Home > Uncategorized > Truth and science

Truth and science

from Lars Syll

What is truth in economics? | LARS P. SYLL In my view, scientific theories are not to be considered ‘true’ or ‘false.’ In constructing such a theory, we are not trying to get at the truth, or even to approximate to it: rather, we are trying to organize our thoughts and observations in a useful manner.

Robert Aumann

What a handy view of science.

How reassuring for all of you who have always thought that believing in the tooth fairy make you understand what happens to kids’ teeth. Now a ‘Nobel prize’ winning economist tells you that if there are such things as tooth fairies or not doesn’t really matter. Scientific theories are not about what is true or false, but whether ‘they enable us to organize and understand our observations’ …

Mirabile dictu!

What Aumann and other defenders of scientific storytelling ‘forgets’ is that potential explanatory power achieved in thought experimental models is not enough for attaining real explanations. Model explanations are at best conjectures, and whether they do or do not explain things in the real world is something we have to test. To just believe that you understand or explain things better with thought experiments is not enough. Without a warranted export certificate to the real world, model explanations are pretty worthless. Proving things in models is not enough. Truth is an important concept in real science.

  1. Ikonoclast
    October 17, 2020 at 3:19 am

    “In my view, scientific theories are not to be considered ‘true’ or ‘false.’ In constructing such a theory, we are not trying to get at the truth, or even to approximate to it: rather, we are trying to organize our thoughts and observations in a useful manner.” – Robert Aumann

    In this statement, Aumann accidentally or deliberately conflates mere initial hypotheses on the one hand with tested and established scientific theories on the other, with their robust credentials and their extensive and successful records of explanatory consistency, predictive success and practical applications. He also stays stuck at the early point of theory-making. Yes, at the very outset of creating a theory, be it an attempted theological, economic or scientific theory, we may well be “trying to organize our thoughts and observations in a useful manner”. But if we remain stuck at this point, we remain stuck in metaphysics, in metaphysical hypothesizing, and do not proceed to science at all.

    Aumann’s conflation seems to be a conscious or unconscious rhetorical device used to deny the difference between (conventional) economic theories and genuine and successful scientific theories. This conflation is a dishonest rhetorical move without any valid justification. The attempt at justification turns on a misunderstand of the terms “true” and “truth”. Aumann makes an implicit assumption and perhaps even an explcit assumption (I would have to check his text again minutely) that a “truth” must an absolute and perfect “truth”. If a “truth” ascertained by humans cannot be an absolute and perfect “truth” then the standard of “truth” must be thrown out altogether. This is Aumann’s contention. He disallows approximations of truth and the approach to truth. I guess this is the standard practice when your discipline if going nowhere.

    There is no point, theoretical or practical, to philosophy or science if one does not posit, accept in principle and discover in practice, instances of truth. In turn, there must be a Theory of Truth, meaning one theory or a set of theories, even if that theory be encapsulated only in pragmatic method like basic scientific method. A “Theory of Truth” means a theory of how signs are related to the signified and how models – which are also signs – are related to the signified systems they model. This equates to the correspondence theory of truth. Such a theory of truth must hold that there can be true representations and false representations. This is so even if, unavoidably in practice, a “true” representation is always an approach to truth rather than a complete, final or perfect truth.

    Some philosophers argue that there cannot be a general a “Theory of Truth”. This is the view held by adherents to the Deflationary Theory of Truth and it seems to be Aumann’s position. “According to the deflationary theory of truth, to assert that a statement is true is just to assert the statement itself.” One might wonder how “Deflationists” would deal with the following obvious rejoinder. A statement that there cannot be any general theory of truth IS a general theory of truth. Their general theory is that there can be no general theory!

    Deflationists seem not to be talking about any global theory of practical, system-related truths, as acts of successful ideational representation or modelling, but of each isolated “truth” as truth-in-itself (conflating “true” with “real”). They appear to be talking about the “truth” (actually the reality) of each isolated existent such that the notion of ontological truth in each case, in each example of an existent, deflates into nothing more than an identity equation. That is, the existent (or the statement) is identical to itself and is true or “truth” in this sense only. This fundamentally fails to take account of the relational nature of all existents in a complex, interconnected system. I think tagging Aumann, as essentially a “deflationist” and a post-modern denier of the possibility of approaches to truth and objectivity, is a reasonable characterization. The nonsense he peddles and writes apologia for is philosophically and politically-economically dangerous: highly dangerous as we see now with the impending collapses of human economy, climate and ecology all being driven by neoliberal, market fundamentalist, economics. Aumann understands nothing, it seems, of a philosophically and scientifically defensible theory of truth. as truth approximation and truth approach.

    “That truth is the correspondence of a representation to its object is, as Kant says, merely the nominal definition of it. Truth belongs exclusively to propositions. A proposition has a subject (or set of subjects) and a predicate. The subject is a sign; the predicate is a sign; and the proposition is a sign that the predicate is a sign of that which the subject is a sign. If it be so, it is true. But what does this correspondence or reference of the sign, to its object, consist in?” – Charles Sanders Peirce.

    (Peirce’s last question is crucially important but answering it comprehensively is outside of the scope of this post.)

    A philosopher working within the correspondence theory of truth, operating with a theory and method of analogical or homomorphic correspondence (between signs or models and the signified or modelled), expands the ambit of logical operations to the pragmatic level and permits truth to be a workable correspondence between an idea (an ideational formal system or model) and a real system. A given model idea seeking empirical truth correspondence is a model intended to represent some real system, or aspects thereof, outside the idea itself. By this definition, neither an idea of a unicorn nor Auamann’s ideas are ideas seeking empirical truth. The intention of “unicorn” ideas is usually that of fantasy story-telling. The intention of Aumann’s ideas likewise seems to be that of fantasy story-telling. Otherwise, why the ideological and fantasist appeal to the “invisible hand”.

    There is a difference between gross error and an approach to pragmatically useful objective truths. This is what Aumann denies. Gross error is an unworkable correspondence or a non-correspondence, in practice, between an idea and the external system to which it refers. In science, margins of error, operational tolerances and operative practicalities as everyday and scientific realities in theory and practice, are permitted and admitted within the pragmatic approach to truth as correspondence in the analogical, isomorphic or homomorphic sense. The map is not the territory. However, a map may be isomorphic or congruent with a territory (meaning the real terrain) to a sufficient degree such that one can navigate the real terrain by using the map. Only the most impractical of theoreticians, and one with a secret ideological axe to grind, would define truth so narrowly that operational, practical levels of truth approximation would be excluded in favor of an unapproachable ideal which is deflationary in theory and practice and denies the difference between metaphysical fantasy and science.

    Aumann is living in a fantasy land. Poor fellow, perhaps it is emeritus disease and thus deserving of sympathy for him while yet maintaing our condemnation of such dangerous and reality-denying fallacies.

  2. pfeffertag
    October 17, 2020 at 3:32 am

    Economics is the only social science to use the scientific method and it is the only social science to build a body of theory. However meagre its effectiveness, economics is, relatively speaking, an enormous success.
    I posted at length about this at
    My diagnosis of the problem is not that economics uses the wrong approach but that it needs to expand the scientific approach to encompass interactions that are not narrowly economic.

    • October 17, 2020 at 6:59 am

      I challenge your concept of the scientific nature of economics. It is becoming abundantly clear that the ‘scientific’ models are not working and the reason for this is that economics is primarily about human behaviour and therefore is a study of a social nature which the models ignore to their detriment.

  3. Yoshinori Shiozawa
    October 17, 2020 at 8:43 am

    This is a part of my essay “Between Math and the Occult” written in 2002 for a journal named “The pleasures of mathematics” (in Japanese) No. 30.

    There is an interesting parable that points out this. Originally of Arab origin, it was introduced and disseminated in economics by writers like Martin Shubik and Thierry de Montbrial.

    At midnight, a drunken person was wandering under a lonely streetlight. He seemed to be looking for something. When a passerby asked, the drunken person said he was looking for a key. The passersby joined in the key hunt, saying it would be a problem. But they couldn’t find the key. After a while, the passerby asked for the drunkard. “Did you really drop the key here?” The drunkard replied calmly. “No, it’s over there where I dropped the key, but it is pitch-dark there and I can’t see anything. That’s why I’m looking for the key in the light.”

    Mathematics is a strong light. But, if we consider mathematics as the only tool of theory, it can be the same as the above drunkard. Only problems on which mathematics can spot the light are prone to be studied, leaving other, more important tasks untouched. For 100 years until 1970, theoretical economics was mostly advanced by the light of mathematics. Mathematics is not to be blamed by that fact. However, there are some frameworks that are easy to shed light of mathematics and others that are not. Optimization and equilibrium is such a framework. As a result, huge progress has been made in the logical level. But it was also the cause that economics began to wander in a wrong direction. Ironically, the mistake was hidden as long as light of mathematics continued to shed light on economics. This is also one of the dangerous paths hidden in economics.

    Those who can read Japanese can read the whole text here.

    • Yoshinori Shiozawa
      October 17, 2020 at 8:47 am

      Sorry. Blockquote tags did not work. The quotation starts with “There is an interesting parable” and ends with “dangerous paths hidden in economics”.

  4. pfeffertag
    October 17, 2020 at 11:54 am

    To AustrianPeter: no challenge! I agree with you. Indeed, I think the failings of economics became clear around 1970. Yes, human behaviour! That’s what I meant by “interactions that are not narrowly economic.”
    But “not working” is relative. Compared with the other social sciences economics is a blazing success. Economics may be bad but the other social sciences—which purport to study human behaviour—have nothing.
    As Yoshinori Shiozawa says, logic and mathematics shed light on economics. Not enough, we all agree. But some. You want models to include human behaviour? Exactly!! Let human behaviour be modelled the way economics is.
    If the scientific tactics of economics were applied to those other social sciences they might, like economics, make some progress.

    • Yoshinori Shiozawa
      October 18, 2020 at 9:24 am

      Dear pfeffertag, if you are asking how I characterize human economic behavior, please read my Chapter 1 that has the same tile with our book: Microfoundations of Evolutionary Economics. Unbounded rationality, foresight, or maximization are not necessary to show that a large economic system as big as national or world economy works This is the very topic that we attack from Chapters 2 to 6. In this sense, the neoclassical economics (represented by Arrow-Debreu theorem) is a wrong theory, because human capability is bounded by three kinds of limits: limited sight, limited rationality and limited range (in time and space) of effects.

    • October 18, 2020 at 10:17 am

      Good to have your agreement and thank you for the link which I will follow up. We all seem to be on the same page but I wonder if anything will change in the time I have left! – (I am 76 now). Must leave it to the younger ones to sort out, I am merely an observer but I do like Jim Rickard’s book explaining his views on economics as a highly complex system problem.

      • Yoshinori Shiozawa
        October 18, 2020 at 2:51 pm

        It seems that we are now living in high-aged society, at least at this blog site. I am one year older than you.

  5. Jorge Buzaglo
    October 17, 2020 at 5:17 pm

    Has Aumann a ”narratives” theory of science and truth? Is he a “terraplanist”?

    Cf. the following assertion:

    “… [T]he story about the roundness of the world [is] one way of viewing the world. An evolutionary geological perspective is one way of viewing the world. A different way is with the six days of creation. Truth is in our minds. If we are sufficiently broad-minded, then we can simultaneously entertain different ideas of truth, different models, different views of the world.” (https://www.lesswrong.com/posts/o7K9PSRShHPKaQ3iS/robert-aumann-on-judaism)

    • October 17, 2020 at 5:55 pm

      Thanks for that quote. Says it all!

    • Ikonoclast
      October 18, 2020 at 12:17 am

      With such a theory, namely “a narratives theory of science and truth”, one’s metaphysics become infinitely malleable. One can “prove” anything at all which means of course proving what one wants to prove. It’s an open license for motivated reasoning.

      Motivated reasoning tends to use narrative “logic” rather than standard logic or empirical science to reach conclusions. In this context, it’s probably worth looking at the narrative paradigm theory of Walter Fisher.


      The book “Morphology of the Folk Tale” by Vladimir Propp is also likely important in this context. It analyses the classic narrative structure of folk tales. “Folk Tales” may be taken to mean all mythic tales, including religious stories. They all conform to a standard narrative structure with some switchable elements. I suspect that the standard “folk tale” of capitalism could be analyzed in Propp’s morphological terms. One would have to find a quintessential “tale” of capitalism and then analyze its elements. The quintessential “tale” of capitalism usually begins with the folkloric exposition of barter and the arising of money from that process. I think people on this blog know the kind of tales I am referring to. In addition, we can probably find the quintessential tale justifying private property in the philosophy of John Locke. The analysis would probably proceed in terms of distinguishing the narrative logic in Locke on the issue of private property from the rational logic (if any).

      For those who might think such analysis would simply be an academic sidelight I would say this. You might be right but there is also a possibility of some considerable illumination in this direction. If Walter Fisher is correct and narrative logic persuasion is more powerful than rational logic persuasion (for most people) this raises a real problem for rational and scientific advocacy. I could write more but a reply should be shortish.

      • Yoshinori Shiozawa
        October 18, 2020 at 5:04 am

        Ikoooclast, you are right to say that (by narrative theory of science) “One can “prove” anything at all”. To prevent this, we use logic and math and even models. To “organize our thoughts and observations” (Aumann) is a important part of scientific efforts.

        But, no method is perfect. As Lars Syll argues, models and math are often used in wrong way. Narrative side of theory making is a counterbalancing effort against excess demand for logical correctness. Please also see my comment below on
        October 18, 2020 at 4:49 am,

    • Yoshinori Shiozawa
      October 18, 2020 at 4:49 am

      The last sentence of iwaraini1 in his post on 22nd Aug 2015 (cited by Jorge Buzaglo) was

      Please don’t strawman his arguments or simply dismiss them!

      Doesn’t the article above by Lars Syll “strawman” Robert Aumann? Truth and usefulness are two different aspects of a science. Truth, or the relation between a theory and the real world, is difficult to establish. What is “real explanations”? It depends on persons and problems we face. One of scientific endeavor is “to organize our thoughts and observations” as Aumann points. It is so basic that we easily forget it. Milton Friedman completely ignored this basic efforts of scientific working in his The Methodology of Positive Economics (Essays in Positive Economics. University of Chicago Press, 1953, pp. 3-4). Aumann is presenting a powerful logic to counterattack Friedman’s methodology, which was once dominated (and remain even now retain substantial influence on) mainstream methodology thinking.

      Let me add that a science can be useful without being true or irrespective of its truth. A typical example is astrology. It was as important “science” (not in the modern sense) as economic policies in our days. How do we learn from our theoretical opponents is an important method to improve our (anti-mainstream) economics and methodology. Aumann’s arguments are not complete. Why cann’t we argue in a positive or constructive way to supplement Aumann’s argument defects?

  6. ghholtham
    October 19, 2020 at 10:56 pm

    “Model explanations are at best conjectures, and whether they do or do not explain things in the real world is something we have to test. ” – Lars Syll.
    Sounds right. So how do we test them?

  7. pfeffertag
    October 22, 2020 at 12:00 pm

    To Austrianpeter: I had a read of the Rickards book you linked to. Good, crisp writing. The grand plan is to eliminate the use of cash; very interesting.

    I, too, am 77 so will not see anything sorted out. But I might die happy if I saw economists recognise the inappropriateness of complaining that economic theory does not match reality.

    No one thinks homo economicus, perfect competition, cleared markets, etc really exist, so why go on about it? The reason is to claim the theory is wrong. This is not valid. Deviation from reality is not grounds for complaint. There is always a difference between theory and reality; that is what makes science science. It has been this way for four hundred years. This kind of knowledge has made the modern world.

    In Galileo’s day pendulums were the in-thing. Galileo produced some mathematics to say that whether a pendulum swings only millimetres or it swings through a larger arc, the time it takes to make a complete swing stays the same. It was pure theory. His friend and patron, Guidibaldo del Monte, made some pendulums, collected some data and told Galileo his theory was incorrect. Galileo replied that his theoretical pendulum had no friction, a string that weighed nothing and weight that had no size. Guidibaldo made fun of him—but whom do we remember?

    Galileo also theorised gravity with a perfect sphere touching a perfect plane at a single point and rolling friction-free. If physicists carried on like some economists they’d be pointing to the reality of landslides and complaining that you could never predict landslides using this theory.

    Science theory interrelates idealised, extreme concepts; theory pays no attention to reality. The fact that economics does the same should be cause for celebration. No other social “science” does it and no other social science has built a body of theory. Over the centuries economics has managed to accumulate knowledge—unlike psychology, sociology, political science and anthropology. A good century of intensive research has accumulated mountains of data from which we learn—what? At best, a few disparate factoids.

    Economics has problems but it does work to some extent and there is no evidence the theory is incorrect. The problem is the theory is incomplete. It is as if that drunk had lost many keys through a hole in his pocket. One or two fell under the lamp while others lie in the darkness waiting to be illuminated. Additional theory is needed, not correction to existing theory.

    The economics approach of interrelating idealised extremes is certainly right. This has allowed economists to build coherent theory because idealised extremes allows everyone to have the same understanding of basic concepts. Perhaps ironically, this feature also allows people to complain about them. No such complaining goes on in the other social sciences; it cannot because common understanding is lacking.

    The first step to solving a problem is recognising it. Economics will bust out of its half century of stagnation when further, idealised theory is added. Such added theory would have to be about behaviours other than the purely economic, i.e., behaviours that assumed social environments other than purely competitive. For some general suggestions see
    Since the behavioural sciences have failed at this, it is obviously a big project. Then again, nowhere in those sciences do we find genuine scientific theorising, i.e., theorising of interrelationships between idealised concepts.

    • October 24, 2020 at 12:12 pm

      Pfeffertag, I’m 83, so perhaps not so good as you at saying what I see, but perhaps in my extra six years I’ve seen how information science idealises your ideal. Not only economic theory does not match reality, but spoken language merely represents it. Theories at best outline hidden structural skeletons on which reality has to hang, generating enough space for the vascular and neural systems within which internal motions generate external behaviour. Economics and behavioural theories have generally failed not because of lack of precision but because they haven’t even looked for what they can’t see. David Hume’s epistemological arguments have diverted scientific attention from Francis Bacon’s advice to “take things to bits to see how they work”.

      Now we realise that even the stars we see on a dark night are representations of what has already happened – perhaps billions of years ago – the system analysis of information science has gone beyond Bacon, revealing the joining of circuital paths of converging flows of sets of data surgeons can see physically provided for and contained in the skeletons and vascular systems of our bodies. Shannon’s measure of information capacity indicates the possibility of spare (“redundant”) capacity which can be used to audit and correct deviations: exactly the sort of function physically realised in our immune systems. What it also indicates is that data can misrepresent and thus misinform, as has been done deliberately in wars and monetarised economics. The whole association of game theory with economics, for example, is misinformation. As a game Monopoly is fine, but in real life it is not.

      • October 24, 2020 at 12:35 pm

        More haste, less speed! I forgot your comment on Rickard’s book: “The grand plan is to eliminate the use of cash; very interesting”. I’ve been advocating a “credit card” economy as a way of ensuring everyone can have what they need, which folk here have been taking as my proposing a cashless society. This despite my specifically saying it wasn’t, because cash is needed, particularly for educating children, trading in the absense of hi-fi and by those acting habitually in their second childhood. What I am arguing for is the need for children to be weaned off cash as they learn to see the need to repay in real terms what their accounts represent as monetary debt; that cash withdrawn from an automated teller represents credit already accounted for, so its habitual use in second-hand and otherwise sustainable local trading need not be affected.

  8. pfeffertag
    October 25, 2020 at 10:05 pm

    For Davetaylor. I am making a different case. Bacon was wrong. It isn’t possible to take gravity to bits to see how it works. Same applies to stars. Our understanding there comes from theory and a scientific theory comes from a human mind. How that happens (induction, inspiration, argument with a colleague, a dream) is an open question. Presumably it is a mind steeped in the relevant field.

    Language is a problem and social researchers (excepting economists) long for agreed definitions. People complain that theory is remote from reality, however the benefit of theory is independence from the reality of language. In science and in economics, theorising makes definitions irrelevant. Everyone in the field has the same understanding of a theory, which is to say the theory is unaffected by theorists’ various opinions of what its concepts actually are. Independence of language arises because the theory is a relationship between extreme idealised concepts. This common understanding may be remote from reality but it allows a body of theory to be built. The other social sciences don’t so theorise and cannot build a body of theory.

    In so far as game theory is applied to economics it is not misinformation. Galileo’s perfect sphere on a perfect plane is not misinformation about landslides; it is insufficient information. Galileo’s theory is needed to understand landslides. I sometimes think left wing economics is nothing but a gigantic diatribe against economic theory. This is misplaced. Economics rules the world and it is successful because it has theories, be they ever so insufficient.

    Data is another matter; data requires definitions. No science theories have ever come from data. With the advent of computers the social sciences became enthusiastic about data but counting things cannot deliver theory. Data collection is vital for administration (planning, selling, informing, indoctrinating) but science theory interrelates measurements, not counts.

    The Rickards book is not so much about credit cards. It is more about minimising bank account cash balances via zero or negative interest rates. The book wanders about somewhat but it holds your attention.

    • October 27, 2020 at 4:14 pm

      Pfeffertag, while I don’t entirely agree with you, I can understand why not. Francis Bacon was the originator of modern science, proposing what we now call “normal” or applied science as against the more “revolutionary” or fundamental science: this (like Newton’s mechanics) uniting different points of view which hadn’t arisen in Bacon’s time. Newton’s mechanics is idealised in neglecting friction, but Shannon’s information theory is all about the equivalent of friction in communication and Keynesian economics: uncertainty due to “noise”.

      If Rickard’s book is about minimising bank balances (as Keynes praised Gesell for doing) then I’m with him. My “credit card” version gives you credit (i.e. recognises your credit-worthiness) which indebts you only insofar as you spend it, this being written of by earning our keep. (There is more to that than a one-line summary, of course).

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.