A link to Nature which has a summary of the primary candidates' stance on science issues. (HT: Mike)
Bora just announced the selections for Open Lab 2007, the best science blogging of the year (which I was a judge, although Bora may have forgotten ...)
The Edge asked its illustrious members, which include some of the smartest scientists and thinkers on the planet "WHAT HAVE YOU CHANGED YOUR MIND ABOUT? WHY?" There is also a nice sub-question:
Science is based on evidence. What happens when the data change? How have scientific findings or arguments changed your mind?
Speaking of science & faith, Dan of (Migrations and Bitesize Bio) and I have been compiling some links on this subject:
- a video clip of Francis Crick discussing the relationship between science and truth.
- Another clip. This one is part 2 of a long discussion between Dawkins, Dennet, Harris and Hitchens. Dennet and Dawkins make interesting points when contrasting religious faith with science.
- An article by Dennis Overbye. I didn't like it, but Dan pointed out this passage:
We don't know, and might never know, if science has overbid its hand. When in doubt, confronted with the complexities of the world, scientists have no choice but to play their cards as if they can win, as if the universe is indeed comprehensible. That is what they have been doing for more than 2,000 years, and they are still winning.
- A post of Science and Truth at Reflections, Ideas, and Dreams. Here's a relevant passage:
The falsificationist description of scientific inquiry, on the other hand, is not concerned with whether or not we can know Truth-even if we stumble upon a theory that is True, we could never recognize it as such! Falsificationism itself is a logical process, not one based on the faith-based belief in the attainability of Truth, and thus can be reasonably preferred when comparing with certain religious interpretation.
And since we are talking about science, here's a link to an ongoing CBC radio series on the anthropology of science. (HT: Nisbet)
Moving along to something different ... a world map of governments who infringe on their citizen's privacy. As the WIRED blog says the map is
rating various nations for their civil liberties records.
(map is below the fold)
Note that blue = good, yellow = not so bad and black = very bad. Click here for a larger version.
That's all for now. I'm off to figure out the results of this year's Iowa Caucuses ...
- Log in to post comments
I'm in the middle of reading a really interesting book by Deborah Mayo that critiques Karl Popper. She, and others, think that his theories don't reflect with what scientists do in real life.
Popper holds that no scientific theory should be held above others, but obviously as scientists, we do hold some theories above others, namely the ones that have predicted results and had them verified very precisely. If a theory is corroborated many times by different experiments, we don't just say it's "not false"; we say it's "a pretty good theory", depending on how rigorous and stringent the experiments were. For things like evolution and the theory of general relativity, they're not just good theories; they're damn good theories. Surely they're much better theories than, say, the theory that life on Earth came from outer space on an asteroid, right?
Every experiment tests a main theory along with a group of auxiliary hypotheses (e.g. the instrument works, there wasn't an arithmetic error, the ozone in the atmosphere isn't oxidizing the dyes in my microarrays), and if the experiment contradicts the predictions, falsifiability only states that something was wrong, but not what. For a well-supported theory or paradigm, we'd be inclined to blame the instrument, the person doing the experiments, reagents, cell lines, protocols, whatever, everything but the theory. And, of course, there are those professors that cling to their pet theories until their eventual, inevitable demise.
I agree with that. The way I see the issue was that Popper was partially right, but in reality scientists follow probabilistic thinking. In other words, given all the results (some which may contradict any particular theory) which model has the best predictive power. If you don't mind my asking, what is the name of the book? I'll have to add it to my list (along with Leviathan and the Air Pump - see the CBC radio link.)
Since you brought up faith again, I'm going to repeat my point that the religious do not have ownership of faith. If I may quote your fellow Montrealer Leonard Cohen (from 'Villanelle For Our Time'):
This is the faith from which we start:
Men shall know commonwealth again.
From bitter searching of the heart
We rise to play a greater part.
Faith like this is something that drives people whether they be agnostic scientists (say like us) or monotheistic religious (say Desmond Tutu), etc.
Anyways, it's sad to see faith used as a bad word. Both sides of the political debate on this issue are to blame for this. Personally, I would say that in the evolution vs. ID debate, for instance, it is the ID people that lack faith: they think that God is no smarter than a human in his creation. For the faithless to claim ownership of faith is a joke.
The book is called "Error and the Growth of Experimental Knowledge."
"Every experiment tests a main theory along with a group of auxiliary hypotheses (e.g. the instrument works, there wasn't an arithmetic error, the ozone in the atmosphere isn't oxidizing the dyes in my microarrays), and if the experiment contradicts the predictions, falsifiability only states that something was wrong, but not what..."
I think you've hit on the crux of the issue here. In the real world, it's indeed impossible to be completely certain of any one particular experimental result, beyond statistical probabilities.
Falsifiability is still an important prerequisite of any useful theory however. All this means is that in theory, there should be a conceivable experiment, that, in a world of perfect, error free experiments, is capable of falsification. I don't think it really applies to the actual process of doing experiments. It's just a useful guideline for eliminating new theories that are total garbage. Once you have a theory that can in theory be falsified, you're in different territory - the experimental world - and you deal with the probabilities. (Of course you still have do decide, at the end of the day, do you believe the result/theory or not - we wouldn't be very effective if we carried p-values around in our heads for every single fact or hypothesis encountered).
For example falsification is a useful test to reject useless theories such as:
- Living beings have a supernatural essence or "soul" that is real but physically undetectable.
- The universe was created by an all-powerful being who left no evidence of her involvement within it.
etc., etc.