I appreciate what you are trying to do here. We don't want to say that any valid arguments is question-begging because its conclusion is contained in the conjunction of its premises. But I wonder if the criteria you suggest to distinguish question-begging from non-question-begging arguments makes it too easy to avoid the charge. I mean, if the mere possibility that someone might be swayed by the argument is sufficient to avoid the charge, that is a very low standard. And if the argument in question is for the most part dialectically useless because most people who antecedently rejected the conclusion have no inclination to accept a key premise, is the mere fact that we can find an accountant in Cleveland who is swayed by the argument that significant? Shouldn't the arguer, who is after all affirming their conclusion on the basis of their premises, still recognize that they have failed to make their case and that they have, at least in relation to their opponents, begged the question? Maybe you would say that in such a case, the arguer is assuming too much but has not begged the question. That's fine, I guess. I don't want to get into a mere verbal dispute (there are too many of those in Philosophy). But if we define "begging the question" narrowly, we may need to be careful to fill out our taxonomy of argumentative vices to include so many arguments that are flawed in ways that tend to be characterized as begging the question. I could mention many examples, but I don't want to pick on anyone.
I would have thought that whether an argument "begs the question" should be an intrinsic property of the argument. It's very different from the extrinsic sociological charge of "appealing to premises that not many people in your local environment currently accept".
That's not to say that "mere possibility" of persuasion is enough. (An argument could be blatantly question-begging, even though a *previously very confused* accountant in Cleveland didn't realize this, and accordingly learned something from it.) I think what we want is to appeal to a standard of "reasonable interlocutors". If a sufficient proportion of *reasonable, philosophically competent and well-informed fence-sitters* could be expected to find the argument informative and persuasive, then it isn't question-begging.
One way to make my substantive point is that referees mistake *themselves* as being the (sole?) target audience of the paper. "I am your opponent, and you haven't persuaded *me*, so your argument is dialectically useless," seems to be the prevailing attitude. I think that's wrong. Referees are not so special. The paper wasn't written just for them, but for the broader philosophical community (which is inevitably far more diverse than they realize). Instead of asking whether they were personally convinced by it, they should ask whether the paper *shed new light* on its targeted terrain.
Of course, it's always possible that the answer is still "no". There are plenty of boring, trivial papers in philosophy. And one way to be boring and trivial is to rely on assumptions that seem broadly unmotivated, such that it's hard to see why anyone would even be interested in what follows from them. But I'm not talking about cases like that.
How would a referee know whether a "sufficient proportion of *reasonable, philosophically competent and well-informed fence-sitters* could be expected to find the argument informative and persuasive"? And why would a fence-sitter find the argument persuasive if a sufficient number of their equally competent philosophical peers reject the relevant premise, especially if the disagreement boils down to a mere difference in intuitions? Such a fence-sitter would be philosophically incompetent.
By assessing the intrinsic a priori credibility of the premises (i.e., the range of credences that a reasonable person could assign)? If you can't do that, you have no business doing philosophy at all. There's no point to deferring to peers who aren't themselves competent at assessing intrinsic credibility. That's just a recipe for arbitrary path-dependent groupthink, not any kind of truth-tracking.
For background, see my post 'Philosophical Expertise, Deference, and Intransigence':
I agree, which is why I am claiming only that, in a case where one's "equally competent philosophical peers" have a different intuition, one shouldn't draw any conclusions based on your own (contested) intuition. It's fine to point out the choices, i.e., reject one of these premises or accept this conclusion, but intellectual humility often requires not affirming a conclusion that rests on an intuition that many of one's epistemic peers do not share. And doing that when all of one's opponents simply don't share your intuition is often dialectically useless in a kind of obnoxious way. (I confess that I have committed this sin many times.) Anyways, I will read the post you linked to. I really enjoy "Good Thoughts" and learn a lot from your posts.
(1) That linked post explains why I don't think we should necessarily give any epistemic weight to peer intuitions/judgments (because we often shouldn't view them as truth-indicative). I also don't think that intellectual humility (of the sort you describe) is a virtue. We should aim to be accurate, not humble. If we had to suspend judgment whenever one's "equally competent philosophical peers have a different intuition", we wouldn't be left able to affirm much at all!
(2) As far as papers are concerned, the structure of the inferential relations into a directional "argument" is somewhat superficial. But it can make the paper more engaging than presenting the logical content as a totally neutral "trilemma" (or whatever) in logical space. Since the logical content is the same either way, I don't really see how the more engaging presentation can be "dialectically useless". My suggestion would be for readers to learn how to extract the logical content despite disagreeing with the directionality. (This ought to be a trivial task for philosophers.)
I think discussions of question-begging are often hampered by a focus on deductive logic rather than bayesian reasoning. We should take into account the independent (im)plausibility of both premises and conclusions, and use deductive relations to update our prior credences into posteriors.
Admittedly, this gets complicated due to our vast, interconnected web of beliefs, but still.
I don't feel that this is persuasive because it hinges on a subjective sense how bad a premise is. If you're going to make that concession then you might as well use the conventional definition of question begging but adopt the norm that a reader asserting "question begging" reflects a difference in the writer and reader's belief in the plausibility of the premises. From this perspective, there is no great need to assert a distinction between contestability and Q-begging - although you could do so to reflect the difference in the intensity of beliefs of writer and the reader. I interpret you as saying Q-begging should reflect a "big" gap in plausibility while contestability should reflect a small gap, but who will respect that norm? Also, is there any conceptual history here to show that Q-begging's meaning has shifted over time? My impression was that it always referred to rejecting the premise and hinged on both parties being good faith.
No, I take myself to be explaining the conventional meaning. (Though I care less about convention and more about inferential role: given that "begging the question" is used to mark an argumentative vice -- or a reason to reject a paper from a journal -- it had better refer to something that is actually vicious in this way.)
It's not about "how bad a premise is", but about how the premise relates (epistemically) to the conclusion: in particular, whether the plausibility of the premise depends upon *already accepting* the conclusion.
Note that you could have an argument with extremely implausible premises that nonetheless is *not* question-begging, because what (minimal) plausibility the premises have does *not* depend upon already accepting the conclusion.
Thank you. It seems you're right, and Aristotle complains about the same overloaded definition.
From Wikipedia:
The Latin phrase comes from the Greek τὸ ἐν ἀρχῇ αἰτεῖσθαι (tò en archêi aiteîsthai 'asking the original point')[7] in Aristotle's Prior Analytics II xvi 64b28–65a26:
Begging or assuming the point at issue consists (to take the expression in its widest sense) [in] failing to demonstrate the required proposition. But there are several other ways in which this may happen; for example, if the argument has not taken syllogistic form at all, he may argue from premises which are less known or equally unknown, or he may establish the antecedent utilizing its consequents; for demonstration proceeds from what is more certain and is prior. Now begging the question is none of these. [...] If, however, the relation of B to C is such that they are identical, or that they are clearly convertible, or that one applies to the other, then he is begging the point at issue. ... [B]egging the question is proving what is not self-evidently employing itself ... either because identical predicates belong to the same subject, or because the same predicate belongs to identical subjects.
"The conclusion [of a question-begging argument] is so transparently contained within the premises that there is no conceivably 'neglected' consideration there to highlight"
Of course, the conclusion of any valid argument must be logically deducible from its premises. In the understanding of this layman, an argument is question-begging if it takes a contestable premise (stated or unstated) for granted. What you're describing sounds more like what's commonly known as a "circular" argument: one in which the conclusion merely restates a premise in different terms.
The relevant relation is epistemic rather than logical: could each premise seem independently plausible or appealing, even to someone who doesn't *already* accept the conclusion. If so, the argument is *not* question-begging.
(This is not the same as logical circularity, since a question-begging premise need not be literally logically equivalent to the conclusion. Again, the relevant form of transparency is instead epistemic.)
"In the understanding of this layman, an argument is question-begging if it takes a contestable premise (stated or unstated) for granted."
That view entails that almost every argument is question-begging. So it's not a good view. There's nothing inherently wrong with taking a contestable premise for granted: almost every good argument in the entire history of philosophy does so. (That's why no interesting philosophical argument ever secures absolutely universal assent.) What matters is instead whether the argument is substantively illuminating, as I've described. Since it is possible (indeed: common) for a contestable argument to still be substantively illuminating, it would be a bad norm to regard contestability as an argumentative vice.
What I meant by "contestable premise" is a premise that is not self-evident as a matter of common knowledge (e.g., Boston is farther from the equator than Miami) or common sense nor readily ascertainable from observation or authoritative sources of information.
To refine what I said about my understanding of the term "question-begging argument," I would not include in that category hypothetical arguments wherein the proponent acknowledges that a premise is contestable but assumes its validity "for the sake of argument" -- i.e., to elicit provisional agreement on some issue.
Can you try presenting an argument for the conclusion that only bad arguments have contestable premises? I can guarantee that you will need to appeal to contestable premises, and so the argument would be self-condemning.
I appreciate what you are trying to do here. We don't want to say that any valid arguments is question-begging because its conclusion is contained in the conjunction of its premises. But I wonder if the criteria you suggest to distinguish question-begging from non-question-begging arguments makes it too easy to avoid the charge. I mean, if the mere possibility that someone might be swayed by the argument is sufficient to avoid the charge, that is a very low standard. And if the argument in question is for the most part dialectically useless because most people who antecedently rejected the conclusion have no inclination to accept a key premise, is the mere fact that we can find an accountant in Cleveland who is swayed by the argument that significant? Shouldn't the arguer, who is after all affirming their conclusion on the basis of their premises, still recognize that they have failed to make their case and that they have, at least in relation to their opponents, begged the question? Maybe you would say that in such a case, the arguer is assuming too much but has not begged the question. That's fine, I guess. I don't want to get into a mere verbal dispute (there are too many of those in Philosophy). But if we define "begging the question" narrowly, we may need to be careful to fill out our taxonomy of argumentative vices to include so many arguments that are flawed in ways that tend to be characterized as begging the question. I could mention many examples, but I don't want to pick on anyone.
I would have thought that whether an argument "begs the question" should be an intrinsic property of the argument. It's very different from the extrinsic sociological charge of "appealing to premises that not many people in your local environment currently accept".
That's not to say that "mere possibility" of persuasion is enough. (An argument could be blatantly question-begging, even though a *previously very confused* accountant in Cleveland didn't realize this, and accordingly learned something from it.) I think what we want is to appeal to a standard of "reasonable interlocutors". If a sufficient proportion of *reasonable, philosophically competent and well-informed fence-sitters* could be expected to find the argument informative and persuasive, then it isn't question-begging.
One way to make my substantive point is that referees mistake *themselves* as being the (sole?) target audience of the paper. "I am your opponent, and you haven't persuaded *me*, so your argument is dialectically useless," seems to be the prevailing attitude. I think that's wrong. Referees are not so special. The paper wasn't written just for them, but for the broader philosophical community (which is inevitably far more diverse than they realize). Instead of asking whether they were personally convinced by it, they should ask whether the paper *shed new light* on its targeted terrain.
Of course, it's always possible that the answer is still "no". There are plenty of boring, trivial papers in philosophy. And one way to be boring and trivial is to rely on assumptions that seem broadly unmotivated, such that it's hard to see why anyone would even be interested in what follows from them. But I'm not talking about cases like that.
How would a referee know whether a "sufficient proportion of *reasonable, philosophically competent and well-informed fence-sitters* could be expected to find the argument informative and persuasive"? And why would a fence-sitter find the argument persuasive if a sufficient number of their equally competent philosophical peers reject the relevant premise, especially if the disagreement boils down to a mere difference in intuitions? Such a fence-sitter would be philosophically incompetent.
By assessing the intrinsic a priori credibility of the premises (i.e., the range of credences that a reasonable person could assign)? If you can't do that, you have no business doing philosophy at all. There's no point to deferring to peers who aren't themselves competent at assessing intrinsic credibility. That's just a recipe for arbitrary path-dependent groupthink, not any kind of truth-tracking.
For background, see my post 'Philosophical Expertise, Deference, and Intransigence':
https://www.philosophyetc.net/2018/02/philosophical-expertise-deference-and.html
I agree, which is why I am claiming only that, in a case where one's "equally competent philosophical peers" have a different intuition, one shouldn't draw any conclusions based on your own (contested) intuition. It's fine to point out the choices, i.e., reject one of these premises or accept this conclusion, but intellectual humility often requires not affirming a conclusion that rests on an intuition that many of one's epistemic peers do not share. And doing that when all of one's opponents simply don't share your intuition is often dialectically useless in a kind of obnoxious way. (I confess that I have committed this sin many times.) Anyways, I will read the post you linked to. I really enjoy "Good Thoughts" and learn a lot from your posts.
Thanks! Two last points to emphasize:
(1) That linked post explains why I don't think we should necessarily give any epistemic weight to peer intuitions/judgments (because we often shouldn't view them as truth-indicative). I also don't think that intellectual humility (of the sort you describe) is a virtue. We should aim to be accurate, not humble. If we had to suspend judgment whenever one's "equally competent philosophical peers have a different intuition", we wouldn't be left able to affirm much at all!
(2) As far as papers are concerned, the structure of the inferential relations into a directional "argument" is somewhat superficial. But it can make the paper more engaging than presenting the logical content as a totally neutral "trilemma" (or whatever) in logical space. Since the logical content is the same either way, I don't really see how the more engaging presentation can be "dialectically useless". My suggestion would be for readers to learn how to extract the logical content despite disagreeing with the directionality. (This ought to be a trivial task for philosophers.)
This essay begs the question in favour of the view that contestable arguments aren’t question begging.
I think discussions of question-begging are often hampered by a focus on deductive logic rather than bayesian reasoning. We should take into account the independent (im)plausibility of both premises and conclusions, and use deductive relations to update our prior credences into posteriors.
Admittedly, this gets complicated due to our vast, interconnected web of beliefs, but still.
I don't feel that this is persuasive because it hinges on a subjective sense how bad a premise is. If you're going to make that concession then you might as well use the conventional definition of question begging but adopt the norm that a reader asserting "question begging" reflects a difference in the writer and reader's belief in the plausibility of the premises. From this perspective, there is no great need to assert a distinction between contestability and Q-begging - although you could do so to reflect the difference in the intensity of beliefs of writer and the reader. I interpret you as saying Q-begging should reflect a "big" gap in plausibility while contestability should reflect a small gap, but who will respect that norm? Also, is there any conceptual history here to show that Q-begging's meaning has shifted over time? My impression was that it always referred to rejecting the premise and hinged on both parties being good faith.
No, I take myself to be explaining the conventional meaning. (Though I care less about convention and more about inferential role: given that "begging the question" is used to mark an argumentative vice -- or a reason to reject a paper from a journal -- it had better refer to something that is actually vicious in this way.)
It's not about "how bad a premise is", but about how the premise relates (epistemically) to the conclusion: in particular, whether the plausibility of the premise depends upon *already accepting* the conclusion.
Note that you could have an argument with extremely implausible premises that nonetheless is *not* question-begging, because what (minimal) plausibility the premises have does *not* depend upon already accepting the conclusion.
Thank you. It seems you're right, and Aristotle complains about the same overloaded definition.
From Wikipedia:
The Latin phrase comes from the Greek τὸ ἐν ἀρχῇ αἰτεῖσθαι (tò en archêi aiteîsthai 'asking the original point')[7] in Aristotle's Prior Analytics II xvi 64b28–65a26:
Begging or assuming the point at issue consists (to take the expression in its widest sense) [in] failing to demonstrate the required proposition. But there are several other ways in which this may happen; for example, if the argument has not taken syllogistic form at all, he may argue from premises which are less known or equally unknown, or he may establish the antecedent utilizing its consequents; for demonstration proceeds from what is more certain and is prior. Now begging the question is none of these. [...] If, however, the relation of B to C is such that they are identical, or that they are clearly convertible, or that one applies to the other, then he is begging the point at issue. ... [B]egging the question is proving what is not self-evidently employing itself ... either because identical predicates belong to the same subject, or because the same predicate belongs to identical subjects.
— Aristotle, Hugh Tredennick (trans.) Prior Analytics
"The conclusion [of a question-begging argument] is so transparently contained within the premises that there is no conceivably 'neglected' consideration there to highlight"
Of course, the conclusion of any valid argument must be logically deducible from its premises. In the understanding of this layman, an argument is question-begging if it takes a contestable premise (stated or unstated) for granted. What you're describing sounds more like what's commonly known as a "circular" argument: one in which the conclusion merely restates a premise in different terms.
The relevant relation is epistemic rather than logical: could each premise seem independently plausible or appealing, even to someone who doesn't *already* accept the conclusion. If so, the argument is *not* question-begging.
(This is not the same as logical circularity, since a question-begging premise need not be literally logically equivalent to the conclusion. Again, the relevant form of transparency is instead epistemic.)
"In the understanding of this layman, an argument is question-begging if it takes a contestable premise (stated or unstated) for granted."
That view entails that almost every argument is question-begging. So it's not a good view. There's nothing inherently wrong with taking a contestable premise for granted: almost every good argument in the entire history of philosophy does so. (That's why no interesting philosophical argument ever secures absolutely universal assent.) What matters is instead whether the argument is substantively illuminating, as I've described. Since it is possible (indeed: common) for a contestable argument to still be substantively illuminating, it would be a bad norm to regard contestability as an argumentative vice.
What I meant by "contestable premise" is a premise that is not self-evident as a matter of common knowledge (e.g., Boston is farther from the equator than Miami) or common sense nor readily ascertainable from observation or authoritative sources of information.
To refine what I said about my understanding of the term "question-begging argument," I would not include in that category hypothetical arguments wherein the proponent acknowledges that a premise is contestable but assumes its validity "for the sake of argument" -- i.e., to elicit provisional agreement on some issue.
Can you try presenting an argument for the conclusion that only bad arguments have contestable premises? I can guarantee that you will need to appeal to contestable premises, and so the argument would be self-condemning.
I don't agree with the proposition that only bad arguments have contestable premises.
It's OK to base an argument on a contestable premise if you acknowledge its contestability and state your reason(s) for relying on it.