" ‘Falsifiable’ does not mean ‘false’. For a proposition to be falsifiable, it must be possible in principle to make an observation that would show the proposition to be false, even if that observation has not been made. For example, the proposition"
This agrees with what I posted.
By the definition I gave (which I took from Anthony Flew) and your Wiki quote, this statement is falsifiable.
No, I misread the Wiki quote. Still, given that I have encountered at least one other who defines the term differently, I’m not willing to regard it as authoritative.
Still, I think we can agree that there is a difference between a statement whose truth is not determinable, and a statement whose truth is know but could be different in another possible world. If not falsifiable, what would you call the latter?
Well, the way I’m thinking about it, I’m presupposing that I correctly picked a red object. Otherwise, I’ve made an error in my experiment, and it’s not going to prove everything. But since demonstrating falsifiability only requires me to have described an experiment that could exist in principle, saying “I’d start by selecting a red object,” and assuming the selection was made correctly doesn’t seem to be a problem.
But at any rate, I think the problem boils down to how we define falsifiability. It doesn’t work to just say “A claim is falsifiable is it is possible to prove it false”, because there are lots of true claims that are falsifiable, but none of them can be proven false. (Otherwise they wouldn’t be true.)
The way I’m defining falsifiability, and the way I think Blake is defining it, is something like “A claim is falsifiable if you can propose an experiment which could in principle be conducted, and for which a certain outcome, if it occured, would prove the claim false.” Whether that outcome could actually occur has no bearing on whether the claim is falsifiable. It couldn’t occur for any true claim – the only difference with tautological claims is we can prove the outcome couldn’t occur by means of pure logic, making the experiment a fairly pointless exercise.
The way I suspect you are defining falsifiability, Thudlow, (and please correct me if I’m wrong) is probably something more like “a claim is falsifiable if there could in principle exist a logically consistent universe in which the claim could be proven false.” With that definition, tautologies wouldn’t be falsifiable, since actually proving one false would violate the rules of logic. On the other hand, the existence of gravity would be falsifiable, because even though it exists in our universe, you could imagine a logically consistent universe that had no gravity, and an experiment that would prove gravity doesn’t exist that could be conducted in such a universe.
I can’t say with complete certainty that my way of defining falsifiability is the most widely accepted one. (I’ve usually just heard “a claim is falsifiable if it could be disproven in principle.” But that just transfers the confusion from what you mean by falsifiable to what you mean by in principle.) But I do have some problems with the second definition:
(1) If it’s a matter of proposing a hypothetical universe in which the claim is false and that falseness could be demonstrated, why couldn’t you say something like “Well, I imagine a universe containing a lightbulb which turns on whenever God exists. If you look at the lightbulb and it’s off, then that proves God doesn’t exist. Therefore, God’s existence is falsifiable, and thus it is a question of Popperian science.” This strikes me as utter rubbish. My preferred definition of falsifiability doesn’t have that problem, since you could respond, “there’s no reason to think such a lightbulb could be constructed in our universe, even given unlimited resources, time, energy, etc.” Whereas when we bring possible universes into it, then who knows what is or isn’t allowed.
(2) What if scientists were to discover that the existence of gravity is necessary for the existence of matter? Then how could you even in principle talk about a universe in which you could disprove the existence of gravity – there wouldn’t be any matter to conduct the experiment with! And if you said that in that universe the existence of gravity isn’t necessary to have matter, then you’re changing the laws of physics, in which case what you’re testing for might not really be gravity anymore. So does that mean that such a scientific discovery would cause me to say that the existence of gravity is no longer a question of Popperian science? I don’t think it should. My prefered definition of falsifiability doesn’t have this problem, because all that matters is that an experiment to test for gravity’s existence could be conducted in our actual world. The fact that that experiment would never actually prove the non-existence of gravity doesn’t matter.
So at any rate, I think it makes a lot more sense to define falsifiabilty as I’ve defined it, and as Blake seems to be defining it, where all that matters is that an experiment can be proposed for which a certain result would disprove the claim, rather than saying anything about it actually being possible to disprove the claim. Otherwise you’d need to find some way that true claims could be disproven, since there are lots of claims that everyone considers falsifiable and everyone also assumes are true (gravity existing, the world being round . . . )
No it doesn’t.
You said that for a statement to be falsifiable there have to be actual circumstances under which it is false. The Wikipedia quote says that there has to exist a potential, in principle to make an observation that would show it false.
That’s a huge difference.
You said that there had to actually be times/circumstances when it was false.
The Wikipedia quote says that if there are times/circumstances when it’s false they have to be observable. There’s no requirement that such times/circumstances ever occur as you claim.
So can you provide a reference for your claim that falsifiable statements have to actually be false under some circumstances? I have asked once and you didn’t respond?
And can you explain the logical paradox that results form such a belief, which is that no true statement can ever be scientifically valid?
That’s an interesting assertion. Please tell us all what potential observations you can make that would be incompatible with the existence of an omipotent God that can create any observations he likes at will?
Missing word in my post. I should have written:
“possible” here should be taken in the sense of “logically possible”.
Could you tell us who this other is so we can judge how much value to give to their dispute?
The universe.
As tim just showed so eloquently everything in the universe fits into the latter category.
The problem with making such a distinction is that I see no practical way of establishing it. It fails the duck test. The difference between a statement that has been proven false here but that could be true in a parallel and a statement that is false in all universes is non existent.
In response to ultrafilter:
If “God exists” is falsifiable, then falsifiability is a lousy criteria for whether or not something is a question of science. That certainly isn’t what Popper meant by “falsifiable”, and I doubt it’s what most people mean.
If what you mean is that the truth or falsehood of the statement isn’t required by logic alone, then why not just say it’s “not purely a question of logic.”
I would go further and say physically and practically possible. A theory that posits the existence of Pink Unicorns doesn’t become scientific just because I say it can be falsified if the Pope flies to the moon on Easter Sunday. It’s certainly logically possible for the Pope to fly to the moon on Easter Sunday but physically and practically it’s not possible to actually conduct any such experiment.
Flew, as I mentioned above.
Au contraire: the latter is a logical contradiction, and the former is not. For a logician, it’s a very important difference.
Can you provide a little more information please. There are probably a hundred thousand people of that name in the world. Can you narrow it down , or is he just some guy you know?
What is logically contradictory about a statement that is true in all universes?
- And can you please support or retract your claim that the existence of God is falsifiable? Can you please explain how, by your own definition, there are circumstances that would be incompatible with an omnipotent deity that can manufacture any circumstances he likes at will?*
I just looked up [“Anthony Flew” falsifiability] on Google.
“His principle of falsifiability’ … says that any statement with real content will, if true, make certain other statements false.”
That isn’t in contradiction to the definition used by Popper and others. It’s saying the same thing another way and without the caveat that there must be objective evidence that the other statements are false.
It’s somewhat looser than Popper’s definition but it doesn’t contradict it.
No, I’m happy with your definition of falsifiability; I’m on your side on that one. I’d say “a claim is not a tautology if there could in principle exist a logically consistent universe in which the claim could be proven false.” Falsifiable statements would, I think, be a subset of such non-tautologies.
What I’m having trouble with, then, is the idea that a tautology could be falsifiable, or “checkable.”
This does seem to me to be a problem: you’re assuming that the experiment will turn out a certain way, so for it to turn out a different way (and prove the claim false) is logically inconsistent with that assumption.
Or in Blake’s formulation,
I think that in a sense, it does. I wouldn’t call it a check if there is logically only one possible outcome, which is the case if you’re just re-checking the exact thing you’ve already checked.
Thus, “There is no life on Uranus” is a falsifiable statement. “There is no life on Uranus given that there is no life on Uranus” is not a falsifiable statement.
Tautologies are not falsifiable.
A tautology is a a compound proposition that is true irrespective of the truth values of the atomic propositions that it consists of.
Testing the atomic propositions is irrelevant to the truth value of the tautology, hence it is not falsifiable.
The example has been given of checking whether all red objects are red.
That is a compound proposition. It can be written:
“If A is red, then A is red.”
As you can see, whether or not A is red has no effect on the truth value of the tautology. If A is red, the tautology is true. If A is not red, the tautology is true. So what is the point of checking? Checking to see if an object is red is irrelevent; the tautology is not falsifiable.
Call it a check, call it an experient, call it a gnork. It doesn’t matter. What it is is a means of potentially obtaining observations that would contradict the hypothesis. Hence it constitutes falsifiability.
So you are saying that if we observed life on Uranus that statement would still be true? How can that be the case? How can it be true that there is no life on Uranus despite our having observed life on Uranus?
I think part of the problem you are having is that you think that all parts of a hypothesis need to be demonstrated false to prove that it is false. That’s not the case. If any part of itis proven false then it has been falsified.
I say “Given a brick weighs 1/1000th of an ounce and a house is made up of a thousand bricks, a house weighs an ounce”. You don’t have to actually wiegh an entire house to prove that false. You can falsify it equally well by weighing a brick.
It’s an easily falsifiable statement simply because there are observations that are easily collected that should be impossible is the statement were true.
It’s a little harder to collect observations of life on Uranus but nontheless the statement remains falsifiable. If the statement is a tautology then its won’t be falsified of course, but there exists a way to potentially observe contradictions to it if they such contradictions.
Cite please.
There is no point at all. That is one point that everyone in this thread has agreed on from the outset. Testing a tautology is redundant. It’s good to see that you too agree and have restated it for about the 20th time.
The mistake you are making is thinking that falsifiability means that something has to actually be proved false. We’ve already established that is not the case.
You do realise that’s a total non sequitur don’t you?
Just because an experiment is irrelevant/redundant doesn’t mean that it’s impossible. As we established much earlier in this thread, all that falsifiability requires is that the experiment be possible.
Wherever did you get the idea that the experiement also had to achieve results that weren’t previously known? Whatever makes you think that an experiment that observes something that was predicted with perfect accuracy beforehand isn’t an experiment?
No, you can’t.
This can be written as:
“If a brick weighs 1/1000th of an ounce and a house is made up of a thousand bricks, then a house weighs an ounce.”
For this statement to be false, you need the antecedent to be true, and the consequent to be false. If a brick does not weigh 1/1000th of an ounce, then the antecedent is false, and the statement is therefore true. You didn’t falsify it at all.
Here is the most important point: if the antecedent is false, then the statement is true, even though the consequent can be false.
For example, the statement:
“If A and not A, then God exists” is a tautology.
Poisonheart you are engaginging in hair splitting. When I say “A brick” I mean it in the same way as if i say “A dog” or “A person”.
Bricks do not weigh 1/100th of an ounce. Houses do not weigh an ounce. I’m sure you realise that. Please stop splitting hairs.
Let me rephrase it in a manner that you won’t be able to split so easily.
I say “People have 10 legs each. That Unit has 10 members thefore the unit has 100 legs.”
You don’t need to actually count all the legs in the unit to prove that statement false. You can simply count the legs of one person.
Exact same deal.
I find it mildly annoying when people hair split in that way and demand that everything be phrased in an unreadable manner to stop them doing it. Surely you realised that bricks do not weigh 1/100th of an ounce?
The statement as an entity is false. Your rubbish about the antecedent only applies to artguments. This is not an argument we are discussing, it is a statement. Surely even you realise that any stement containing the phrase “houses weigh an ounce” is false.
Specifically, this is asynchronous tautological entailment.
I agree that this is falsifiable.
It is also not a tautology.
No. That is what I am trying to explain to you.
A compound proposition can be true even when the consequent is or could be false.
I gave one example:
“If A and not A, then God exists.” That is a true statement.
“If A and not A, then houses weigh an ounce” is also a true statement.
In fact, both are tautologies.
Think about it; how could either statement be false?
If either of those statements is false, then I am Mickey Mouse.
(also a true statement)