Monday, December 5, 2011

What is Science?

In http://sigma5.blogspot.com/2010/11/is-science-religion.html I asked "Is Science a Religion"?  Actually, most of that post is dedicated to the question "Is Science Scientific?"  The answer I came up with was "No!  Science is not Scientific".  I then went on to do some hand waving on the title question, namely "Is Science a Religion".  Upon further reflection, answering this latter question doesn't get you anywhere.

In 1964 Supreme Court Justice Potter Stewart famously opined "I know it when I see it" when talking about Pornography.  The question of "What is a religion" is even slipperier.  Certainly lots of people would claim to "know it when they see it" when it comes to what constitutes a religion.  And I am sure this is a firmly and sincerely held belief.  But there are a lot of things out there that someone thinks of as a religion.  And there is little or no commonality.  How many gods are there?  The most popular modern religions will answer "exactly one".  But the three civilizations that underpin modern western culture (Egypt, Greece, and Rome) all subscribed to polytheistic religions.  And Atheists gets lumped into the category of religion by many.  Atheism is the belief that "there is no god".  So there is no common answer to a simple question "how many gods are there".  Religions are also all over the map on "appropriate" marital practises.  Some believe monogamy is the only way to go.  Others are pro polygamy.  Some are even down with group marriages or prohibit marriage in any form.  And so it goes.

I have characterized the business of religion as "behavior modification", e.g. trying to convince people to do more "good" things and fewer "bad" things than they would otherwise do.  But even this idea is not universal.  Some religions are of the "chosen people" model.  If you are one of the "anointed" you are good to go.  If not, bad things are going to happen to you and there's nothing you can do about it.  Upon reflection I have been forced to conclude that saying something is a "religion" tells you nothing.  So, if you are trying to decide what Science is, knowing whether it is a religion or not doesn't help any.

So if Science is not "scientific" and deciding whether it is a religion is useless, what is it?  I have decided that Science is a popularity contest.  Now there are lots of kinds of popularity contests and the kind of popularity contest Science is is peculiar.  The popularity contest that science represents is not about people.  It is about ideas or, more specifically, the argument in favor of specific ideas.  And the scoring is odd.  More popular ideas in science are more popular because smart people and, more specifically, fair people come to support them.  It's not about the numbers.  It's about the quality.  So how does this work?  How does one construct a popular argument in the Scientific sense?  Let's first consider some approaches that don't work.  And lets consider them in a context.

I want to convince you that my idea is the best one.  What do I mean by best?  I mean "most likely to be true or, if no contender seems to be completely true, then truer than the alternatives".  I think everyone would consider that a fair objective.  So let's say I said "I'm right and you are an idiot if you don't agree".  How likely would that argument be to sway you?  It's "argument by insult".  Most people don't like to be insulted so that argument is a terrible one.  So let's improve it and go with "I'm right - trust me".  The insult is now gone and that's a good thing.  And if I am known as fair, knowledgeable, smart, and nice, it might be pretty effective.  But is it the best argument?  No!  Everybody is wrong some of the time.  And, by adding in the "fair, knowledgeable, smart, and nice" part I am really adding a component to the argument.  More generally, this enhanced argument is just a version of "Accept my argument because this important/smart/powerful person agrees with me".  Important/smart/powerful people are more likely to be right than unimportant/dumb/powerless people but this is just a "statistical" argument.  The argument is more likely to be right but there is still a good chance it is wrong.  Can we improve our odds more?  Can we get to 100% right all the time?  If we could, that would be ideal.  Before moving on, let me spend a little more time with the "powerful" version of this argument.

In practical terms as measured by the success religions have in converting people from some other religion to their own this argument is remarkably successful.  Why so?  Well, let's hear what Chief Seattle said on the subject:

Your God loves your people and hates mine, he folds his strong arms lovingly around the white man and leads him as a father leads his infant son, but he has forsaken his red children, he makes your people wax strong every day . . . while my people are ebbing away like a fast-receding tide, that will never flow again. [“Eyewitness to the Old West”, Richard Scott, ed., page 129]

If you look at successful mass conversions of peoples from one religion to another the argument that worked is the "your God is stronger than my God" argument that Chief Seattle is making. And this follows from the justification of many religions, namely that "my God is very powerful".  If he likes you he will work miracles that will help you (e.g. heal the sick) and if he doesn't he will work miracles that will hurt you (e.g. rain fire and brimstone upon you and yours).  When the whites moved into the Americas they proved more powerful than the locals.  Therefore the gods of the whites must be more powerful than the gods of the locals.  So the "powerful" argument works.  But then lots of arguments work at least some of the time.  And there are other arguments that work better.  So let's return to the mainline thesis of this post and get on to them.

For instance:  "I have done this and here's what happened".  Here I am demonstrating to you in a manner that is convincing to you how the world works.  And there is a better version of this argument.  It goes "Do this with your own hands and see what happens.  Notice that the same thing happens as if I and others do it."  Now I have eliminated the "trickery" problem.  If you and lots of other people do it then I haven't done something to trick you into believing something that I know is false.  This approach is far more convincing than any of the above arguments.

Well, not always to everyone.  We have all run into people who hold to a belief that we know to be false even after we have abundantly demonstrated that it is false.  But is other person being fair?  We all know the answer is no.  My point is that there is no way to convince all the people all the time that a specific truth is true.  So requiring all people to be convinced by an argument is an impossibly high standard.  That's why I have thrown the word "fair" in a few times now.  Since it is impossible to convince everyone, Science settles for convincing "fair" people, people that are open to being convinced by a well constructed argument.

And in fact it is usually too much to require that everyone do the demonstration themselves.  So the Scientific "standard" is repeatability.  The more people that do the demonstration and get the same result, the more convincing the demonstration is.  And Scientists are well aware that this weakens the power of their argument.  So they try to take precautions.  What is important is that the demonstration is done properly, in a way that avoids the possibility of trickery.  And "trickery" turns out to be quite a complex problem.

Consider "Clever Hans".  Clever Hans was a horse who in the early 1900's demonstrated that he could do arithmetic.  So was trickery involved?  A lot of people investigated and could find no reason to believe so.  The horse's owner was honest as the day was long.  And no one could figure out how the trickery was done or even by who, assuming trickery was involved.  Then in 1907 Oscar Pfungst made in interesting observation.  Hans did not always get the right answer.  And in each case where Hans got the wrong answer his owner got the wrong answer to the same problem.  Hans' owner was able to transmit cues to the horse completely unconsciously and in a manner that was not apparent even to supposedly trained observers.  Clever Hans and many other examples have shown over the years that even the most honest person can be fooled.  Scientists have taken this to heart in two different ways.

The most obvious way is by using "double blinds" in experiments.  In a double blind experiment neither the testee nor the observer knows whether the thing being tested for (e.g. a drug effect) is actually there or not (e.g. in the case where a placebo has been substituted). That way a subconscious cue can neither be sent nor received.  The other way Science has been affected is the substitution of observations by instruments for observations by people.

A famous and not well known example of this is the work of Edward Muybridge (a photographer by trade - not a scientist).  For years a controversy had raged within the horsey set as to whether a horse lifted all four feet off the ground at the same time while galloping.  In 1872 Leyland Stanford (Governor of California and founder of the University of the same name) decided to put up the money to find out the answer.  He hired Muybridge to find a way to settle the argument "once and for all".  Using human observation was out of the question.  Learned experts could be found in large numbers on each side of the controversy and they all supported their position by citing "personal observation".  After much experimentation Muybridge was able to take a series of photographs of a galloping horse that were closely separated in space and time so that all aspects of a gallop cycle were shown.  Next, Muybridge developed a method of flashing the pictures one after another quickly.  This short demonstration consisting of only 12 frames of film is considered the first movie.  What the movie did, when shown at speed was convince people they were seeing a horse galloping in the normal way.  But one of the frames showed all four feet off the ground at once.  This combination was utterly convincing, and a convincing demonstration was something that had been impossible until human observation was replaced by instrumental observation.

Both of these changes in how Science does things, the "proper scientific technique", are driven by the same desire, the desire to make scientific arguments more compelling.  "How can we make our arguments more compelling" is an impulse that has driven science for centuries.  Put another way, scientists continuously ask themselves "what possible flaws exist in our arguments and how do we eliminate them?"  Scientists strive to both remove flaws in specific arguments about specific items and, by improving "proper scientific technique" to remove flaws in the methods Scientists use to create scientific arguments.  So what are some of these possible flaws?

One example is the counter example.  For many years many people believed that all swans were white.  This was based on observing many swans for many years and finding that they were all white.  Then famously black swans were found in Australia in 1790.  Arguing that "all swans are white because most swans we see are white" is not convincing.  "All swans are white" is true only if ,well, there is no such thing as a swan that is not white.  So proper scientific technique requires that your argument take into account all the data.  A single counter example (e.g. an Australian black swan) is completely fatal to the truth of the argument.

In a similar but more general vein, mathematics is critically important to science.   Why?  Because mathematicians have been trying to "break" (e.g. find flaws in) mathematics for about 3,000 years.  Over these Milena mathematics have developed a very reliable sense of what works and what doesn't.  Now there are many shortcomings and limitations to various mathematical techniques (e.g. dividing a number by zero results in nonsense).  So it is important to use mathematics properly.  But mathematicians have documented these shortcomings and if you have demonstrated that you have avoided these shortcomings then your argument is on solid ground.

Another general scientific principle is "it's not who does it, it's how they do it".  A poorly done argument is not a "properly done scientific argument" even if it is done by the most preeminent person in the field.  Contrarily, if the argument is done well by an "amateur", gifted or not, it is a proper scientific argument.  (See the Muybridge discussion above).

Another general scientific principle is to base an argument on clear agreed upon statements.  Imagine you are having an argument with someone about the color of a specific fruit.  Now imagine you are talking about apples (red, yellow, green, not orange) and the other person is talking about oranges (normally orange but green if unripe).  If the two of you don't set about to come to clear agreed upon statements about what the object of the argument is then both of your arguments may be ineffective in winning over the other (fruit is red - no fruit is orange).  You may even agree inappropriately (e.g. if I am thinking of a ripe Granny Smith (green) and you are thinking of an unripe Orange (also green)).  Scientists go to a lot of trouble to make sure everyone is taking about the same thing.

Many other general scientific principles have been developed.  They all have the same objective.  If you construct an argument using those principles you will end up with a proper scientific argument.  And that proper scientific argument will be convincing to fair people.  And, as a result will become part of the scientific consensus (e.g. those arguments that are very popular - a whole lot of people agree that they are right).

So Science is a popularity contest.  And Science has evolved its general principles to result in proper scientific arguments.  And proper scientific arguments are constructed to be popular.  If this is all so then why do a large segment of the general public find Scientific pronouncements unconvincing?  Why is Science so unpopular outside the scientific community?  There are, of course, the unfair people.  Those that are unwilling to be convinced no matter what.  But there is more to it.

I think Science is a victim of its own success.  Until about 1900 most scientific arguments could be understood by the general public.  In 1704 Isaac Newton published a book called "Opticks".  It was written in Latin.  But, when translated into English, the experiments he describes and the conclusions he reaches are easily understood by anyone.  In the latter half of the 1800's many lecturers traveled the country doing scientific demonstrations.  These were very popular and people came away with a real feeling that they understood the scientific principles being demonstrated.  And they found the demonstrations convincing.  Even if they personally did not attend the demonstration they read about them in the newspaper or talked to a friend that had attended.  People felt that they were a part of the scientific world.

But by about 1900 Science started getting weirder and weirder.  Much of Science was still perfectly understandable.  But more and more of it and especially some of the parts of it that scientists considered the most important got really weird.  Maxwell's equations (actually developed in the early 1860's) that described electricity and magnetism involved an imaginary number (the square root of -1).  Michaelson and Morley did an experiment in 1887 that proved that the speed of light was a constant.  In 1900 Max Planck started talking about the "quantum hypothesis".  Then in 1905 Einstein published a series of papers postulating really weird stuff like what later became known as "Special Relativity".  General Relativity followed roughly a decade later.  Quantum Mechanics was developed in its early form in the 1920s and has since evolved over the decades into the "Standard Model" of particle physics.  It features a long list weird features.  But one of the weirdest is something called "entanglement".

So through most of the 1800s most of the latest scientific results were accessible to the average people.  They might not have understood them in detail but they got the general idea and often could understand the reasoning behind scientific thinking.  But as the 1900s have progressed the distance between cutting edge science and what the general public could understand in some detail has gotten large.  Scientists have tried to bridge the gap.  Steven Hawking famously published "A Brief History of Time" in 1988.  It became a best seller.  But talk to most anyone who has tried to read it.  People start out enthusiastic and getting it.  But at some point in the book they can no longer follow what Hawking is saying.  And Hawking completely omits all of the mathematics that underlay what he is talking about.  We have to take what he is saying, and what he is saying is pretty weird, on faith.  If he had put the mathematics in so that theoretically we wouldn't have had to take him on faith, we still would have had to take him on faith.  The mathematics behind what he is saying is beyond even people with a standard 4 year college degree in the sciences or engineering.  You have to be able to understand graduate level mathematics and graduate level physics to be able to understand what Hawking is talking about "in the raw".

Much of Science is still understandable to anyone who is willing to put a little time and effort into it.  But much of Science is not.  In many cases Science has fallen victim to the Holmsean curse.  As Sherlock Holmes opined "[o]nce you eliminate the impossible, whatever remains, no matter how improbably, must be the truth".  The scientific version of this curse is even worse than the original Holmsean one.  Take the example of the composition of light.  In "Opticks" (see above) Newton lays out the results of a number of experiments.  The obvious conclusion from these experiments is that light consists of waves.  But Newton did not put all the experiments he had done in "Opticks".  And the results of those experiments, and for other reasons, Newton concluded that light actually consisted of particles.  For hundreds of years scientists wrestled with this wave/particle dilemma.  The problem was there was no "whatever remains".  All the possibilities had been rendered impossible.

Finally in the 20'th century the conundrum was solved.  But the solution was weird and makes no sense to ordinary people.  How can light be both a particle and a wave at the same time?  And the mathematics that describe light are beyond the capabilities of most people.  Scientists did not opt for this complex and unnatural solution to the "light problem" out of anything except complete desperation.  All the sensible and simple solutions were known to be wrong.  Unfortunately, Scientists in situation after situation have been driven, usually completely unwillingly, to the weird and the complex for the simple reason that the ordinary and simple is known to be wrong.

So with Science we have this paradox.  Science is built on simple and easily understandable principles.  Scientific arguments are constructed to meet common sense requirements.  The objective is to construct an argument that is utterly convincing.  And, for those who understand the arguments and are fair, the arguments are utterly convincing.  But Science has been too successful.  These techniques have allowed scientists to construct utterly convincing arguments about how the world works.  And in many cases these arguments can be followed by the general public.  But Science has gotten so good at this sort of thing that they have been able to successfully investigate situations and construct convincing arguments about what's going on.  But what's going on turns out to be utterly weird.  And it sometimes takes talents far beyond those possessed by average people to understand and appreciate these arguments.

This has put a lot of people off on Science.  And this gap between what average people can understand and what Science tells us about the world around us has weakened the connection between Science and ordinary people.  And unscrupulous people have taken advantage of this.

There is nothing weird about Global Warming or the Science that underlays it.  It is not difficult to understand that glaciers are shrinking around the world.  There are "before" and "after" pictures of many of these glaciers.  A casual glance at these pictures leads inescapably to the conclusion that the pictured glacier has shrunk, usually by a lot.  No arcane scientific knowledge is required.  There are many other lines of evidence for Global Warming that are as simple and direct as the glacier story.  But people take advantage of the weakened connection between Science and ordinary people that exists in areas like particle physics and apply it to things like Global Warming.  They pretend that there is some kind of great scientific controversy about whether Global Warming is real and whether part of it is man made when there is no scientific controversy.  Unfortunately, many people do not make the effort to dig into something like this for themselves and the result is they are convinced that the science is unreliable when it is not.  The same kind of situation applies about the "controversy" surrounding Evolution.  Here too the Science is easily understood by ordinary people and is extremely sound.

   

No comments:

Post a Comment