Yes, even though it’s been accepted by the Greeks as a perfectly valid argument for over two millenia…
The underlying flaw of syllogism is decently simple: the validity of the syllogism relies on the validity of the major premise. But the major premise cannot be proved until the minor premise is proved; but to prove the minor premise is the purpose of utilizing syllogism.
An example: all men are mortal. Socrates is a man. Therefore, Socrates is mortal. Right?
Not necessarily. The fact that all men are mortal cannot be established until we know for certain that all men are, indeed, mortal–we would need to know whether Socrates is mortal or not before we can know if all men are truly mortal. (Unless being mortal were specified in the definition of being a man; but it’s not, as man is not an artificial term.)
Of course, this is only considering the argument in purely logical/philosophical terms, removed from all practical implications. In real life syllogis would remain safe to use.
You are confusing validity (if the premises are true, the conclusion must also be true) with soundness (the premises are true and the argument is valid, and hence the conclusion is true). The validity of the syllogistic form ‘All A are B; x is an A; therefore, x is B’ is indisputable. You wish to argue that, validity notwithstanding, any instantiation of this syllogistic form is unsound based on the (lack of) evidence for the major premise. However, your concern is of a limited scope. Obviously, it can only apply to syllogisms where the major premise is known based on inductive reasoning. Any case where the major premise is known a priori escapes your objection. Ex. All bachelors are unmarried. Gorsnak is a bachelor. Therefore, Gorsnak is unmarried. We do not need to know whether Gorsnak is married in order to know all bachelors are unmarried. We know this as a matter of understanding the concept of bachelor.
Furthermore, your objection is only as worrisome as general doubts about induction. Hume and friends aside, induction is a perfectly acceptable means of coming to know something. The whole ‘any conceivable room for doubt entails a lack of knowledge’ thing makes for a couple extra chapters at the beginning of epistemology textbooks, but no one seriously questions the possibility of inductive knowledge.
Anyways, given the level of confidence that we have in our inductively-formed belief that all men are mortal, were we presented with evidence that Socrates is not mortal, we would conclude not that all men are not mortal, but that Socrates is not a man.
In other words, you’re confusing validity with truth.
This is valid, though it contains no truth:
All men are green. This computer is a man. Therefore, this computer is green.
This is invalid, though every statement is true:
A square has four corners. the Earth is the third planet from the Sun. Therefore, Lincoln was assassinated.
Whether it rains this afternoon or not, we will go to the movies.
We will not go to the movies this afternoon.
Therefore, my uncle Bernie will die of a heart attack next week.
This is a valid syllogism (if the premises are true, the conclusion must be true as well), but it is not sound (as the premises can’t all be true).
The idea that a cognative structure such as a " syllogism" has a more than incidental value in determining the truth of a statement is a bull-shitting tip.
Less than competent modern teachers like to teach this trick to young folk as a short cut to actual thought.
The latter-day ancient Greeks were wonderful geeks who sat around thinking about idle things like golden rectangles and squaring circles and syllogistic word games.
I love and admire the ancient Greeks.__ (major)
This is the 21st Century._____________(minor)
Therefore…_______________________(conculsion)
I concede. Should be debating the soundness of arguments based on syllogism and not the validity of the arguments. (Having never taken any philosophy courses…)
Gorsnak: yes, my comments were directed only toward major premises based on induction. The case where “all bachelors are unmarried” was alluded to in the thread–"…unless xxx were specified in the definition of zzz." In this case, being unmarried is entailed by being a bachelor.
The above is NOT a valid syllogism. Syllogistic validity and logic in general have NOTHING AT ALL TO DO WITH WHETHER OR NOT SOMETHING IS TRUE.
A syllogism is as follows:
Axiomatically state that all members of set A have trait B.
Axiomatically state that C is a member of set A.
Therefore, one concludes that C as an individual has trait B.
Note that the first two statements are AXIOMATIC. Their real-world truth is irrelevant. Truth is irrelevant to logic.
The syllogism would epistemologically be a priori as a logical tool. It would be no surprise that we would need complete induction (an impossibility) to apply it in the Real World[sup]TM[/sup], but it wouldn’t counteract the validity of the syllogism.
An example: (On preview, I see an example already extant. Ah well.)
1: It is raining. (Premise)
2: Either it’s raining or bears don’t shit in the woods. (Entailed by 1)
3: It is not raining. (Premise)
4: Bears don’t shit in the woods. (Entailed by 2 and 3)
1 and 3 are obviously contradictory, and just as obviously anything could be put in the place of ‘bears don’t shit in the woods.’
Here’s one way of explaining why this happens - the truth conditions for ‘if A then B’ are for either B to be true or A to be false. The definition of validity is “If the premises are true, then the conclusion is true.” But in the case where the premises are contradictory, by necessity the premises are not true - they can’t all be true simultaneously; that’s just what it is for them to be contradictory. And, if the premises aren’t all true, then the conditional ‘if the premises are true, then the conclusion is true’ is itself true, and hence the argument is deductively valid (albeit in a very uninteresting way).
It was not a syllogism in the technical sense, as my previous post quite clearly stated. Quite arguably there is a non-technical meaning of ‘syllogism’ by which it would qualify (my dictionary gives such a definition, but it’s not the OED either). Given that some readers are likely to be unfamiliar with the nitpicky usage of logical terminologiy, that your comments appeared, to me at least, to be attacking the whole of ultrafilter’s post, and that this thread is in no small portion concerned with validity, I thought clarification would not be amiss, in the interests of avoiding the misleading of those readers without training in formal logic. Try not to get your knickers in too much of a knot over the driest subject known to man.
