Is the Scientific Method Flawed?

[Thanatos34]

Hey, I was in the middle of writing a paper, and I thought, what the heck, let's see what others think! This is a "critical thinking" paper, so if you can prove a point with logic, you do not need to reference sources. (I do like these papers, normally, but this one is throwing me for a loop.)

The topic:

Is the scientific process inherently flawed?
(Based on the fallacy of affirming the consequent.)

For those who are unaware of what said fallacy is, let me explain:

A > B (A implies B.)
B (B is found to be true.)
/ A (Therefore A is true.)

^^^ This is the fallacy of affirming the consequent. The conclusion is incorrect. If you do not find B, you can conclude that A is not true, but the reverse is not the case.

But, this is what the entire scientific method is based on, is it not? One forms a hypothesis that states a certain scenario, (A), is true. The only real way of figuring this out is to go looking for evidence, (B). One finds B, lots of B. So one concludes that A must be true!

To give a real-life example: Let's say that my theory, (A), is that all geese are white. I find evidence, (B), namely white geese. A lot of white geese. However, no matter how many geese I find, unless I find every goose in the world, (unlikely), I cannot conclude that my theory is correct. I can say that it is probably the case, but I cannot say that it is 100% correct.

Of course, in my opinion, this is exactly what the scientific process is supposed to look like, it is never supposed to prove something with 100% accuracy. My teacher, however, seems to think that this means something is wrong with the scientific method, (yes, he's a little wacky). Anyone got any other thoughts on this?

 

[bodyklok]

Everyone knows the scientific method is flawed, but out of all the methods it's probably the best.

 

[garfoldsomeoneelse]

It's the best we've got. Don't knock it.

 

[crepesack]

I suppose in science, it doesn't matter if you can find the absolute or not. Science is constantly changing. In science you never say ALL or EVERY or NONE. Because of our infinite universe and because of our limited understanding of nature we are left with either "some" or "most".
A-->B
B, Therefore A.

Not necessarily. In science there is more than one cause to a single effect.
if
A--->B
C--->B
D--->B
B is true then what happens?

That is in my opinion what the scientific Method.
A control, and variables.
If A didn't make B happen then C and/or D must have, but C occurs 1 out of 20million times. so for all intensive purposes, D must be more true than C.

Another analogy, we could all agree that a lobster is red. Oh wait what? Does that not mean we are implying all lobsters are red? Yes, it is common practice to assume the most common but this is not the case. Albinism, and different color morphs occur within lobsters. So would you say "most" lobsters are red afterwords? No you wouldn't because the majority are red.

Then take another example, An electron has mass it is around 9.10938188 × 10^(-31) kilograms but for all intensive purposes it weighs in at 0.

 

[Weaver]

You're slightly misunderstanding. In affirming the consequent you're simply suggesting the notion that A->B, there is no actual proof for this. I realize most critical thinking does not teach formal logic very well, but the scientific method is more like this:

You have A
You make a single and isolated as best change as you can to A
Event B occurs.
We can hypothosize that A->B, and thus if more testing keeps supporting this result we can find that A->B.

However, in affirming the consequent, you never HAVE A, you simply pull A out of thin air, for lack of a better term.

 

[Alex_P]

It's not classical logic, it's Bayesian inference.

-- Alex

 

[Weaver]

Perhaps I should describe it like this, if you have
A->X
B->X
C->X
D->X
E->X
X

event X occurs, thus you assume event C is true. That is affirming the consequent, since why is it C that is true and none of the others?


In the scientific method you would have more like this:
test1
----------
A
X
/A->X

test2
----------
A
X
/A->X
----------

.
.
.

etc.

If tests for B-E did now seem to cause X, then we could assume A implies X.
This is sort of auspicious since in the scientific method all events are observed and controlled. Affirming the consequent is more like if in life you saw a smashed up car on the side of the road you assume it hit another car, but in reality it hit a moose or something.

 

[Thanatos34]

Perhaps I should describe it like this, if you have
A->X
B->X
C->X
D->X
E->X
X

event X occurs, thus you assume event C is true. That is affirming the consequent, since why is it C that is true and none of the others?


In the scientific method you would have more like this:
test1
----------
A
X
/A->X

test2
----------
A
X
/A->X
----------

.
.
.

etc.

If tests for B-E did now seem to cause X, then we could assume A implies X.
This is sort of auspicious since in the scientific method all events are observed and controlled. Affirming the consequent is more like if in life you saw a smashed up car on the side of the road you assume it hit another car, but in reality it hit a moose or something.

Hmmm, so you would argue that the scientific method does not actually involve the fallacy of AtC at all?

I have taken symbolic logic, and your definition of AtC seems a little off. You do not need multiple implications to have the fallacy, you only need the one.

A>B
B
/A

Was always a fallacy, if our teacher ever caught us using it, he would dock us points.

 

[Weaver]

Perhaps I should describe it like this, if you have
A->X
B->X
C->X
D->X
E->X
X

event X occurs, thus you assume event C is true. That is affirming the consequent, since why is it C that is true and none of the others?


In the scientific method you would have more like this:
test1
----------
A
X
/A->X

test2
----------
A
X
/A->X
----------

.
.
.

etc.

If tests for B-E did now seem to cause X, then we could assume A implies X.
This is sort of auspicious since in the scientific method all events are observed and controlled. Affirming the consequent is more like if in life you saw a smashed up car on the side of the road you assume it hit another car, but in reality it hit a moose or something.

Hmmm, so you would argue that the scientific method does not actually involve the fallacy of AtC at all?

I have taken symbolic logic, and your definition of AtC seems a little off. You do not need multiple implications to have the fallacy, you only need the one.

A>B
B
/A

Was always a fallacy, if our teacher ever caught us using it, he would dock us points.

This is correct, but I was simply demonstrating WHY it is a fallacy, since A->B in itself is a fallacious supposition, as there could be other things that AREN'T A which caused B.
And yes, I would argue the scientific method does not have this fallacy, since hypothesis generally are not made up simply from imagination, or at least have good cause of reason for them.

So, most hypothesizes development will be something like:

A
B
/A->B

We thus hypothesize that A causes B. We do A a bunch more times to see if B happens again and again. If it does, we can say we've developed a scientific theory, and that event A correlates with event B.

The purpouse of the scientific method is to test the hypothesis, A->B in this case. Thus, even if you just started out with the hypothesis A->B with no reasons, you would still have to produce event A to see if event B occurs.

Much of science was always "I see event B occurred, WHAT is the event A causing it?" and then they go on their lives trying to find the event A which produces B. Thus I can see why you would confuse it, but the hypothesize found false are not kept

This is a sort of... formal logic version of the scientific method, the real method is naturally more complicated, but you probably get the drift.

 

[UsefulPlayer 1]

hmmm?

Sounds like thesis writing to me. Anyway, the only thing that comes to mind is that Scientific Revolution dude I don't really remember.

See what your doing is going form General to Specific. In which yes, you can leave out information, like saying nerds are smart and since Tommy is a nerd he is smart, and come to a conclusion. A better method would be to start with Specific to General, like Tommy is a nerd and then applying Tommy to all the other nerds to make a comparison and all that jazz.

Oh and I have no idea what I'm talking about and just wanted to feel smart...

 

[Thanatos34]

hmmm?

Sounds like thesis writing to me. Anyway, the only thing that comes to mind is that Scientific Revolution dude I don't really remember.

See what your doing is going form General to Specific. In which yes, you can leave out information, like saying nerds are smart and since Tommy is a nerd he is smart, and come to a conclusion. A better method would be to start with Specific to General, like Tommy is a nerd and then applying Tommy to all the other nerds to make a comparison and all that jazz.

Oh and I have no idea what I'm talking about and just wanted to feel smart...

Deductive and Inductive Methods.

And it's not thesis stuff, just a paper for a philosophy class I am taking.

 

[Thanatos34]

Perhaps I should describe it like this, if you have
A->X
B->X
C->X
D->X
E->X
X

event X occurs, thus you assume event C is true. That is affirming the consequent, since why is it C that is true and none of the others?


In the scientific method you would have more like this:
test1
----------
A
X
/A->X

test2
----------
A
X
/A->X
----------

.
.
.

etc.

If tests for B-E did now seem to cause X, then we could assume A implies X.
This is sort of auspicious since in the scientific method all events are observed and controlled. Affirming the consequent is more like if in life you saw a smashed up car on the side of the road you assume it hit another car, but in reality it hit a moose or something.

Hmmm, so you would argue that the scientific method does not actually involve the fallacy of AtC at all?

I have taken symbolic logic, and your definition of AtC seems a little off. You do not need multiple implications to have the fallacy, you only need the one.

A>B
B
/A

Was always a fallacy, if our teacher ever caught us using it, he would dock us points.

This is correct, but I was simply demonstrating WHY it is a fallacy, since A->B in itself is a fallacious supposition, as there could be other things that AREN'T A which caused B.
And yes, I would argue the scientific method does not have this fallacy, since hypothesis generally are not made up simply from imagination, or at least have good cause of reason for them.

So, most hypothesizes development will be something like:

A
B
/A->B

We thus hypothesize that A causes B. We do A a bunch more times to see if B happens again and again. If it does, we can say we've developed a scientific theory, and that event A correlates with event B.

The purpouse of the scientific method is to test the hypothesis, A->B in this case. Thus, even if you just started out with the hypothesis A->B with no reasons, you would still have to produce event A to see if event B occurs.

Much of science was always "I see event B occurred, WHAT is the event A causing it?" and then they go on their lives trying to find the event A which produces B. Thus I can see why you would confuse it, but the hypothesize found false are not kept

This is a sort of... formal logic version of the scientific method, the real method is naturally more complicated, but you probably get the drift.

In your development

A
B
/A-> B

What is A? Can you give me an example? I'm having a hard time trying to figure what A is that wouldn't be some sort of theory with a testable result. (Thus, it turns out that step 1 is actually A-> B.

 

[NeutralDrow]

Of course, in my opinion, this is exactly what the scientific process is supposed to look like, it is never supposed to prove something with 100% accuracy. My teacher, however, seems to think that this means something is wrong with the scientific method, (yes, he's a little wacky). Anyone got any other thoughts on this?

I'd agree with you on the teacher. That the scientific method is based on inductive rather than deductive reasoning (assuming that all geese are white, for example) isn't a flaw with the method itself, but with human limitations.

 

[Weaver]

Perhaps I should describe it like this, if you have
A->X
B->X
C->X
D->X
E->X
X

event X occurs, thus you assume event C is true. That is affirming the consequent, since why is it C that is true and none of the others?


In the scientific method you would have more like this:
test1
----------
A
X
/A->X

test2
----------
A
X
/A->X
----------

.
.
.

etc.

If tests for B-E did now seem to cause X, then we could assume A implies X.
This is sort of auspicious since in the scientific method all events are observed and controlled. Affirming the consequent is more like if in life you saw a smashed up car on the side of the road you assume it hit another car, but in reality it hit a moose or something.

Hmmm, so you would argue that the scientific method does not actually involve the fallacy of AtC at all?

I have taken symbolic logic, and your definition of AtC seems a little off. You do not need multiple implications to have the fallacy, you only need the one.

A>B
B
/A

Was always a fallacy, if our teacher ever caught us using it, he would dock us points.

This is correct, but I was simply demonstrating WHY it is a fallacy, since A->B in itself is a fallacious supposition, as there could be other things that AREN'T A which caused B.
And yes, I would argue the scientific method does not have this fallacy, since hypothesis generally are not made up simply from imagination, or at least have good cause of reason for them.

So, most hypothesizes development will be something like:

A
B
/A->B

We thus hypothesize that A causes B. We do A a bunch more times to see if B happens again and again. If it does, we can say we've developed a scientific theory, and that event A correlates with event B.

The purpouse of the scientific method is to test the hypothesis, A->B in this case. Thus, even if you just started out with the hypothesis A->B with no reasons, you would still have to produce event A to see if event B occurs.

Much of science was always "I see event B occurred, WHAT is the event A causing it?" and then they go on their lives trying to find the event A which produces B. Thus I can see why you would confuse it, but the hypothesize found false are not kept

This is a sort of... formal logic version of the scientific method, the real method is naturally more complicated, but you probably get the drift.

In your development

A
B
/A-> B

What is A? Can you give me an example? I'm having a hard time trying to figure what A is that wouldn't be some sort of theory with a testable result. (Thus, it turns out that step 1 is actually A-> B.

I can try
You've probably noticed I'm not a very good teacher which is why I always loose marks on "explain this: " questions on exams even if I understand it perfectly in my head lol.

Well, this is pretty much where treating it in a formal logic sense breaks down, because in formal logic A->B means just that, every time A occurs, B occurs. In the scientific method this ("this" being A->B) would be a hypothesis, thus we could not know it for certain.

So, if we had our hypothesis as step 1, summarized by this:

A->B
B
/A

This is clearly confirming the consequent and WRONG TO DO. Since we are taking the occurrence of event B in support of event A, but we don't know if A occurred at all. Thus, a more scientific approach, and the approach they would use is like this:

A->B (our hypothesis)
A (we cause event A to occur)
B (we find that event B occurred!)
/A->B (we have strengthened the support of our hypothesis!)

Now, not to complicate things, but in most studies they try to disprove the null hypothesis. This means they try to disprove that A does not imply B, but you don't really need to worry about that. For an example... lets see...

If we make our hypothosis that "Apple trees drop apples to the ground" with
A = Apple trees grow apples; these apples drop from these trees.
B = These apples come to rest on the ground.

Scientifically we hypothesize that A->B

So, we thus scientifically do not go look for apples on the ground (B) but we rather go and look for (A) apple trees!

So we get 50 apple trees with our hypothesis that A->B
We observe A, yes apple trees grow the apples.
And then we observe B! Ha it seems like A->B. We do this for all 50 trees and we have a sustainable scientific hypothesis. "Apple trees grow apples and these apples drop to the ground and come to rest"

Thus
A->B (we ASSUME, so lets test!)
------
A
B
/A->B

In a form of affirming the consequent, you could, even rightfully assume A->B, but lets say you're just walking around and find an apple on the ground. It would be wrong to 100% conclude an apple tree dropped it there, as a person may have placed it there, or perhaps an animal moved it, or the wind blew it away or whatever.

This example kind of sucks, but I'm not very creative

 

[DestinyCall]

Another analogy, we could all agree that a lobster is red. Oh wait what? Does that not mean we are implying all lobsters are red? Yes, it is common practice to assume the most common but this is not the case. Albinism, and different color morphs occur within lobsters. So would you say "most" lobsters are red afterwords? No you wouldn't because the majority are red.

Actually the majority of lobsters are brown. The bright red lobsters people typically think of when someone says "lobster" is a cooked lobster. When alive and uncooked, the lobster's carapace is a much darker brownish color. This is actually an excellent example of how sample bias can introduce error when conducting an observational study. A appears to be B, but is actually C ... and all that nonsense.

 

[Thanatos34]

Another analogy, we could all agree that a lobster is red. Oh wait what? Does that not mean we are implying all lobsters are red? Yes, it is common practice to assume the most common but this is not the case. Albinism, and different color morphs occur within lobsters. So would you say "most" lobsters are red afterwords? No you wouldn't because the majority are red.

Actually the majority of lobsters are brown. The bright red lobsters people typically think of when someone says "lobster" is a cooked lobster. When alive and uncooked, the lobster's carapace is a much darker brownish color. This is actually an excellent example of how sample bias can introduce error when conducting an observational study. A appears to be B, but is actually C ... and all that nonsense.

Nice first post, and welcome to the Escapist. I was going to correct him myself, considering I worked for three years at the Monterey Bay Aquarium, but I figured it was superfluous to the point he was making, and I'm tired from writing this paper. (It's one of two I have to do for this professor. Two different classes. I'm sure I'll be on here tomorrow night with the other topic. Yes, I have a bad habit of waiting till the last minute. I don't always do so, but I did for these two.)