Tuesday, September 11, 2012

Science, well sort of...the 6 things I want our kids to know about science


As I write about science, how it’s understood and its impact, I often get caught in the weeds of one issue or another…but the more I look at this, the more I’ve come to believe that we all need to back away a bit, and look at science as a meta issue (not really sure what that means, but it sounds good).  And frankly, this will be lost on most of you reading this because you’ve already solidified your views on both the meaning and implications of science.  So that’s why I’m focusing on our kids…here are Uncle Alex’s 6 rules to know about science…you are not allowed to play science before you commit these to memory.

1. Improperly using words that start with “C” can get you into trouble
I’m speaking of course about causation, correlation and certainty.   Causation is something that happens because of something else, and correlation is something that happens next to something else.  The significance of the distinction is if you know what causes something else, you can change the outcome, like when we understand what causes diseases; But if it only correlates to something, not so much.  Pretty straight forward on the face of it, but the misunderstanding of the distinction between the two, and the nominal desire for certainty leads to endless endlessness.  A great example is the claim, “smoking causes cancer”.  Well, yes it does, but darn, not always.  So is less than 100% causation only a correlation…again, not really.  It’s those pesky variables such as genetics, amount and length of smoking, etc…Further, to figure out when something is causation or correlation does not follow a simple black and white rule.  Therefore, we need to look at the bigger picture…here for instance we have decided that smoking causes cancer and other health effects sufficiently to warrant social and political action to try to limit smoking, which is bad for people, bad for health care costs, etc.  So as individuals and as a society we have an interest in accepting the results of the science, even though it is not absolute and determinative.  My dear old grandmother smoked filter less Pall Malls until she was 86, but that does not justify changing efforts to reduce smoking.
As for certainty, well science simply can’t provide it. In fact, rather than seeking certainty, science seeks to constantly test and question previous assumptions of fact.  And while we think we desire certainty, we really don’t need it in nearly all cases…this tends to really bum people out sometimes.

2. All science is not the same, nor are all scientists
This is a two parter that leads to lots of confusion when trying to figure out what to believe.  Part one is about science itself.  We hear daily about this study or that which seems to conflict with other previous studies.  How could this be? Lots of reasons…the variables or testing conditions were slightly different, inadequate sampling methods, spurious data, miss calibrated instruments, stoned grad students…and the list goes on.  The point is this is tricky stuff and we should never read all results as equal.  And this is even more the case when the word “social” precedes science.  The second part goes to the scientists themselves.  Frankly, just because someone has a PhD in a discipline does not mean they are able or qualified to opine on topics outside their very narrow expertise.  Nor does a graduate degree confer objectivity, lack of bias, or ideological purity.  Heck, they may not even be decent scientists, so be careful kids.

3. It’s impossible to prove that something won’t happen
This one is closely linked to the fact that science can never provide absolute certainty.   In a nutshell, while science is excellent (but not perfect) at predicting what will happen, it is unable to prove something will not happen.  For example, science cannot prove that something cannot harm you, it can only predict that it might…This one is a bit of a mind bender, but always be suspicious when folks demand proof that something is safe…that proof cannot be supplied by science, period.

4. The “Ant Colony Problem”
Science is by its nature reductionist, taking various elements of complex systems and testing hypotheses about them piece by piece…however, this may or may not tell us meaningful things about the systems as a whole.  My favorite example of this is ant colonies.  The fundamental unit of the colony is the ant.  However, no matter how much we study any individual ant, how much science we throw at it, it will never be able to tell us how the colonies function as a whole.  Therefore, we need to be really careful about taking narrow results and using them to explain complex systems.

5. Science on the way to or from church
Science is not the opposite of faith, religion, spirituality or whatever term your parents are comfortable using.  However, a key conflict between the two does exist.  While science can never proclaim certainty, faith often does.  Therefore, scientific thinking can be perceived as threatening simply because it requires an openness to revise thinking base on new data and experimental outcomes, as opposed to faith, which is often held has certain, absolute and immutable.  So it’s not that science provides alternate answers, but rather that any answer must remain open to further examination and revision.
Science is glorious, and helps us understand the world and our place in it.  But it will never provide all the answers, only more data from which we are challenged to arrive at broader conclusions and which will always require further research.

6.  …and just to mess with your young minds
There are many unknown things… there may be unknowable things…it is unknown whether there are unknowable things.  (Credit to my good friend Leonard Tramiel for this one…not sure if he made it up, or like me, stole it)
Now go out and goof off….you’ll probably learn something important in doing so.

-------------------------------------------------------------------------------------------------
Link for earlier blogs on ...the Science...well sort of:  http://www.chabotspace.org/open-science.htm