Q&A: Regarding the fallacy of “begging the question,” about the discussion you had with Aviv Franco.
Regarding the fallacy of "begging the question," about the discussion you had with Aviv Franco.
Question
I recently watched your discussion with Aviv Franco on the channel “Head to Head.”
The discussion was very interesting. As a non-believer, I found your views very interesting, and honestly I think Aviv somewhat missed some of your arguments, which is a shame.
But something in your opening remarks really stuck in my throat, and I wanted to ask you directly what you meant, because maybe I didn’t understand you correctly.
I’m referring to your claim that “every valid argument begs the question, by definition.”
From what I know—and correct me if I’m wrong:
An argument that “begs the question” literally assumes within its premises the conclusion it is trying to prove.
From what you explained, a logically valid argument must contain within it the conclusion it wants to prove, otherwise it would not be valid.
You gave the classic example:
Premise 1: All human beings are mortal.
Premise 2: Socrates is a human being.
Conclusion: Socrates is mortal.
Now, as I understand it, this argument does not beg the question, and there’s a pretty simple way to check that.
Premise 1, by itself, does not beg the question, because it could be that Socrates is not a human being, and then he would not necessarily be mortal.
Premise 2, by itself, does not beg the question, because it could be that Socrates is a human being and yet not mortal.
Only when premises 1 and 2 are assumed together does the conclusion follow.
That is, the argument does not assume the conclusion, but rather assumes various things such that if all of them are true together, the conclusion follows from them.
Therefore the argument does not “beg the question” in the sense of the logical fallacy of “begging the question.”
Just to explain this better, I want to give another example:
Premise 1: All triangles have 3 sides.
Premise 2: X is a triangle.
Conclusion: X has 3 sides.
This argument really does beg the question!
Premise 1, by itself, does not beg the question, because X might not be a triangle, and therefore not necessarily have 3 sides.
But premise 2, by itself, does beg the question, because by definition a triangle is a polygon with 3 sides, so necessarily X has 3 sides if we assume that X is a triangle.
And therefore premise 1 is superfluous in this argument; that is, there is one single premise on which the whole argument relies, and therefore it necessarily begs the question.
So I wanted to ask in order to understand:
Are you claiming that an argument that is logically valid “begs the question” in the sense that “it commits the logical fallacy of begging the question”?
Or did you perhaps mean to say something closer to the idea that a logically valid argument must assume premises that, taken together, “contain the sought conclusion”?
Because that is different from saying that a logically valid argument commits the logical fallacy of “begging the question,” since that fallacy means that you literally use the conclusion you want to prove within the argument.
Answer
Of course I am not claiming that every logically valid argument contains the conclusion as one of its premises. What I said is that every logically valid argument begs the question. Whether that is a fallacy or not is not really interesting. There is no point arguing over what is called a fallacy and what is not, because that is semantics.
In my opinion, begging the question is not a fallacy at all in the substantive sense, for the reason I’m about to explain. But even if you call it—or one subtype of it (the one that assumes the conclusion as one of the premises)—a “fallacy,” it is clear that both of these types share two characteristics: 1. An argument that begs the question (in both senses, including what you call a “fallacy”) really is valid. If you assume X, then the conclusion X follows from it necessarily. It’s just not very interesting. 2. Such an argument (in both senses) adds nothing beyond what is already in its premises, and that is in fact where its validity comes from. That is, if I prove some conclusion to you by a logical argument, it was already latent in its premises (either in one of them or in the combination of all of them). That is what I wanted to claim, in order to clarify the framework of the discussion (namely, that whenever there is a disagreement about the conclusion, we will find a disagreement in one or more of the premises). Now you can choose for yourself whether to call that a fallacy or not.
Hope your throat has been cleared 🙂
Discussion on Answer
You are mistaken. In just a random search I did now, see for example here: https://www.kshalim.com/post/petitio-principii-%D7%94%D7%A0%D7%97%D7%AA-%D7%94%D7%9E%D7%91%D7%95%D7%A7%D7%A9-%D7%98%D7%99%D7%A2%D7%95%D7%9F-%D7%9E%D7%A2%D7%92%D7%9C%D7%99
I know it is surprising to discover that every valid argument begs the question, and that is why people argue about it. But they are mistaken. But I’m not going to argue with you about semantics.
As for the principle of causality, I would refer you to David Hume. He was the first to point out that the principle of causality is a priori. And Kant went on and explained that it is synthetic a priori. Note that there is no proof that it is true a priori, only that it is a priori. You can argue and say that it is not true. But you cannot claim that it is a posteriori.
The fallacy of “begging the question” is an informal logical fallacy in which an argument literally uses the conclusion as part of its premises.
Either as a premise in itself, or as an assumption hidden inside one of your premises.
If you are not assuming the conclusion in its entirety in one of your premises, then you are not “begging the question,” by definition.
In the Wikipedia entry “question-begging assumption,” there is an explanation of “circular reasoning,” which is also called “begging the question”:
https://he.wikipedia.org/wiki/%D7%94%D7%A0%D7%97%D7%94_%D7%98%D7%A2%D7%95%D7%A0%D7%AA_%D7%94%D7%95%D7%9B%D7%97%D7%94
“Circular reasoning (Latin: circulus in probando), also called ‘begging the question,’ is an argument that assumes in advance what it seeks to prove. *The assumption is used to prove itself,* a tactic that in its simplest form is not especially persuasive, though it may wear more sophisticated disguises.”
Did you read the examples you gave in the link? Because all of them commit this fallacy clearly.
I’ll give you, specifically, the first argument as an example:
1. “The Bible was written by God” — assumes that God exists, and also that He wrote the Bible.
2. “God tells the truth” — assumes that God exists, and also that He speaks truth.
What you’re trying to prove is “God exists,” but you really do have to assume that as a premise, as part of premise 1, and also as part of premise 2.
So in this case both of your premises “beg the question.”
I apologize, it’s hard for me not to respond to you on this because you are simply objectively wrong…
By definition—which is definitely a matter of semantics, but a matter of semantics that is easy to check.
Thanks for referring me to David Hume.
I want to add: why is it even important to me to argue about this?
First of all, a logical argument that commits a logical fallacy (formal or informal) ultimately fails in its attempt to prove its conclusion because it makes some move within it that is agreed to be invalid.
Usually, for very obvious reasons, in the case of begging the question, because the argument is circular.
There is definitely a reason to talk about what counts as a “fallacy” and what doesn’t.
The simple reason is that if a logical argument commits a logical fallacy, then it is not sound in terms of its structure, and therefore the argument can be rejected purely because it makes a move that is logically invalid.
That is a more objective level that can actually be checked, even before we disagree about the premises.
This is not just semantics; proper logic depends very much on this semantics.
Therefore, to say that “every logical argument begs the question” when that is simply not true at the level of the definition of “begging the question” is a very serious error in logic.
It means entire arguments that fail purely because of their structure might be accepted as sound when they are not sound.
And yes, an informal logical fallacy is also a fallacy *in the structure* of an argument, in the logic behind an argument.
The fact that technically an argument committing an informal logical fallacy is “valid” in the sense that the conclusion follows from the premises still does not mean the structure of the argument is sound.
Begging the question is a classic example of this kind of logical fallacy: even if the argument is valid, you haven’t “proven” anything.
You are repeating the same thing again, and again you are mistaken. The circularity exists in both cases. There is no defect in circularity. At most, a circular argument is not useful and not interesting, but it does not fail. The argument X and therefore X is valid. There is no substantive defect in it. It is not fundamentally different from the argument about Socrates. How is it different from the argument ” ‘X and Y’ and therefore X”?
Let me formulate it differently. Look at the argument about Socrates. Instead of presenting the two usual premises, I will present a completely equivalent set of premises. I’ll start by unpacking “All human beings are mortal.” This is really just a shorthand way of saying that so-and-so is mortal and such-and-such is mortal and Socrates is mortal and Muhammad is mortal, etc. You understand that this is a formulation entirely equivalent to the two premises at the base of the argument. But then you immediately see that this is begging the question even by your definition. You are simply assuming that Socrates is mortal. This is circularity in the strictest sense and meets all your requirements. Therefore the second formulation, which is logically equivalent to it, is also circular and begs the question.
In short, every valid logical argument—and specifically the Socrates one—is circular, and according to your wording you have not proved anything with it. But it seems to me we’ve exhausted this. It really is just semantics.
It’s easy to show that the Socrates argument does not beg the question.
I’ll repeat what I wrote in the first response:
Premise 1, by itself, does not beg the question, because it could be that Socrates is not a human being, and then he would not necessarily be mortal.
Premise 2, by itself, does not beg the question, because it could be that Socrates is a human being and yet not mortal.
Only when premises 1 and 2 are assumed together does the conclusion follow.
It is simply wrong to say that this argument “begs the question,” because none of its premises contains the assumption that Socrates is mortal.
You do not use your conclusion, “Socrates is mortal,” at any stage of the argument.
What you do assume is several different premises from which the conclusion follows when taken together, and of course that means that if you assume all the premises together you *also* get the conclusion.
But this is not a circular argument, because you do not need the conclusion in its entirety at any stage of the argument.
In other words, you do not need to assume the conclusion as a premise in your argument for the argument to be valid.
That is exactly the difference between “if X then X” — circular.
and “if X and Y then Z” — not circular, because Z is not a datum you rely on in the argument.
We’ve really exhausted this; it’s a shame you’re arguing about something that really is Logic 101…
You could simply have written goto the previous messages. Einstein already said what he thinks about someone who repeats the same thing over and over and thinks that in the end it will work.
I explained why Socrates satisfies exactly your definition. I gave the argument x and y therefore x (not therefore z, as you quoted), and you didn’t address it. The examples you gave, like God wrote the Bible, are exactly of that type. And Socrates is also like that, because there too you unpack the premises. When you do that for Socrates, you get the same thing.
You are simply captive to the confusion I encounter many times among those who run into the circularity that exists in every valid argument for the first time, and you refuse to part with this formal and unimportant distinction. But definitions are no problem. Let everyone define it however he wants. That’s it.
An argument of the form “if X and Y then X”
of course begs the question; that’s what I said earlier, in the example *you gave* and that I quoted…
I thought you got confused because that’s exactly what I said, sorry for the mistake.
The Socrates argument simply is not of that form…
It is of the form “if X and Y then Z.”
I’d like you to show me how the Socrates argument is of the form “if X and Y then X,” and how I get the same thing,
but I believe you’re already tired of me haha
I get it, I’m stubborn. It’s okay if you don’t want to answer me anymore; don’t feel like I’m forcing you to respond.
I refuse to part with this formal, well-defined distinction precisely because it is important…
It’s a shame that you don’t see that.
Sorry for butting into the discussion, but I didn’t manage to understand the difference between the two arguments you presented in the question.
Regarding the triangle, you wrote that the argument begs the question because saying that something is a triangle necessarily entails that it has three sides, and that is based on premise 1. Meaning, here too the reason the second premise begs the question is because of the first premise.
Alternatively, if we examine the argument about Socrates, on the basis of the first premise, the second premise already contains the conclusion. Just as the definition of a triangle is that it has 3 sides (which is really the first premise), so too one could say that the definition of a human being is that he is mortal.
No, wait.
Premise 2 begs the question because by definition, a triangle has 3 sides.
That is, to say that X is a triangle is like saying that X is a polygon with 3 sides.
In order to assume that X is a triangle, according to the definition of the term triangle, you have to assume that X has 3 sides.
Otherwise it won’t be a triangle.
So premise 2 literally assumes your conclusion in full, and you use it in the argument to prove your conclusion.
You can write the argument like this:
Premise 1: X is a triangle.
Conclusion: X has 3 sides.
And the argument is completely valid, by the definition of the term triangle.
If you define the term ‘human being’ to mean ‘mortal,’ then yes, the above argument begs the question.
But since the term ‘human being’ is usually defined as ‘a species of mammal with certain characteristics,’ as long as being ‘mortal’ is not built into the definition of the term ‘human being,’ you are not assuming that all human beings are mortal by definition.
If the definition of the term human being explicitly stated that it is mortal, then you would indeed have begging the question here.
Exactly like the definition of triangle, which is that it has 3 sides.
Here is another argument:
Premise 1: For all triangles, the sum of the angles is 180 degrees.
Premise 2: X is a triangle.
Conclusion: The sum of X’s angles is 180 degrees.
A valid and proper argument, which does not beg the question, since premise 1 by itself does not assume what the sum of X’s angles is; since the definition of the term triangle is ‘a polygon with 3 sides,’ there is no reference to angles here at all.
Here is a completely valid argument:
Premise 1: All human beings are bald.
Premise 2: Socrates is a human being.
Conclusion: Socrates is bald.
The above argument does not beg the question.
Not only because premise 1 is simply not true, but because the conclusion does not follow from either premise separately.
In other words, you are not using ‘Socrates is bald’ as some sort of datum inside the argument itself within the premises; rather, you are assuming 2 different premises whose truth together also necessitates the conclusion.
By the way, it’s easy to see that the argument is valid, because if premise 1 were true and premise 2 were true, the conclusion would necessarily be true.
Gil, thanks for all the examples. I learned a lot from you about logic (I got here through your post on the atheism forum). In my opinion you should join the atheist hotline’s team of guides.
Gil, I think (not 100% sure) that Michi’s point was that when you assume “all human beings are mortal,” you are already assuming that Socrates is mortal precisely because he is a human being. Therefore, begging the question. You are simply saying “all human beings are mortal”—that’s the premise and that’s the result.
So every logical argument clearly begs the question. Logical arguments are ultimately just a description of reality in the form of a syllogism. And there is nothing wrong with that, as Michi said. At worst, the arguments simply add nothing new.
In every logical argument, the same thing happens: you assume the conclusion as a question-begging premise, and some people accept that assumption and some do not. The dispute is always about the premises, not about the conclusion. Ever think about why? Exactly for this reason—because the conclusion is already hidden in the premises! No wonder people argue about them. Because you are begging the question. You are simply assuming the conclusion *even if you split it into two different premises*—as in the example with Socrates.
Now maybe your next question comes up: if all arguments beg the question, why not simply assume any conclusions we want? For example, “God exists” or “God does not exist.” And that is exactly the point: we really do this in every logical argument, until eventually we arrive at a factually correct premise and the argument becomes valid.
In short, in the Socrates argument you are splitting things across premises, when each premise is in any case true because the second premise is also true, and then the conclusion necessarily follows, so we have assumed the conclusion. Logical arguments are ultimately just a description of reality based on factual and accepted premises.
When does begging the question become problematic and count as a fallacy? Very simply: when the premise is unfounded and untrue. Then the conclusion also will not follow from it.
What you were arguing about here was terminology, nothing more than that.
Thank you very much for the answer!!!
What you’re describing is not called “begging the question,” and of course this is a matter of semantics, but logic requires us to deal with semantics; if we are not precise in our definitions, we may commit the fallacy of “confusing language.”
Even so, in practice I really agree with the idea you were trying to convey (now that I understand what you meant), just not with the way you say it.
The term “begging the question” has a specific definition in the context of logic, and you’re simply using it to talk about something that is not “begging the question” according to the definition of the term in that context.
By the way, it is an “informal” fallacy precisely because on the technical level the argument is logically completely valid, but it fails because you really do have to assume the conclusion you’re trying to prove within the argument, so in effect the conclusion follows purely because you assumed it.
Personally, I think this can open the door for people to say, “So what if I beg the question? Every valid argument begs the question,” when their argument is literally using its own conclusion inside the argument.
Unrelated, I have a follow-up question to your discussion with Aviv:
How do you prove that the principle of causality is true at the level you were talking about? Meaning, not just empirically.
If you can point me to something to read on the topic, I’d be very happy.