"Logical" objection to Empiricism?

[Speakpigeon]

A logical truth is a statement or logical expression which is necessarily true. That is to say, a logical truth could not possibly be false.

Or perhaps less metaphysically, there doesn't seem to be any conceivable logical case or situation in which we would assess the expression as being false.

Now, the mere existence of logical truths seems to be a problem for Empiricism...

Logical truths
https://en.wikipedia.org/wiki/Logical_truth

The existence of logical truths has been put forward by rationalist philosophers as an objection to empiricism because they hold that it is impossible to account for our knowledge of logical truths on empiricist grounds.

Any rational view or comment on this interesting piece of Wikied trivia?
EB

 

[Baldeee]

I think the empiricist would classify logical truths as analytic knowledge, as opposed to synthetic knowledge.
Analytic knowledge, the argument goes, merely analyses concepts but doesn't actually provide any additional knowledge.
Synthetic knowledge goes further than simply analysing concepts and does provide additional knowledge.

So a logical truth is simply an analysis of the relationships between concepts.
It only becomes knowledge, though, when those concepts are given meaning through synthetic knowledge.

Or something like that.

Further, the empiricist might argue that while logical truths and mathematics might become known as a result of reason, it at least starts with experience.
We know that 2+2=4, for example, because we first experience it and then analyse it to arrive at that truth.
We know valid logic because we first experience examples of it, then analyse.
To arrive at knowledge through reason alone, they might argue, requires ideas to be innate.
If so then all rational people should hold the same ideas, and acknowledge them.
Some members on this forum could be taken as adequate proof that we don't all hold such innate ideas, thus knowledge can not be achieved through reason alone.

Or something like that.

The ultimate claim of the empiricist is that where there is no experience there can be no knowledge.
We must have experience of relationships between things before we can even start to reason, etc.

 

[Speakpigeon]

I think the empiricist would classify logical truths as analytic knowledge, as opposed to synthetic knowledge.
Analytic knowledge, the argument goes, merely analyses concepts but doesn't actually provide any additional knowledge.
Synthetic knowledge goes further than simply analysing concepts and does provide additional knowledge.

So a logical truth is simply an analysis of the relationships between concepts.
It only becomes knowledge, though, when those concepts are given meaning through synthetic knowledge.

Or something like that.

Further, the empiricist might argue that while logical truths and mathematics might become known as a result of reason, it at least starts with experience.
We know that 2+2=4, for example, because we first experience it and then analyse it to arrive at that truth.
We know valid logic because we first experience examples of it, then analyse.
To arrive at knowledge through reason alone, they might argue, requires ideas to be innate.
If so then all rational people should hold the same ideas, and acknowledge them.
Some members on this forum could be taken as adequate proof that we don't all hold such innate ideas, thus knowledge can not be achieved through reason alone.

Or something like that.

The ultimate claim of the empiricist is that where there is no experience there can be no knowledge.
We must have experience of relationships between things before we can even start to reason, etc.

Sure, but.

As I understand it, that doesn't quite address the initial claim, except perhaps in the possible dismissal that knowing a logical truth is to know anything at all.

So, what would you say? Is a logical truth something that we can appropriately say that we know?

Because, if so, then the OP's claim still seems to stand.

What do you think?
EB

 

[sideshowbob]

What good is "truth" if it isn't empirical?

 

[Speakpigeon]

What good is "truth" if it isn't empirical?

I'm not quite sure what may be your point here in relation to the OP but I would assume it's very much common sense that logical truths are not only useful but very useful, and not just on occasions but pretty much throughout our lives from petty concerns about price and familial relations to highfalutin theoretical work, including in science. I believe there's just such an example given by another one of those grumpy posters in my first thread here, something about Quantum Physics. "Bell's Theorem" I think it's called.

Still, how does that relate to the OP exactly? Could you be asking me to convince you that this question somehow matters? That should do it, no?

Still, maybe you're absolutely right.

So, now, rather than me telling you what new angle follows from this very real possibility, how about you doing it? It's easy.

Your brain just has to apply a bit of logic for you, here. I even suspect it's just another one of those logical truths the OP is talking about.
EB

 

[iceaura]

Any rational view or comment on this interesting piece of Wikied trivia?

Since logical truths rest on assumptions - explicit ones - and these assumptions are by definition not logically deduced or established independently of empirical reality, no conflict or problem appears to intrude.

There seems to be no difficulty in knowing that if A then B. As a guide to research, it can be very useful knowledge.

 

[Sarkus]

Since logical truths rest on assumptions - explicit ones - and these assumptions are by definition not logically deduced or established independently of empirical reality, no conflict or problem appears to intrude.

What assumptions do you think logical truths rely on? As I understand them, they are necessarily true due to their logical components rather than the subject matter. E.g. No X is also not-X. It doesn't matter what you input for X, it will always be true due to the logic alone. E.g. "no unmarried man is also married" (or "no bachelor is married").
How does empiricism know that "No X is also not-X" is always true? Empiricism can certainly provide confirmation for given specific examples but how does it provide knowledge that it is always true?

 

[iceaura]

What assumptions do you think logical truths rely on? As I understand them, they are necessarily true due to their logical components rather than the subject matter

When one reasons from axioms, the axioms are assumed. Likewise premises, less fundamentally. Sometimes they are "self-evident" (i.e. evident to oneself), other times ad hoc or hypothetical, but no logical reasoning can begin without them. Syllogisms are built of premises - no premises, no syllogism. That's often the purpose of a chain of logical reasonings - one gains knowledge of the effects or consequences or implications of certain assumptions, by installing them as premises in syllogisms and inspecting the conclusions.

"A proof tells one where to concentrate doubt"

How does empiricism know that "No X is also not-X" is always true?

It isn't, necessarily. It is assumed, for the argument. So in your example: "If and only if No-X implies Not-X, then - - - - "

E.g. "no unmarried man is also married" (or "no bachelor is married").

An illustration of a common assumption, the law of the excluded middle. If you make it a matter of the definitions of the words, then it is not very useful - you gain no knowledge except of the proper deployment of your own vocabulary.

 

[Sarkus]

When one reasons from axioms, the axioms are assumed. Likewise premises, less fundamentally. Sometimes they are "self-evident" (i.e. evident to oneself), other times ad hoc or hypothetical, but no logical reasoning can begin without them. Syllogisms are built of premises - no premises, no syllogism. That's often the purpose of a chain of logical reasonings - one gains knowledge of the effects or consequences or implications of certain assumptions, by installing them as premises in syllogisms and inspecting the conclusions.

But the truth of the logical truth is not from premises, and the only axioms are the axioms that underpin the logic itself. A logical truth is necessarily true by dint of the logic alone. Only the specific examples, the confirmations, are from premises. The axioms themselves can not be known to be true empirically - only confirmed on a case by case basis.

It isn't, necessarily. It is assumed, for the argument. So in your example: "If and only if No-X implies Not-X, then - - - - "

You misunderstand... I didn't write the equivalent of "No-X = Not-X" but was rather stating the law of noncontradiction within classical logic... i.e. there is no X that is also not-X.
This can not be known empirically. Yes, it is an axiom of logic but that is what determines logical truths.

An illustration of a common assumption, the law of the excluded middle.

I know.

If you make it a matter of the definitions of the words, then it is not very useful - you gain no knowledge except of the proper deployment of your own vocabulary.

I agree. Being useful or not is not in question, though, but simply whether one can know a logical truth with reason alone.

 

[iceaura]

You misunderstand... I didn't write the equivalent of "No-X = Not-X" but was rather stating the law of noncontradiction within classical logic... i.e. there is no X that is also not-X.

Ok, then in your example: "If X and Not-X is (axiomatically) False, then - - - - - ".

I agree. Being useful or not is not in question, though, but simply whether one can know a logical truth with reason alone.

And the answer is yes, as you establish. I'm not seeing the problem - maybe it's here:

Only the specific examples, the confirmations, are from premises.

The premises are assumed. One reasons from them, and without them there is no logic involved. Logical truths are not "confirmed" by examples, but by valid reasoning from the premises, the assumptions.

 

[Speakpigeon]

Since logical truths rest on assumptions - explicit ones - and these assumptions are by definition not logically deduced or established independently of empirical reality, no conflict or problem appears to intrude.

There seems to be no difficulty in knowing that if A then B. As a guide to research, it can be very useful knowledge.

Nobody knows that if A then B since it's certainly not the case that if A then B unless you specified A and B.

I don't see that we necessarily use any assumption to recognise a logical truth but maybe I'm missing something here.

Well, now I hate to ask you for an example of that. The last time I offered you this danse it didn't go too well. So, feel free to decline.
EB

 

[sideshowbob]

I would assume it's very much common sense that logical truths are not only useful but very useful

Logical conclusions based on real-world observation are useful. We can use them to understand and manipulate the real world. But what good are "logical truths" that are not based on real-world observations? How can you even know they are "true"?

 

[Speakpigeon]

Being useful or not is not in question, though, but simply whether one can know a logical truth with reason alone.

Do we actually always need reason?

I take reason to be a thinking process which is essentially conscious, and at least minimally articulated and deliberate. Do we always need such to recognise a logical truth?
EB

 

[Speakpigeon]

Logical conclusions based on real-world observation are useful. We can use them to understand and manipulate the real world. But what good are "logical truths" that are not based on real-world observations? How can you even know they are "true"?

???

How would you get anyway near any useful logical conclusion based on real-world observations without using at least one logical truth?

If only the bridge can get you to the other side of the river I think it's fair to assess the bridge as very useful.
EB

 

[sideshowbob]

How would you get anyway near any useful logical conclusion based on real-world observations without using at least one logical truth?

What the @#$% is a "logical truth"? How do you know it's "true"?

If only the bridge can get you to the other side of the river I think it's fair to assess the bridge as very useful.

It's also fair to assess the bridge as empirical.

 

[Sarkus]

Do we actually always need reason?

I think it depends on what you define reason to be. To wit:

I take reason to be a thinking process which is essentially conscious, and at least minimally articulated and deliberate. Do we always need such to recognise a logical truth?

I am not sure we could recognise a logical truth without reason, as logic is purely a matter of reason... and if a logical truth is one that is entirely self-contained within that logic, how else, other than reason, would one establish that it is true, and thus know it?
A rationalist would say that you couldn't, that you would need reason. An empiricist, as Baldeee mentioned, would probably assert that any claimed knowledge based solely on reason is not adding anything new to what is provided by experience.

One could argue that we need experience to be able to apply meaning, and that without meaning we can not know anything, hence experience is the root of all knowledge. But I don't think we can know anything without there being some reason involved. If we take knowledge as a justified true belief, then the belief element surely contains some element of reasoning, even if justification is "I experienced it". If the justification is pure reasoning, then you're into knowledge from reason alone.

Unfortunately I'm not sure what school I tend toward, if either, so I'm very much trying to think things through as I respond, so please don't take anything I say as being a dogmatic insistence.

 

[Sarkus]

What the @#$% is a "logical truth"? How do you know it's "true"?

A logical truth is a truth that is true purely because of the logical form of the sentence, rather than relying on that which the logic is applied to.
E.g. a true statement would be that Pete Sampras won Wimbledon a few time.
However it is not a logical truth, because the form of the statement is "A won B a few times", and if we replace A and B with other things it may not be true. I certainly haven't won Wimbledon a few times, and that's just changing one of the items.

A logical truth is one that is true irrespective of A and B. So "all men are male" might be considered a logical truth. Since "men" could be equated to "male human" and thus you're basically saying that all "male humans are male". The form of this might be seen to be "something having the property of A is said to have the property of A".
This is now true regardless of what A might be - and is thus a logical truth. Note that "all male humans are male" is but one example of this form, and while not proving the logical truth at least confirms it (see earlier post on the distinction).

It's also fair to assess the bridge as empirical.

I rather see the sides your crossing from/to as being empirical. The bridge might be pure reason. But since the sides are already there, and known, the empiricist might say the bridge is simply helping you to what is already there, rather than reason creating the other side (new knowledge).

 

[Sarkus]

How would you get anyway near any useful logical conclusion based on real-world observations without using at least one logical truth?

But I think the question of "how do you know it's [the logical truth] is true?" is a good one. As iceaura argued, all logical truths are only as sound as the veracity of their axioms/premises. A logical truth is only true in so far as the truth-value is defined by the same logic. How do we know that the circle of logic is itself true, in an objective sense, rather than just true with reference to that which is calling it true? Can we know that? Or do we just make the assumption - even go so far as making it an axiom - that such logic produces truth in every instance?
I mentioned how empiricism can at least confirm the truth but cannot prove it - so how can we know it without proof? What can provide proof? Is it again reason alone? I wish I knew enough to know the answer.

 

[Speakpigeon]

What the @#$% is a "logical truth"? How do you know it's "true"?

You should have read the OP. All has already been revealed.

It's also fair to assess the bridge as empirical.

???

This was just a metaphor. That you can't get yourself to understand a straightforward metaphor suggests to me there's nothing much to talk about.
EB

 

[Speakpigeon]

I think it depends on what you define reason to be. To wit:
I am not sure we could recognise a logical truth without reason, as logic is purely a matter of reason... and if a logical truth is one that is entirely self-contained within that logic, how else, other than reason, would one establish that it is true, and thus know it?

I believe human beings have a sense of logic, with "sense" here to be taken literally, i.e. as an awareness, a native, intuitive understanding which is a basic capability of the mind or brain. So, I would assume that most people can recognise at least some logical truths simply by taking the time to consider the proposition and let their brain or mind do its magic. I know people are divided on this, and maybe it is a false opposition, but I don't understand how Aristotle could have provided a first formalisation of logic if he didn't have a sense of logic to tell him where he should find his footing. I personally haven't found anything wrong in Aristotle's logic and I have no good reason to think there may be something wrong. And nobody seemed to have improved on his work until modern logic at the end of the 19th century, 2,500 years later.

A rationalist would say that you couldn't, that you would need reason. An empiricist, as Baldeee mentioned, would probably assert that any claimed knowledge based solely on reason is not adding anything new to what is provided by experience.

I think that all reasonable people who are also pragmatic recognise not only the usefulness but the necessity of both reason and experience.

One could argue that we need experience to be able to apply meaning, and that without meaning we can not know anything, hence experience is the root of all knowledge. But I don't think we can know anything without there being some reason involved. If we take knowledge as a justified true belief, then the belief element surely contains some element of reasoning, even if justification is "I experienced it". If the justification is pure reasoning, then you're into knowledge from reason alone.

Personally, I don't take any belief, justified or not, to be actual knowledge. I can conceive, although I can't imagine, that a belief might well be true, but that still wouldn't amount to knowledge in my view. Still, I'm also pragmatic and recognise that you have to trust at least some of your beliefs as if they were true, which is exactly what I think most people do and, so far, it seems to have worked well enough on the whole minus a few casualties in the course of human history.

The question rather, is whether we know anything at all. Well, I would have said that we must know at least some logical truths, somewhat in the same way that we know redness or pain when we experience it. But I understand this may be controversial.

Unfortunately I'm not sure what school I tend toward, if either, so I'm very much trying to think things through as I respond, so please don't take anything I say as being a dogmatic insistence.

Not to worry, I have another "sense" beside logic telling me who's dogmatic and who isn't. What you've said so far hasn't triggered the damn thing.
EB