On Sat, Jan 28, 2012 at 1:56 AM, Charles Haynes <hay...@edgeplay.org> wrote: > > I would argue for the scientific method
*curses* I would ignore Silk right now for lack of time, but this claim is an important point to nuance, else it is dangerous. Forgive me, for I am going to telescope a bit because I have little time, however I think you can fill out the missing bits - please ask if something isn't clear. My responses maybe slower than usual in coming. The scientific method has delivered us a lot, no doubt about that. However the method has also been reaching diminishing marginal returns in many areas. It is being increasingly successfully circumvented (not necessarily willfully or deliberately) in ways that are as yet not widely understood. For example, today any clever person, and not just big tobacco, can fund a research using the best of science to prove opposing conclusions rather easily. (For lack of time I can't point you to the recent Harvard(?) Medical Journal paper where a jaded doctor proves conclusively that smoking improves athletic performance) So, scientific proof is sometimes an unreliable compass, and more so today when manipulation and bad science are more common. There hasn't been enough (mostly any?) serious debate on this. We are basing our knowledge increasingly on very feeble foundations, and this is dangerous. This isn't the fault of the method or the proof, but of the tests and to some degree a problem of ethics. Next... The way the sci. method works is that its gains are typically incremental - bold leaps of logic are not how scientific method functions. This is neither good nor bad, I'm just observing the nature of the thing. However, let's bear in mind that humans have accumulated useful knowledge for millennia and often in bold leaps, and it has not always been through luck (to claim so would be supreme hubris of this era of science), and the absence of the scientific method then did not prevent the occurrence of good, useful inventions. However these pedagogical/inquiry styles have sadly not been codified the way the scientific method has. An example: You can prove through double blind tests and other proven logical and statistical methods that Yoga has a beneficial effect on the human body and mind. However it is not clear to me that one can use the same scientific method to invent a rather complicated thing like the Yogasutra in one stroke. Thus, Pathanjali's Yogasutra was not a product of the scientific method. Why is this so? The way I see it, scientific method operates on human logic which requires data; however this has two problems. 1. There is a limit to how much data individuals can hold in their heads and computer models at any one time. 2. Data is not always available, and the scientific method deals badly with missing information The way the scientific method gets around the missing information is to propose a model, even one known ab initio to be broken, and iteratively discards this model for a newer (usually better) one over time. This is a great, robust, reliable way of doing things, however we don't always need or desire this level of precision. Human decisions themselves are not always so precise and expensive. Daniel Kahneman's work on decision making that won him the Nobel Prize for Econ. last year touches somewhat on this. At any decision making point we have both expensive decision alternatives and cheap decision alternatives available to us. We often choose to make the approximately correct cheap decision rather than the verifiably correct expensive decision. ex: "Would you like some salad dressing with that? French or House?" - what decision do we most commonly make here? Expensive or Cheap? We can't launch space crafts on a series of cheap decisions, but we can get to say a holistic model of healing rather quicker with fuzzy decisions. Fuzzy optimization and decision making is still only being superficially understood in today's scientific world, but the collective split brain of the world today would like to believe and behave as if the scientific method is infallible, which it isn't. My purpose isn't to discredit the sci. method - it has won us a lot, but it is important to point out that this isn't something we have fully understood, and there are problems with focussing on a very step-by-step approach. Sci. method does not encourage big picture thinking enough. So we have market models, economic models, medical models, even value and ethical models based on myopic thinking. Problem: Efficient calorie intake Bad decision: Farm factories for meat -> factory produced burger -> pre-frozen to McDonalds -> minimal skill wage slave -> incredibly cheap food (only because you ignore the hidden transaction costs) -> indigestion and ailments -> pharma offers endless solutions to fix this -> they have side effects -> more medicines -> more side effects -> more burgers and so on. Better choice: Don't eat stupidly Why do so few people make bad choice over better choice? The data is all there, but our mental models don't see far enough. More later.
