Agrippa's Trilemma
Quoting Stanford Philosophy Encyclopedia
[b]How would you argue against this position?
- Curious to know a different perspective.[/b] :lol:
Before presenting a reconstruction of Agrippa’s trilemma we need to introduce some definitions. Let’s say that a belief is inferentially justified if and only if it is justified (at least in part) in virtue of its relations to other beliefs. A justified basic belief, by contrast, is a belief that is justified but not in virtue of its relations to other beliefs. An inferential chain is a set of beliefs such that every member of the set is allegedly related to at least one other member by the relation “is justified by”. Agrippa’s trilemma, then, can be presented thus:
1. If a belief is justified, then it is either a basic justified belief or an inferentially justified belief.
2. There are no basic justified beliefs.
Therefore,
1. If a belief is justified, then it is justified in virtue of belonging to an inferential chain.
2. All inferential chains are such that either (a) they contain an infinite number of beliefs; or (b) they contain circles; or (c) they contain beliefs that are not justified.
3. No belief is justified in virtue of belonging to an infinite inferential chain.
4. No belief is justified in virtue of belonging to a circular inferential chain.
5. No belief is justified in virtue of belonging to an inferential chain that contains unjustified beliefs.
Therefore,
There are no justified beliefs.
[b]How would you argue against this position?
- Curious to know a different perspective.[/b] :lol:
Comments (26)
In particular, what is the alleged justification for the premiss that there are no justified basic beliefs?
Right, if one says that belief in the conclusion advocated by the Trilemma (“Therefore, no beliefs can be justified”) is justified by the fact that logic leads to the conclusion that no belief can be justified, then that would contradict the conclusion reached, since there would be at least one belief that can be justified, namely: The belief that no beliefs can be justified.
If it's not justified, that means the Trilemma doesn't (by itself) give us a good reason to accept the conclusion, even if it were true.
Quoting aletheist
On the other hand, it seems that the justification would be to say that if a belief were justified, then it would not be a basic belief (since basic beliefs are supposed to be unprovable, like the Law of Contradiction for example, and therefore can't have a belief that would justify/prove them), and that the belief that justifies it would in that case be the basic belief, unless it were also justified by another one, in which case that one would be the basic belief that justifies the other 2 beliefs, and so on.
But if we said that that justifies the premiss, it would again contradict the conclusion.
Others have objected to Agrippa's trilemma as flawed for the following reason. However, first impressions may be deceiving.
Agrippa's argument's conclusion, A = There are no justified beliefs
Counter-argument to Agrippa
1. IF A THEN A is not justified
2. A
So,
3. A is not justified = U [1, 2 Modus Ponens]
Agrippa's conclusion is self-refuting. However, the counter-argument itself fails because it has as its premise, A = There are no justified beliefs, and if so, U = A is not justified, is itself not justified. This counter-argument refutes itself too.
Try something else please.
If you want justified knowledge, that's a bit impossible. As per the proof. In empirical cases. Although "cogito ergo sum" is a justified empirical statement proving the self's existence to the self, and to nobody else.
But why is "Albert not the same as Albert" an unjustified wrong statement? Language perhaps, as the unjustified conveyor of thoughts, is the culprit that we object to? Is language an unjustified conveyor of thoughts? Well, the Chinese room experiment is a proof of that. I mean... you need language to disjustify the valid existence of a priori proofs expressed in language.
That's a bit too rich. Because then we get to the paradox (and not merely a contradiction) of @Amalac. The proof of proving all proofs wrong proves itself wrong... therefore all proofs are not wrong... including the proof in the trialemma; then it means the trilemma is not wrong... which means it renders itself wrong if it's right, and it renders itself right if it's wrong.
Truly a clear paradox.
So if yo don't want to go insane following a paradox, don't start reading this thread.
Quoting TheMadFool
(Edit: It seems I misunderstood what you meant, I thought you meant the counter-argument to the counter-argument refutes itself too. Still, the conclusion of the counter-argument is “ If A, then A is not justified”, which is true. So I think what I say here is still relevant).
It doesn't quite refute itself, that just means that if it were true, it could not be justified/proved. It may be one of those “true but unprovable” statements.
Just as the Law of Contradiction is not justified, and cannot (according to philosophers like Aristotle) be justified, since all justifications assume its truth. (Unless we say it is justified by it's selfevidence, in which case we can either say: that that belief (“The Law of Contradiction is selfevident”) is what justifies the Law, which Aristotle won't accept, or we can say that it justifies itself/ doesn't need justification, which can be seen as either foundationalism or coherentism).
The distinction here is that between truth and justification: As Plato understood it, knowledge is justified true belief. It is not enough to believe something that happens to be true by mere chance, rather one must have a justification (reason) or proof to believe it.
For example: Suppose a child has been taught bad arithmetic, and that after being taught addition, he is also taught that A+A=A•A. When asked what's 2•2, he answers correctly: 4.
But when asked to give a justification for his belief, he answers: Well, 4+4=8=4•4, 3+3=6=3•3, Therefore 2+2=4=2•2.
It is clear that he has no justification to believe that 2•2=4, although by pure chance he answered correctly.
Carneades.org has an interesting video on the subject: https://youtu.be/_Kzhae8sFfY
Because it violates the Law of Contradiction, unless the word “Albert” at the beggining of the sentence doesn't mean the same as at the end of the sentence. So unless you are a LNC sceptic you must admit that it is false and unjustified, it would seem.
Quoting god must be atheist
It's not “the proof of proving all proofs wrong” but rather “the proof that no claim can be proved”. It may be true despite being itself unproved. And if it were true, then It would be impossible to prove it.
It would be something like Gödel's statement.
1. X = [Every argument must have one of the following formats:
a) It must be circular i.e. the conclusion is part of the premises.
b) It must regress into infinity i.e. each premise requires an argument, the premises of that argument, each must itself require another argument and so on.
c) It must be such that the premises are axiomatic i.e. assumed to be true without proof.]
2. IF X THEN there are no justified beliefs
Therefore,
3. There are no justified beliefs [1, 2 Modus Ponens]
The proposed counterargument @Amalac
3. There are no justified beliefs
4. IF there are no justified beliefs THEN there are no justified beliefs is (also) not justified
Therefore,
5. There are no justified beliefs is (also) not justified
Statement 5 undermines Agrippa's argument as it casts doubt on Agrippa's conclusion 3. There are no justified beliefs.
However, the counterargument to Agrippa's argument is itself an argument and one of the premises is Agrippa's conclusion, 3. There are no justified beliefs. This premise needs to be true if the counterargument to Agrippa is to work but then, if 3. There are no justified beliefs, is true, the counterargument itself fails as a justification for the conclusion that 5. There are no justified beliefs is (also) not justified.
Put differently, the counterargument offered by @Amalac fails for the reason that it assumes 3. There are no justified beliefs, but when that's done, it self-destructs.
What's the best way to resolve the matter?
We can't use Agrippa's trilemma on Agrippa's trilemma argument because when we do that we're conceding that Agrippa's trilemma argument is sound or that Agrippa is justified in claiming 3. There are no justified beliefs (based on Agrippa's trilemma).
We can't also try @Amalac's technique because it fails for the reasons shown above.
It appears we have no choice but to agree with the great Agrippa.
Of course, Agrippa's trilemma argument applies to this very argument I'm making. Wow! Mind blowing!
But it doesn't self-destruct, it proves that “There are no justified beliefs” would not be justified if it were true. (As you yourself seem to understand) which is true. That does not contradict Agrippa's argument, but does show that Agrippa's argument doesn't prove that its conclusion is true. (As you yourself point out when you say it casts doubt on it). That is why in my first post on this thread I said that it was the simplest way to start arguing against it, since it's the most commonly used argument against the Trilemma, but in fact I don't intend to refute the Trilemma.
Furthermore, if the conclusion of Agrippa's argument were true, then the premises would not be good justification for believing the conclusion, and so it would seem that as an argument it would only be good enough to show that it's conclusion might be true. But the same could be achieved if Agrippa had only stated the conclusion and we had considered the implications of it being true. There is then no need for the premises, except to show that a seemingly sound use of logic can lead to strange results.
So we have 2 possible scenarios, either:
1.“There are no justified beliefs” is false, and that is why it can't be justified.
2. “There are no justified beliefs” is true, but unjustifiable, in the same way Aristotle thought the Law of Contradiction is unjustifiable, or in the same way as Gödel's “true but unprovable” statement.
So my intention with the counter-argument was not to refute Agrippa, but rather to reach a state of equipollence, where we should suspend judgement as to whether scenario 1 is the case or scenario 2 is the case, we have no way to tell which of them is true since they look identical.
I think our disagreement is purely verbal though.
I don't mean to be rude or anything like that but I'd like to refer you to the reasons why your argument "...does show that Agrippa's argument doesn't prove that its conclusion is true". These reasons are Agrippa's trilemma. Basically, you're using Agrippa's trilemma against Agrippa. Doesn't that prove Agrippa's point? The situation is analogous to the statement, "this sentance has three erors". The statement is right because it's wrong. Agrippa is right because Agrippa is wrong; Agrippa is wrong because Agrippa is right. :chin:
Let's see: I say that it doesn't prove that its conclusion is true, not that it (the conclusion) isn't true. It may be true, but if it were true then, by it's own implications, the premises (the horns of the Trilemma) would not prove its conclusion (otherwise it would contradict itself).
I'm saying that if the conclusion were true (which is possible), then the premises wouldn't prove it (since in that case it would be justified by them, which would be impossible if the conclusion “There are no justified beliefs” were true).
Quoting TheMadFool
I'm not using it “against itself”, I'm pointing out the logical implications that would result if the conclusion of Agrippa's argument were true. You can see it as a sort of Reductio Ad Absurdum, but one that instead of proving the conclusion false, proves that if it were true, then the premises wouldn't justify the conclusion (for that is what the conclusion implies).
Basically, if the conclusion were true, then the premises wouldn't give us a good reason to accept the conclusion, even if the conclusion were true.
Do you disagree with this?:
Quoting Amalac
Quoting TheMadFool
The statement is meaningless, because “sentance” and “erors” don't mean anything.
After correcting the words it's no longer the same statement.
Quoting TheMadFool
If Agrippa is right, then no belief can be justified (including belief in Agrippa's conclusion). That doesn't mean that Agrippa is wrong (as in: Agrippa's conclusion is false), it means that Agrippa's conclusion can't be justified, even if it were true. That means it can't be justified by any premises, not even by the horns of the Trilemma that are the premises in Agrippa's argument.
If it were true, and also could be justified, then it would contradict itself. If that were true, then that would imply a logical contradiction. Therefore, using Proof by Contradiction, we conclude that if Agrippa's conclusion were true, then it would not be justified.
If Agrippa is wrong (meaning Agrippa's conclusion is false) then his conclusion can't be justified (since if it were justified, then it would be true, although the opposite is not true, that is: it's not the case that if it were true then it must be justified. An argument's conclusion may be true despite it being impossible to justify it with any premises.)
This video makes matters quite clear:
Quoting Amalac
In case you are wondering, I agree with the video's conclusion.
Quoting TheMadFool
Agrippa's Trilemma= the premises of Agrippa's argument (but not the conclusion), right? If so, then no: my reasons are not Agrippa's Trilemma.
I start by assuming that the conclusion is true, not that the premises are true.
Then I say: If the conclusion (No belief can be justified) is true, then the belief in that conclusion cannot be justified by any premise or set of premises.
This means we have 2 options if the conclusion were true:
a) The conclusion (no belief can be justified) does not follow from the premises(premises=Agrippa's Trilemma).
Or:
b) At least one of the premises is false.
This is because if the conclusion of the argument:
Quoting Stanford Philosophy Encyclopedia
Followed from the premises (and also assuming that they were all true):
Quoting Stanford Philosophy Encyclopedia
... Then it would be justified by those premises. But that is impossible if the conclusion is true, since it would contradict the proposition which states that “no beliefs can be justified” (the conclusion of the argument), since it also asserts that belief in itself can't be justified either.
And the same holds true for any other premises besides Agrippa's Trilemma that would try to justify the belief in the conclusion that “No beliefs can be justified”.
If the conclusion were true, and neither a) nor b) were true, then we could only conclude that the conclusion can't be true, for if it were true, then that would imply a logical contradiction. Unless we are to be also sceptical about the truth of the Law of Contradiction.
This is the logical consequence that results if “No beliefs can be justified” were true. In either case, the conclusion of the Trilemma can't be justified/proven.
That does not mean, however, that the conclusion of the argument is false, only that if it is true, it can't be proven true by any argument, since if the conclusion could be proven, then it would contradict what is logically deduced from its truth, and thus the proposition “no belief can be justified” would contradict itself if it could be justified/proved by any premises, such as the horns of the Trilemma.
Hopefully that clears up the misunderstandings.
Agrippa's conclusion, A = There are no justified beliefs. If A itself were a justified belief, that would contradict A. Ergo, A must be unjustified but doesn't that fit neatly into Agrippa's claim that A = there are no justified beliefs? That Agrippa can't justify his own claim which is A = there are no justified beliefs, proves his point, no?
There are two sets of beliefs to consider here. One, Agrippa's conclusion, A and two, all other beliefs. That all other beliefs can't be justified follows from Agrippa's trilemma. That A can't be justified follows from the contradiction that arises when we assume A is justified (kind courtesy of your keen observation). Ergo, ALL beliefs can't be justified. Am I justified in believing this?
Thus,
Agrippa's justification
1. Agrippa's trilemma [premise]
Ergo,
2. There are no justified beliefs = A [conclusion]
However, the conclusion A applies to itself as A is a belief and if so, Agrippa's justification for A is unacceptable. In other words, A isn't justified because A claims "there are no justified beliefs".
One is tempted on that score to reject Agrippa's claim, A = there are no justified beliefs, since the justification for A is, by its own admission, inadequate. Agrippa, by claiming that A = there are no justified beliefs, has shot himself in the foot because he can't justify A.
However, notice something interesting here: Agrippa failing to justify A doesn't contradict A i.e. instead of Agrippa's inability to justify A working against him, it actually is an instance of Agrippa's claim that A = there are no justified beliefs. What this means is that A isn't justified can't be used against A for it isn't a counterexample; to the contrary, it's a confirmatory example of A.
I agree, that's one of the points I was making, but although it doesn't contradict A, it also doesn't confirm A, since there is another possible explanation that could account for the fact that Agrippa failed to justify A: That A is false.
Quoting TheMadFool
Like I said, it was not my intention to prove that A is false. However, you say Agrippa's inability to justify his conclusion is an instance of A. But that's only if A is true.
How can we tell if A is true or not? By your own admission, if A were true, it could never be justified. But then, how can we distinguish the scenario where A is true, but not justifiable, and the one where A is not justifiable because it's false? The way we usually have to distinguish true beliefs from false beliefs, when they have to do with abstract reasoning, is to see whether they are justified or not, (except in cases like those of the so called 3 Laws of Thought) but that won't help us here. Since we have no way to tell, I say we should suspend judgement as to the truth of A. I think Agrippa himself would agree with my suggestion.
On the other hand, if A is true, and also not justified by the Trilemma, then that seems to mean that either:
1. A doesn't follow from the Trilemma
Or
2. At least one of the premises is false
Or
3. Both 1 and 2 are true.
Otherwise, A would be justified by the Trilemma. Which of those 3 options is the right one then?
If it's none of them we are in trouble: since according to logic A would be both true and justified, which contradicts A itself. And yet according to logic, A must be justified if 1 and 2 were false.
So if we are to avoid such a self-destruction of logic and reasoning, we must say that one of those 3 options is true, it would seem.
The puzzling part, however, is that the premises seem to be true, and the conclusion seems to follow from the premises, and if that's true A must be both justified and true according to logic, which contradicts A itself. In that case,sound logical reasoning would imply the truth of a claim that contradicts sound logical reasoning.
On the other hand, it could be said that sound logical reasoning could only imply anything at all if it were completely trustworthy, in which case nothing could contradict sound logical reasoning.
I see only 2 ways out of this problems: Suspend judgement or hold that either 1, 2 or 3 are the case.
Quoting TheMadFool
It could be a confirmatory example, but only if A is true, otherwise it is a confirmatory example of A's falsehood. And like I said, we have no way to tell if A is true or not.
That does seem like a plausible option, the Law of Contradiction for example can't be justified, but that is because it does not need any justification, Aristotle would argue.
The problem is that we then have to rely on criteria like selfevidence, and some beliefs which appear self-evident to one man may not appear that way to another. Some mystics may say, for example, that God's existence does not need to be justified because it's as obvious as the Law of Contradiction. For some, this is good evidence for the claim that foundationalism shouldn't be accepted.
There is of course an argument against such mystics: They may be deceiving themselves, and in fact are wrong in thinking that God's existence is that obvious or that they really experienced God and not something else.
But there is no way to tell if they are in fact deceiving themselves and/or others, or are actually telling the truth.
The only reason, given a valid argument, for that argument to fail is if the premises are false. That out of the way, focus on the Agrippa's argument below:
Agrippa's justification/argument
1. Agrippa's trilemma [premise]
2. If Agrippa's trilemma then there are no justified beliefs [premise]
Ergo,
3. There are no justified beliefs [conclusion, modus ponens 1, 2] This is Agrippa's conclusion
Your counterargument:
3. There are no justified beliefs [Agrippa's conclusion]
4. If there are no justified beliefs then there are no justified beliefs is not justified
Ergo,
5. There are no justified beliefs is not justified [your conclusion, 3, 4 modus ponens[
Notice that 3. there are no justified beliefs is a premise in your argument i.e. 3. There are no justified beliefs has to be true; your argument has to be sound to make an impact and in addition to being valid (it is, modus ponens) the premises need to be true. In other words, for your conclusion 5. There are no justifies beliefs is unjustified, 3. There are no justified beliefs must be true [it's a premise that needs to be true]. If not, you wouldn't be able to justify 5. There are no justified beliefs is not justified. In the end by having to assume 3. There are no justified beliefs, you destroy your own counterargument which is a justification, no? To top it all off, you're agreeing with Agrippa that 3. there are no justified beliefs. The bottom line - your counterargument requires Agrippa to be right on the money viz. 3. there are no justified beliefs.
Quoting TheMadFool
Yes
Quoting TheMadFool
Yes
Quoting TheMadFool
Yes
Quoting TheMadFool
Yes
Quoting TheMadFool
Yes.
Quoting TheMadFool
Ah ok, so you are saying if I assume Agrippa's conclusion, then I can't even conclude that “There are no justified beliefs is not justified”, because in order to do so, I would have to justify it and then the conclusion would be justified, which contradicts 3, right? In that case though, you would be unable to justify the claim that my counter-argument destroys itself, since you are also using an argument which pretends to be a justification of the conclusion that my counter-argument destroys itself, which can't be the case if Agrippa's conclusion is right, correct?
Let's assume the opposite were true, that is: “There are no justified beliefs” is justified: Then it's a self-contradictory statement, since it itself is justified, and therefore its truth implies its falsehood, therefore via Reductio ad Absurdum we conclude that it can't be justified. The only option that we have left is to say that it is not justified.
But I suppose you will reply that even to say this is to give a justification for the claim that it can't be justified, which can't be the case if “There are no justified beliefs” is justified (and therefore true). But then Agrippa himself would have to admit that he is also not justified in thinking that my conclusion: “There are no justified beliefs can't be justified” is wrong.
...And so on, this kind of self-reference paradox will go on forever, and so my advice would be to suspend judgement.
And then I ask you: Is the Trilemma justified or not? My answer would be to say: I don't know (though even this may not be justified).
Perhaps somebody else can help us with this matter:
[quote=Leszek Kolakowski]It was known to the ancient sceptics and it has been known ever since that any epistemology, that is, any attempt to build universal criteria of validity for knowledge, leads either to an infinite regression or to a vicious circle or to an invincible self-reference paradox (invincible, that is, unless it is spuriously solved by being converted into an infinite regression). The most vexing side of this old insight consists in that once it is stated it falls prey to its own veredict, which means that a sceptic is inconsistent by the very fact of preaching the sceptical doctrine.[/quote]
Kolakowski sums up the situation nicely:
Since you made the first move with a justification, I followed suit with my own justification. On the whole though you're right on target - I'm not justified in saying you're not justified IF Agrippa is correct but, by the same token, neither are you justified in saying Agrippa is not justified and, to further complicate matters, neither is Agrippa himself justified but that doesn't contradict Agrppa's claim that there are no justified beliefs and that's perfectly ok.
Quoting Amalac
Music to a skeptic's ears. I'm sure Agrippa would've been delighted.
Here's some food (argument) for thought.
If one believes, asserts, maintains that J = There are justified beliefs, we must accept Agrippa's argument/justification; it is, after all, a justification. Agrippa's argument/justification proceeds as follows:
1. J = There are justified beliefs [Agrippa's trilemma doesn't matter]
2. If J then we must accept Agrippa's justification/argument as good [Agrippa's trilemma doesn't matter]
3. If we must accept Agrippa's justification/argument as good then we must accept Agrippa's conclusion, A = There are no justified beliefs
4. If J then we must accepy Agrippa's conclusion, A = there are no justified beliefs [2, 3 HS]
5. We must accept Agrippa's conclusion, A = There are no justified beliefs [1, 4 MP]
If one assumes that there are justified beliefs and since Agrippa's argument/justification is good (it's a sound argument since Agrippa's trilemma doesn't matter), it follows that we must accept Agrippa's conclusion which is A = There are no justified beliefs.
Note how Agrippa manages to use the proposition J to ensure that his argument/justification is good/sound. Basically, Agrippa's claim A = There are no justified beliefs, is justified if we assume J = There are justified beliefs.
[quote=Cypher]Jesus! What a mind-job![/quote]
1. Agrippa's trilemma doesn't matter = ~M [Assume for reductio ad absurdum]
2. If ~M then Agrippa's argument/justification that Agrippas's trilemma matters is good
3. If Agrippa's argument/justification that Agrippa's trilemma matters is good then Agrippa's trilemma matters
4. If ~M then Agrippa's trilemma matters [2, 3 HS]
5. M = Agrippa's trilemma matters [1, 4 MP]
6. M & ~M [1, 5 Conj]
7. ~~M [1 to 6 Reductio ad absurdum]
8. M = Agrippa's trilemma matters [5 DN]
9. If M then there are no justified beliefs
10. There are no justified beliefs [8, 9 MP]
Basically, if we assume Agrippa's trilemma doesn't matter, we can infer that it matters. If Agrippa's trilemma matters, there are no justified beliefs. It all boils down to the first assumption we make. When I assume that Agrippa's trilemma doesn't matter (line 1) what I actually mean is that there are justified beliefs. If there are justified beliefs/Agrippa's trilemma doesn't matter, Agrippa's justification/argument that there are no justified beliefs is good/sound. If so, we must accept that there are no justified beliefs.