Wednesday, September 14, 2016

Ramanan — Physics Envy


Ramanan schools Paul Romer.

The Case for Concerted Action
Physics Envy
V. Ramanan

19 comments:

Ignacio said...

Bullshit, string theory has many flaws and problems and has been critiqued by many physicists, there is a lot of literature on that. The fact is that even nowadays string theory has become more a field for mathematical research than physics. It has produced some good thought, I don't think many would disagree with that, but it's connection to real physics has faded more and more and phenomenologists have been increasingly critical of it.

Appeals to authority are crap, because there are 'authorities' both supporting and not supporting the theory, and all of them have 'reasons'.

The problem comes when a discipline becomes a circle jerk, go figure that most breakthroughs in any discipline come from thinking outside of the box, but when the whole discipline becomes an echo chamber and drags out for decades it's helpful start rising an eyebrow at the lack of empiricism.

So given a choice between continuing a line of thinking which has cornered itself with a lack of empiricism and branching out and allowing some fresh air, continuing trying to approach a problem the same old way because 'reasons' and 'promises' it the worse of both options. This has a terribly detrimental aspect on the discipline, specially on how bad are research positions nowadays on the academe due to influence of neolib zealots.

Young people is discouraging from pursuing their own paths or trying to come with refreshing theories, all to follow in line and keep the echo chamber resounding in exchange of tenures and a research vacant.

Matt Franko said...

Rationalist theories are all BS....

Matt Franko said...

To me Romer is arguing against rationalist theories in general... But he doesn't frame it that way...

Tom Hickey said...

Rationalist theories are all BS....

All theories are rationalistic. They are formal systems whose consistency ensures the logical necessity of theorems entailed by axioms or postulates (stipulations).

If the starting points are assumptions about empirical reality, then the theory is a scientific theory. To the degree that the starting points (assumptions) are realistic, then the theory will generate a representational model that is a general description of reality from which specific descriptions can be formulated as testable hypotheses. This is what makes a scientific theory empirical as well as rational as a combination of a formal model and observables.

Special case studies are not general theories and while they may generate significant results that are useful they are not general explanations based on causation.

Scientific theories generally arise over a long period from a combination of observation and explanation. Formalizing a theory as set of laws has generally come late in the game and most developed scientific theories are recent, since the 18th century.

See for instance Timeline of electromagnetic theory

Tom Hickey said...

He seems to be arguing that economics is not a science like physics in that it cannot (yet) formulate a general theory. As Jason Smith has pointed out, there is no framework. Given a framework competing theories can be developed and tested to arrive at the best explanation. That is not (yet) possible in economics or the other social sciences, or psychology. These field can only study special cases. Business schools realized this some time ago and teach based on the case method that originated at Harvard Business School.

The problem in economics is that there is no such thing as "the economy." It's not constructed as far was we can tell on laws like physics. Similarly, there is no such thing as "society" with respect to iron laws either. Nor has psychology come up with a framework. Same with history. So these fields are left studying particular types of behaviors in particular types of context in order to identify regularities that may be useful.

MRW said...

Paul Davidson gives the history of this. And he wrote two recent books about it, one general, the other more technical. Here is a video of Davidson trying to explain to undergraduates at the Univ of Chicago Econ dept how Paul Samuelson was first to create a "physics" model so that economists would gain respectability as scientists. Davidson says Samuelson's position is bogus. And that Keynes, whose authority Samuelson claims, wrote that such a position was ridiculous. Davidson reveals in the "general" book that Samuelson admits in an interview in the latter part of his life that he never understood Keynes (in the Appendix, worth buying the book alone).
https://www.youtube.com/watch?v=31wjPE-mUb4

Ignacio said...

You don't need a "general theory" for something to be a science.

There are no general theories except those ruling basic physical laws, everything else is constructs we build to abstract over more complex phenomena we cannot explain to a level of degree consistent with more basic physical laws.

What you require for something to be a science is to be grounded on empirical facts, that you can make predictions about and can checkout if those predictions happen or not. Is that simple, cut through the bullshit and it comes down to that.

Now, you can be doing philosophy, I've no problem with that, but don't pretend it's science. Both things are different things, and policy should not be based on philosophy.

It's a matter of confidence on your predictions. Science is not perfect, but good science is as perfect as it comes when making predictions. We can discuss theories all day (like SWL and his complain for "no theory"), but: can you make predictions? If so, with what confidence?

I can bet my life on the laws of classical mechanics working fine at a defined domain (we do this all the time whenever we use a car, for example), I won't bet my life on what a macro-economists says EVER.

Ignacio said...

Tom this has nothing to do with 'general theories'... There is no science which has a 'general theory' that explains it all. Physics doesn't either (hence all the fuss about strign theory or competing ones).

All sciences explain 'particular cases of some domain and make predictions about it and there is nothing wrong about that because our knowledge will be always incomplete.

Matt Franko said...

Tom,

This is what I mean:

http://plato.stanford.edu/entries/rationalism-empiricism/

The two forms can be seen as in competition with each other ....

Survival of he fittest! ;)

Matt Franko said...

Here from there::

"The dispute between rationalism and empiricism concerns the extent to which we are dependent upon sense experience in our effort to gain knowledge. Rationalists claim that there are significant ways in which our concepts and knowledge are gained independently of sense experience. Empiricists claim that sense experience is the ultimate source of all our concepts and knowledge."

And:

"Their attitudes toward science can be further illustrated by two anecdotal sayings. Hegel notoriously maintained that if facts contradict theory, then “um so schlimmer für die Fakten”—so much the worse for the facts. This can be seen as indicative of the paramount arrogance of a philosophy that takes no notice of such trivialities as empirical data."

To me Romer doesn't evidence knowledge of this larger conflict between empiricism and rationalism.... But to me that is what he is arguing

Matthew Franko said...

Ignacio,

"There are no general theories"

The people who look for these general theories seem to have a continuous view of time... iow they see time as one big long continuum...

This is like the people in religion who say "for ever and ever" or "eternity", etc.. where the word there was aeon which is a (long but) finite interval of time...

There must be something in their cognitive abilities where they tend not to be able to break up time into intervals... or re-set t=0 and just go forward from there as is done in the sciences all the time... same thing for their ability to do spatial analysis...

they seem to have trouble with time domain and spatial domain... they look for "general" theories... cant seem to break things up into components...

Matthew Franko said...

Here:

"Hegel notoriously maintained that if facts contradict theory, then “um so schlimmer für die Fakten”—so much the worse for the facts."

Evidence that there are people out there actually walking around among us thinking this way....

S-C-A-R-Y !!!!!!!!!!!

Tom Hickey said...

There is a difference between a science and a scientific approach. Physics his a science, engineering is not a science. Engineering is applied science based on the science of physics. Biology is a science. Medicine is applied science based on the science of biology. Pharmacology is not a science either. It is a scientific approach based on the sciences of chemistry, biochemistry and biology.

What is economics? Many economists want economics to be a science. But there is no general thereby or even a framework for a theory. Economics takes a scientific approach as so the social sciences and psychology, which are all still trying to become sciences. But they remain piece as explanations with many different approaches contributing to them.

I think Romer gets this. He is saying to scale down the claims and admit failure in the attempts based on them, like DSGE as a macro modeling method capable of being a foundation for policy science, which is the way it is being used. This is of practical significance since Romer has just been appointed the next chief economist of the World Bank.

Philosophy is like the photo-sciences in some respects, e.g. those that rely heavily on formal logic, and it is like the humanities and overlaps with them in attempting to portray the human conditions.

Systematic philosophy, which is no longer in favor or being seriously practices, is an attempt to develop a theory of everything. However, it's assumptions are non-empirical and it doesn't aim at making predictions but rather is aimed at prescription.

Economics shares a lot with philosophy and it grew out of philosophy. Adam Smith and Karl Marx were both philosophers, for example. Neoclassical economists tried to make economics more scientific by adding more formalism but in doing so they moved away from empiricism, generating theories based on assumptions that were only corroborated in special cases and failed as general explanations based on economic laws.

Business recognized this and moved toward applying scientific method to special cases, developing the case method rather than focusing on a general theory of management, for example.

The reason for the lack of success of philosophy, economics and applied economics in finance and business have failed to achieve general explanation is that there is no scientific basis for this in the study of human motivation and behavior. No general laws of human motivation or behavior have been discovered that have proved useful in constructing a general theory generating laws that can be corroborated by empirical testing.

So the best that can be attained in these fields at present is special cases studies that are based on use of statistical methodology rather than inductive-deductive methodology used by the hard sciences to arrive at general explanation based on scientific laws.

Tom Hickey said...

There is no science which has a 'general theory' that explains it all. Physics doesn't either (hence all the fuss about strign theory or competing ones).

No exactly the case. Physics is pursued in with a specific framework (conservation) that makes room for developing competing theories. Some theories have sown themselves to be the best explanations wrt the subject matter. Classical physics is fairly complete and most theoretical physicists are not trying to push out the envelope in classical physics. The horizon of physics has moved on.

I doubt any scientist thinks that one day their will be theory of everything so complete that no one will feel the need to bother with pursuing the development of theory anymore and all scientific development henceforth would be applied rather than theoretical.

The problem with economics and the social science is that they don't have a framework and no best explanation has been forthcoming. These fields are still groping for a framework in which a best explanation could be developed.

Leaders in these field often try to limit competition by declaring the methodological debate to be over, but that is based on politics rather than on scientific methodology.

The predominant approach in these field remains narrative, even when it is dressed up in the cloak of formalism.

Tom Hickey said...

Matt, the debate in Plato's time was realism vs. idealism. The rationalist-empricist debate doesn't begin until Descartes and ends with Hume in modern philosophy. Kant proposed a way to synthesize them.

Using "survival of the fittest" for competition conflates the sue of "fit" in evolutionary theory with ordinary language. They are different. Natural selection is genetic, based on "fitness" of inherited traits for specific environments rather than in-group or inter-group competitiveness.

Tom Hickey said...

The people who look for these general theories seem to have a continuous view of time... iow they see time as one big long continuum...

Ergodicity is a requirement for a scientific general theory.

Tom Hickey said...

"Hegel notoriously maintained that if facts contradict theory, then “um so schlimmer für die Fakten”—so much the worse for the facts."

Evidence that there are people out there actually walking around among us thinking this way....


"Wenn die Fakten nicht zur Theorie passen, ändere die Fakten"
Albert Einstein, Begründer der Relativitätstheorie

Tom Hickey said...

"Philosophy is like the photo-sciences in some respects, e.g. those that rely heavily on formal logic," should be "proto-sciences

jrbarch said...

I do prefer the ‘human narrative’ approach. For a long time we sat around campfires and told our stories of the heavens and earth, and human deeds, both foul and heroic. Made memorable in music and rhythm. Nowadays the campfire is digital.

For me, it is very simple. The heart wants fulfilment. Mind jumps in, and says – ‘Oh, let me provide you a solution’. The heart doesn’t listen.

The heart wants substance, not vapour. Mind is curious and wants to understand everything. The heart wants to experience, for itself, the experiences that it seeks. It cannot be fooled. Only then, does the mind understand.

That is the oldest oldest story; written in the stars. We have never learnt how to feel it, and follow it to our core!