Scientific Conclusions

[_MrStakhanovite]

So in a response to a challenge, let’s tackle this statement:

A scientific conclusion is holding something to be true because of strongly corroborated evidence which does not allow for an alternate interpretation.

(size in the original)

So this is pretty much doesn’t work and to demonstrate this, I’ll walk through two standard paradoxes that exist. To begin with, let’s start with a basic rule from inductive reasoning:

(P1) A Generalization is confirmed by any of its instances.

So if you generalize that all tigers have stripes, and any tiger you find that has stripes confirms your generalization and any tigers found without stripes disconfirms your generalization.

To be kind, I’m going to skip all the problems with inductive reasoning and assume for argument’s sake that there is genuinely legitimate inductive reasoning without all the hoo-ha. Next Premise:

(P2) If two hypothesis can be known a priori to be equivalent, then any data that confirms one, confirms the other.

Something can be known a priori based on reflection and reasoning, not by any experience. So let’s take on another premise:

(P3) All ravens are black

The following premise is logically equivalent to (P3):

(P4) Everything non-black is a non-raven.

So anything that confirms (P4) also confirms (P3), and visa versa (by P2), Next premise:

(P5) This non-black soda can is a non-raven.

(P5) is an instance of (P4) and by (P2) is also an instance of (P3), but this is absurd, how does data from a Coca Cola can have anything to do with Ravens? But the problem still gets compounded:

(P6) X is bellow if and only if X is black and has been examined or X is yellow and has not been examined.

Now (P6) is really screwed up, it tells us all ravens can be considered ‘bellow’ and that all unexamined ravens are yellow and not black.

So according to Beefcalf, there can’t be any scientific conclusions, because it would be easy to come up with a massive (but denumerable) list of alternative theories that on the face of it, appear silly, but in fact marshal just as much evidence as the accepted conclusions.


Q.E.D.

[_Aristotle Smith]

Goddammit Stak, people don't take this stuff seriously. There's no such thing as philosophy of science, and there is especially no such thing as the problem of induction. Am I supposed to get PhDs in astrophysics and philosophy before I can say I know anything about science? I mean what the hell. Stop sodomizing Ray Comfort and get on board with reality man!

Science is utterly transparent and true to everyone. Just like Mormonism was for me before I realized that it was utterly opaque and false to just about everyone. What the hell is wrong with you?

[_beefcalf]

Ok. You lost me at P6. Please go over that one in more detail, please.

by the way: I didn't challenge you. I accepted your invitation to be schooled.

[_Blixa]

Goddammit Stak, people don't take this stuff seriously. There's no such thing as philosophy of science, and there is especially no such thing as the problem of induction. Am I supposed to get PhDs in astrophysics and philosophy before I can say I know anything about science? I mean what the hell. Stop sodomizing Ray Comfort and get on board with reality man!

Science is utterly transparent and true to everyone. Just like Mormonism was for me before I realized that it was utterly opaque and false to just about everyone. What the hell is wrong with you?

You are on a roll that I hope never stops...

[_zeezrom]

Stak,

The usage of "corraborated" was new to me. I don't have much to say about it. All I know is you have a hypothesis, you do an experiment, you gather data, you analyze the data, you determine whether the data supports the hypothesis and therefore whether you can conclude it to be true. If the experiment does not support your hypothesis, you go back and see if your hypothisis needs to be adjusted and so forth.

Here is an example. I hypothesized that if I prayed to God asking if the Book of Mormon is true, historical scripture, I would begin to feel a mood change/bubbly feeling/goosebumbs within 5 minutes or so of my prayer. I recorded my feelings in my journal and then I analyzed the data. The data supported my conclusion.

[_Sethbag]

There is a reason that "induction is the glory of science, and the scandal of philosophy".

The reason this doesn't trouble me, even in theory, is that science doesn't claim absolute knowledge of things. When looked at up close, with a microscope, even the strongest scientific claims look more like statements of confidence or probability, and are always subject to further data changing what we must think of any given claim. We may be comfortable knowing that the process has worked well in the past, and tends to correct itself over time, such that any error in the theories we now have are likely to be weeded out over time, with better information. See, an inductive reason for the non-catastrophic nature of the problem of induction for science.

[_Sethbag]

Another comment: if we didn't operate through inductive reason all the time, every day of our lives, we wouldn't get out of bed in the morning.

Indeed, it's probably been argued by some PhDs in some natural science that our brains evolved with built-in inductive circuitry. It's how we learn. We see someone do something, and we see the results, and we figure out that if we do that thing, we'll probably get that result too. It's a purely inductive process.

How do you know the sun will rise in the morning? If you say "because it always has in the past" then you've just used an inductive argument. I'd argue that it would be an argument backed by some pretty good hypotheses, but if you were to dig down deep enough, you could probaby show that many of those hypotheses were based on inductive arguments themselves, so you can't really escape it. But does the fact that even predictions such as that the sun will rise being inductive renders them somehow useless? Not at all. And if it ever doesn't, we'll know we were wrong. ;-)

[_mikwut]

Be careful Sethbag, your getting awful close to defending a crude form of the principle of sufficient reason.

[_The Dude]

I have a hard time remembering the difference between inductive and deductive. Seriously, I always have to Google it. I learned early on that you will not be taken seriously if you talk about such things in the wet laboratory where actual experiments take place.

[_beefcalf]

Ok. So maybe you're busy. Maybe you're waiting for my due diligence...

(P1) A Generalization is confirmed by any of its instances.
(P2) If two hypothesis can be known a priori to be equivalent, then any data that confirms one, confirms the other.
(P3) All ravens are black
(P4) Everything non-black is a non-raven.
(P5) This non-black soda can is a non-raven.

(P5) is an instance of (P4) and by (P2) is also an instance of (P3), but this is absurd, how does data from a Coca Cola can have anything to do with Ravens? But the problem still gets compounded:

(P6) X is bellow if and only if X is black and has been examined or X is yellow and has not been examined.

Now (P6) is really screwed up, it tells us all ravens can be considered ‘bellow’ and that all unexamined ravens are yellow and not black.

So according to Beefcalf, there can’t be any scientific conclusions, because it would be easy to come up with a massive (but denumerable) list of alternative theories that on the face of it, appear silly, but in fact marshal just as much evidence as the accepted conclusions.

I'm fine with P1 through P5, although I do not agree that P5 is absurd.

So where does P6 come from? Why are we inventing an adjective? What is the justification for introducing the conditions of 'examined' and 'non-examined'. You've made a leap here that I haven't been able to make.

by the way, the tendency for a logical paradox to bring me to my intellectual knees decreased somewhat after I read Hofstadter's GEB and had Zeno explain why Achilles could never, ever, ever win a race against a tortoise, as long as you gave the tortoise a head start, no matter how small. Why should a layman like me ever consider these type of arguments as anything other than smoke and mirrors?