The governance of theoretical physics

[SimonA]

It seems we have come to a point where Einstein's concerns regarding the loss of the "broad horizon" has become a reality. There is a fractal like dispersion of tenable theories to such an extent that there is no solid ground in theoretical areas in general - only in each specialised area.

The fact that there is no governance of theoretical physics as a whole is a far more serious problem than many seem to realize. Researchers have to choose from so many theoretical first points of principle on which to base their theories that the wood often gets lost for the trees. So in some areas researchers assume a given number of dimensions, in others they assume a multiverse. There is no tangible evidence for any of these. We could well be living in a universe with 12 dimensions, but as soon as that is a valid preposition, its then suddenly assumed that these dimensions are rolled up spatial dimensions - "cause we can't see 'em".

Likewise, the multiverse seems to be a valid starting point to a theory even though its a wildly speculative theory which only has other wildly speculative theories as its basis.

The only ownership of governance in these matters seems to be peer review and Nobel prizes.

Is it any wonder that so many of our theories are like plasters on a gaping wound ? Why can't we use basic governance techniques to rate every theory based on the evidence that supports it ?

Or we can just become mechanics with hammers banging the ground and waiting for a hole to open up that we can follow like Alice in Wonderland...

 

[apeiron]

It does seem a fair question whether bleeding edge physics is sufficiently self-critical.

It seems too easy for groups to splinter off with their own jargon, some "promising" idea, and carve out a nice little academic niche for themselves. And on a larger scale, to persuade tax-payers to fund supercolliders and space telescopes to chase Higgs particles and inflation events which may be illusions of such pet theories.

What are the right analogies to draw?

You could say the situation is good like the rich ecology of a rainforest. So many niches large and small. A more vibrant world (if apparently less productive) than of the days of Einstein, Schrodinger, et al.

A post-modernist would celebrate the diversity of it too. The exhuberant play of ideas, ever less weighed down by mundane concerns. Like they say about maths, why do we need to talk about reality? We are now up another level where we are theorising about realities in general.

In Einstein's time, there were only a few hundred professors, a few thousand papers published a year. Now there are tens of thousands and millions.

Once, the necessary self-critical mechanisms were easy to instutionalise. There was an established empirical philosophy. There was a lack of money and research centre empires to build. There was real peer review in that the whole of the scientific world could hope to be up with the whole of the ideas being generated.

Now who is keeping the game honest? What would modern self-regulation of theoretical physics look like?

 

[Coin]

I am not really sure what you mean by "governance". You mean centralization?

The problem you describe seems to be one of theory-- we have no single clear path forward, only a series of "splinters", each so specialized that few people can credibly master more than one of these splinters.

But you seem to be suggesting we address this problem with a sociological solution-- "governance". But governance would not change the basic theoretical problem we face-- it would not give us a clear path forward! Worse, without such a clear path forward, how can such "governance" function? If we had some clear method to judge between the good theories, then such "governance" could apply that method. But we don't have such a method-- that's the whole problem in the first place. We have only the splinters, we do not know which splinter is right, and with theoretical ideas as estranged from experiment as they tend to get in the current field of "Beyond the Standard Model" physics we do not have any way to grade the splinters against each other.

It seems like without objective signposts, if we attempted to apply "governance" that would lead only to centralization behind whichever one of the splinters can garner the most popular support-- a few years ago that would probably have been string theory. What if the splinter that winds up dominating the governance is not the correct one?

 

[marcus]

Another term for governance is guidance. The Greek root of govern is steering a boat, if I remember right. Let's suppose SimonA is
Greek and that he really means guidance.
Historically, theoretical physics has been guided by philosophical analysis of the concepts it uses---what is time, space, motion, matter, ...?

The philosophy of science, and particularly the philosophy of physics, may have been neglected in recent decades.

People like Boltzmann, Maxwell, Einstein, Bohr may actually have been philosophically more sophisticated---more reflective and analytical about the fundamental concepts they were using---than today's theorists. In a naive state there can be a tendency to build more elaborate virtuoso mathematical structures on possibly inadequate conceptual foundations.

Quantum field theory, for example, is still not general-relativistic. It still requires a fixed geometric background (which GR teaches us is unrealistic.) Only a small minority of the theory community actually is trying to find out how to construct a quantum field theory without a background spacetime. I suppose that more philosophical sophistication would lead to re-consideration of theorists' priorities.

There is also the failure of empirical discipline that you mentioned: Another disease that a proper grounding in the philosophy of physics could help prevent. Unless theorists take seriously the responsibility to make theories testable, theory becomes a self-indulgent pursuit of fantasy---Landscapeology being one example.

 

[apeiron]

Choice of language is important here. Governance and guidance are not terribly "scientific" notions. But self-regulation comes from cybernetics/system science and leads to concrete models of optimality. We could have a theory of how physics should be run and measure its performance accordingly - predict and test.

My question was whether anyone had actually modeled the current situation to give us some objective sense of whether things are well run or not.

Landscapeology is one good example of where physics may have tipped over in post-doc career manufacture. Too hard for anyone outside to understand, so therefore safely beyond criticism from the larger science community (but thanks for the grants).

But how do we know? What is the theory by which we would measure the activity and come to a defendable opinion rather than just an enjoyable prejudice?

The search for quantum gravity is another story. Here there has emerged a strong constraining consensus about what is the right question for theoretical physics to be asking. It is the main game and so every avenue is worth exploring.

My own opinion, based on certain logical considerations (from a systems perspective) is that this is a false goal. Or at least, the only way to merge QM and GR is through an "asymmetric" transformation mapping of some kind.

But now I would be arguing the consensus is too strong and causing people to ignore anyone taking this kind of "fringe" view. The landscape is not large enough to have considered the available systems science alternative to banging its head against a particular brick wall.

Again, this is a prejudice. That does not mean it is also untrue, but it would be a comfort to know that theoretical physics as a field is so well run - it has concrete models of its own social behaviour - that its effectiveness is being measured and its activities adjusted from the feedback gained.

I don't have an insiders view of theoretical physics, but I did have one of the birth of brain scanners - neuroimaging - in the late 1980s. I saw the eyes of a lowly teaching hospital radiologist light up after visiting the splendid physics palace of Cern.

And lo, a decade later, this radiologist had shifted from a scummy hospital next to slums and a prison to a most prestigious address in central London. He could employ the brightest post-docs by the score. The world's press fawned upon his establishment. Yet the science produced was frankly pretty crap.

Again the same thing happened with the rise of evolutionary psycholgy a few years later - though minus the lavish instrumentation, just the same extravagant level of hype and career building.

So self-regulation of science is a general problem. Who is attending to it?

 

[marcus]

...
So self-regulation of science is a general problem. Who is attending to it?

Science as we know it is a tradition and an elite community where the reward is status and peer-recognition.
Thus it is a self-selecting aristocracy somewhat feudal in character. The PhD student and the protected postdoc owe a kind of allegiance to their supervisor. There is some carry-over from the monastic tradition as well, I suspect.

This is OK. In fact it is wonderful. Science has been at its best when it's community was most vital, and the sense of honor (an aristocratic feature) was strongest.

I wish to say that the enterprise of understanding nature empirically and mathematically is not a function of the state. The sci community either has the ethos or it does not have it. It either raises heros (according to it's heroic ideal, another aristocratic feature) or it doesn't. If "governance" means control by the state, forget it.

The ethical soundness and vitality of the science community is up to us, the civil society. It is our job to celebrate what is great, in our time. And not be taken in by hype.
It is our job, the responsibility of all of us, to recognize what is exceptionally fine and scorn what is bogus.

BTW apeiron, I agreed wholeheartedly with a lot of your post, just couldn't take the time to say so in detail.

 

[apeiron]

And I agree with your sentiments, but still would feel they may be just a prejudice.

What is the rightful business of science? Is it to seek the kind of knowledge to deliver control over nature? Arguably this is what modelling relations and other systems science approaches to "knowing" would say.

And yet I personally am more interested in "truth" than control. Which may be a luxury response, a merely philosophical dream.

 

[SimonA]

Thanks for your replies. I'm certainly not suggesting more control of science by the state!

And yet I personally am more interested in "truth" than control. Which may be a luxury response, a merely philosophical dream.

As am I. Unfortunately the post modernist slant that seems to have evolved is hidden a little in theoretical physics simply because we are at a point where there are so many possible roads and each new theory is like another path off one of the many roads.

The reality of the search for truth seems to be relative to which road that researcher is on to start with. So for example it may well turn out that everyone working on theories which involve a multiverse has no hope whatsoever of finding the truth because of their presumption that there are more than one universe, and all their presumptions on based on another initial presumption. Or all those working on dark matter theories could very well end up finding out that dark matter is simply a place holder concept for our lack of understanding of the nature of gravity.

I realize of course that we need to explore these avenues, but it does seem that the amount spent on research within each 'road' doesn't match the degree of evidence supporting the idea that the road will lead to a deeper truth.

I don't have any easy solutions - its just a feeling I have that we could have a less scatter gun approach. Maybe a team of the best people from each sub field could work together to come up with a percentage probability (to the best we are able) of each basic presumption being correct, and the main spending then focusses on the highest probabilities.

This of course has the failing that all the money gets spent on those that are seen as the most likely to be true and through that they can produce the most supporting evidence that is is likely to be true! It would probably have given Special relativity a high score but General relativity would possibly have been laughed off the pitch...

So anyway I have no real solutions and the current system does seem to limp along fairly successfully, albeit somewhat lost in details and applying plasters wherever the data doesn't match the theory.

Regards

Simon

 

[apeiron]

A radical outcome of a proper analysis might be that we ought to trim the number of free research posts back to 1930s levels. Fewer voices would create greater coherence. All the lesser minds could stick to teaching and application.

Again, I feel that in the mind sciences I was witness to a dumbing down of the academic discourse as the number of publishing authors accelerated.

Go back to the 1930s and mind science seems a little like physics in the high quality of the thinking. These days we have types like Pinker - good in his own small discipline, but not a serious contributor on the wider ideas front.

However, he brings things down to a level which the majority of his fellow academics are comfortable with. They do not feel out of their depth. If Pinker can make it, there is hope for them too.

I think there is a general problem with the culture of academia now. Theoretical physics would be an example of a wider story.