Right. But you will notice that logical form is not the only concern.
>This is not to say, however, that the
>conclusion is false, nor does a sound argument automatically mean that the
>premises ARE true,
That is exactly what a sound argument means. Soundness absolutely requires
the truth of the premises. It is *validity* which is unconcerned with
truth, and focusses only on logical form.
Just to restate for the sake of clarity (I will ignore inductive logic for
the moment):
Valid argument: an argument such that, if the premises are true, the
conclusion is guaranteed to be true.
Sound argument: an argument which is valid, and its premises and conclusion
really are in fact true.
>only that IF they are all true then the conclusion would
>also be true if the argument is valid. With this kind of language it sounds
>like your building is made of mist rather than concrete.
If soundness meant only that "if the premises are true, then the conclusion
would be true", then soundness would be indistinguishable from validity.
In fact, validity is a precondition for soundness -- but not the only one.
>"Philosophy is still concerned with what really is the case. If pink
>elephants do not exist, then your 'pink elephant syllogism' is not
>satisfactory for *either* a philosopher or a scientist. Why would you think
>your pink elephant example *would* be satisfactory for philosophers?"
>
>Because it is valid, if unsound.
But if it is unsound, then it is not satisfactory for philosophers.
>If there are no sound arguments to
>contradict it, however, then it is better than an unsound invalid argument,
>which by the way can also have all premises true with a true conclusion.
>Your logical form/rationality dualism sounds like it could get into trouble
>in that case; it might even declare such an invalid argument sound if
>rationality took precedence over form.
No. There is no "precedence". Truth and validity are both necessary
conditions for soundness. (Rationality just refers to the way we decide
whether the premises are true.)
>However, since in fact form is the
>necessary condition and rationality only a minor consideration,
No! They are both equal considerations for soundness.
>a completely
>true argument that is nonetheless invalid would be superceded by any valid
>argument even if it had at least one false premise.
>
>"Do you think philosophy is content with false arguments?"
>
>First of all, only statements and propositions can be true or false;
>arguments are either valid or invalid, with the truth-values of the
>statements determining soundness.
Of course you are correct here. I should have said, "do you think
philosophy is content with unsound arguments?"
>Secondly, an unsound argument only means
>that the conclusion has not been established to be true, not that is has
>been established to be false.
The unsoundness of an argument means one of two things:
1) The argument does not have the correct logical form;
2) At least one premise (or the conclusion) is false.
It is possible to have unsound arguments with all true premises and a true
conclusion, e.g.:
1. Dallas is in Texas.
2. The Great Pyramid is in Egypt.
3. Therefore, marmots are mammals.
... and it is possible to have unsound arguments which are logically correct:
1. If squirrels have three ears, then whales are reptiles.
2. Squirrels have three ears.
3. Therefore, whales are reptiles.
I hope this is just restating what you have said; I just want to make
really clear that neither logical correctness nor true premises are *alone*
enough to establish soundness. They are *both* required.
>Thirdly, a valid argument nearly always
>supercedes an invalid argument, even if it is unsound. The only exception
>might be a valid argument with at least one false premise and a false
>conclusion versus an invalid argument that is completely true, but I doubt
>it.
I'm not sure what you mean by "supercedes". The point of arguments is to
establish the truth of the conclusion. Arguments either do that or not; if
they do, they are worthwhile arguments, and if they don't, then they are not.
>If you might mean empirical fact rather than soundness or validity,
>tell me then, how do you determine which (if any) of the statements in the
>following argument are (factually) true?
I'm not sure what the "rather than" is supposed to mean. Validity denotes
logical form. Soundness denotes truth + logical form. Empirical facts
might be used to support the truth of premises, and thus may be part of a
philosophical argument (rather than an alternative to one, as you seem to
suggest).
>If the universe is finite, then it has a physical boundary.
>The universe is finite.
>Therefore it has a physical boundary.
>
>Believability cannot be a factor in determining truth because it is too
>subjective.
The factor is not mere "believability", but rational warrant. That is, I
may consider it believable that Dallas is in Egypt or some such thing (who
knows, maybe I'm insane), but this is not to say that this conclusion is
rationally warranted. Rational warrant, in turn, is not even necessarily
an infallible way of arriving at truth; it is simply the best method
available to we non-omniscient human types.
>Structurally, this argument is valid, but is it sound? So far,
>everything we know of the universe would tell us that premise 1 and the
>conclusion are false, but that premise 2 is true.
You can stop right there. If any of the premises are untrue, then the
argument is not sound.
>So we could claim that
>the argument is unsound. But new evidence might show that premise 1 is
>true. That would then make the conclusion true, which would make the
>argument sound. Therefore, using only empirical fact we cannot say whether
>the argument is sound or unsound.
Of course using facts is required! I never claimed otherwise. Remember,
*I* am not claiming that philosophical inquiry is uninformed by facts!
>However, a philosopher could construct
>another sound argument that concludes with premise 1. That would then make
>premise 1 true, which would in turn render the conclusion true, making the
>argument sound. If, on the other hand, the argument concluding in premise 1
>was valid but had one false premise itself, yet the argument propositions
>had an evidential relationship that made the conclusion true, then even
>though the argument was unsound it would still assert the truth of premise 1
>and thus be an important part of the entire composite argument.
I don't understand this part.
<snip>
>"How does this impact on the nature of philosophy in general?"
>
>It doesn't; it merely demonstrates what usually happens when an exercise in
>pure reason, with no empirical evidence to support it, based on premises
>that
>are not themselves grounded in scientific reality, tries to describe a
>phenomenon in scientific reality that can be directly tested: it falls flat
>on its face.
If it doesn't impact on the nature of philosophy in general, what is the
point of bringing it up...? At best, you're just arguing against the use
of pure logic, and even your argument here is unspecified. That is, it is
not enough to simply say "one case where pure logic was used ended up being
false".
>"At one point in the history of science, Lamarck thought that, if someone
>lost an arm, that characteristic could be passed on to the next generation."
>
>That is a gross oversimplification of Lamarck's hypothesis. Besides, his
>theory of acquired traits was a testable attempt to explain the fact of
>evolution. It was based on empirical data to start with and it was
>empirical data that eventually refuted it.
Of course. My only point was that showing one failed example of a
methodology is not a refutation of the methodology in general.
>"Therefore, the proper realm of science is to make silly claims."
>
>Not silly; preposterous. Because it is only by making claims that seem like
>nonsense, then testing them against scientific reality with empirical data
>guided by reason, does science advance.
Right. And only by making arguments which are both logically valid and
sound does philosophy advance.
>"Surely you don't agree with that conclusion; [see above] therefore, you can
>see why I don't agree with your conclusion about philosophy."
>
>Philosophy as a whole does not make silly claims, but philosophers who
>believe they can explain scientific reality without recourse to observation,
>empirical data or experimentation very often do, because science invariably
>proves them to be silly.
What is "scientific reality"? There is only one reality; philosophy and
science are simply different ways of inquiring about it. With that in
mind, I don't really understand your argument here.
>Now, regarding Hegel specifically, I think you may actually be incorrect in
>describing his views. The Oxford Dictionary of Philosophy has this to say:"
>
>Where? I can't find it.
It is on p. 168 of my edition, under "Hegel". Edition details: Blackburn,
Simon. *Oxford Dictionary of Philosophy*. New York: Oxford University
Press, 1996.
>"I haven't read *De Orbitis Planetarium* myself. But if you want to use
>that example, you might want to give the issue a second look."
>
>Gauss did; so did Bronowski; so have others. Hegel's defenders say the
>claim is false; these people say that it is not.
So clearly, it is a very contentious issue. It probably isn't a great
example for exactly that reason... unless, of course, you want to inquire
into the issue yourself. (I.e., actually read *De Orbitis Planetarium* and
see what Hegel said.)
>What is clear at least,
>however, is that Hegel made some claims, based solely on pure reason, that
>nonetheless were testable against empirical data and turned out to be wrong.
That isn't clear at all, since the exact point of contention is whether
Hegel made any such claims, or whether that is simply a misinterpretation
of his work.
This much can be said: if an exercise in pure logic is *a priori*
impossible or not conducted correctly, then its conclusions will of course
be false. If someone believes that producing truths by an exercise in pure
logic *is* possible (e.g., the proponents of Ontological Arguments for God)
then they will respond to your "Hegel" example simply by saying "well,
clearly Hegel made some error in logic". In other words, even if you are
right about Hegel, there is no apparent sense in which this falsifies the
use of pure reason in general.
Moreover, it is worth remembering that "empirical data" is *less*
compelling than pure logical inference. If it is possible to produce
truths by exercises in pure reason, and a philosopher does this correctly,
then his conclusions are true. If a scientist has empirical data which
appears to contradict his conclusions, the empirical data *must* be
misleading. This is because a proposition whose truth is logically
necessary (supposing any such propositions to be possible) *must* be true,
and all evidence to the contrary must be false. Now, note that this is not
an argument *I* am making. I personally am dubious about the use of pure
logic in gaining truth. But if you want to take on a "critique of pure
reason", you should take note that your opponents have many apt responses
to make to the arguments you've presented thus far.
>That Hegel believed his conclusions were right I think should be obvious;
>otherwise he never would have made them. Why did he believe they were
>right? Because he believed his philosophical system could determine
>scientific truth without the need for empirical data. Well, he was wrong,
>as has been everyone else who has made the same claim.
Whoa! "As has everyone else who has made this same claim"? How on earth
can you support that conclusion? That is exactly analogous to saying,
"Lamarck believed he could determine truth through scientific fact, and he
failed, therefore everyone who uses scientific fact will fail"...! But
surely you can see the fallacy there. Basically, you need some argument
supporting this leap from particular instance to universal generalization.
>"Absolutely, many philosophers do believe this. And there *is* potentially
>a better road to knowledge than observation: pure logic."
>
>Except that each case, where pure logic (or rather pure reason) was used to
>create a hypothesis in the complete absence of empirical data, has failed.
How do you know this? How do you know that pure logic is inferior to
science? Apparently, because science tells you so... but that begs the
question, does it not? Again, for the record, I am not necessarily
defending the use of pure logic in gaining knowledge. But all the same,
your arguments against it are currently not compelling in the slightest...!
>Why? Because the universe does not conform itself to our demands of how it
>should be structured.
This is a misunderstanding of logic. Logic is not an attempt to impose
conformity on the universe; it is an attempt to impose conformity on what
we can validly say about it. This can be understood by realizing what a
logically improper assertion *is*: it is not an assertion which just
happens to be false, but rather an assertion which negates itself, and
consequently is *necessarily* false. Similarly, if something is a
logically necessary truth, then to suggest it *untrue* would be to commit a
logical error -- i.e., to propose something which is self-contradictory,
and thus necessarily untrue.
For this reason, if there *are* such things as logically necessary truths
(which are the kind of truths that exercises in pure logic attempt to
discover), then you *cannot* meaningfully say "well, but those aren't true"
any more than you can meaningfully say "a have a square triangle" or
"squarbling bloborfs splink marvellously".
>It has its own structure, but we can only discover
>that structure by testing our demands against empirical data collected by
>studying the universe.
Empirical data alone does not tell you anything about the structure of the
universe. Structure is discovered by *inference* from empirical
observation. What empirical data does tell you is the particular facts of
the observable universe -- and it does not even do that infallibly.
Empirical observation of something is not 100% deductive proof of its
reality, but rather probablistic (inductive) evidence for it. This is
because observations may be incorrect or misleading.
>99 times out of 100 our demands are wrong and need
>to be changed. As such, any demand that claims it does not need to be
>tested has only a 1% chance of being right.
Huh? Think carefully about all the things you are presupposing by saying
the above.
>"Why is pure logic a potentially better road to truth than observation?
>Because observation is necessarily *inductive* in nature. That is, no
>matter how many times you can observe something, the most this can justify
>is that it is *likely* to be true. You can observe light going at the same
>speed a hundred times... a thousand... a million; but you cannot state with
>deductive certainty that it always travels at the same speed."
>
>Philosophically, yes; scientifically, no.
That's right, because science, being empirical, is *necessarily* inductive
in nature. This makes the point that science can never give 100% deductive
certainty. (This may be true of philosophy too, it depends on your
philosophical position regarding exercises in pure reason.) Anyhow, my
point stands. Pure logic, if it works, discovers truth with 100%
certainty. Science cannot even theoretically make that same claim to
absolute certainty. (This isn't to say science is irrational, of course.
If something is very very likely to be true, then it ought to be believed
in the absence of more compelling evidence to the contrary.)
>The reason is because, based on
>our current theories of light, there is no known mechanism by which the
>speed of light may be changed.
"We haven't yet found any evidence that light changes" is, however, not the
same as "we have found evidence that light never changes". The former is
no kind of evidence at all that light does not change.
>Also, observation is not entirely inductive;
>it depends upon what you are observing. Much of what I do yields results
>that do not need to be interpreted, because they are so unambiguous. Under
>those circumstances my observation are entirely deductive, because I simply
>collect and chart the data, then use it to design new experiments.
The inductive nature of science does not result from *ambiguity*, but
rather from the inherently imperfect nature of observation. It is possible
to hallucinate. It is possible to see something as one thing, when in fact
it is another. (For example, if I see a cylinder from a certain angle, I
can see it as a cube.) It is possible to miss things. Being
non-omniscient, humans can and do commit error. And then there is the
question of whether observation really puts us in contact with an external
world at all. Saying things about reality based on observation is never
100% guaranteed to be true (and remember, deductively sound arguments
always guarantee 100% certainty).
>In fact
>I tend to suspect that observation is inherently deductive in any
>discipline; the inductive part is when you use that observation to create a
>general theory.
Why?
>"(This is not to say that it is unreasonable to believe that light always
>travels at the same speed. Of course it is a reasonable belief. But it is
>not a 100% deductively-provable belief, nonetheless.)"
>
>Nothing is 100% deductively-provable;
Then why did you just say that "observation is inherently deductive"? All
things proven by deductively sound arguments are 100% proven.
>the scientific question is not whether
>you can absolutely prove that something is true, but whether you understand
>it enough to be able to define what it is, describe how it works and use it
>to help you understand something new.
Right. But the point of pure reason *is* to absolutely prove something is
true.
>"I would say that is a bit misleading. Science is best regarded not as
>something wholly different from philosophical inquiry, but rather a very
>specialized kind of philosophical inquiry: empirical & naturalistic inquiry.
>Philosophy can, and often does, try to acquire knowledge by *using* the
>results of science... just as science itself is ultimately based on
>foundations of philosophy."
>
>There is still one major difference between the two: science tests all its
>claims against empirical data, whereas philosophy test is claims against
>logic.
This is *very* misleading. In fact, both philosophy *AND* science test
their claims against both evidence and logic. That is, an argument which
is logically flawed or evidentially unsupported is not satisfactory for
either philosophers or scientists. The exception here would be exercises
in pure reason, which have certain unique properties.. but that is not by
any stretch of the imagination the only kind of arguments philosophers use.
Many philosophers don't use them at all.
>Philosophy can only use empirical data to test claims when it exists
>(which it usually doesn't) and science avoids pure logic whenever possible
>(which is usually always). As long as this simple fact is true, I stand by
>my statement
Science avoids exercises in *pure* logic (i.e., deducing the truth of
propositions without the use of contingent propositions), but does not
disregard logic in general. Philosophy can use empirical evidence when it
exists, as well as any other kind of evidence which is available; if no
compelling evidence is available, then philosophers have no reason to
consider the claim correct. Basically, at this level of verifying claims,
there is no great difference between scientific and philosophical method.
Both methods ascertain the logic and truth of the claim. The only
difference is that science necessarily tests truth through empirical
methods, whereas philosophy is not necessarily limited that way.
>"Both science or philosophy may address abiogenesis."
>
>But philosophy won't tell you whether it can happen, how it could happen or
>even if it did happen, which is ultimately all we on this list group are
>interested in finding out.
Sure it can. Here's an argument I could advance in a philosophy class:
1. If life has not always existed, then at one point life had to come about
from non-life.
2. Life has not always existed.
3. Therefore, at one point, life came about from non-life.
I just gave a valid argument for abiogenesis. Is it sound? Well, I'd have
to substantiate some of my premises further, of course. And I could do
that by appeal to empirical evidence (including science), or indeed any
kind of compelling evidence which might be applicable. So what is the
problem?
>"For example, philosophy might instead examine whether abiogenesis is a
>logically consistent or meaningful concept."
>
>Then let me ask you this: what would you say if a follower of Hegel wrote a
>thesis that concluded that abiogenesis was both historically and
>mechanistically impossible, therefore it wasn't a meaningful concept,
No, you have misunderstood me. The question of whether something is
logically consistent or meaningful is the question of whether it makes
sense *in principle* to suppose it. For example, "the earth is a cubic
sphere" is not a logically consistent or meaningful claim, because it
contains a self-contradiction (something cannot be both a cube and a
sphere). Similarly, a philosopher could consider whether there is any
logical incoherence implicit in the concept of abiogenesis.
>then
>the very next year a research team announced that not only had they
>discovered conclusive evidence that abiogenesis had occurred, but they also
>knew the precise mechanism of how it worked?
This suggests one of two things: either that I committed some error in my
argument, or that the research team committed some error in theirs. Which
of those is the case depends on the exact arguments we have advanced.
Simply "your argument is philosophical, theirs is scientific" says nothing
about the accuracy of either, unless you are assuming without argument that
science is always right and philosophy is always wrong.
>My point is that any
>philosophical claim about whether a scientific phenomenon is meaningful
>based solely on pure reason is highly unlikely to be worth the paper it's
>printed on once that phenomenon has been verified and explained, as has
>happened so often in the past.
Well, meaningfulness has nothing to do with whether a claim is *true*, just
whether the claim makes sense. Comparing a philosophical argument about
the meaningfulness of a claim and a scientific argument about the truth of
a claim is a category mistake.
>When a scientific theory makes that many
>failed predictions, scientists discard it for a new one, yet no matter how
>many times they fail, there is always some philosopher who will try yet
>again to develop a method of acquiring perfect scientific knowledge without
>empirical data.
You know, the attempt to reach truths through empirical observation (i.e.,
science) has failed many times in history. Scientists don't always get it
right. Therefore, scientists are stupid to keep practicing science. When
will they ever learn? (Spot the fallacy.) This seems to be what you are
saying of attempts to use pure reason. They have failed often; therefore
they will always fail. Now, there *are* some compelling reasons to reject
pure reason. Read Kant's *Critique of Pure Reason* for some of them. But
this argument you keep advancing is *not* a compelling reason at all.
Just for the sake of completeness, I will give you the argument *I* think
refutes pure reason.
1. All reasoning is either analytic or synthetic.
2. Analytic reasoning is such that it simply analyzes a concept without
adding to it; i.e., any truth brought about by analytic reasoning must have
been implicit in the concept to begin with. Only synthetic reasoning
actually adds any information to a concept.
3. Any use of pure reason is necessarily analytic, because to do otherwise
would be to introduce synthetic elements (i.e., contingent propositions).
4. Therefore, the application of pure reason to a concept cannot *add* the
property of "truth" to it. Truth must be implicit in the concept to begin
with.
5. Truth cannot be predicated. That is, a concept describes the nature of
some state of affairs *if* it were true; adding truth to the concept itself
is invalid. In other words, truth cannot be implicit to a concept.
6. Therefore, truths cannot be discovered by use of pure reason.
This is a minimally-developed argument, and there is much debate about
premise 5 in particular. But this at least is an attempt to show why pure
reason is in principle impossible.
>"I think the problem is that you hold the layman's view of philosophy."
>
>I don't agree, as I hope I have demonstrated in this post. My problem is
>certain philosophers who claim that, based on pure reason, they have a
>better
>understanding of the origin, structure and mechanisms of the universe than
>the scientific understanding that has been developed over centuries of
>empirical and rational investigation coupled with direct testing of ideas.
>It took science over a 1000 years to crawl out of that abyss; I have little
>tolerance for people who seek to drag us back into it again.
IF pure reason is possible, then the results of a correct exercise in pure
reason would be 100% likely to be true. The results of science are never
100% likely to be true. Therefore, if pure reason is possible, and it is
exercised correctly, and science contradicts the result... science is
wrong. This is the argument you will face if you advance your argument to
a philosopher who affirms that pure reason is a valid way to reach truth.
In order to argue against this, you cannot simply say "the results of
science contradict your results, therefore science is right and you are
wrong" -- that is a blatant fallacy, and commits the same "arrogance" you
have accused these philosophers of. You would have to advance a
philosophical argument (like the one above) showing that exercises in pure
reason cannot even in principle arrive at truth.
Regards,
Mike Hardie
<hardie@globalserve.net>
http://www.globalserve.net/~hardie/dv/