1.01 - izvirni znanstveni članek
Danilo Šuster
Filozofska fakulteta Univerze v Mariboru
Begging the Question - Proper Justification or Proper Conversation?
Petitio -kršitev pravil upravičenja ali pravil dialoga?
Aristotel ponuja dve veliki pojasnili zmote petitio (krožno sklepanje): napačna poteza v argumentativnem dialogu ali pa epistemska napaka (premisa je manj "utemeljena" od sklepa). V drugem primeru vedenja o premisi ne moremo imeti neodvisno od vedenja o sklepu. Dialektični pristop uporablja pojem sprejetja (angl. cmmittmenf)\n se sklicuje na normativnost pravil dialoga. Predlagam hibridni model, ki temelji na teoriji F. Jacksona: razlog za sprejetje obvez v dialogu je epistemski.
Ključne besede: argument, peto, epistemski model, dialektični model, sprejetje.
»I think, therefore I am.« Descartes has begged the question here, because when he said »I think,« he'd already implied »I am« (or how else could he think?). Yet his fallacy continues to persuade people, over three hundred years later.1
1.
An influential paper starts with a remark (Sinnott-Armstrong, 1999: 174): »No topic in informal logic is more important than begging the question. Also, none is more subtle or complex.« Not just informal logic, reasoning and argumentation, the circles of justification and inference in general - all these topics have to address this fallacy, if the fallacy it is. The issues are complex and the literature is large and growing, but there remains something deeply puzzling when arguments are discredited as being circular, or question-begging, or being a petitio principii - the labels are different, but I will assume in this paper that they describe more or less the same phenomenon.
1 Mission Critical - an old web page, probably no longer active, though still accessable: http://missioncritical.royalwebhosting.net/part2/circular.html, January 8th, 2021.
AnaliZA 01 2020
37
Danilo Suster
Let me start with a quick textbook definition: to beg the question is to assume the truth of what one seeks to prove, in the effort to prove it, according to Copi and Cohen (1991: 102). In a later edition petitio is described as an informal fallacy in which the conclusion of an argument is stated or assumed in any one of the premises (Copi et al., 2013: 140). Textbooks will often say that an argument begs the question if any doubt about the conclusion would equally infect the premises. A question immediately arises: is this a logical fallacy at all? Mill reminds us that any valid argument is such that if one doubts the conclusion, one ought to doubt the premises. So every valid argument begs the question? Some tricks are needed to avoid this generally unwanted conclusion. Hurley states that the fallacy of begging the question is committed whenever the arguer creates the illusion that inadequate premises provide adequate support for the conclusion, for instance: »Clearly, terminally ill patients have a right to doctor-assisted suicide. After all, many of these people are unable to commit suicide by themselves.« But the argument »No dogs are cats. Therefore, no cats are dogs« commits no fallacy because no illusion is created to make inadequate premises appear as adequate (Hurley, 2013: 163). No illusions, no disappointments according to the Japanese proverb?! I nevertheless think that the »no dogs« argument remains a disappointment if the argument is used to remove anyone's doubt about the conclusion. And the first argument will remain fallacious even after omitting seducing words »clearly« and »after all« .
Since perfectly valid arguments can beg the question, the mistake involved cannot be of a logical kind. A very radical diagnosis was proposed by Robinson (1971). If the fallacy is not one of deductive logic then, as matter of fact, no arguments beg the question. »The prohibition of begging the question is not a law of logic, nor a maxim of good scientific method. It is merely a rule of an old fashioned competitive game.« And he adds: »There are only two proper ways of condemning an argument. One is to say that the conclusion does not follow from the premises. The other is to say that you do not accept the premises as true ... Begging the question appears to be neither of these. So it is not a proper accusation.« (Robinson, 1971:114)
Let me first note that the accusation of begging the question is really often misused so as to cover any assumption found problematic in the argumentative exchange. You disagree with your opponent? You find her conclusion implausible? There must be something wrong with her premises, most likely she already assumes the conclusion in one form or another. Augustus De Morgan has already remarked that:
38
Begging the Question - Proper Justification or Proper Conversation?
There is an opponent fallacy to the petitio principii which, I suspect, is of more frequent occurrence: it is the habit of many to treat an advanced proposition as a begging of the question the moment they see that, if established, it would establish the question. [...] Are there not persons who think that to prove any previous propositions, which necessarily leads to the conclusion adverse to them, is taking an unfair advantage? (quoted by Walton, 1991: 257)
I have actually heard an academician argue in the following way: »So you argue for, say, decriminalization of (soft) drugs. I disagree strongly! Let me see, hmm ..., your conclusion follows from the results that show that the strict policy of prohibition is unsuccessful. But we all know that you are usually biased in your interpretations of data. Your premise is thus unacceptable, petitio!«
Following Robinson one could say that this accusation merely marks a disagreement with one of the premises. But how about circularity proper as the criterion? It is often said that »begging the question« occurs when the same proposition is asserted twice, both as a premise and as a conclusion. But it would be too restrictive to limit the defect to cases when the conclusion appears explicitly as a premise for that would allow to avoid the charge by a mere reformulation of premises. Surely, the following argument begs the question: »Since firefighters must be strong men willing to face danger every day, it follows that no woman can be a firefighter.« One could try to amend the criterion by saying that the defect resides in the fact that the conclusion appears implicitly in the premises. And what does that mean? Do one's premises include implicitly one's conclusion if they cannot all be true unless the conclusion is? Well, then all valid arguments beg the question, »so it is not a proper accusation,« just as Robinson has said.
The equivalence conception of the fallacy is that some premise of the argument is equivalent to the conclusion (explicitly or implicitly). I think that a more lax, dependency conception is the right way to go (cf. Walton 2006). We should start with a probative function of an argument - the premises provide evidence of a kind that gives the respondent a reason to accept the conclusion. The conclusion depends, justificatorily, on the premise, the »flow of inference« is the »flow of evidence« and it has to go from the premise to the conclusion. The premises are used to remove the respondent's doubt about the conclusion. But where it is also required that (some kind of) inference be made in the other direction, from the conclusion to the premise, the argument begs the question. In a petitio the premise depends on the conclusion in a way that undermines the probative purpose of arguing. Justification is an epistemic notion and I am inclined to accept broadly epistemic criteria of begging the question and arguments in general. Within a broadly epistemic theory the principal goal of argumentation is, roughly, to induce belief or elicit a reasoned change in view. In the case of petitio arguing is epi-stemologically unsuitable for the purpose of proving the conclusion in that parti-
AnaliZA 01 2020
39
Danilo Suster
cular discussion. But we should also acknowledge that wrongness ofpetitio stems from a pragmatic and contextual notion of how an argument is used in an argumentative dialogue. Can the two approaches be reconciled?
The split between »dialectical« and »epistemic« approaches to begging the question goes back to Aristotle. In the Prior Analytics (64b 33) petitio is characterized as the attempt to prove what is not self-evident by means of itself. But demonstration proceeds from what is more certain or better known: if a man tries to prove what is not self-evident by means of itself, he begs the original question (64b 37). To beg the question in this sense is to violate the epistemic principle of the priority in knowledge of the premises over the conclusion in a demonstration. Aristotle uses the same terminology in the framework of the dialectical or conversational account of the fallacy. In the Topics the account is set in terms of contentious disputation between two or more parties. Begging the question is said to occur when the party who is supposed to be arguing for a certain thesis asks to be granted the thesis as a premise to be conceded by his opponent. In Sophistical Refutations Aristotle discusses petitio in the context of »arguments used in competitions and contests« where one party has the task of proving a proposition (the »question« to be proven) to another party (165b 12). To carry out his task, the first party will have to ask the second party to grant or concede premises that the first party can use in his argument. However, if the first party were to ask (beg for) the very conclusion as a concession, without doing the required work of proving it, then he would be »begging for the question at issue.« This would not be allowed, because the prover would be avoiding the task of proving the proposition at issue.
I have sketched some main dilemmas about petitio: there are deflationary views (begging the question is not a fallacy at all) and inflationary views (any argument that contains an unwarranted presupposition or inadequately supported premise is a petitio). Moreover, how to avoid the verdict that every valid argument begs the question? And which is better, the epistemic or the dialectical approach?
Let me start with The Bank Manager Example (»a staple of many textbooks«):
Manager: Can you give me a credit reference?
Smith: My friend Jones will vouch for me.
Manager: How do we know he can be trusted?
Smith: Oh, I assure you he can.
In this dialogue one person is supposed to vouch for the reliability of the other. The reliability of the vouchee is in doubt and some secure source is needed to re-
40
Begging the Question - Proper Justification or Proper Conversation?
assure this doubt. But if the reliability of the voucher is questioned, the reliability of the vouchee cannot be used to reassure this doubt, because it is itself in doubt, in the first place (cf. Walton, 2006: 248). One could say that Smith falsely presents his reliability as an accepted starting point in a dialogue. Circular arguments, in general, are fallacious because they violate normative rules of dialogue which demand consensual starting points according to modern defenders of pragma-dialectical approach (van Eemeren and Grootendorst, 2004).
Argumentation, according to this approach, is understood as an effective means of resolving a difference of opinion in accordance with discussion rules acceptable to the parties involved. Fallacies are illegitimate moves in a given discourse context, they violate rules of a critical discussion. There are eight rules or commandments and the sixth rule (the starting point rule) states (van Eemeren and Grootendorst, 2004: 193):
Discussants may not falsely present something as an accepted starting point or falsely deny that something is an accepted starting point.
By falsely presenting something as a common starting point, the protagonist tries to evade the burden of proof; the techniques used for this purpose include advancing argumentation that amounts to the same thing as the standpoint (petitio prin-cipii). Circular arguments are fallacious because they violate normative rules of dialogue which demand consensual starting points.
Begging the question according to this account is a dialectical mistake that depends on the violation of some general rule of dialogue. Let me use another evergreen (two sentences from this passage are given as an exercise by Copi and Cohen, 1991: 110):
To know or tell the origin of the other divinities is beyond us, and we must accept the traditions of the men of old time who affirm themselves to be the offspring of the gods - that is what they say - and they must surely have known their own ancestors. How can we doubt the word of the children of the gods? Although they give no probable or certain proofs, still, as they declare that they are speaking of what took place in their own family, we must conform to custom and believe them. [Plato, Tymaeus 40d-e]
Like the Bank Manager Example this case is on the borderline with some other fallacies. It has to do with trustworthiness, so it relates to ad verecundiam and ad hominem that also have to do with matters of the reputation of an agent and the trustworthiness of a source. On the face of it we have inductive reasoning »from tradition«:
AnaliZA 01 2020
41
Danilo Suster
We must conform to custom and believe the men of old time.
The men of old time declare that they are speaking of what took place in their own family.
The men of old time must surely have known their own ancestors. The men of old time affirm themselves to be the offspring of the gods.
So,
The men of the old time were the children of the gods.
Not blatantly fallacious (the power of custom carries at least some probative force, one usually knows one's ancestors), but the problem is created by the very inductive nature of the argument: is it not possible for the premises to be true and the conclusion false? Could men of old times err? But then we get the rhetorical question: How can we doubt the word of the children of the gods? The possibility of error is removed, but the question is begged. Here is a reconstruction proposed by Iacona and Marconi (2005, 33):
1	We cannot doubt the word of the children of the gods. P
2	[The men of the old time were the children of the gods.] P (?)
3	We cannot doubt the word of the men of the old time. 1, 2
4	The men of the old time affirmed themselves to be
the children of the gods.	P
5	The men of the old time were the children of the gods. 3, 4
The intended conclusion (5) is just the missing premise (2) which the arguer inserts in order to make the argument valid. Iacona and Marconi describe the case as a rhetorically efficient covert petitio, an invalid argument which hints at reasoning that can only be carried out once a premise is added, but the integrated argument is used to make patent petitio (in this case one of the premises being equivalent to the conclusion). But Marconi and Iacona ignore the larger inductive setting. To make the example even more vivid I will consider a dialogue between Achilles (supposed to be the son of nymph Thetis) and the Tortoise inspired by the classic Lewis Carroll (1895).
Tortoise: Can you, Achilles, state some credible evidence for your being the son of the sea nymph Thetis?
Achilles: I am speaking of what took place in my own family and I surely know my immediate ancestors.
Tortoise: How do we know you can be trusted?
42
Begging the Question - Proper Justification or Proper Conversation?
Achilles: Oh, I can assure you. How can you doubt the word of a child of the gods?
In The Bank Manager Example when Smith gives Jones as a reference who can endorse him, the argument presumes that Jones is trustworthy because of his reputation, or because he is a member of a profession that is trustworthy, etc. But if one person endorses another, then the second cannot be used as a reference to endorse the first. Similarly, Achilles is trustworthy because he is a member of the club where membership requires special qualities. And how do we know that he really is a member of this club? Because members of the club do not lie. Membership in the club guarantees trustworthiness, but we have to accept his trustworthiness in order to count him as a member of the club.
On the conversational approach in the event of a difference of opinion one discussant puts forward a standpoint (»I am a child of gods«) and the other discussant calls that standpoint into question (»A son of gods, I really doubt that!«). The discussants are not in agreement on the acceptability of a certain standpoint. If any attempt to resolve this difference of opinion by means of a regulated discussion is to have any chance of success, it is necessary for the discussants to adopt a number of propositions accepted by both parties as their starting points. The best method for judging cases of begging the question is to keep track of the commitments in the dialogue in relation to the theses to be proved by both sides. So let us include this dimension:
Achilles (the proponent): I really am a child of gods.
Tortoise (the respondent): Force me, argumentatively, to accept the conclusion as true. Pick up your note-book again and kindly enter mutually accepted propositions.
Achilles: Proceed! And no tricks this time - I will be on watch!
Tortoise: Tell me whether you agree with the following proposition: There are people.
Achilles: How trivial! Tortoise: So true, write it down. Achilles: Done. Tortoise: There are gods. Achilles: How trivial! Tortoise: So true, write it down.
Achilles: Done. And you should also add: Gods are trustworthy.
AnaliZA 01 2020
43
Danilo Suster
Tortoise: Well, they have their weaknesses, but I am, for the sake of discussion, prepared to grant that gods do not lie.
Achilles: I will also write down: We must conform to custom.
Tortoise: Granted, provisionally, for the sake of discussion.
Achilles: According to the custom the word of the children of the gods should not be doubted.
Tortoise: Accepted.
Achilles: One knows what took place in his childhood in his own family.
Tortoise: Well, not always, in my case there were those hundreds of eggs and ....
Achilles: All right, all right, I just want to rationally convince you, the Euclidian iron logic (hmm ...) is not really required. So let us say: A person (human or god-like) usually knows the nature of his mother and father. And I certainly know them!
Tortoise: Usually is not always. Maybe you err?
Achilles: You can trust my word.
Tortoise: Hmm, let me have a look in your note-book. Sorry, trust is not enlisted.
Achilles: But I told you I know my parents and Thetis, a divine being, is my mother!
Tortoise: Let me have a look ..., sorry, not on the list! We only have: a person (human or god-like, and why not a turtle-like?) usually knows her mother and father. Maybe you do know your mother, but you wrongly think she is a goddess?
Achilles: But look, you agreed that gods are trustworthy. Tortoise: Indeed.
Achilles: And children of gods are speaking of what took place in their own family.
Tortoise: Well, I believe that children usually are honestly speaking of what they think took place in their own family, but . .
Achilles: So how could you possibly doubt my word, a word of a child of the gods?
Achilles has taken on the burden to fulfill a probative function by putting forward an argument to the Tortoise. The initial standpoint (»Achilles is a son of goddess«) cannot, of course, form any part of the list of propositions that are ac-
44
Begging the Question - Proper Justification or Proper Conversation?
ceptable to both parties, otherwise there would be no difference of opinion, nothing to argue for. In the case of begging the question, the error that is made is that the protagonist (intentionally or unintentionally) makes use of a proposition that, as he can know beforehand, is not to be found in the list of propositions that are acceptable to both parties. As I developed the case, the proponent (Achilles) is at least implicitly aware of this dialectical rule and tries to present the contested proposition as a conclusion (»so, ...«). But we get the feeling that the dialogue could go on and on. The Tortoise is never presented with the premise he can commit to, independently of his doubts about the conclusion.
Instead of an accepted starting point one might speak, more generally, about acceptances in a conversation (cf. Hazlett 2006). A proposition p is accepted in conversation c when all the speakers of c are allowing utterances that express p into c (i. e. they are prepared to meet utterances expressing p with positive behavior such as saying 'Yes' and nodding, and are not prepared to meet utterances expressing p with negative behavior such as saying 'No' and shaking their heads). A proposition p is in question in c when neither p nor not-p is accepted in c. On this Gricean analysis an argument then begs the question to the extent that it violates the so called Submaxim of Relation (Hazlett, 2006: 356):
You are allowed to use as premises only what you can reasonably expect your audience to accept based on their contribution to the conversation and whatever else you know about them qua speaker.
Walton (2006), inspired by Hamblin, uses the notion of a commitment in dialogue. What an arguer is committed to is what she has gone on record as saying in a dialogue, judging from the textual evidence in the case. Through arguing we commit ourselves to propositions and the commitments incurred are what we are concerned with when evaluating the argument. The fallacy of begging the question is then a failure that relates to how the respondent's commitments are used by the proponent's argument attempt. To rationally convince the respondent to come to accept the conclusion that she doubts, the proponent needs to use an argument with premises that consist only of propositions that the respondent is committed to, or is prepared to accept, independently of the proposition to be proved by the proponent.
Standpoints, commitments or acceptances - begging the question, according to this approach, is a matter of what your audience has indicated a willingness to accept in a dialogue. But where do commitments come from? Why are some of them included on the initial list and the others are not? What is the rationale for the relevant rule? Let us join Achilles and the Tortoise once again:
AnaliZA 01 2020
45
Danilo Suster
Tortoise: Why do I doubt your word? In words of a famous philosopher yet to be born, we have to consider, whether it be more probable, that you should either deceive or be deceived, or that the fact of your divine origin should really have happened. And I think that the falsehood of your testimony is less miraculous.
Achilles: So you do not believe me? You think I am not trustworthy? Tortoise: I just considered our list and since trust was not enlisted, I ... Achilles: You call me a liar?
Tortoise: I never said that! Look, let us inspect our list ... Achilles (shouts): Beware the rage of Achilles!
Tortoise (hides her head under the shell): But we agreed to use rational argumentation as an effective means of resolving a difference of opinion in accordance with discussion rules acceptable to both parties involved.
(Suddenly a noise is heard, knocking on the door and shouting): Achilles, son of Peleus! Agamemnon, son of Atreus, the king of Argos is asking you to join Achaeans in the just war to effectively and ultimately resolve a difference of opinion with the Trojans.
Tortoise (quietly): I knew they were still slightly primitive and do not understand the subtleties of logic in spite of their philosophers. (Sighs and adds). As a matter of fact, for the next two thousand years the argument of power will prevail over the power of argument.
Achilles (overhears her): I really wonder how could somebody who has doubts about modusponens complain about my logical abilities.
The Tortoise gives epistemic reasons for his doubt in the contested commitment. Divine origin would count as miraculous for him, but, according to Hume (1748/2000, 86-87): »a miracle is a violation of the laws of nature; and as a firm and unalterable experience has established these laws, the proof against a miracle, from the very nature of the fact, is as entire as any argument from experience can possibly be imagined.« Given that he doubts the conclusion, the evidence provided by Achilles for the contested premise is no evidence for him. In general, I think that given the probative purpose of the argument commitments and acceptances are not arbitrary but subject to epistemic constraints.
46
Begging the Question - Proper Justification or Proper Conversation?
Why is a certain proposition on the index prohibitorum as a commitment in an argumentative dialogue? Well, because it does not advance the issue. Arguments have different functions but I will assume that the main use of an argument is probative - an argument should »prove« its conclusion (make it knowable, rationally believable or acceptable, remove doubts ..., - Walton 2006 makes this clear). A probative function of argument uses the premises to provide evidence that gives the respondent a reason to accept the conclusion. The question-begging commitment does not remove the respondent's doubt about the conclusion: the reasons offered by the proponent for the contested premise are ineffective against this respondent. We can agree that when one begs the question in a dialogue one breaks the rules of conversation, but the rationale for these rule is a prohibition of an epistemic deadlock. At least when fulfilling the probative function is required by the type of conversation the participants are engaged in.
I would thus argue for a hybrid model of begging the question, integrating epi-stemic dimension into a dialectical setting. The most promising approach which also provides a plausible solution to the problem of deduction (all valid arguments are petitio) was offered by Jackson (1984). On Jackson's view arguing has two basic functions. One is the 'teasing out' function, which basically amounts to proving something to those who already had the resources to do it themselves in the first place - showing them how propositions that they already believe entail other propositions which they had not previously recognized as consequences of what they already believed:
The act of propounding an argument may have brought a half-buried piece of information to the surface, may have alerted me to the relevance of certain facts to my final concern, or may simply have enabled me to see how to get it altogether, so as to make transparent what I want to know (Jackson, 1984: 27).
I think that the »teasing out« function explains the initial puzzlement about the usefulness of Cogito. How about petitio proper? Here we have to consider the second function of arguing, the evidence-borrowing. In presenting an argument, speakers advertise themselves as having a certain sort of evidence for the premises which the audience may not possess. Hearing the argument enables the audience to 'borrow' that evidence, thereby coming to have justification for the premises, and a fortiori for the conclusion. This purpose of arguing has to do with the new information conveyed by the selection of the premises. Consider the valid argument: »Fred is an expert tax lawyer. Expert tax lawyers are wealthy. Therefore, Fred is wealthy«. Anyone who doubts that Fred is wealthy should doubt that Fred is an expert tax lawyer. Suppose the respondent has doubts about the conclusion. Still, the evidence backing the premise that Fred is an expert tax lawyer might be something »new« for her, so to speak (say, she bases her initial
AnaliZA 01 2020
47
Danilo Suster
doubts on his modest life-style) and not undermined by the evidence offered in the argument. The argument is not question-begging for this respondent (Jackson, 1984: 34). This, in brief, offers a solution to the problem of all valid arguments being a petitio.
In order to understand the structure of petitio we have to go beyond the explicit premise/conclusion structure and consider the evidential support for the premises. It is just too simple to say that the very conclusion of an argument is fallaciously assumed in one of the premises or that the premises presume, openly or covertly, the very conclusion that is to be demonstrated. The argument begs the question for the addressee S if one of the premises, P, which is supported (according to the speaker/proponent) by evidence E, S who antecedently doubted the conclusion would adopt assumptions against the background of which the evidence E would not support P. And an argument is »begging the question proper« when one propounds »an argument such that any (sane) audience which was in doubt about the conclusion would have background beliefs relative to which the evidence provided by propounding the argument has no impact« (Jackson, 1984: 35). This might be the case with »the children of the gods« argument, since any rational audience who antecedently doubted the conclusion (»Achilles is the son of gods«) would have background beliefs relative to which the evidence provided (Achilles is trustworthy because gods are trustworthy) has no impact.
Or perhaps not? The notion of »sane« audience is notoriously elusive. There is a difference between saying that the evidence on display is no evidence for that particular audience (say the Tortoise) and saying that the evidence is no evidence for any (sane, rational, normal) audience. Would anyone who sanely doubted the conclusion have background beliefs relative to which (s)he would not regard all of the evidence implicitly offered for the premises (scripture? clergy?) as evidence? The way the dialogue was presented one might say that some new evidence does »pour« in, after all (the first-person testimony - »I am speaking of what took place in my own family«). This evidence is ineffective against the Tortoise and her Humean scepticism, but it might still carry some evidential weight for a different, not necessarily insane audience.
In any case, what makes an argument question-begging depends on argumentative features of the context in which it is proposed. There are rules against including certain premises (commitments) in the argumentative dialogue. But the rationale for such rules is ultimately epistemic. Let me test this proposal against what looks to be a very clear exposure of the conversational model. Late David Lewis in his recently published correspondence discusses van Inwagen's version of the consequence argument for the incompatibility of free will and determinism. Lewis adopts the compatibilist's position and objects to one of the premises in the argument which is »awfully close to an explicit denial of compatibilism.« And then he writes in a letter to van Inwagen (Lewis, 2020: 90):
48
Begging the Question - Proper Justification or Proper Conversation?
To beg the question isn't to argue from premises; to argue from premises that fail to be neutral with respect to the conclusion; to argue from premises that your opponent is free to deny; or to argue from premises that your opponent probably would deny. (It isn't a fallacy to argue, or to argue va-lidly, or to argue with a free man, or to argue with a stubborn man.) The most promising account of begging the question supposes that there's a dialogue going on and doesn't directly apply to a monologue (or -graph): there's a status of being under challenge, there are ways for a participant to give something that status, and there's a rule against using challenged premises, the point of which rule presumably is to avoid going on and on in a dead-locked condition.
Two parties are involved in dialogue and one is trying to prove something to the other according to some procedural rules. The proponent presents a premise which would automatically come under challenge as soon as the conclusion was in dispute. There is a rule against using such a premise. What might the rule be? Do not use a contested premise if you want to avoid going on and on in a deadlocked condition. But why would the discussion go on and on? Well, the contested premise is supported (according to the proponent) by the evidence such that the respondent who doubts the conclusion has background beliefs relative to which this evidence has no impact. Perhaps a premise under challenge is very close to being just a reformulation of the conclusion or the premise is connected to the conclusion in an epistemically illegitimate way. The rule prohibits endlessly restating the old and ineffective evidence in new clothing.
Epistemic advancement is the name of the game: an argument uses the premises to provide evidence of a kind that gives the respondent a reason to accept the conclusion. I thus agree with Walton (2006: 238) that the failure to provide such reasons, a probative failure, is at the root of the fallacy of begging the question. But I think that a hybrid model, based on Jackson's combination of pragmatic and epistemic components provides the most plausible approach in analyzing the phenomena of begging the question. Not every valid argument begs the question, because we have to consider the role of potentially new evidence revealed by the proponent's selection of premises. In most general terms in the case of petitio a statement (P) is made that presupposes or depends upon the point at issue (Q) (cf. Fogelin, 1987: 95). We can agree that reasoning is embedded in a conversation that represents a goal-directed dialogue (to establish whether Q is the case) that the questioner (proponent) and respondent are taking part in. The proponent should offer an argument with premises that provide evidence that supports the conclusion that the respondent doubts or disagrees with. And there is a rule against using certain premises. But the rationale for the rule should be spelled out in epistemic terms: when Q is in dispute a certain sort of evidence is discredited. Some new or extra evidence has to be introduced in the argumentation. The respondent must be presented with premises (and thereby the sources of evidence)
AnaliZA 01 2020
49
Danilo Suster
she can accept, independently of her doubts about the conclusion (such that her doubts do not block the »flow« of evidence). One could say that begging the question is a violation of some general rule of dialogue. Yet the rationale for this rule is epistemic - sorry, this commitment is unacceptable because of its episte-mic »corruption«, your premise depends on the conclusion in an epistemically illegitimate way. Proper (argumentative) conversation requires proper (epistemic) justification.
Begging the Question - Proper Justification or Proper Conversation?
Since Aristotle there are two main approaches in the explanation of begging the question (petitiO)'- a dialectical mistake (an improper move in an argumentative dialogue) and an epistemic mistake. According to the latter begging the question is committed when the premises of an argument cannot be known independently of knowing the conclusion of the argument. Dialectical approaches use the notion of a commitment (acceptance, standpoint) and rules of dialogue as their basis. I propose a hybrid model, inspired by Jackson- the ratoinalefor introducing commitments and rules is epistemic.
Keywords: argument, begging the question, epistemic model, dialectical model, commitments.
Literature
Carroll, L. 1895. »What the Tortoise said to Achilles.« Mind 4, 278-280.
Copi, I. M. and Cohen, C. 1991. Introduction to Logic 8th. New York: McMillan.
Copi, I. M., Cohen, C. and McMahon K. 2013. Introduction to Logic 14th. Harlow: Routledge.
Fogelin, R. 1987. Understanding Arguments. New York: Harcourt Brace Jovanovich.
Hazlett, A. 2006. »Epistemic Conceptions of Begging the Question.« Erkenntnis 65, 343-363
Hume, D. 1748 et seq. An Enquiry Concerning Human Understanding, Ed. Tom L. Beauchamp. New York: Oxford University Press, 2000.
Hurley, P. J. 2013. A Concise Introduction to Logic 12th. Cengage Learning.
Iacona, A. and Marconi, D. 2005. »Petitioprincipii: What's wrong?« Facta Philosophica 7, 19-34.
50
Begging the Question - Proper Justification or Proper Conversation?
Jackson, F. 1984. »Petitio and the Purpose of Arguing.« Pacific Philosophical Quarterly 65, no. 1 (January 1984): 26-36.
Lewis, D. K., Beebee, H. and A. R. J. Fisher, eds. 2020. Philosophical Letters of DavidK. Lewis: Volume 1: Causation, Modality, Ontology. Oxford, New York: Oxford University Press.
Robinson, R. 1971. »Begging the Question.« Analysis 31, 113-117.
Sinnott-Armstrong, W. 1999. »Begging the Question.« Australasian Journal of Philosophy 77, 174-91.
van Eemeren, F. H. and Grootendorst, R. 2004. A Systematic Theory of Argumentation. The pragma-dialectical approach. Cambridge: Cambridge University Press.
Walton, D. N. 1991. Begging the Question: Circular Reasoning as a Tactic of Argumentation. New York, Greenwood Press.
Walton, D. N. 2006. »Epistemic and Dialectical Models of Begging the Question.« Synthese 152, 237-284.
AnaliZA 01 2020
51