Thursday, October 24, 2013

Can Philosophy enlighten climate science?

In recent weeks several threads here on Klimazwiebel have raised the question if the social sciences and humanities can play a role in the discussion of climate research, and if so, which. I think this is an important question and suggest to discuss exemplary texts in the following months.

First up the list is Nancy Cartwright's work which has been discussed on the previous thread 'Debunking skeptical propaganda'. Cartwright has made an interesting journey from an 'anti-realist' to an 'anti-fundamentalist' position. The anti-realist position was developed in her book How the Laws of Physics Lie, the anti-fundamentalist position is developed in her later works. In Nature’s Capacities and Their Measurement she develops and embraces the concept of 'capacities' which replaces the concept of 'law' (which she rejected). In The Dappled World she tries to marry both approaches from the previous books.

Fundamentalism to her means 'blind faith' in the working of laws of nature, often 'in the face of contradictory evidence'. If you want to read a defence of fundamentalism, here is a nice quote by Carl Hoefer: 'These [fundamental laws of nature] are what physics has been seeking, and getting closer and closer to actually grasping, since the time of Descartes. They are truths, expressible in mathematical language, that accurately describe the behavior of all things in the physical world, at all times and places. This view has been standard among physicists, and most philosophers of science, for at least a hundred years” (quoted in Teel, see below)

As I do not want to suggest an in depth discussion and evaluation of Cartwright's philosophy of science I have chosen one example mentioned by Cartwright (in The Dappled World) as starting point. This example contains an obvious parallel to a problem encountered in climate science. I quote from a recent PhD thesis by Paul David Wilkinson Teel which summarizes the issue nicely (and is a treat to read):
Cartwright refers to a thought experiment put forth by Otto Neurath in his 1933 article, “United Science and Psychology,” in which a one thousand dollar bill is swept away by the wind in [Vienna's] Saint Stephen’s Square. Cartwright says that those of us brought up within what she calls the “fundamentalist canon” know through Newton’s second law that force equals mass times acceleration. This law applies to falling objects; the thousand dollar bill is a falling object; therefore, the physicist should be able to predict where the bill will land. But, in practice, the physicist cannot. 
Cartwright says that Newton’s second law does not apply to this situation, because (like all scientific laws) it applies only in models that have been deliberately set up so that, all things being equal, the law holds. These models invariably involve shielding, which keeps out things like wind (or even air), magnetic fields, sound, or whatever else might interfere with the particular law at work. Cartwright writes that many scientists would object here and say that “there is in principle . . . a model in mechanics for the action of the wind, albeit probably a very complicated one that we may never succeed in constructing.” But Cartwright says this objection is based on a fundamentalist faith rather than on any evidence we actually have.
This is why Cartwright argues that the world, rather than being homogenous under a universal rule of law, is dappled. She acknowledges that natural laws are real and are useful. But she insists they are limited to shielded models that do not always match real-life circumstances. Others, she says, disagree: 'Fundamentalists want more. They want laws; they want true laws; but most of all, they want their favourite laws to be in force everywhere. I urge us to resist fundamentalism. Reality may well be just a patchwork of laws.'
The analogy to climate models should be obvious and the following two questions arise: How fundamental are the 'laws' or 'principles' which govern the models, and how much 'dappledness' does climate science allow?

Note: Strictly only comments on this topic. Others will be deleted.


Unknown said...

I think that although philosophy can play a minor roll in climate, it should be mostly up to science and hard core facts.

MikeR said...

I think the issue is that anywhere you look where you can constrain the conditions enough that you can calculate what the results should be, either (a) that is what the results are, or (b) you then realize that your theory is wrong, work out a new theory, and that works.
The author is suggesting that since there are lots of areas we can't calculate because they are too complicated, it is an act of faith that scientific theories would explain those areas if we ever could calculate them.
But it seems to me that faith is based on a lot of experience. Why should it be that (almost? most?) all the areas we have calculated do work according to calculation?

Heber Rizzo said...

So, it would be something like: "lacking science, we should resort to belief".
That is post-normal climatic science, really.

Karl Kuhn said...

Auweia, this is in danger of becoming a pretty confused discussion. Maybe it would be useful to first clarify what the lady meant before starting to discuss links to climate science and, yeah, modelling, right?

Let me try an own interpretation of Cartwright with which I can work further:

"Physical 'laws' do not necessarily work [visibly] in very complex situations, so we should not expect that these laws reign supreme."

This would be a stupid argument. If the money note does not drop, the Law of Gravity cannot really be a law, because reality does not obey it. I could similarly claim that the functioning of ships or airplanes puts the law of gravity into question. Let us assume that this is not what Cartwright really meant. But I have no idea what else it could be, what problem she wanted to address, and what her alternative explanation would be.

Anonymous said...

I appreciate you initiating a dialogue between the humanities and climate science. But I am a bit puzzled. Would this be a correct interpretation/extrapolation of what you say?
The weather forecaster tells me that it will rain tomorrow - I take an umbrella - am I a 'fundamentalist'?
I take a plane - assuming that it will normally stay in the air - am I a 'fundamentalist'?
I am told by epidemiologists that there will be a severe outbreak of flu this winter - I get myself vaccinated - am I 'fundamentalist'?
I am told by climate scientists that greenhouse gases contribute to global warming - I reduce my greenhouse gas emissions - am I a 'fundamentalist'?
I bet I got something wrong here ....

Howard said...

This is a critically important topic. Laws, like Gods are fictions made up by mankind to help explain and understand the world. That is not to say that scientific laws are equal to religious doctrine. However, one religious law I can think of is pretty close: human pride (vanity, ego, call it what you will) is the #1 source of depravity and suffering.

We all know A=A and that if A=B and B=C then A=C. Except that in our universe, there are no unit things and nothing is equal. What equal really means is that, to a close practical approximation, these things are near enough indistinguishable from each other.

This philosophy boils down to all models (or physical laws) are wrong, but some (or almost all in the case of physical laws) are useful.

To answer Raffa, I would say that there is no such thing as a fundamentalist, however, some beliefs are based more on faith than others. Comparing an insignificant weather precaution and modern aviation, which is falsifiable and tested millions of times per year, to the poorly supported efficacy of flu vaccine and the unfalsifiable theory of CO2 induced global warming is a silly trick of sophistry. However, that is due to the unstoppable pride most people feel in their certainty of how the world works.

Since climate science is a small part of geology and geology is ungodly complicated, it is useful to recognize that it is easy to miss important factors that result in the drilling of dry holes.


@ReinerGrundmann said...

Cartwright points to the difference between an idealized (closed) system and empirical reality. She says no model can predict where the dollar bill will land (although we may identify the forces that are at work to make such a prediction). A 'fundamentalist' says we could construct such a model if we wanted because we know the relevant forces. Cartwright retorts that because such a model does not exist the fundamentalist position is expressing a belief.

Anonymous said...

@Reiner Grundmann

I don't really understand what you try to say.

Is the model to complex or do we interpret the results of a model as reality, which it is not, even if the result were 100% correct?

Models in physics can have different meanings. In meteorology we have nice results even if the models are not 100% correct. Some physical laws may not describe reality, but their results are 100% correct.

If climate models are not = real, this may just be a lack of understanding every bit of the

But these are two very distinct problems. The one is philisophic, the other a lack of knowledge (so far).

Imo Karl Kuhn has pointed out the problem. How can we discuss if we don't know what we are talking about.

The problem rises again and again.

Best regards


Karl Kuhn said...

"She says no model can predict where the dollar bill will land (although we may identify the forces that are at work to make such a prediction)."

The analogy to climate models is there, but not very strong. Climate models do not claim to make prescise 'point predictions'. They claim to include all relevant shift factors (forcings that are subject to a trend, most of them anthropogenic), while treating a lot of natural variation as noise. The more detailed these models become, the more of natural variation is included - hopefully, increasing the predictice power of such a model on smaller time scales. So justification to continue with such models is that they are getting better with time, and that they are able to tell us already today about long-term trends in the global energy balance. That climate models are not able to precisely predict the global average temp anomaly some 40 years ahead (a sceptic argument of the stupid sort) is no reason not to use them, as long as they get long-term trends right.

"A 'fundamentalist' says we could construct such a model if we wanted because we know the relevant forces."

Indeed, climate modelers sometimes claim they know all the relevant forces, and that climate simulation models are elegantly built on 'basic physical principles', which is why it would be childish to question them. This argument assumes that the climate system is of limited complexity, and very well udnerstood already. A letter to the Economist by some PIK guy (reacting on the 'faltering climate consensus' article in summer 2013) is a nice example of this: 'If we question climate models, we have to question basic physics!' Given the uncertainty in many areas of the climate system, particularly water vapour and cloud formation, this argument is rather a smokescreen to hide uncertainties and preoccupations (with the role of CO2) in climate modeling.

"Cartwright retorts that because such a model does not exist the fundamentalist position is expressing a belief."

I do not believe that our problems to (1) statistically analyse and, based on that, (2) simulate complex processes of climate allow the conclusion that basic laws of physics are not reliably working. My gut feeling is that this is rather an ideological attempt to 'delegitimize' or 'deconstruct' natural sciences, something which has become fashionable in leftist academic circles in recent decades.

Howard said...


All of science deconstructs and de-legitimizes itself from within by emphasizing the publication of positive or sexy or controversial results to increase Q ratings and grant funding.

No one is paying attention to lefty academic social science deconstructions that has become fashionable, as you say. I suspect their deconstruction of *hard* science is a defensive response to their systemic impotency.

@ReinerGrundmann said...

I want to appeal to the good sense of the commentators before the levels descend to new lows. Please show a bit more discipline when discussing this topic. The fact that you do not agree with what you read does not mean that you should rubbish it -- and with it the whole of social sciences. This looks like the worst caricature of scientists engaging with the philosophy of science.

To avoid misunderstanding: I am not advocating Cartwright's position but am presenting it for discussion. So please do not confuse Grundmann and Cartwright. I might play devil's advocate, especially when I don;t see genuine engagement with the arguments. To repeat: please address the points of substance and try to engage with her argument. She has a quite clear exposition if you care to read her. If you disagree, please try to make your case with arguments, not with innuendo and broadsides against social sciences. Do I sense some uncomfortable message in her position which some commentators find difficult to accept? Is this the reason for the emotive reaction?

Karl Kuhn:

you exemplify nicely one point made by Cartwright when you express your 'belief' in your last two sentences. Should this not give you pause to think?

Karl Kuhn said...

Dr. Grundmann,

no, my last sentence has nothing to do with what Cartwright says. It is, as I said, my gut-feeling, a political speculation based on occasional observations of what has been going on in post-modern cultural and social science. If someone titled a book with 'How physic's laws lie' in the 1980s, it is likely that she also wanted to send a political message. I will think again once you or others prove me wrong in whatever way. Psychologizing me will not do the job.

Moreover, there are lots of thoughts and arguments to be found in this comment section, maybe not all of them very polite. Sorry for not having read the original book - I guess most of us are dependent on your summary. You may appeal to the readers of that post first to read the book before joining the discussion.

Günter Heß said...

@Reiner Grundmann

I don’t quite understand Cartwright, because I was never taught in physics that Newton’s law is more than a basic equation that is used to describe a set of phenomena.

So its limits is made obvious to scientists who study physics.

Cartwrights argument therefore seems to me a “straw man” argument about fundamentalists or believers whom I did not encounter in physics throughout my scientific career. Or maybe her argument is directed to other sciences like philosophy or social sciuences.

Carl Hoefer seems not to be a physicist.

However, according to your source, Cartwright writes:
„Cartwright says that Newton’s second law does not apply to this situation, because (like all scientific laws) it applies only in models that have been deliberately set up so that, all things being equal, the law holds.“

How does Cartwright know that Newton’s law does not apply?

Best regards
Günter Heß

Pekka Pirilä said...

More or less every working physicist has trust that basic laws of physics are valid over a wide range of applications. They know also that many of those laws are violated under some other conditions, as Newton's laws are violated when Special Relativity gives more correct answers. Philosophy of Science is not likely to provide much additional understanding here.

Cartwright may have somewhat different situations in mind. Most practical problems are too complex for being solved exactly based on the basic laws discussed in the above paragraph. Thermodynamics and Statistical Mechanics are examples of physics where the laws have a different nature. The Classical Thermodynamics is a separate mathematical theory built on abstract postulates. Its original justification came largely from empirical success, but it was not an accurate theory for real world systems, only an extremely useful abstraction. Statistical Mechanics is another mathematical theory that can be used to justify Classical Thermodynamics. The connection is convincing, but some fundamental issues remain, some of them are close to philosophy.

Proceeding to Fluid Mechanics, the difficulties grow. The fundamental equation of Fluid Mechanics, Navier-Stokes equation is again a mathematical idealization as it is written to be valid down to zero distance, but real physics deviates from that at some point. The problems of N-S equation are large enough to make a Millenium Problem of mathematics. Real physics is well behaved at short and large distances, but N-S might fail. Even so it and the methods based on it are useful. They are useful even though calculation of turbulent flows has not been successful at fundamental level.

The case of N-S equation might be a case close to the problems Cartwright brings up. Similar issues come up in other complex problems. They are not solved from basic equations of physics but using some other methods believed to be applicable. But here we come back to the question "Can Philosophy enlighten physical sciences?". It's not at all obvious that the answer is positive.

(Other issues come up with Quantum Mechanics and its interpretation, but I don't go further in that direction.)

Pekka Pirilä said...

In my previous comment I didn't discuss Climate Science. The Earth System and its long term behavior are definitely complex problems that cannot be solved fully and with high accuracy. The complexity leads to both purely practical and more fundamental problems.

Fundamental problems apply to what's ultimately possible to reach in making climate predictions (or projections), practical problems to what it's possible right now, and how correct the present "best scientific understanding" is. I would not classify even the fundamental problem as philosophical but as an issue best answered by natural sciences.

Choosing climate policies is a different issue. Weighing present and future, well-being of our own children and of the present or future people of poor countries are issues of ethics, not of climate science nor of economics.

eduardo said...

very interesting post, thank you.

I just wanted to underline that most of the real problems in physics cannot be solved exactly. A archetypical example is three bodies mutualy atracted by Newtonian gravitation. This means that even assuming that Newton's laws are correct, they provide an excat solution only in the most simple problems. Thus the p4oblems faces by climate models are notat all unique. They are very common. In this sense, physics is an aproximate science, not an excat science

Anonymous said...


Can the problem not be solved exactly? Is it to complex to be handled and calculated exactly? Or do we ignore some influencing factors?

Or do we just confuse model with reality. If we are simplifying a problem in a way that it can be solved but does not correspond any more to reality, is that what she is talking about?

Or are climate scientists overconfident about their knowledge?

I really don't know what she tries to say, nor what you try to say.

Bitte entschuldigen Sie mein furchtbares Englisch.


eduardo said...

@ Yelph,

Sorry for the lack of clarity. There are two different levels of 'complexity'. Let me give you two examples:

In one example we know the governing laws, but the solution of these laws is mathematically too difficult. Three bodies, for instance the sun, the Earth and the Moon, attract one another via the well known Newton law of gravitation. The exact prediction of their trajectories is impossible because, although the laws are known, they are in this case mathematically intractable. Thus, further assumptions that render the problem easier are needed. In this case, that one of the bodies is much larger than the other two. This further assumption allows to determine the trajectories, but then this prediction is only approximate.

The vast majority of real problems in physics are of this kind. The laws are known, but intractable. The confusion may arise because at school we only learn of those problems that are solvable (the harmonic oscillator, two bodies under gravitation, etc). But this short list very much contains all that can be solved exactly (= an exact prediction is possible)

A second example is when the physical problem itself is complex that the laws are not known. For instance, the flow of fluids. In this case, we only know approximate laws, which in turn are also difficult to solve in general and require further simplifying assumptions. However, you have to consider that, even with these uncertainties, we can build planes and ships. The key aspect here is , however, that in these cases we can perform experiments, test our assumptions, correct if necessary, etc. In climate science this is, obviously, much more difficult.

Additionally, you have to be aware that all 'computer' solutions are intrinsic approximate. s illustration, computers do not know what the numbers pi or square root of 2 are. these are irrational numbers that require infinite digital representations (pi=3.1415....etc). A computer can hold only a truncated, finite, representation of pi, and thus it is only approximate.

This is what I was trying to say. Physics and even more so computerized physics is an approximate science. Cartwright, as I see it, refers to other, more fundamental, issue (next comment). Hvw may see it differently, though.

eduardo said...

I try to illustrate my interpretation of Cartwright. Newton's law of gravitation is that two point masses attract each other with a force proportional to their mass and inversely proportional to the squared distance. We encounter here already two problems: one is that the law is formulated using 'objects' that clearly do no exist (nobody has seen a point mass; what is a force? ). Thus, to achieve a prediction that can be compared to real observations, the law has to be augmented with further 'interpretations' and attached strings. For instance, that the forces exerted by the infinite number of 'point masses' that form a whole solid body can be added. In other words that the attractive force with which of a minuscule portion of the Earth at the equator attracts a minuscule portion of the Moon at its pole, can be added to all other minuscule forces to calculate the total attractive force.
One can see that a physical law is a law with a lot of strings attached to it. It is formulated in terms of abstract concepts that do not exist and have to be some how translated (mapped) to the real world.

The second, more fundamental problem, is the following: where does this law come from ? Obviously, it comes from Newton's brain. How can we tell that this law 'exist' in nature and is universal, and not simply a formula that Newton concocted and that happened to match the phenomena he observed at his time ? Newton though his law was fundamental and universal. we know now that it is not the case - as there are observations that do not match his predictions, mostly at large cosmological scales or in high precision measurements close to the Earth. For that we need the theory of general relativity - which is formulated din totally different terms and using totally different concepts. Now imagine that Einstein and Newton had lived in the same epoch and that both had put forward their theory at the same time. The available observations at that time could not tell which theory was correct, and so we would think that both were just concocted formulae to describe the observe phenomena, but being so different, at least one could not be fundamental.
You can ask your self the following question. When we eventually meet inhabitants of other galaxies, will they have the same 'fundamental' theories as we have ?

Cartwright argued that our theories are just formulae that fit the observations, but they are no way fundamental - other civilizations will have other formulae. They are useful, because they allow us to make predictions .. until they fail.

hvw said...

Cartwright, as I see it, refers to other, more fundamental, issue (next comment). Hvw may see it differently, though.

No, I see this pretty much the same way. The only disagreement with you seems that I do not share your interpretation of Cartwright that predictive power stems (only?) from the "fundamental" laws.

I think this discussion is a distraction from the more relevant, yet related questions, which are important for the climate change debate. Other commenters have mentioned a number of environmental models as examples. I would think it more interesting to ask 1) What makes us trust (or distrust) predictions of environmental systems, and 2) How is the level of trust related to our willingness to act on those predictions?

Weather prediction stands on epistemologically much more solid feet than climate prediction, because we know its performance very well, in probabilistic terms. And we act on severe weather warnings. An epidemiological model about a newly discovered virus might be as shaky as climate prediction, yet we likely act on it. What is the difference? Isn't there room for the social sciences to explain something? Has it to do with the personal responsibility of decision makers if they ignored the epidemiologist’s warnings? Maybe we overestimate the relevance of more reliable predictions for political decisions (I guess that is what von Storch and Krauss say)? Should that then not be reflected in research priorities? If a researcher had the possibility to bracket climate sensitivity more closely, say from a range of 3 to 1 degree C. she would definitively go for it (probably does it right now) and count that as enormous success. But would that scientific breakthrough be reflected in policy? Would that change anything at all?

eduardo said...


I do not really understand your first point, but maybe we are talking past each other. For me, if a law is just phenomenological (my understanding here is an equation that fits the available observations) , its predictive power is very much in doubt. It may be right outside its rage of 'interpolation' or maybe not. There is not solid ground to claim it has predictive power. This is different for a fundamental law, as Newton probably understood.

Concerning your second points, we have other examples in which predictions are clearly very bad, and yet policy is based on such predictions, usually with huge sums involved. Economic predictions are very wobbly; many of the mechanism that drive economics are not well understood. For instance, it is not certain how Quantitative Easing acts on aggregate demand, and yet the FED pumps huge amounts of money. Angela Merkel navigates 'auf Sicht'. My impression is that policy makers do not pay much attention to scientific predictions. They act only if the scientific argumetation help them win elections

Karl Kuhn said...

an interesting post over at Judit Curry:

She quotes Briggs:

"if the predictions derived from a theory are probabilistic then the theory can never be falsified. This is so even if the predictions have very, very small probabilities. If the prediction (given the theory) is that X will only happen with probability &epsilon (for those less mathematically inclined, ε is as small as you like but always > 0);, and X happens, then the theory isnot falsified."

So we have to settle on a minimum value of epsilon above which we still believe the model?

Another interesting quotation from this post:

“In then addressing the question of how GCMs have come to occupy their dominant position, we argue that the development of global climate change science and global environmental ‘management’ frameworks occurs concurrently and in a mutually supportive fashion, so uniting GCMs and environmental policy developments in certain industrialised nations and international organisations. The more basic questions about what kinds of commitments to theories of knowledge underpin different models of ‘complexity’ as a normative principle of ‘good science’ are concealed in this mutual reinforcement. Additionally, a rather technocratic policy orientation to climate change may be supported by such science, even though it involves political choices which deserve to be more widely debated.”

I think this is all spot-on what is discussed here by hvw and Eduardo.

Anonymous said...


Thank you very much for your reply. Now it is more clear what you try to say.

I wonder if Cartwright is so clear in what SHE tries to say. I like to read Karl Popper and I had to read Kant. But what Richard Feynman says is what I can live with. Imho Cartwright has nothing new to say.

Nobody really believes anymore in natures laws the way she pretends they/we do, imo.

If YOU are so straightforward in your thinking, why would you need philosophy? The complexity and inaccuracies are part of the process of understanding, not an obstacle.

Kind regards


Peter Bobroff said...

The involvement of the social sciences and humanities in support of CAGW provoked me to look at the extent of the scientific method and integrity in those disciplines. To my horror, CAGW looks good in comparison.

Perhaps CAGW and similar memes flourish in the generations of social science and humanities students turned out by the universities. For the skeptics, CAGW may just be a short term skirmish with the major campaign being against academic disciplines not strongly embracing the scientific method.

Perhaps the scientific method is the major defence against all too common human failings, but that is probably a question for the social sciences and humanities.

@ReinerGrundmann said...


thanks for your elaborate comments of examples which I take to be rather supportive of Cartwright's position.

You express, however, a belief that such problems in science will best be solved by scientists, not philosophers. I think no one doubts this. The question is if philosophy can provide a conceptual toolkit to understand these problems.

Pekka Pirilä said...


I have not read enough of Cartwright's writing to tell how supportive or contrary I would consider my views to be. What I tried to do is to describe thoughts of a (former) theoretical physicist who likes to ponder also the philosophy of science.

My views may often be at least partially supportive as my experience is that many issues can be described correctly in many different ways that appear contradictory in a superficial consideration. That's true within the "exact science" of physics to much larger degree than most realize. In the case of well understood physics all alternatives must, however, lead ultimately to the same results for concrete well defined cases.

hvw said...

Eduardo, #21
I do not really understand your first point, but maybe we are talking past each other. ...

Yes, I think we are. Maybe because you are a scientific realist and I am leaning towards instrumentalism and have a hard time to identify and distinguish these "fundamental laws". However, examples where prediction based on "phenomenological laws" is superior are abundant. Most of hydrology and its successful predictions are based on Darcy's law, a posterchild for a "phenomenological law". How much "fundamental" law goes into tomorrow's weather forecast (given that the primitive equations do not count as "fundamental")?

My impression is that policy makers do not pay much attention to scientific predictions. They act only if the scientific argumetation help them win elections

Given the poor performance of politics to act on results from climate science, I can very much understand your frustration, which must be the reason for this nihilistic comment. Yet, you can't deny that fortunately we live in a society where many political decisions are strongly influenced by rational considerations based on scientific results. Maybe the reason for this is that a politician who obviously (to the electorate) acts irrationally will loose elections. That'd be an argument to push outreach activities in climate science though.

Your example of "economic predictions" appears misplaced to me, because these "predictions" are of a fundamental different sort than even the most "wobbly" predictions in the natural sciences. Economics seem to be always based on an ideological fundament and that seems to be true for most of the other social sciences too. That explains how they arrive at the same time at completely contradictory predictions based on the same data. [This is a topic fitting the next thread, maybe answers should be placed there] In most economics the underlying ideology is clear. However, when it comes to work related to climate change, the climate-economists and -sociologists frequently "forget" to identify their normative assumptions, which makes them often appear so ridiculously hypocritical when, at the same time, they complain about the IPCC not reporting in a "value-free" manner.

@ReinerGrundmann said...


'Economics seem to be always based on an ideological fundament and that seems to be true for most of the other social sciences too. That explains how they arrive at the same time at completely contradictory predictions based on the same data.'

Do you have evidence for singling out the social sciences?
Recently Hans von Storch posted on the concept of 'underdetermination' which is a concept that applies to all sciences.

hvw said...


The majority of scholary work in economics seems to be based on, or accept, a particular school of economic thought. There are many, and many of them seem to employ fundamental normative values and/or beliefs that have no epistemic base. Neo-liberals and Marxists won't settle their dispute by an elegant experiment, as string-theorists and proponents of loop quantum gravity might, one day.

As for other social sciences, just look at your own publication list :). While your latest article seems to be an example of "trying to keep the values out", the preceding four could not have been written without axiomatically adopting certain values and beliefs and a lot of it actually consists of stating such beliefs. That is not bad in any way, but it is the privilege of the social sciences.

I guess the answer to that one has been provided recently by Gregor Betz?

@ReinerGrundmann said...


I am disappointed. You should try harder: you make very strong claims ('Economics seem to be always based on an ideological fundament and that seems to be true for most of the other social sciences too') without a shred of evidence. In my scepticism about mainstream economics I might agree with you in a facile way but the onus is on you I am afraid. Did you hope for an ideological agreement here?

You point to 'articles' of mine to substantiate your claim about the 'social sciences' which honours me in a way (who would have thought that my publications could serve as a proxy for the state of social science!). But these are blog posts (or did you really mean you read my past four journal articles and your comment was about them?)

With regard to the difference between economists and physicists you suggest the latter would 'settle their dispute by an elegant day'. So there is hope for this tribe, why not for others?

Do you think current economic debates are led by Marxist economists fighting the Neo-liberals? Or was it Keynesians...?

Not sure I understand your point about underdetermination and maybe I have not expressed myself clearly when pointing to the discussion about this concept. The underdetermination thesis holds that no theory is completely determined by empirical evidence, and that there is always more than one theory that is able to explain a given dataset.

hvw said...


maybe you should concentrate less on finding imperfect qualifiers of statements but rather try to understand. I said "The majority of scholary work in economics seems to be ..". Feel free to add any amount of "often", "not always", "with exceptions", "to my knowledge", etc. instead of grabbing an "always" from a preceding post.

["You point to 'articles' of mine ..."]
I was indeed referring to your publication list on your university homepage.

["With regard to the difference between economists and physicists .."]
You disagree that (in principle, almost, as commonly accepted) all questions in physics can be settled by experimental evidence, while this is not true for Economics? Interesting.

"Do you think current economic debates are led by Marxist economists fighting the Neo-liberals? Or was it Keynesians...?"
Wonder what makes you think this?

I was referring to your quote of Betz in the other thread here: "The common core of both versions is a kind of underdetermination thesis." because it is related to the claim of the necessary involvement of non-epistemic values in scientific results.

@ReinerGrundmann said...


even with the less absolute qualifiers you are making sweeping statements. It is perfectly OK to say that this is what you believe but you cannot accuse social scientists of inherent bias (thus exempting yourself) but then go on and profess in an opinionated way about the social sciences.

I am on record of saying that in some instances we can see motivated reasoning among scientists and social scientists but one needs to document them.

It is true that some sciences have the methodological option of performing experiments but there are caveats. One is 'experimenters' regress', the other is that not all sciences can use experiments to settle disputes. Climate scientists have one planet (no 'control' planet), the same applies to social sciences. They try to find or construct 'natural' experiments, mainly by using comparative methodology.

Regarding underdetermination, you seem to assume that Betz sneers at the concept. He does not. This is what he has to say about underdetermination reigning in climate science:

'Different methodological approaches which avoid the underdetermination problem have been developed in the philosophy of science during the last century... these approaches, though possibly yielding insights when applied to other disciplines, are inapplicable to climate science—a failure which explains why underdetermination reigns in climatology.'
'Underdetermination, model-ensembles and surprises: On the epistemology of scenario-analysis in climatology', J Gen Philos Sci (2009) 40:3–21 DOI 10.1007/s10838-009-9083-3

Karl Kuhn said...

As an economist, I do not take offense at HvW's characterisation of scholarly work in economics ("The majority of scholary work in economics seems to be based on, or accept, a particular school of economic thought. There are many, and many of them seem to employ fundamental normative values and/or beliefs that have no epistemic base."). I am aware that this is quite representative for the perception of economics by other disciplines. But this perception is basically about normative economic analysis which explicitly (!) contains value judgments, otherwise it would not make sense. Positive economics - on which normative economics are then based - by contrast, should not contain value judgments ... at least that's the idea.

Nicely explained here if you can sacrifice the 5 min:

Economists are well aware of the problem of inherent value judgments in economic analysis, maybe because they have been in the policy advice business quite some time already. And we don't mind being called ideologues or heartless penny-pinchers and the like as long as we are allowed to respond (ie not demonized).

@ReinerGrundmann said...

Karl Kuhn

'Economists are well aware of the problem of inherent value judgments in economic analysis...'

The problem may start even before analysis, through the definition and framing of an issue. Many concepts have a built in 'positivity' (here used as 'normativity'). Think of employment (good)-unemployment (bad).

Climate change is regarded by many as 'bad'. Those who refuse to do so point to climate as a natural process, and we usually find nature 'good', or indifferent (can be 'good' and 'bad').

Terms like climate catastrophe and Klimaschutz (climate protection) are inherently value laden. If this is the case, the nice distinction between positive and normative science crumbles. Scientists would need to police their language, trying to find neutral concepts which then could be used for evaluation (in terms of supporting or opposing them, seeing the as 'good' or 'bad' ).

The same problem appears with descriptors of discussants in the climate debate, skeptics are not happy being called 'deniers', advocates sometimes object being called 'advocates', and mainstream scientists (and 'advocates') somehow want to share in the virtue of skepticism, thus denying the term 'skeptic' to 'deniers'.

Perhaps the philosophy of language (or plain linguistics) can be helpful here?

Nicholas Shackel said...

I think you should know that 97% of philosophers of science think Cartright's position is false (OK, I chose the percentage for its resonance with another claim, but it is true most philophers of science, and also metaphysicians, are what she calls fundamentalists). Fundamental laws can easily give the appearance of her dappled world just through complexity.

@ReinerGrundmann said...


not sure I get your point(s). You seem to be saying something about other scholar's views of Cartwright, and Cartwright's characterization of others. Is this correct?

Can you provide some evidence for the 97% claim?

And what do YOU make of her propositions?

Nicholas Shackel said...

On the whole we don't do surveys of opinion in philosophy (although I suspect if we could aid a stream of funding worth billions of pounds by doing so we would) but if you look at current literature in the philosophy of science and metaphysics it will be evident that Cartright's view is not one people are bothering with or bothered by. Furthermore, I gave you a direct reason for why her view is false: any world of with fundamental laws (to use her terminology) but sufficient complexity will give an appearance as of her dappled world. Really she is just running a very old kind of anti-realist argument based on under-determination by evidence. In the end all these arguments just come back to inductive logic doesn't guarantee the truth of the conclusion given the truth of the premisses. So I really don't think you or anyone should be basing anything on her philosophy of science. Complexity and the mathematical chaos of non-linear dynamical systems alone suffices for the kind of points being made.

@ReinerGrundmann said...


"if you look at current literature in the philosophy of science and metaphysics it will be evident that Cartright's view is not one people are bothering with or bothered by."

What do you mean by that? That I should look at the citation index? At a selection of top journals? And what would these articles tell me? That Cartwright is criticised a lot, or that she is not mentioned at all?

Still, I am puzzled by your statement that "it is true most philophers of science, and also metaphysicians, are what she calls fundamentalists"

As you will have perhaps noticed, Philip Moriarty vehemently denies this.

Nicholas Shackel said...

I shouldn't have said 'what she calls fundamentalists' but just 'fundamentalists' in the sense of believe in fundamental laws. I don't see Moriarty denying anything I am saying but rather agreeing with exactly my point when I said that any world of with fundamental laws but sufficient complexity will give an appearance as of her dappled world.

Will you please get rid of the stupid proof you're not a robot nonsense in your commenting system. You should be wasting our time like that.

Nicholas Shackel said...

'What do you mean by that?'

I meant that if you look at current literature in the philosophy of science and metaphysics it will be evident that Cartright's view is not one people are bothering with or bothered by.

Hans von Storch said...

Just for technical info - this blog is normallly not moderated, but after some time, postings are closed for comments -and this one is closed- so that at all additional comments must be ok'd by hand. This takes a little time, but all comments are published after no more than, say 18 hours. Sorry for inconvenience.
It happens quite often that ad-like comments are sent to old postings.

@ReinerGrundmann said...

You are repeating yourself without answering my specific questions.