify the findings if they spot a chance. No matter how many times you confirm something, it is never proved. Psychologists presumably believe, with Blind Freddie, that there is an individualism-collectivism difference between Japan and the US (and elsewhere) but they haven’t the clarity or the confidence to set out to try to falsify it.

    They might say they want to show IND-COL exists before they attempt falsification and are chagrined that attempts to show it keep failing. They fail because the way they go about it is rubbish. What theory, what typology, has ever been developed in natural science by factor analysis of mass data? Is there a single one, anywhere at all? Could you measure everything associated with fluttering autumn leaves and put it into a computer and arrive at a theory of aerodynamics? Could you measure everything you can think of associated with avalanches, do a linear regression, and derive Galileo’s or Newton’s theory of gravity? Can you take Tycho Brahe’s observations and get a factor analysis to spit out elliptical orbits? Could you take Mendeleyev’s data, or even its exhaustive modern equivalent, and expect a computer to print the periodic table? It is ludicrous; there is not a shred of evidence that the approach could ever work. Yet thousands of social scientists round the world are trying to do this every day. Theory comes from a human mind, not from a statistics model.

    Moreover theory theorises ideal-types which are unrealistic extremes. A statistics program working on data in effect generates averages; it smoothes out the extremes. If IND-COL were to be of any use, they would have to be theorised as total individualism and perfect collectivism. This would be the necessary starting point.

    The connection and the divide between the individual and the collective is quite basic to understanding human behaviour. Probably everything, including a theory of taste, would depend on sorting out this elementary binary division of the social world. So far there has been no progress whatsoever. If there is ever to be any, in this or any area of social science, it will require theorising idealised relations between ideal-types ? the way the natural sciences do, the way economics does.

  63. Pingback: cognitive psychology key concepts | COGNITIVE PSYCHOLOGY

  64. Pingback: Club Troppo » Introducing Julia Thornton

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.