Thursday 22 September 2011

Peer-reviewed research - turtles all the way down...


Peer-Reviewed Research (PRR) is the lineal descendant of Real Science (RS) in the same way that liver flukes are the lineal descendants of free-living nematode worms: PRR represents the parasitic, pathological and degenerate form into which Real Science evolved in response to a changing environment.


In other words, despite some residual structural similarities, Peer-Reviewed Research is essentially a different entity than Real Science.

Just as when something looks like a duck and quacks like a duck, then it is a duck; so when something looks like a bureaucracy and communicates like a bureaucracy, then it is a bureaucracy.

What is calling itself science looks and behaves nothing like Real Science, so we should stop calling it Science.


Whereas Real Science works by obvious, in-your-face and common sense criteria (the problem being merely to understand how it works), in contrast ‘modern science’ (PRR) does not work – or rather, the only evidence that PRR works is the propaganda and publicity of scientists and the agencies who fund them – and these need to expend vast resources on persuading the public (and themselves) that they are doing anything at all but consume and mislead.

Old science obviously worked – there were frequent breakthroughs in solving longstanding problems; Old medicine obviously worked with new cures for previously incurable diseases and traumas. But very little modern science and medicine gets validated in this way – instead there is a profession of evaluation which is variously psychological, marginal, probabilistic, statistical, economic… but not obvious, in-your-face common sense.

Indeed, the main strategy is that of the Texas Sharpshooter, who fires his pistol into the barn door then draws a target around the bullet holes: modern ‘science’ does what it does, looks at the result, then spins the outcome as a ‘breakthrough’.


Real science is primarily theoretical – by contrast with technology which is practical and mostly operates by trial and error within the theories of science.

The way it works is (roughly) that science makes breakthroughs in theory which are then linked (via chains of causation) to predicted real world consequences. The scientific theory is checked against observations of the real world, including prediction, experiments, and attempts to manipulate (and change) the real world.

This link between science and technology is loose, and there is no specific set of rules which govern it - i.e. there is no such thing as ‘scientific method’, nor legitimate inference (above and beyond that of normal human reason) and all attempts to define such have failed.

Nonetheless, this link between science and common sense observation – for all its informal and subjective qualities - is the basis for evaluating science. Without such a link science is not really science, but a detached, free-spinning intellectual activity: a ‘glass bead game’ played for the amusement (and/ or enrichment and status) of intellectuals.


In other words, when science is not validated by technology, and technology validated by common sense, then science is merely validated by scientists – by scientific opinion.

Nowadays, scientific opinion is formalized in peer review committees – so without the link between science, technology and common sense then science is merely committee voting, just like any bureaucracy.


(Committees are committees are committees – they may operate by formal procedures or by no procedures and it makes no difference – because procedure is arbitrary and conjectural in its effect; no procedure has any intrinsic validity.)


So PRR marks the takeover of Real Science by a wholly different evaluation system: the change from validation from theoretical science linked to technology linked to common sense, into whatever happens to be the outcome of a system of interlinked committee voting:

government committees, employment appointment committees, promotions committees, grant awarding committees, corporations run by committees, media and publicity committees, committees for investigating the ‘integrity’ of science, committees to decide what gets published where and in what form, committees to decide who gets to talk at conferences, Nobel Prize committees...

Like the mythic world which stands on a turtle standing on a turtle standing on a turtle, and so on: modern ‘science’ is committees all the way down



Gabe Ruth said...

This is excellent. Another variation on the journey or process being elevated above the destination or result. Very well put.

What do you think of this faster than light particle? Am I correct in thinking that the speed of light being unsurpassable has been taken as an axiom for the last several decades of theoretical physics? Meaning its broad acceptance as fact is mostly due to the assumption yielding useful results and committee votes?

Bruce Charlton said...

I am reflexly mistrustful of modern 'scientists'. especially in big science like particle physics. It could take decades to sort out - even if the people involved were being as honest as they can be - which they are not.

Anonymous said...

I would say - as a physics person - that this thesis is VERY domain sensitive. Experimental sciences have repeatable experiments, repeatable data and the theoretical explanation of why that data should be present and not other data.

The Peer Review part of physics, is the repeating of the experiment, that re-capturing of the data and the assessment of the credibility of the data from the experiment.

For the theorist, is means having the theory fit inside of other theories as well as explain the current experimental data.

Bruce Charlton said...

@GBA -

"The Peer Review part of physics, is the repeating of the experiment, that re-capturing of the data and the assessment of the credibility of the data from the experiment."

No - that isn't peer review. Peer review is the - relatively recent - a few decades - procedure of reviewing/ evaluating science and scientists by the *opinions* of their 'peers' (who are supposed to be working in the same field).

In a sense peer review is the opposite of what you describe - actual testing, replicating, checking and so on