- “Myth is truth (p < .05)”
- “The functional neuroanatomy of science journalism”
- “Gay or straight, it’s decided at birth”
- Homosexual behavior due to … ?
- “New Scientist Needs a Reality Check”
- Civil Discussion
- Intelligent Design’s “Negative Science”
Language Log takes frequent note of strange things science journalists say. Their most recent is about the neuroscience of mothers watching children in distress. Here is part of what LL’s Mark Liberman’s had to say:
It’s rhetorically interesting that Ms. Parker-Pope takes the existence of brain differences observed by fMRI as evidence that the reactions in question are “hard-wired”, i.e. innate. No doubt the ability to recognize one’s children and the impulse to empathize with them have a substantial evolved biological substrate. But the fact that the psychological states in question are distinguishable in fMRI scans tells us nothing whatsoever about the balance between Nature and Nurture, in this case or in any other.
….
I guess that it’s the bizarre inference from observation in fMRI scans to innateness that makes this story at all newsworthy.
This is akin to the inference neuroscientists have made (examples here and here) that because they see no soul in their scans, therefore there is no such thing. (The Language Log posts notes later that the researchers themselves were partly guilty for the “bizarre inference.”) There’s an unjustified logical leap in both instances.
In the case of the soul, I suspect this reflects a bias that “if it isn’t science, it isn’t true,” or at least, “if it isn’t science, it isn’t knowledge.”
I thought it was more that because they don’t see one, or the action of one, that they couldn’t say one exists. There’s always a chance that new instruments or insights will show some “soul”-like component to the brain. It’s simply that so far, there has been no evidence supporting one, and no need to invoke one.
How about “If you can’t check it (ie. test it), then even if it is true, you can never know that”
Then–your final sentence–is there no knowledge except what can be scientifically verified? (I’m not sure I got your meaning exactly so I’m double-checking.)
Comment deleted: off topic
Great Site! have you read “Your Mind Matters” by John Stott? I just finished reading it not long ago and he backs up much of what you are saying. I have it listed on my “Current Reading List” page of my blog “An Uncommon Grace”. here is the link if you want to check it out:
http://anuncommongrace.wordpress.com/current-reading-list/
No, I’m saying that for something to be knowledge it must be able to be objectively tested - something that is true for you is not knowledge, it’s simply a personal feeling/belief.
The existence souls is a claim concerning reality. If that claim cannot be tested, then while it may be true (and many people are certain that it is), we can’t know it to be true - it isn’t knowledge.
Havok,
To have knowledge that this is true (rather than a personal feeling/belief), you must have objectively tested it. Did you?
SteveK,
Following that path we end up realising we can’t know anything.
Accepting non-objective (eg. personal revelation) information as knowledge leads, as far as I can see, to accepting absurdities such as the invisible pink unicorn as truth. So, though I believe the IPU exists and he is true for me, I wouldn’t expect others accept it without some kind of evidence which could be objectively tested.
I’d be interested in your thoughts, assuming you disagree, and weren’t simply being contrary.
Havok,
I’d be interested in getting your response to this. Moreland makes a good case for a different view of knowledge.
And to pick up from SteveK here: what is the objective test for your knowledge that there is no knowledge where there is no objective test for it?
That’s not a trick question. What Steve and I are getting at is that your position seems self-contradictory at the outset, that there is no objective test by which that statement’s proof can be established.
Objectivity itself is a very tricky term requiring definition. We got into a discussion here once with Paul (I think) saying that my knowledge that I am hungry is not objective because nobody else can confirm it. Nobody else knows what I’m thinking; does that mean I don’t know what I’m thinking?
This is some of what Moreland gets at in his short article.
Let’s not lose sight of the main point in the process: there are those who say that scientific knowledge is the only real knowledge, but this is a self-contradictory belief.
Edit: we cross-posted here. I wrote this without having seen your 6:50 comment. I’ll leave this for you to look at for now anyway.
Tom,
The article on knowledge is interesting. I’m just wondering if, as seems to be the case, I should accept his knowledge that God exists, whether he accepts my knowledge that the invisible pink unicorn exists, or a muslim’s knowledge concerning allah?
If you wouldn’t accept them, why not?
There’s this interesting paradox (maybe mystery is a better word) when it comes to knowledge. You have to know something about what you are looking for before you begin so that you will recognize it when you find it. Does that make sense?
SteveK, yeah it does.
I suppose I’m trying to find a method to identify method, which has few initial assumptions, and results in coherent knowledge which corresponds to reality.
The scientific methods seems to me to be the best method of achieving this we (as humans) currently have, though I’d be happy to be enlightened as to any other methods which can be shown to provide knowledge.
Regarding whether knowledge must be tested or verified, I’d like to discuss a specific example.
So, can someone state some knowledge about the outside world, beyond our minds, that hasn’t been tested and verified? I know I’ve qualified the question (I’m excluding qualia), but bear with me anyway.
Paul,
In the general sense, qualia encapsulates all perceptions that are untestable so if you exclude them then there is nothing left. If that is not what you intended then please clarify.
Havoc,
I think we are all trying to find this.
I think you’re throwing the baby out with the bath water. That might be a phrase known only in the USA, I don’t know. Said another way, your method filters out too much of reality. You said you wanted a method that results in coherent knowledge which corresponds to reality.
The scientific method (SM) is good, but it isn’t everything - not even close. I think you know that. If you depended solely on the SM for knowledge then every bit of perceived meaning would be stripped from your knowledgebase. That’s a MASSIVE part of your reality - and mine.
You could never say that you know the SM is important (!!), you know you love your wife, you know your thoughts** (!!!!), you know what hope is, you know what a book is about, etc.
It’s clear to me that knowledge requires much more than the SM. Can I know that is true without the aid of the SM? Yes! Logic dictates it.
** Required for the SM to even work. Think about that.
Hi Paul,
I think I can.
1) If all (outside world, beyond our minds) As are Bs and all Bs are Cs then all As are Cs.
2) Nothing can be both A and Not A in the same manner at the same time.
3) Human reason gives us an approximate if imperfect representation of the way the world is.
4) We are not brains in vats.
5) We our universe is not one of an infinite number of universes of varying physical laws such as satisfy the anthropic principle.
Charlie, I’m not sure where to take this, so allow me several approaches, I may not hold to one approach or the other, depending on where things wind up, I’m just thinking out loud.
I think we know the transitive property and the principle of non-contradiction (#1 and 2) by testing.
8 = 4+4 = 2+2+2+2.
If I put 2 apples four times into a bag, and put four apples twice into a bag, and have a bag with 8 apples, and then compare the number of apples in each bag, and I find it’s the same number, I have tested and verified the idea that if A = B (8 = 2+2+2+2) and if B = C (4+4=2+2+2+2), then A = C (8=2+2+2+2). If this didn’t work, the prinicple of A=B, B=C, so A=C wouldn’t hold. So the principle is dependent on that test, quite directly.
A similar test is possible for A cannot = not-A.
That human reason mirrors reality is founded on testing, too, Charlie’s #3 is not different in principle from his #1 and 2; #3 is just a generaly statement of the specific logical principles in 1 and 2.
I have no way of knowing that I’m not a brain in a vat. But it doesn’t much matter anyway, does it?
I don’t see how #5 isn’t founded on testing, if only through logic, but logic is founded on testing anyway, see above.
SteveK, are you saying that, aside from qualia, everything we know is tested and verified? That’s how I read your coment, and that idea is what I was trying to hypothesize.
Hi Paul,
I still disagree with you all-round.
Logic is not known by testing and verifying but by reason. If you now want to call reasoning testing and verifying then I see no point in your asking your question in the first place.
1) This truth is not known because of a test but because of logic and reason.
2) There is no such empirical, verifying test.
This statement is proven true by the fact that its opposite is self-refuting. One cannot truthfully say that human reason does not give a representation of the universe: if the statement is true, it is false.
You do have a way of knowing. If you were a brain in a vat you could not say “I am a brain in a vat” and have that statement be true.
If you were a brain in a vat your Reason would not be a reflection of reality, but it is, because the opposite is self-refuting.
It is tested by logic. If your claim is now that we can know things tested by logic and reason rather than observation, weighing and measuring then welcome to the club.
The principle is in no way dependent upon that test. You may perform that test if you like, for whatever reason, but it does not verify the principle. Putting apples in bags doesn’t even tell you what 2, 4 or 8 are or mean. I could say kuka apples plus kuka apples equals zipo apples. This proves neither logic or math.
However, I can use such imaginary terms for truth #1 and still have knowledge based upon logic. If all kukas are zipos and all zipos are tangs then all kukas are tangs. You can not test this (you couldn’t test it in the A, B, and C example either) but it is true and known nonetheless.
Since logic and reason are synonyms, you’re saying that logic is known by logic?
I showed you a specific example of how logic is verified by observation. I’ll repeat: if the observations I laid out didn’t verify the logic, we wouldn’t believe in the logic, so the logic is directly dependent on those observations.
The rest of your critiques all rest on the above point, I think.
But there’s no way I can tell if that statement (”I am a brain in a vat.”) is true or not. So I still don’t have a way of knowing.
You can tell.
1) If you are a brain in a vat then what you are calling a brain has no reference, nor does a vat. You cannot be that which you are talking about because that which you are talking about doesn’t exist to you.
2) If you are a brain in a vat your thoughts do not match to the outside world. But your thoughts do match (to a degree) to the outside world.
No you didn’t. Show how observation verifies the true point, that if all As are Bs and all Bs are Cs then all As are Cs.
Show me with math with apple-bags that this string does not entail that all Cs are As.
Yes, that was a sloppy sentence (but your supposed rebuttal is false, not all reasoning is logical). Logic is not known at all, is it? The validity of arguments and the consistency of propositions are known implicitly through reason, or, as you like, explicitly through logic, and are not known through empiricism, observation, verification, etc. (unless you are going to continue to shoot your thesis full of holes and insist that this very process is what you mean by “verification”.
So, what method(s) do you use to provide a coherent picture of reality?
What do you mean by perceived meaning? If you mean my perceptions, you’re wrong. I’m simply taking about things which are accepted as true. If I think purple jeans look good (a perception) I wouldn’t expect that to be some kind of truth, accepted by anyone. Is that about what you mean by perceived meaning?
I know I love my wife because I feel it, but wouldn’t expect anyone else to accept that assertion as truth. I know my wife loves me due to her behaviour - I guess you could say that the love is tested through behaviour. All of the things you list can be tested in some fashion (apart from my own thoughts and feelings, but I wouldn’t expect someone to take those as being truth without some further verification - would you?)
So, how do you know truth without being able to verify it in some manner?
I might be slow on the uptake, but it seems you’re said that such a manner exists, but you haven’t mentioned what it is.
Charlie,
You understand that the “brain in a vat scenario” entails a matri like situation, where all of your sensory inputs are generated right? It seems, from you’re points, your thinking of an isolated brain in a vat.
There are very many variations of the brain in a vat scenario which far predate the Matrix. I prefer the one where the brains are natural elements of the universe, uncaused and uncreated by anything.
Nonetheless, the position holds. If you are in the matrix what you think is a brain, or a vat, is nothing of the kind. It is a representative symbol. As you said, all your inputs are generated. Therefore, there is no “real” brain correlated with what your input perceives of as a brain, and you cannot be that thing.
Charlie,
I think the arguments generally have all of the inputs simulated, but to be a flawless simulation. In such an event, how can you know that you’re a brain in a vat, when what you experience is a coherent reality? I think that is the basic question, and the general answer is, you cant. Have I missed something?
It is a question of causation between the thing known/described and the thing itself. No matter the complexity of the inputs they are not the thing itself. The thing called a brain by the brain in a vat is not a brain. Therefore, the brain in the vat cannot be a brain in a vat.
What do you mean here? If we replicate the nerve signals to the brain, such that there is no different between the signal for say holding an apple, and the simulated signals for such an act, how can any difference be distinguished by the brain (and by extension the mind)?
The difference can’t be distinguished by the mind because there are no two things between which to differentiate. There is only what the brain in the vat thinks brain is, but it isn’t that thing. The signal may be the same, or it may not, as the brain wouldn’t know, but in the one case there is no brain causing the signal. The thing causing the signal is not a brain. But it is the thing the brain in the vat would refer to as the brain, were it to contemplate its vatness. Obviously, if it were a brain in a vat it wouldn’t be the brain it knows as a brain.
For that matter, neither does the vat exist.
Ah, so we agree on this. That’s something.
I’m not sure what you’re saying here?
Are you saying that because the simulated reality being fed to the brain is indistinguishable from what the “actual” reality would be, that as far as the brain in the vat is concerned, there is no vat, and that brain isn’t in it?
If so, then we’ve agreed all this time - The brain in the vat cannot tell that it is a brain in a vat - though someone external to the vat could.
Is that what you meant?
No, I am not saying that that the simulated reality is indistinguishable from what the actual reality is. I’m saying we are talking about two very different realities, and that one is completely disconnected from the other.
No, it’s not that the brain in the vat can’t tell its a brain in a vat. It’s that the very proposition “I’m a brain in a vat” makes mincemeat of the words. They do not correlate to a reality in which there could be a brain in a vat.
You can do this by presuming that you are the brain in the vat. What would it mean for you to say “I’m a brain in a vat” and have that statement be true? For starters, you wouldn’t know what a brain was.
Then I’m not sure what you’re saying. How are they distinguishable?
I thought we were talking about a brain/mind being fed simulated sensory information, and whether there was a way for the brain/min to distinguish that the sensory information was simulated?
Again, you’ve lost me. The brain in the vat could think it’s a brain in a vat - it may find itself, interacting in it’s simulated world on a blog much like this.
Again, you’ve lost me. I can, sitting here at my keboard, imagine that everything around me was simulated, ala the matrix. In such a situation, why would I not be able to conceive of what a brain was - I’d be/have one, for a start. Am I missing something?
You are missing something. Sitting at your computer, imagining that you are a brain in a vat, you are not at a computer or being or having a brain. You are being a sim having a sim. Your experience of a brain is that of a signal, a simulated program, not a brain. But that is not what you would be if you were, in fact, a brain in a vat. The two cannot be the same and cannot refer to the same thing.
Now I have no other way of saying this so I’m going to have to let it rest there for now.
But even if you can’t buy my argument or accept it, or even understand what I am trying to say, we have a simple fact here: I know I’m not a brian in a vat (and you do, too) and I have no way of testing this or verifying it.
I actually know I’m not a brain in a vat, as well.
You, on the other hand, might not be sure that I’m not really Brian - pseudonyms what they are, and all.
Wow–busy night!
Paul, you said we know the law of non-contradiction by testing. Not true, not true at all. I’ll show it to you quite easily. Let’s take some things that we might think we know by testing:
“This banana cannot be a BMW.”
“The sun cannot be an oak tree.”
“Odd integers cannot be even integers.”
How have we observationally tested these? By induction, I suppose: we’ve never seen a banana that could be a BMW, or the sun vacationing as an oak tree. All the odd integers we’ve seen have remained stubbornly non-even. (Recall that we’re talking about empirical testing, not logical testing, because Paul said he thinks we can establish non-contradiction by experience rather than by logic.)
Let’s take this to the most generous possible extreme, and suppose that we repeat this kind of observation to the furthest reaches of reality. We observe every A and every non-A*, and we note that they are not equal. Have we proved the law of noncontradiction?
At this point I would stand up and say, “No! Absolutely not! The law of noncontradiction is a pile of hooey!” And you would say, “but we’ve tested everything in the universe and we’ve shown that A is always unequal to not-A.” To which I could reply, “No you haven’t;” and then you would probably say, “But of course we have!”
That’s when I would say, “Are you contradicting me? What if we’re both right–just this one time? What if this is the one exception you’ve been searching for all this time? How do you know that it isn’t?”
I think the answer is that you just know, as Charlie said above. The infinite search was unnecessary from the start.
There are similar problems with your claim that we know the transitive property by testing, but I won’t go into them.
(*There’s also a question-begging entailed in “observing every A and every non-A” during an empirical test for the truth of non-contradiction; but I thought this other line of argument was more interesting.)
Regarding the brain in a vat: have you read this Moreland piece yet?
Tom, one point before I get to your real point. I didn’t claim to prove absolutely that the law of non-contradiction worked, I only set out to check it out by a certain method (observationally). That is, proving it absolutely concerns the *extent* to which the law holds (100% of the time, or maybe only 90$ of the time, etc.); and seeing to what extent it does hold concerns the *method* by which check it out.
I don’t have to claim that the law of contradiction holds 100% of the time for all time and throughout the universe in order to observationally verify it to some extent.
OK. What you’re doing is assuming that the law of non-contradiction is a thing out in the world like an apple is, whose existence is subject to verification just like an apple. But that is not necessary for my argument. The law of non-contradiction can be viewed merely as a summary of the results when we test and verify claims about apples, atoms, ideas, etc. It does not need to be a conclusion, but merely a convenient way of talking about many conclusions. So to turn the law upon itself introduces an unnecessary category mistake.
Tom, I missed the part of the Moreland piece that related to the brain in the vat. What part did you think related?
Hi Paul,
No you didn’t. You set out to demonstrate that we can only know the truth of the law of contradiction because we can test/verify it. This is false.
It holds 100% and verification and testing can’t tell you this. More knowledge, unverified.
And unicorns can be viewed as a convenient way to discuss horses without horns and I can say that I’ve seen scores of them. You can’t redefine terms and beg the question to make your point. The law of non-contradiction is not the summary of results of tests. It is a law of proper reasoning.
Havoc,
No. If the SM is the only path to knowledge then you don’t know you love your wife. The SM requires objective, external verification that can be confirmed by anyone. You admitted this can’t be done when you said that others aren’t expected to agree with you.
How does the SM lead you from not knowing what love is, to knowing what love is? I’m curious to read your answer.
Paul,
I was going to leave this for a response to your next response, but I have to leave momentarily.
You keep saying things like “we can test the law of non-contradiction” by observation and verification. But you can’t even use tests of observation and verification without the law of non-contradiction. Tom pointed this out and you changed his point. How can you claim a result holds if that result could be something else?
“Look, I have concluded four apples!”.
“No, you have conclude three bananas”.
“No, look, count, these are four apples.”
“I am lookjng and counting. They are three bananas.”
Nature provides us countless examples of apparent contradictions to our knowledge and our model of the world. What do we do when we find these apparent contradictions? We theorize and experiment to change our way of looking at nature, to amend our knowledge. What we don’t do is say “well, that’s no problem, because we can just scrap the law of non-contradiction”. The apparent violation must be explained so that it is not a violation, but the law is not changed.
Paul,
The parts about
1) The anxious quest for certainty: even though there’s a trace possibility that you are a brain in a vat, you have sufficient information to be able to know that you are not.
2) Knowing without necessarily knowing how you know: you don’t have to be able to describe the test you’ve performed that proves you’re not a brain in a vat. You can know it without having to explain how you know.
Further on Charlie’s last comment. Paul, you wrote yesterday,
That was after Charlie had responded to your question,’
Now you’re saying this:
If the LNC does not hold 100% of the time it is not the LNC, it is something else entirely. The LNC is an absolute statement regarding reality.
I don’t see how I made that assumption; I just gave us a thought experiment in which “you” (as I put it, the hypothetical experimenter) summarized your results when you tested and verified claims about apples, atoms, ideas, etc., and then I showed you in the end that your summary was invalid.
I didn’t draw the argument out in complete detail. I’ll try to do that now. In order to summarize your findings, you have to assume the LNC is true; in fact, you have to know the LNC is true. You can’t even undertake the procedure you have described without the LNC’s being true.
To extend the example: I could go back through your list of findings and disagree with every other finding. I could say you were just wrong when you determined that apples are not bananas, and that ancient Greece was not south of ancient poetry, and that you were never Napoleon. In order for you to say “hang on a minute, I wasn’t wrong!” you would have to assume the LNC. If I say the items in every other case were identical, and you say they weren’t, you are contradicting me. In order for your contradiction to have any force, the LNC must be true.
In the end, I could just say, “you haven’t proved a thing, in fact, you haven’t even demonstrated the LNC works more than half the time.” (I could actually dispute every instance, not just every other one.) So how will you show me that you actually have demonstrated the LNC in action, unless we agree the LNC rules our discussion from the start? So you have to use the LNC as part of your method for your inductive investigation; otherwise you end up with just disputable findings.
Here’s an alternate version: suppose I follow you through your process from the beginning, and we do this:
You: Apples are not bananas.
Me: Yes they are.
You: Houses are not earthworms.
Me: Yes they are.
You: The sky is not my daughter.
Me: Yes it is.
You: You know, all your contradictions are getting annoying!
Me: I’m not contradicting you!
Finally: imagine that instead of this being you and me, imagine it being you and you. Without the LNC that would be entirely possible.
I’ve probably gone on far too long with this and over-stated my point, but it was kind of a fun exercise, so I hope you’ll forgive me.
Charlie wrote:
1. How do we know this?
2. How do we know it stands up to Tom’s March 15th, 2008, 6:54 am challenge.
3.
Says you. ; )
Then I don’t believe in it as a law, but merely as a summary. I may, however, believe in it as a law when I see your answer to my #1 and 2 above.
Regarding if there are 3 or 4 bananas, (1) the conflict between how many bananas are in the bag is a problem with perception (the mere counting of how many items are in a bag). I counted 3 bananas, you counted 4 bananas, in another bag or not.
The mere fact that people can disagree about perceptions doesn’t invalidate using observation as an empirical means. It doesn’t have to work perfectly 100% of the time for all people in order to work well enough.
Tom, I’ll get back to your last post soon.
Perception is not an escape from the law of non-contradiction.
“Paul perceives 3 bananas” and “Charlie perceives 4 bananas” do not contradict each other.
Tom, I haven’t forgotten about my response to your last post. But what I have to do to reply is to summarize your argument, which will take a bit of time for me.
But, in the meantime, can you tell me how we know that the law of contradiction is true, 100%?
Tom: As I summarize the discussion, I have a question about your argument. How much does it rely on your idea that my observations that verify the transitive property rely on induction? I don’t see my idea as relying on induction. I only meant it to be an example of one particular case, and whether we can apply induction to other cases is a separate point. That is, for any one specific set of bags of apples, the foundation on which we can apply the transitive property is the observation that, for those bags, they have the same number of apples. Induction isn’t needed here, only observing 3 particular bags.
Also, I think we need to be clear as to whether we’re talking about the transitive property or the LNC.
I’ve been talking about the LNC, with just a brief and minor reference to the transitive property.
Hi Paul,
I’d like to see some responses to some issues that you’ve left on the table.
Can you show me the test please? Show me, without assuming the LNC, that A is not not-A.
Have you decided that testing with logic is “testing and verifying”, as per your original challenge? Have you determined that we can know that which reason and logic tell us?
Can you show me the test for this?
Or this?
Can you address this?
I only ask because if these keep piling up I’m afraid you will become overwhelmed and have to leave the conversation.
To do my part, on my assertion that the LNC holds 100% of the time, you asked:
Because we know that we can reason and reasoning demands it. As we’ve discussed before, once you doubt reason you have nothing left. If you put yourself outside of reason there is no non-question-begging way back in. If we are to know anything then we know that valid arguments coupled to true premises cannot but give true conclusions. If we don’t know this then there is no point doing a single one of your experiments, as you can’t demonstrate or conclude anything from them (think 4 apples v. 3 bananas) and you can’t even know anything - but you do think we can learn from experiments - therefore, you know the LNC holds. One would then necessarily become the radical skeptic of a couple of centuries ago and would have to retreat to such thought experiments as the possibility of vatness as evidence and argument.
New question:
A summary of what, though? And signifying what?
As for apples and bananas, as Tom said, it’s not a matter of perception. If the LNC doesn’t hold then 4 apples=3 bananas cannot be said to be wrong.
Charlie, we *assume,* a priori, that we must accept reason and rationality, right? Why should we otherwise trust in reason? Because without it, all is lost, including discourse, communication, etc. Wouldn’t you agree?
If reason and rationality are assumptions, then, they are not things that we *know,* but things we assume. I wouldn’t call what we assume or take for granted knowledge, which was the original question.
If they are not necessary starting points, then how do we know that reason must be accepted without using reason in our justification? I don’t think I’m moving the goal post here, because the LNC and the transitive property are merely details of reason and rationality, it turns out to be the same issue, just in more detail.
Your argument an exercise in question-begging.
You say “what can we know without verification?” and I say that we can know that A=C.
We see that is a product of logical analysis.
And then you contend that such results are not “known” because our trust in logical analysis is axiomatic - assumed.
Even if we accept your premise about reason your conclusion is a non-sequitur, but worse yet, look what it does to your criteria for “knowing”. If the fact that reason is accepted a priori as a starting point invalidates our claims to knowledge based on it then you have defeated all knowledge. That includes knowledge gleaned from experiments, observations and verifications.
Your skating around the issue is doing no good but further entrenching you in your radical skepticism and bringing you within a hair’s breadth of solipsism.
You are painting yourself into the same post-modernist corner that had us asking previous po-mo participants why they even bother to try to communicate if they can’t know anything.
I do wish you’d answer those points in my previous comment. I know how you hate when they pile up and become too much to handle - procedurally, I mean, not substantively.
Charlie, I’ll do my best to hit as many of your questions as I can tonight.
First: My hypothesis is that, like science, all knowledge is tentative and not absolute. My empiricism says that we first must assume rationality, and then conclusions based on observation follow, just like proofs of geometry follow from postulates. It doesn’t destroy knowledge, it only makes it useful, not some absolute.
Here’s my observational test that A is not not-A. Let A = an apple. Now, in a bag I have an apple. I call it A. In another bag there is a banana. Is it an apple? No, it’s not. So it’s not-A. I then check to see if what is in the bags are the same. I look, and they are different. So bag A is not bag B, so A is not not-A.
The spirit of this test is a bit along the lines of the operation in mathematics and set theory by which the true way in which we know that two sets have the same number of elements is not to count each and compare the abstract numbers, but to pair up each element in both sets, and if there are no members left over, then the sets are equal. Are you familiar with this operation? If so, it might place my attempt here in a more valid context.
I was not trying to use the words testing and verifying with the breadth that you not unreasonably took them. I meant to limit the testing and verifying to empirical observation. But, honestly, I’m not sure about the implications of this, I don’t know where you’re going, so I reserve the right to change my ideas if a certain context you might identify demands it.
For testing Kukas and Zipos, I’d say that the reason that we can know someting about them is only because we’ve found out certain things about apples and bananas and then generalize and abstract and summarize those observations, but we don’t need to elevate those summaries to a higher ontological status (like the LNC exists somewhere, perhaps Platonically).
At least I tried, gotta run now.
Paul, you’re fudging:
Is it knowledge or is it not? We know it’s based upon our axiomatic acceptance of reason, so what does that say about your previous point?
Thanks for your attempt, but you missed this one again:
—-
Yes, it is. Apples can be bananas.
But the banana is not not an apple, bag A is bag B, and A is not A. There’s no reason why not - especially since your reasoning here is supported only by a presumption and the LNC is only a summary.
That’s exactly how I took it. You’ll notice the countless times I said “empirical, observational”, etc.
I only came to suspect otherwise when you said “we tested it, even if only through logic”.
We already saw the context that caused you to change from empirical testing to “testing” logically. (Paul March 14th, 2008 at 10:19 pm)
At the time it looked like you were trying to subsume all knowledge under your “testing” criteria, just as you have previously with statistics, and “verificationism” …
Now maybe that’s not necessary as it looks as though your ideas will cause all knowledge to slip away.
Oh yes, and remember this one:
Since all As are Bs and all Bs are Cs then all As are Cs.
You tried to resort to apple math bags to show an equality.
Now, use your apple math bags to show that this relationship does not show the opposite equality, ie: not all Cs are As.
Do you see the equivocation in your answer?
Sounds like “Is it moral or not.”
Then so is Karl Popper.
Remind me to never ask you make a banana split for me.
Can you support that via observation? If so, you should increase the dosage on your meds.
Sorry, it’s getting late, both in the evening and in the conversation.
Here’s a rundown on a debate Bill Craig recently had in Vancouver.
Some of you will notice some trends replaying here. Some of you won’t.
http://www.reasonablefaith.org/site/News2?page=NewsArticle&id=6097
Hi Paul,
Sounds like not an answer to me.
Is it knowledge or not?
That’s a popular name with you. What is it supposed to mean here?
Sure. All I have to do is look and say so. That’s what observation gets you absent logic. Especially when, as you’ve insisted before, we can’t trust our own perceptions (that applies to you as well, you know) without getting somebody’s agreement and you admit that your perceptions might just as well be fed to a brain in a vat - in which case bag A is exactly what you think it is (and exactly what I think it is) and nothing more. And, of course, anybody verifying it might just be a sim fed to a vat-brain.
Sounds like you’re about to run short of time again without facing the implications of your thoughts. More later, I suppose?
Hi Paul,
While I wait on your thoughts about Popper, here’s the name of another late-great philosopher of the 20th century with whom I imagine you share some sympathies: Bertrand Russell.
He had this to say on empiricism and a priori truths:
Charlie, did you miss it when I said earlier that logic was an assumption? The issue is now whether an assumption is knowledge. In your quote of Russell above, he notes that logic must be self-evident, which is an assumption, a necessary starting point.
Charlie, your last paragraph in your March 15th, 2008 at 6:55 pm comment only pointed out the necessity of logic in order to talk at all, with which I agree, but necessity is not verification. Even if we absolutely need logic and reason, we adopt them on as assumption, not something proven, either by logic (which would be circular), and the only thing left is by observation, and if you’re willing to adopt that stance (thought I doubt it), then we’d really have no argument.
Your last paragraph I mentioned above did not answer my question “HOW do we know that logic is true 100%,” it only spoke to the need for it. So I’m willing to be convinced and retract my position here if you can answer that question beyond anything that reduces to positioning logic as an assumption.
Then, we’ll talk about whether an assumption is knowledge. I’ll grant you right off the bat, though, that conclusions, given a set of assumptions (as is Euclidean geometry) are a type of knowledge, so if our disagreement hinges on me missiing that point, I plead guilty and apologize.
Hi Paul,
I just erased my response to you, so my rebuilding of it may miss something …
Please don’t ask pretend questions. A cursory glance at my responses will show many instances where I most evidently did not miss that you said logic was an assumption.
But Russell does not call logic self-evident.
But it is not “logic” which is in question. Your pedantry busted me on this mistake earlier and now your last defence requires the equivocation. My answer was that we can know that A is not not A, and that (through transitive relations) A=C. We know these things: we do not “know” Logic. These are the propositions which, as Russell says, are affirmed [or] obtained from particular premises by means of logical principle. And this is knowledge.
Your last defence against this was that because we assumed Reason or the truth of logic, that these propositions were not, then, knowledge. But that means that nothing is knowledge, nothing can be known. Which led to your invoking the name of Karl Popper.
As you can see from above, the question isn’t about whether or not that which you say must be assumed is known: “Logic” is not the thing “known” in the examples.
So my repeated question “but is it knowledge?” seems to be answered (yes, with your usual qualifier). Your qualifier, of course, applies to all knowledge given that everything depends upon logic and reason, as you’ve stated above.
So now, I’m sure you’re satisfied that your original question is answered - and has been from the start:
Charlie,
I skimmed over a lot of your conversation with Paul. I think I’ve grasped the main points, but I’d appreciate you (and Paul) summarizing in a short paragraph….for posterity’s sake.
Are assumptions knowledge? If they’re solid, then why not?
As to the empirical test for the law of non-contradiction, I’m surprised this is still alive. Here’s another version of the refutation:
You begin your observations, with a non-committal attitude toward the LNC because you are trying to determine empirically whether it’s true or not. Your test begins as follows, which I’ve quoted from you above, except for changing the first person pronouns to second person:
In a bag you have an apple. You call it A. In another bag there is a banana. Is it an apple? No, it’s not. So it’s not-A. You then check to see if what is in the bags are the same. You look, and they are different. So bag A is not bag B, so A is not not-A.
At this point I speak up, and this conversation ensues:
Tom: “The object in Bag B is an apple.”
Paul: “You’re wrong, Tom, it’s a banana.”
Tom: “Well, of course it is.”
Paul: I see we agree now. Thank you.”
Tom: “But just because it’s an apple doesn’t mean it can’t be a banana!”
Paul: “But apples and bananas are not the same,”
Tom: “Right, right right; absolutely right! But just because they’re not the same doesn’t mean they’re not the same, does it?”
Paul: “But look–I have an apple here, and a banana here–”
Tom: “Sure! That item in bag B is an apple, and that item in bag A is a banana. And the one in bag A is an apple and the one in bag B is a banana. And of course they’re completely different from each other in every conceivable way, and they’re absolutely identical in every way besides.”
Paul: “That’s impossible, Tom.”
Tom: “On what basis is it impossible?”
Paul: “Well, you’re contradicting yourself every time you open your mouth!”
Tom? “So what? You say you have an empirical instance in your two bags that demonstrates the LNC is true. I’m giving you an empirical instance out of my mouth that it isn’t. You don’t have a way to decide which to rely on–not unless you know in advance that the LNC is true.”
On the very first step of your empirical inquiry into the LNC, you rely on the LNC. Without it you cannot disagree with my conclusion that an apple is a banana even while apples and bananas are different kinds of things.
In short, you cannot demonstrate the LNC empirically without assuming, before you begin, that it is true.
Again, this is knowledge. You have not, and cannot, make an empirical case for the LNC, but you know it anyway…
Hi Steve,
I think I will give a short summary in a little bit.
Since I think it would be self serving, however, I think I’d prefer to give Paul the first crack at describing our short journey.
Are you game, Paul?
Argh, the response I tried to post earlier today never made it. Here’s what I remember:
Tom, how do we know the LNC is valid? Charlie’s answer was that it was necessary, but that’s not an argument for validity. If it’s an assumption, what makes an assumption solid (=valid)?
Also, Tom, I see your point about the observation of the LNC. And I had some really great point to make about it in response that has escaped into the ether. Must not have been that good, eh?
Not that that was exactly my defence, Paul, but wouldn’t you say that the fact that we could know nothing without the LNC is a pretty good argument for its validity?
Can we know anything?
If not, why bother asking us about any kind of knowledge at all? Why waste anybody’s time arguing on a blog about anything?
If so, you know the LNC is true - and you didn’t test it empirically in any way.
And isn’t that the point, in any case? That we know it without verification? Even you know that a thing cannot both be A and not A at the same time. You know this, oyu rely upon it, you base arguments upon it, you propose experiments requiring it, etc.. And you can’t observe, verify or prove it. That is the point.
Any chance you’ll accommodate Steve’s request for a summary?
I’d be interested to know where you think we are and where you stand now.
http://en.wikipedia.org/wiki/Law_of_noncontradiction#Alleged_impossibility_of_its_proof_or_denial
According to the link, there have been logics proposed which weaken or deny the law. This formal logic stuff isn’t my forte, but if it’s the case that systems of logic exist in which the law does not hold, then Paul is right in saying the LNC is an assumption - we base our knowledge upon the assumtion of the law holding. Is this incorrect?
Charlie,
I haven’t read the WLC article yet, but will get around to it.
Given this statement from WLC:
“We’ve already said that it’s the Holy Spirit who gives us the ultimate assurance of Christianity’s truth. Therefore, the only role left for argument and evidence to play is a subsidiary role.”
Why should he be taken seriously, when regardless of the evidence shown him to the contrary, he would still take the testimony of the Holy Spirit (whatever that is) to be the “truth”.
To me that seriously impacts a persons credibility.
Hi Havoc,
Your comments seem less and less about being real here.
We can trust Craig’s arguments when they are logically sound, contain true premises, comport with fact, etc.
If you’d been kind enough to supply a link with that damning quote I’d show you what Craig meant, why the context is important and why a man whose life is about evidence and positive argumentation ought to be taken seriously. I’d show you why your conclusion “regardless of the evidence…” does not follow from his quote.
Well-Googled, Havoc,
I’ve read your wiki link previously.
Here’s a note from a few paragraphs above your link:
Done and done.
Charlie, if you can offer no reason why the LNC is true other than the alternative is unacceptable, for whatever reason, then we don’t know it’s true, we only have to proceed as if it were. It’s hard to imagine how it could be false, but that’s not a proof or a reason for validity. Part of the problem here is that we’re trying to use logic and reason to justify the LNC, but the LNC is part of logic and reason. Of course I use the LNC in my everyday life, but that’s not the same as providing a rigorous reason for why it’s true. I often act like Newtonian physics is says everything about physics, but we know since Einstein that that’s not true.
That’s why I say we don’t know that the LNC is true, but we have to assume it.
I think we’ve come to the heart of the matter. Charlie, it seems to me that you’re willing to claim that something is true merely because the alternative is unacceptable. That’s a position I won’t adopt.
Then it’s settled…we don’t know if ANYTHING is true. So much for all that talk about verification and the scientific method.
Thanks, Steve!
Thanks, Paul, for affirming what I’ve been asking all along.
So, once again, bearing Steve’s comment in mind, what was the point of your initial question?
“Won’t”. That is the key. You make a conscious choice to maintain this (faux) radical-skepticism. It is not that the alternative is unacceptable, but that it is impossible if we are to know anything. If we are not to know anything then why, why, why do you keep posting?
The fact that I can know something, and I know I know something, tells me that I know the LNC is true.
You don’t (today, and for whatever reason).
Paul,
When you feel hungry do you know you are actually hungry, or do you just assume that you are?
Do you know anything about the assumption you’ve made? Do you know that you just made an assumption?
We don’t know anything in geometry, either, Euclidean or not, until we start with some postulates. Then, given those postulates, we can know some things about that geometry, but even than that knowledge rests on the postulates, and we can’t say that we *know* the postulates; rather, we just postulate (assume) them, and then make conclusions.
Remember, too, that my original question was what *knowledge* do we have that isn’t observed, verified, etc. We may have the LNC as a necessary assumption, but we don’t know it. Any geometry, Euclidean or not, requires postulates, but we don’t say that we *know* the postulates; rather, we postulate them, or, we assume them. *Then* we can begin to know some thing about that geometry, *give* the postulate.
A postulate or an assumption is not knowledge.
Also, my original question was limited to the outside world, and not about qualia. We have direct knowledge of qualia (I think).
Hi Paul,
You are presenting geometry as though it is not true but merely consistent within arbitrary bounds. This seems to me the equivalent to saying you know you can’t take three steps carrying a basketball because of the rules of basketball.
But this makes no sense to me. Even if you say we don’t know the postulate (which does not follow just because we can’t prove the postulate, we can still know that knowledge built upon these is true. I say it is objectively true that two line segments perpendicular to a single line segment are parallel to each other. I say you can know that the square of the hypotenuse of a right triangle is the sum of the squares of the other two sides.
Can you please tell me a postulate such as you describe and show that the knowledge thus derived is not actually knowledge?
And then can you show how this means that knowledge about the real world derived through logical reasoning is likewise not knowledge?
Thanks.
Oh, and finally -
Can you tell me how you can have knowledge about whether or not Mary showered yesterday, given that you can not observe, test or verify the event but based upon the statistical finding that people shower 9 days out of 10? It would seem that the information rests on a foundation far less secure than that of logical necessity or geometric postulates. So how was it you could contend that by “knowing” a statistical likelihood we had “knowledge” about a particular event?
Which geometry are you talking about? Euclidean? Or one of the others? http://en.wikipedia.org/wiki/Non-Euclidean_geometry
As I said, knowledge built upon them is true, *given* the postulate. Two straight lines *do* meet in some geometries, and they don’t in others.
I think this is not true in some geometries.
I’m fairly sure that this is not true in some geometries.
If we accept that knowledge must be based on at least one assumption (the LNC), and maybe others, then we can know lots of things with that qualified definition of knowledge, but you can’t then say that the LNC or those other assumptions is/are something we know without observation because we’ve already called it a necessary assumption.
So we need to keep the following straight:
Postulates: like the LNC and maybe some others, these are necessities in order to even discuss anything;
Knowledge: a conclusion based on observation and verification and also relies on some necessary postulates or a few assumptions;
Qualia: (I’m not sure I can define qualia strictly here).
Lastly:
We wouldn’t have knowledge as to whether Mary, in particular, showered yesterday, beyond the statistical, including when we talk to Mary tomorrow and she tells us she showered, we have statistics built from our past experience with Mary when we have directly observed her showering (”Oops, I didn’t think anyone was in the shower!”) or based on how often she lies to us.
Paul,
So far you don’t *know* anything (trust me, I know). Everything is based on assumptions that you have no real knowledge of. Your knowledge of the assumption is not real knowledge - it’s an assumption that you may or may not (or both) have had, which was based on reasoning or sensory input that you may or may not (or both) have had.
To tweak a popular phrase: it’s ‘assumptions all the way down’ with no real knowledge whatsoever….and you can’t even *know* that.
To repeat what Charlie said, if we are not to know anything then why, why, why do you keep posting?
EDIT:
If you change your mind and conclude that you really *know* (not assume) you made an assumption then we can talk.
SteveK, your characterization of my idea ignores what I’ve already said about the parallel to geometry, so why won’t you acknowledge that in your posts?
Would you say that we don’t know anything about geometry, too, even though that is exlicitly based on postulates?
HI Paul,
I’m on my way out the door so I’ll have to hit most of your points later.
However, on “which geometry?” , well, you choose, as you seem very knowledgeable on the subject (remember to refer to the actual question when you do).
But when I address the rest of your post I’m going to be referring to Euclidean geometry.
Why must we keep straight your question-begging? If you were just going to define “knowledge” as that based upon observation and verification why did you bother to ask in the first place whether any knowledge could be otherwise? You already knew the answer. Given your own invented parameters you can have anything you want.
On What About Mary - thanks for finally admitting the failure of your previous standard to address the question and provide knowledge in that situation. I’ll emend our previous discussion on the topic this evening, and link it for you.
We do advance around here once in awhile.
Paul,
I did considered what you said about geometry. My question is easy…
Do you have actual knowledge of the postulates?
SteveK, I don’t know what you mean about knowledge of the postulates. Can you explain?
Charlie, I didn’t just define knowledge arbitrarily, as a form of question-begging, but developed that definition based on the ideas that we threw back and forth, came up with that definition as the only one that made any sense, or at least that is my claim.
And, Charlie, is it only other people who advance, who modify their ideas on occasion, or do you do that too?
SteveK, please answer my question about geometry. Do you think that we know something about geometry, given that it is based on postulates? If you say yes, then why is not that an analogy that can help one to understand the idea that I’m claiming for assumptions like the LNC?
Paul,
You speak as if you have actual knowledge of the postulate details. To be clear, I’m not talking about the truth/false nature of the postulates for that is a separate question.
If you do, then you have knowledge without the need for any assumptions. At some point, the assumptions stop and the knowledge just IS.
If not, then you are on the road to an infinite regress without knowing ANYTHING because everything is based on an assumption that you have no actual knowledge of. Make sense?
I’d say we know geometry because we were made to know it. Same with logic.
If I can know my thoughts without the need for assumptions (which are just more thoughts) then it seems obvious, per logic, that assumptions aren’t required for actual knowledge.
Where’s Holo when you need him?
SteveK, I don’t think it leads to infinite regress if I know what the postulate is, I’m not sure what details you mean, but let’s take the LNC. We can say what that “law” (assumption) is. I’ve in effect defined what the LNC is, are you intending that definition of any word would have to be considered in figuring out whether an infinite regress occurred? That sounds wrong to me, but I can’t put my finger on it.
I have to honestly say that I was stunned to read that. That statement ignores the role of postulates in geometry, and apparently ignores the existence of geometries different from Euclidean, including the difference in their postulates. Am I wrong about that? If not, please explain how your statement acknowledges and handles postulates and their differences in different geometries.
As far as knowing thoughts, remember that the limits of this discussion about knowledge was limited by me to knowledge about the outside world, excluding qualia, which I think can include our own thoughts.
Charlie,
Sure we can trust them when they are sound. The question is, should we trust that craig is not simply ignoring evidence and argument to the contrary, given the statement I quoted above.
The damning quote is from his book “Reasonable Faith”.
http://jcnot4me.com/Items/contra_craig/contra_craig.htm
The above link is from a person who presented craig with that quote, asked him to affirm it, and presented craig with a though experiment concerning falsification of the resurrection. If presented with this falsification, craig would continue to assert it’s truth.
Please, in light of the though experiment presented to craig from the above link, show me how “regardless of the evidence” does not follow.
I’ll grant you, Craig is a great debater and can make a persuasive argument, but from my (admittedly limited) exposure to his work, he bases those arguments on assumptions which are not known to hold, even in the fashion we “know” the LNC to hold, which is being argued above.
Charlie, one other thought about definitions.
Would you accept a difference between knowledge and assumption? If so, I presume it’s the one I recognize as well. And, if so, then wouldn’t you agree that the LNC is an assumption (lacking any proof, which would seem to be circular in any form) and not knowledge?
Paul,
If you actually know what the law is, then you can say it. Do you assume to know the law - which is not knowing - or do you actually know it? If you actually know it then, by definition, you know it without any qualifier.
I’m not ignoring the postulates. I know the postulates themselves without relying on more postulates such as: ‘assuming (not knowing) I’m not a brain in a vat’, ‘assuming (not knowing) I’m not hallucinating or dreaming or ‘assuming (not knowing) I can think rationally’.
Assumptions on top of assumptions on top of assumptions add up to NO KNOWLEDGE just like a sum of zeros add up to zero.
Now, if I can actually *know* the assumptions (they’re no longer assumptions are they?) then why can’t I know the truth value of them?
SteveK,
i know what the law means. I assume the truth of the law as we seem to agree it can’t be tested - it’s as assumption in various systems of logic, and also appears to apply in reality.
And here I was thinking this was what we were discussing here with Paul (and myself, though poorly) taking the position that we can’t know the truth value of them and have to assume it.
SteveK, do you think we know the proofs of geometry even though they are completely dependent upon postulates, or do we not know them?
We don’t say that we know a big zero about geometry, but every single thing we know about geometry we know only because of initial postulates.
You’re trying to make the idea of “know” do much, much more than it has to. It only need be limited by being dependent on a few assumptions, like the LNC, as well as observables. That’s all it has to do to work for us.
Havoc
Which means you know the truth value of the question: Can I know what the law means? And you know it without assumption!
I can understand struggling with truth value, but don’t tell me it’s because it can’t be tested. This didn’t prevent you from knowing the truth value above.
Hi Paul,
Sure my views change. But I think it is more than appropriate to point out, on yet another thread where you are impervious