Tai Goo said:
... Isn't it about time we put science itself through the same kind of scrutiny that it puts on everything else?...
/rant on/
"Good" science DOES this. Good science IS this. Science is a methodology for identifying, creating, and communicating information. THAT'S IT.
It is a method of identifying what you do (think you) know in a way that can be represented, evaluated, and repeated by others. It is a methodology for creating new knowledge that can also be represented and repeated and evaluated by others. The scientific method withstands scrutiny by its own methods. This is its great power. Scrutinizing knowledge by non-scientific methods may produce doubt, but not *information*.
A human can learn skills, procedures, art (broadly defined), but cannot transfer those skills effectively to others through modern means (aka 'writing') without the recipient developing an enormous amount of experience on his own. Scientific methods distill what CAN be communicated via writing. Science demands precise definitions and boundaries. This is its great power, as it means we can all be talking about exactly the same thing. It provides a foundation for further inquiry.
Science is first and foremost about INFORMATION.
Art is not about information, it is more than that (and sometimes less). Art is about integration, skill, beauty (to which science is only seldom applied). Art is about decisions, history, foresight, execution, and making the whole more than the sum of the known parts. Science can never take away the ART of execution by a human, because the experience, encoding, skill, and execution are PART of the artist's neural composition. Art is quite literally part of you. Art is application.
Humans develop applied skills of all kinds through training neural networks. The advantage of neural networks is that they can solve incredibly complex problems, optimizations etc. They incorporate non-linear feedback loops, and are exceedingly adaptable. This is VERY COOL. However... You cannot look inside a neural network and get the information OUT. Its encoding is mush - this is true of both biological AND computational neural networks.
The power of neural networks is not to be underestimated. Look at any work of complex art. Look at the action required to produce the perfect Hamon with an interrupted quench. Look at the kinematics at play in fielding an incoming grounder and throwing out the runner at first. It is SO HARD to even APPROXIMATE this sort of behavior in procedural software (i.e. intentionally encoded scientific information) that it's really a boondoggle to even try.
Science is procedural - The information can be identified, extracted, communicated, debugged, and changed. This is HUGELY valuable.
Integrated Human skills are encoded differently. They allow a single organism to solve infinite problems of STAGGERING complexity (including something as 'simple' as picking your nose).
Scientifically obtained information is useful as an input into the Human learning processes.
There IS NO ARGUMENT.
Science isn't wrong because science is a METHOD of producing information. Science DEMANDS the revision of information produced by scientific methods when NEW information is obtained. The information can be wrong - so then fix it! 'Science' insists!
Human skill and development isn't wrong - it's a means of APPLICATION. Claiming that one is fundamentally superior is a basic misunderstanding of the topic.
Applying scientific methods to human art is not demeaning. It is not destructive. It is about deriving INFORMATION from knowledge encoded in a neural network so that SOME of that knowledge may be communicated and applied in more general ways. Science cannot "know everything". It can merely aspire to identify what IS known, so that it can be shared.
Science is information.
Art is application.
They are best friends. I wish everybody would stop trying to get between them.
I blame Descartes.
For a modern version, try Zen and the Art of Motorcycle Maintenance.
/rant off/
-Daizee