So I’ve been watching this YouTube video of Tim Minchin’s science rant titled “Storm“, which got me to think about some of my rants about science. I believe there are several common mistakes made by those who reference science and some pretty big misunderstandings of those who criticise science.
Firstly, let’s define science.
Science is a philosophy. It has several basic principles.
- The Principle of Universal Nature (PUN) – that is, what is fact here is also fact over there
- Induction – that is, what is fact now will be fact later
- Science can only ever disprove a statement, never prove it – that is, a scientific theory must be fallible. Proving the fallibility disproves the theory, but failing to do so does not prove the theory.
- The more the theory diverges from mainstream thought, the more impressive the “evidence” for it must be for the scientific evidence to be credible.
- The greater the sample size, the better the test.
- Scientific knowledge and theory must adapt to evidence.
There are greater levels of complexity than these, but this gives you a start. To use an example of swans is fairly popular. First we make a statement – all swans are white (in Europe, they pretty much are). The fallible aspect of the statement is that if we can find a swan that is not white, then the statement is false. This is where the statement “it is the exception that proves the rule” came in. If I can not find an exception – something about the statement which if found proves it wrong, then it is not a rule, but an all encompassing infallible statement. An example of such a thing is “God is that which there is nothing greater”. There can be no exception to this, for if you do, that thing which you found was greater then becomes “God”.
Now PUN (principle of universal nature) states that swans are swans everywhere in the universe, and the statement that all swans are white must be true everywhere. If you find one non-white swan anywhere in the universe, then the statement that all swans are white is false, therefore scientific philosophy has dis-proven the statement.
This doesn’t stop a non-white swan being found on Jupiter in 2000 years. We didn’t think to look their, then, so we can not state that the statement is true, we can only state that in the samples we looked at, we did not find a contradiction. Until we find a contradiction, we can assume the statement is true enough, but that does not make it true.
If I only look at 5 swans, then the strength of my evidence is fairly weak. If I look at hundreds, my statement of evidence is stronger. If I look at thousands, my statement is stronger again.
A common mistake is to state that if you find something that looks like a swan in all ways but colour, then that is not a swan because it does not meet the definition. This undoes the fallible point of the statement, so does not agree with the scientific philosophy.
Here are the common mistakes people make when citing science:
- “I have scientific proof/evidence of the validity of my theory” – science only disproves, it never proves. Thus the statement should be, “My theory has been scientifically tested”. Again, this does not mean your theory is right, but is has greater strength because the theory has not been dis-proven yet, but has been tested. An untested theory is not necessarily wrong either, nor is it right. It is just untested.
- “Science tells us why”. It doesn’t. It only tests for facts, it doesn’t explain why. It can test all of the parts of an explanation of why, but that assumes a causal chain which can not be proven. The why is not part of science, only that a relationship between two events has not been discredited. For example, when I push the button by the side of the door the light turns on. Science doesn’t tell me why this happens, it just tells me that every time this has been done in this particular way (wires, electricity and a working switch and light globe – that is the induction, or repeatability of the experiment) it has worked. When it doesn’t work, we first look to see if a difference in the parameters of the experiment exist before stating that the correlation between the two events is no longer linked. The explanation for why is something that people come up with to help create fallible prediction to test the statement. A successful test for a statement to prove why does not prove the statement of why to be true, it only indicates that the statement of facts you created from your why has not been dis-proven.
- The nature of the universe is universal. It isn’t. First of all, Cartesian geometry points out that each point in the universe has four components, X, Y Z and time. No two can be the same, so there is no universality about it – however the idea is that the principles of the universe are consistent for all these points. No two points have exactly the same gravity and it is accepted that gravity changes space and time. Therefore no two points have the same properties as the nature of space and time are different for each Cartesian point. However the difference are so subtle that other than at very high speeds or on very small scales, the principle of universal nature is good enough.
- Induction is real. I can do the same test in the same way for millions of iterations. It doesn’t mean that the next time it will be the same. However it is a fair assumption given our experience of the universe to assume that this principle is true enough. After all, our experience of the universe suggests that it is pretty consistent, and if we assume that everything we know now is based on principles that may not be true for the shear sake of randomness rather than our faulty comprehension of the nature of the universe, then why are we even bothering? So while we can’t prove induction, it is a fair assumption.
- There is no magic and science has dis-proven it. Tim Minchin quite rightly points out that Alternative Medicine is defined as methods that have either not been proven by science to work, or methods proven by science not to work. All of the methods ‘proven’ by science to work are given a new name – medicine. (I’m misquoting, but you get the point). Magic is an ambiguous term and is sometimes defined as “how things work that are not scientific or not provable”, which means that when science figures it out it stops being magical. Personally I think this definition is faulty. I would prefer it to mean “Magic is the amazing way that things work, whether we understand it or not”. Thus when science investigates and figures out what actually does work and what doesn’t, it doesn’t stop being amazing just because it become scientific. We understand pretty well the systems in place for human reproduction. This understanding doesn’t make child birth any less miraculous and amazing. We made new life from old life. Wow. Scientifically understood (for the most part), still magic.
- If science has not investigated it then it is crap – Science has not investigated everything and it has not proven everything. If it has, then there would be just a huge tomb of “what is” and no scientists. Scientists keep investigating this thing we call the universe to test for new statements of fact, re-investigate evidence to find new things about the universe and keep testing old ideas to see if our new methods still agree with old statements. There is so much that we just do not know and have not tested. A lack of scientific investigation does not mean it is false. By the same token, it does not mean it is true either. It just means it is untested.
- A correlation that occurs without a provable scientific test is just coincidence. This is not necessarily true, since we do not know everything. By the same token, it is not necessarily false either. This mostly come back under point 6 – if the correlation (that is, the seeming link) keeps happening and we have not yet worked out if there is or is not a link, then there may or may not be one. A lack of investigation or evidence does not make a fact one way or the other.
Hopefully this clears up my position a little. Of course this is my understanding and I am quite willing to adjust to better logic.
