Sunday, August 29, 2004
Science Fact and Ockham's razor
Nowadays, science and the scientific process are largely accepted by the general populace, yet also largely misunderstood. This is in no small part due to the scientists themselves. They're human, they make mistakes, and they even cheat and lie sometimes. What are people to think when a 'study' suggests a link between two things, and another study suggests no link between the same two things?
Unfortunately, many of the studies conducted today are nothing more than statistical treasure hunts looking for correlations between two things. These correlations can be coincidental (blue eyes and blond hair), causal (smoking and emphysema), dubious (Super bowl winners and the economy), or even misleading. The problem is that many news outlets don't care to discuss the nuance behind the links. Often, the journalist reads the abstract of the article and goes no further, jumping to conclusions. In one study of flu shots, the researcher wanted to see if there was a correlation between having a flu shot and preventing a cold, cough, or other non-flu illnes. When there was no correlation, a few news outlets reported this as 'Flu shot's effectiveness questionable'. The science behind it wasn't faulty, but a faulty conclusion was drawn.
So some people don't trust science, and they don't really get what science says is fact. How do we know that what we 'know' now isn't going to be disproved in a matter of decades?
The answer is that given what we know now, science should always be able to give a best guess at something. With more knowledge than in the past, we should always be able to at least do no worse than previous scientists. While this isn't always true (because science is rarely ever untainted by human emotion, politics, stupidity, and outright lies), in theory, we should always be progressing in our knowledge of the universe. While we may be wrong, even blatantly wrong at times, if we follow the scientific process, we should eventually stumble upon the truth. So where does Ockham's razor fit in? Well, it is part of the scientific process, and it's a little part of pop culture that has made it into the fringe of the populace (thanks to Contact), so I feel a need to clear it up a few things. It is named after an English philosopher, William of Ockham (I think), and he said it in Latin first, (I think).
Anyway, Ockham's razor has been quoted and misquoted many times. I won't even try to quote it here, but I will point out that "The simplest explanation is the correct one" is NOT, Ockham's razor.
To paraphrase in plain English, I would say that the spirit of Ockham's razor is if you have multiple hypotheses that explain something equally well, then the hypothesis with the fewest assumptions should be used. It isn't necessarily correct, but the simplicity is important. If you had two hypotheses:
1) The apple fell off the table because of Earth's gravitational pull
2) The apple fell off the table because of Earth's gravitational pull, which is caused by aliens with supermagnets.
Then obviously the first one is preferable. But it isn't just because of the ridiculousness of the aliens hypothesis.
1) The apple fell off the table because of Earth's gravitational pull
2) The apple fell off the table because of Earth's gravitational pull, and because somebody bought the apple and put it on the table.
Does the second hypothesis explain it better? Maybe. But with no evidence to support that, we would have to go with the first one. If we had a receipt of the apple purchase, then the second one would win because of the better explanatory power.
In any set of hypotheses, it is always possible to add more assumptions, but come up with the same conclusion. There are fewer simple hypotheses, so to make it easier on us, we choose the simpler ones first. They can always be modified to be more complex. Just as Newtonian physics gave way to relativistic and quantum physics, simpler theories do NOT equal more correct theories. But it is important not to jump to conclusions, and to always yield to evidence, as long as it is reputable and reproducible.
So there's today's science lesson. Go science!
Unfortunately, many of the studies conducted today are nothing more than statistical treasure hunts looking for correlations between two things. These correlations can be coincidental (blue eyes and blond hair), causal (smoking and emphysema), dubious (Super bowl winners and the economy), or even misleading. The problem is that many news outlets don't care to discuss the nuance behind the links. Often, the journalist reads the abstract of the article and goes no further, jumping to conclusions. In one study of flu shots, the researcher wanted to see if there was a correlation between having a flu shot and preventing a cold, cough, or other non-flu illnes. When there was no correlation, a few news outlets reported this as 'Flu shot's effectiveness questionable'. The science behind it wasn't faulty, but a faulty conclusion was drawn.
So some people don't trust science, and they don't really get what science says is fact. How do we know that what we 'know' now isn't going to be disproved in a matter of decades?
The answer is that given what we know now, science should always be able to give a best guess at something. With more knowledge than in the past, we should always be able to at least do no worse than previous scientists. While this isn't always true (because science is rarely ever untainted by human emotion, politics, stupidity, and outright lies), in theory, we should always be progressing in our knowledge of the universe. While we may be wrong, even blatantly wrong at times, if we follow the scientific process, we should eventually stumble upon the truth. So where does Ockham's razor fit in? Well, it is part of the scientific process, and it's a little part of pop culture that has made it into the fringe of the populace (thanks to Contact), so I feel a need to clear it up a few things. It is named after an English philosopher, William of Ockham (I think), and he said it in Latin first, (I think).
Anyway, Ockham's razor has been quoted and misquoted many times. I won't even try to quote it here, but I will point out that "The simplest explanation is the correct one" is NOT, Ockham's razor.
To paraphrase in plain English, I would say that the spirit of Ockham's razor is if you have multiple hypotheses that explain something equally well, then the hypothesis with the fewest assumptions should be used. It isn't necessarily correct, but the simplicity is important. If you had two hypotheses:
1) The apple fell off the table because of Earth's gravitational pull
2) The apple fell off the table because of Earth's gravitational pull, which is caused by aliens with supermagnets.
Then obviously the first one is preferable. But it isn't just because of the ridiculousness of the aliens hypothesis.
1) The apple fell off the table because of Earth's gravitational pull
2) The apple fell off the table because of Earth's gravitational pull, and because somebody bought the apple and put it on the table.
Does the second hypothesis explain it better? Maybe. But with no evidence to support that, we would have to go with the first one. If we had a receipt of the apple purchase, then the second one would win because of the better explanatory power.
In any set of hypotheses, it is always possible to add more assumptions, but come up with the same conclusion. There are fewer simple hypotheses, so to make it easier on us, we choose the simpler ones first. They can always be modified to be more complex. Just as Newtonian physics gave way to relativistic and quantum physics, simpler theories do NOT equal more correct theories. But it is important not to jump to conclusions, and to always yield to evidence, as long as it is reputable and reproducible.
So there's today's science lesson. Go science!
# posted by Stuart @ 11:35 PM 
