ISSA Proceedings 2002 – The Pragmatic Dimension Of Premise Acceptability
We hold that one factor determining whether or not a premise is acceptable is its cost, more precisely the cost of taking that statement as a premise.This thesis requires some clarification.When critically evaluating an argument purportedly giving us good reason to accept its conclusion, we are taking the role of a challenger in a simple dialectical exchange. The person who put forward the argument is the proponent. His role is to advance an initial claim together with reasons discharging the burden of proof making that claim itself incurs together with any burdens raised the by subsequent premises he puts forward or questions of their adequacy to support the conclusion he alleges they support. Our role as challengers is to raise those questions, to point out that there are specific burdens to be discharged or questions to be answered. We may do this overtly, if we are in a critical conversation with the proponent, or implicitly, should we be considering the proponent’s argumentation in the form of an argument as product. Here we note what burdens have been raised and whether they have been discharged. This dialectical exchange is an example of what Walton calls an asymmetrical persuasion dialogue. See (1989, pp. 11-12).
The question for us as challengers then is whether from our perspective a claim which the proponent has advanced raises a burden of proof or whether there is a presumption for it. We judge this from our perspective, since our awareness of the dialectical situation on the whole gives us information relevant to determining this issue. For example, we may be aware that a proponent’s claim is a matter of personal testimony or expert opinion in an area where the proponent has expertise. We may not be aware of any reason to hold that the proponent’s competence is questionable in this case – that he may be deceived by a perceptual illusion or that his recent scientific work has been criticized for sloppiness – or that his integrity is compromised, such as his speaking from vested interest. Depending on the statement the proponent is putting forward, such information may be germane to recognizing rightly whether we should recognize a presumption for the proponent’s claim or whether we may rightly ask him to provide evidence for it.
Beyond these epistemic conditions concerning the presumptive reliability of the source of a premise, the issue of cost is a factor in determining premise acceptability. The concept is easily illustrated. Keep in mind that the source of a premise need not be an interlocutor other than oneself. My own belief-generating mechanisms may propose a claim for acceptance. The clock in my bedroom makes a loud tick at the time when the alarm would have gone off, were the alarm turned on. I need nothing more to tell me that it is time to get up. One morning I hear what I take to be that tick and get up. Should I be asked to justify my action, the statement that the clock has just ticked would be a basic premise of my argument. Yet I glance at the clock and see that it is an hour earlier than I expected. I had not heard the clock but my radiator expanding because the furnace is now sending up steam. But what are the consequences of my accepting my mistaken belief? They are pretty minimal. My rest has been disturbed for only a few minutes. The incident quickly disappears into the mists of memory. On the other hand, I may be a juror who has just heard personal testimony from one, but only one, witness that he had seen the accused stab the victim, who later died from these wounds. No evidence has been presented that the witness is perceptually compromised in this case or that he may be speaking insincerely. Should I accept the witness’s statement that the accused stabbed the victim as a premise on which to convict of capital murder? If my fellow jurors concur, that could result in terminating a human life – not so trifling a consequence as unnecessarily getting up for a minute an hour early. Here is a statement whose acceptance involves a very significant cost. Although there is a presumption for the general mechanism of coming to hold a belief on personal testimony, given this cost should I accept that statement on the personal testimony of just this one witness?
How may we understand the concept of cost that we have been intuitively employing in this discussion? Following Clarke in (1989), we define first the concept of the cost of an action or state of affairs in general as a binary relation between an action, activity, or state of affairs and a person:
A has a cost for X if and only if X has an aversion to A.
(Compare Clarke, 1989, p. 79). Clarke presents also a ternary comparative relation:
A is more costly than A¢ for X if and only if X has a greater aversion to A than to A¢. (Compare Clarke, 1989, p. 79).
This makes the cost of an action or state of affairs relative to a given person. X might have an aversion to A while Y has no aversion to A. So A will have a cost for X but no cost for Y. If cost is understood in this relational way and cost of acceptance should be a factor in premise acceptability, then acceptability becomes relativized not just to the epistemic position of the challenger but to the challenger’s desires and aversions. But is this consequence acceptable? Suppose X and Y are aware of the same evidence pertaining to a statement p. Should p be an acceptable premise for X but not for Y simply because X has no aversion to taking p as a premise or to the consequences which accepting p may bring about while Y has some such aversion? Should a premise be acceptable for one and not for the other on the basis of their differing attitudes toward accepting p or the consequences of that acceptance?
We need not however define cost in this relativizing way. Why should X have some aversion to A? Presumably, either X finds that A itself has intrinsic disvalue or leads to a state of affairs B which has intrinsic disvalue. But intuitionists such as Ross (1930) have shown that in virtue of possessing certain properties or features, states of affairs are objectively prima facie intrinsically good or bad. For Ross, involving pleasure, knowledge, virtue are prima facie intrinsically good making features of a state of affairs. Their opposites make a state of affairs prima facie intrinsically bad. (See 1930, pp. 134-39.) These states of affairs may be constituents of complex facts or wholes, which may affect their actual intrinsic value. Pleasure may be prima facie intrinsically good, but taking pleasure in the pain of others is not actually intrinsically good. When viewed in the light of the morally relevant wholes to which such states of affairs belong, we may speak of them as being objectively intrinsically good or bad simpliciter. Surely if a state of affairs A were intrinsically bad and X were cognizant of the badness, or of the factors on which that badness supervened, X should have an aversion to A. Hence, we may define cost objectively in terms of intrinsic disvalue.
Now an action or state of affairs can either involve intrinsic disvalue in itself or lead to some further state of affairs B which has intrinsic disvalue. This motivates the following definition:
Where A is an action, activity, or state of affairs, by the cost of A, we mean the amount of intrinsic disvalue of A itself together with the intrinsic disvalue of any consequences B¢ of A.
We may analogously define the benefit of A objectively[i]:
Where A is an action, activity, or state of affairs, by the benefit of A we mean the amount of intrinsic value of A itself together with the intrinsic value of any consequences B¢ of A.
The intrinsic disvalue of A includes the intrinsic disvalue of the effort required to perform A together with the loss of intrinsic value of any benefits we forego in performing A. (Clarke refers to the latter as the opportunity costs. See 1989, p. 79)
In the preceding definition, A ranges over actions or states of affairs in general. But we are interested in the cost of one type of action or activity, that of accepting a statement as a premise. Now it is easy to appreciate that the intrinsic disvalue of accepting a statement p may differ, depending on whether p is true or false. If it is true that Jones stabbed Smith, and should all the jurors accept that he did, a consequence could be their all voting to convict Jones of Smith’s murder and Jones’ facing a capital sentence. This obviously involves the intrinsic disvalue of significant pain (at least psychological) to Jones and the intrinsic disvalue of the termination of human life. But if Jones is guilty, one could argue that the punishment is deserved, that pain or unhappiness here is being meted out in proportion to vice or the viciousness of his action. But it is intrinsically good that happiness be proportioned to virtue and thus that punishment be proportioned to vice. But now suppose that Jones did not stab Smith, even though the one witness testifies that he did. Suppose all the jurors again accept that Jones stabbed Smith on the basis of this testimony and vote to convict. Their acceptance now has the further intrinsic disvalue that Jones is about to be unfairly, unjustly punished, that unhappiness will not be proportioned to vice in this case. For just this reason, the intrinsic disvalue of accepting that Jones stabbed Smith differs in these two cases.
Since our concern is with premise acceptability, unless the evidence for a premise is something to which we have direct or internal access (a self-evident truth of reason or of introspection), the question arises of whether we should risk accepting the premise on the evidence before us even if it is false, or should seek further evidence bearing on the premise. Hence, we have two actions here whose costs can be weighed against each other – the cost of the action of accepting a premise when that premise is false or mistaken versus the cost of the action of seeking further evidence. Does the cost of obtaining testimony from a further witness or of obtaining other pertinent evidence outweigh the cost of accepting that Jones stabbed Smith should that statement be false?
This motivates what Clarke calls the pragmatic condition for premise acceptability. As a first approximation, we can say that if the cost of mistakenly accepting p outweighs the cost of obtaining further evidence, then p is not acceptable on the basis of the evidence e proffered at this point. Clarke formulates this preliminary version of the pragmatic criterion this way:
For X to be justified in accepting a proposition p relative to evidence e as true the cost of acquiring additional relevant evidence e¢ must be higher than the cost of acting on the basis of p that would be incurred if p were to later prove mistaken. (Clarke, 1989, pp. 80-81).
Why should this condition be called “pragmatic”? Why should we say that the issue of the cost of accepting or conceding a statement raises the issue of a pragmatic dimension of premise acceptability? For Clarke, pragmatism is “a theory that claims that the standards used in justifying acceptance of a proposition as rational must include reference to individual or community purposes.” (Clarke, 1989, p. ix). Hence pragmatism insists that “a necessary condition for the acceptance of p” involves “the fulfillment of interests and purposes to which this acceptance is related.” (Clarke, 1989, p. 73). We have already indicated how speaking of the aversions (and thus implicitly of the purposes) of a given individual introduces an unacceptable element of subjectivity into the analysis. But we have also indicated how we can avoid this element of subjectivity by defining cost not with respect to the aversions of a given individual but with respect to the intrinsic disvalue to which a given action or state of affairs leads. The intrinsic goodness of a state of affairs is a reason for acting to realize that state of affairs. Likewise, the intrinsic badness or disvalue of a state of affairs is a reason for aversion[ii]. Hence, by connecting acceptability with cost defined in terms of intrinsic disvalue, we are connecting acceptability with a reason for action and thus maintaining a connection with purpose.
The pragmatic criterion, as formulated, seems well motivated. It certainly captures our intuitions in the contrasting cases we have been considering. Now when thinking that I had heard the clock tick, and before getting up, I could have checked my watch on the table beside my bed to add its testimony to what I have perceived. But why should I seek such corroboration before accepting my belief that the clock had just ticked as a premise for my action of getting up? What value would be jeopardized by my getting up which needs to be safeguarded by ensuring this corroboration? Might my insisting upon having such corroboration before acting betray epistemic scrupulosity, an irrational fear of being in error, of making a mistake? If I were to insist upon this as a general policy, might it not be more trouble – checking my watch does involve some inconvenience – than what it is worth – avoiding mistakenly getting up on occasion? The situation is completely different where I am serving as a juror in the trial for capital murder. The prosecutor’s finding a second eyewitness and the court’s receiving that testimony would involve expending some time and effort, thus involving cost. But surely it would seem that the cost of obtaining testimony from another witness is less than the cost of mistakenly accepting that Jones stabbed Smith.
Our criterion also handles the intuitions behind Blair’s illustration of the pragmatic requirement in (1995, 197):
If one’s child’s life depends on a claim’s being true, and time and resources allow, then one wants to know that the undefended premisses supporting that claim are true. If all that is at issue is finding one’s way to the sea from Amsterdam on a lazy afternoon, then an undefended premise in an argument supporting the recommendation of one particular route need be no more than plausible for it to be adequate.
Suppose when all is said and done that we did not take the most efficient route to the sea from Amsterdam on that lazy afternoon, even though the recommendation was from a presumptively reliable witness. What value was lost? Was that loss obviously greater than the loss we would have incurred had we sought testimony from some further witness? On the other hand, if a premise were false and our accepting it thus mistakenly would lead to the loss of our child’s life, that loss would certainly be greater than any inconvenience we might encounter in attempting to secure evidence for that premise or independent corroboration for it. Seeking corroborating testimony for the best route from Amsterdam to the sea seems scrupulous, but not seeking evidence for a premise whose mistaken acceptance could cost the life of one’s child.
Gaining further evidence need not involve just seeking corroborating testimony, as our discussion so far might suggest. Does a certain object belong to you? Is it yours[iii]? That question might be settled by a cursory perceptual glance or by a more reflective perceptual gaze. In either case, perception is the belief-generating mechanism and there may be a presumption of warrant for that source. But does it matter for acceptability whether the belief has been generated through a perceptual glance or gaze? That depends upon the cost of mistakenly accepting that the object belongs to you versus the cost of examining the object more closely. If I incorrectly identify a simple pencil as mine, what will be the cost? Why should I scrupulously examine a simple wooden lead pencil to determine whether it is mine? What cost of mistakenly identifying it as mine could outweigh the inconvenience of this anxious checking? But if I incorrectly identify a Stradivarius violin as mine, one found in someone else’s possession who is accused of stealing it, the cost of mistakenly accepting that statement could be significant – the person could be convicted of a serious crime and deprived of liberty for a significant amount of time. That the pencil is mine is acceptable on the basis of a perceptual glance, but that the Stradivarius is mine is acceptable only on the basis of a careful perceptual inspection. Indeed, more than a perceptual gaze may be necessary. One may need to check that various criteria have been satisfied, coming to believe these propositions through perception, and infer from them that the Stradivarius violin is mine. Nonetheless the cost of carefully examining a Stradivarius to determine whether it is mine is less than the cost of wrongfully convicting someone of stealing it.
Returning to our previous examples, we anticipate an objection to our discussion thus far. Checking my watch to see if it is time to get up does not seem to involve expending much effort, apparently less than what my getting up itself involved. Did my accepting that the clock had ticked really conform to the pragmatic criterion? On the other hand, suppose a second witness had independently testified that she had seen Jones stab Smith. Suppose a third witness could be identified. Should I insist that this third witness also be deposed before accepting that Jones stabbed Smith? Is the cost of receiving testimony from this third witness less than mistakenly accepting that Jones stabbed Smith? But would we not ordinarily consider corroborating testimony from two independent witnesses sufficient? Again, suppose I had sufficient evidence supporting the claim upon which my child’s life depended, but not a deductively valid argument from incorrigible premises. Suppose further evidence was available. Would I be wrong in accepting that claim before taking account of this further evidence? In each case, the answer depends on probabilities. How often when I take what I hear to be a loud tick from my clock early in the morning am I mistaken? When two witnesses independently give corroborating testimony, how likely is that corroborated testimony false? If I have inductively strong evidence for a claim, how likely is it that the claim is nonetheless untrue? We must incorporate these probabilities into the formulation of the pragmatic criterion.
This brings us to what Clarke identifies as the expected cost of an action or state of affairs. As the calculation of expected utility or expected value involves the product of the return of a given possible outcome with its probability, so expected cost is a function of the probability of the consequences of an action or state of affairs together with their intrinsic disvalue. (Compare Clarke, 1989, p. 81). We do not compare the simple cost of gathering additional evidence with the simple cost of mistakenly accepting a proposition, but the expected cost of gathering additional evidence with the expected cost of mistaken acceptance. This motivates the refined formulation of the pragmatic condition:
X is justified in accepting a proposition p relative to evidence e as true only if the expected cost of acquiring additional relevant evidence e¢ is higher than the expected cost of acting on the basis of p which would be incurred if p were to later prove mistaken (Clarke, 1989, p. 82).
As Clarke points out, we shall ordinarily be quite certain that acquiring additional evidence will incur certain costs. Hence in practice the cost of gathering further evidence does not differ much from the expected cost. The probability that a mistake could occur could vary distinctly from case to case. If my auditory sense perception is presumptively reliable, then the probability should be low that if I hear what I take to be my clock’s loud tick, I am mistaken in believing that the clock has ticked. Given this low probability, the expected cost of my mistakenly accepting that the clock had ticked might very well be lower than the expected cost of checking my watch. On the other hand, there is some non-negligible probability that one witness could be mistaken in the testimony he gives or that he might be testifying disingenuously. Given this probability and the significant disvalue of wrongly convicting Smith, the expected cost of mistakenly accepting that Jones stabbed Smith might be far greater than the expected cost of obtaining testimony from a further witness. But should two witnesses independently give corroborating testimony, the probability that both were unreliable would seem to be much lower than for either singly. Is the expected cost of mistakenly accepting their mutually corroborating testimony greater than the expected cost of obtaining testimony from a third witness? If my evidence constitutes the premises of an inductively strong argument for a certain claim, it would seem that the probability of that claim’s being false would again be low. The expected cost of mistakenly accepting a claim as a conclusion of a strong inductive argument might very well be less than the expected cost of supplementing the premises of that argument.
Several objections still remain. How do we determine or come to know the probability that a possible consequence of a certain action or state of affairs will come about? Likewise, if we cannot assign some numerical value to the cost of a consequence of some action or state of affairs, how can we determine the product of that cost with the probability of the consequence coming about? If we cannot readily determine these values, then it seems we cannot determine the expected cost and thus the pragmatic criterion would be inapplicable generally. We reply by invoking Aristotle’s wisdom in the Nicomachean Ethics,
Our discussion will be adequate if it achieves clarity within the limits of the subject matter. For precision cannot be expected in the treatment of all subjects alike, any more than it can be expected in all manufactured articles (Aristotle, 1962, 5).
In ordinary life, we estimate costs and probabilities intuitively. As Clarke puts it, “We rely on rough, intuitive judgments of our degrees of want or aversion towards consequences of our actions and rough estimates of the probabilities of these consequences coming about” (Clarke, 1989, p. 80). In ordinary life, we do not need to determine numerical values for probability and cost to grasp estimated costs sufficiently to apply the pragmatic criterion.
That we have defined the cost of an action A not with respect to the subjective aversion of the agent towards A but with respect to the objective intrinsic disvalue of A together with its consequences does not affect the point we need to make here, that we may rely on rough, intuitive judgments of degree of cost. For the sake of argument, let us agree with Ross that there are three basic types of intrinsic goods – pleasure, knowledge, and virtue, with loss of pleasure or pain, lack of knowledge or mistaken belief or acceptance, and loss of virtue or vice as the opposites. If, in a given case, cost involved just one of these types of intrinsic value, could we make a rough, intuitive judgment of its degree? Can we make such judgments of degree of intrinsic value based on the intensity and duration of pleasure or pain? Can we make such judgments based on the extent of knowledge gained and the depth of its explanatory power, or the extent to which a body of propositions contains mistaken statements of fact or erroneous principles of explanation?
Virtue, for Ross, is intimately connected to motivation, where the desire to do one’s duty qua duty is the highest desire. Actions proceeding from virtuous desires are themselves virtuous. The desire to obtain pleasure for oneself is morally indifferent but an action motivated by such a desire which excludes the doing of one’s duty or some other virtuous action is selfish and morally bad. Is it not clear that the greater the extent of virtuous motivation and the less the extent of selfish motivation the greater the extent or amount of positive intrinsic value, an extent which might again be estimated in a rough and ready way? Might we not also estimate the extent of intrinsic disvalue for balances of selfish motivation over virtuous motivation? It seems straightforward that in all three cases, we can make rough, intuitive judgments of degree of intrinsic value. Suppose now in a particular instance that cost involves a combination of these basic values. This we expect is typical of acting on the basis of p where p is mistaken, especially where p is a hypothesis. Not only will accepting a mistaken general hypothesis have intrinsic disvalue in itself, as Clarke points out it may lead to the loss of various sorts of intrinsic value. We know that discovering and accepting certain hypotheses have led to the devising of beneficial applications. If some mistaken hypothesis had been accepted instead, these benefits might not have come to light. That would be part of the opportunity cost of mistakenly accepting that hypothesis. Further costs may be involved. If a hypothesis is accepted, it may be used in the testing of further hypotheses. But if the hypothesis is mistaken, these tests may be fatally flawed and the effort expended in carrying them out wasted. Opportunities for increasing knowledge would be missed. Thus if we are to speak of estimating amount or degree of intrinsic disvalue, we must be able to consider the basic types of intrinsic value and disvalue together to arrive at an overall judgment.
In comparing different types of intrinsic value, our intuitions may indicate that virtue transcends other types of value or their combinations, and likewise vice, loss of virtue transcends all others in disvalue. Comparing virtue with pleasure in general, Ross holds his intuitions indicate that “no amount of pleasure is equal to any amount of virtue, that in fact virtue belongs to a higher order of value” (1930, 150). Should one strive for virtue or pleasure? “It seems clear that, viewed in this way, pleasure reveals itself as a cheap and ignoble object in comparison with virtue.” (Ross, 1930, 151). Clearly, a cruel disposition is a vice, but suppose one takes pleasure in one’s cruelty. Could that pleasure ever be intense and enduring enough so that its goodness would outweigh the badness of the disposition and the state of affairs be intrinsically good on the whole? (Compare Ross, 1930, 151). As virtue always ranks above pleasure, so it always ranks above knowledge. “When I ask myself whether any increase of knowledge, however great, is worth having at the cost of a wilful failure to do my duty or of a deterioration of character, I can only answer in the negative.” (Ross, 1930, 152).
We expect, however, that in most cases of estimating intrinsic cost, we shall be dealing with a combination of mistaken acceptance, loss of opportunities for knowledge, loss of pleasure, and encountering certain forms of inconvenience and thus of pain. But our intuitive examples at the beginning of this paper illustrate that comparisons and intuitive overall estimates of these values can be made. Hence, we may meaningfully speak of making rough intuitive judgements of degrees of intrinsic value, where this may involve a combination of types of intrinsic value.
There is a further objection we must address. By supposing that our challenger were a member of the jury, we invested her accepting the witness’s testimony with special consequences. Her accepting that Jones stabbed Smith can be a premise for her action to vote to convict, which will have such grave consequences for Jones if the other jurors concur. But what if our challenger were not a juror and could in no wise affect the outcome of this legal proceeding? How then could she act on her accepting that Jones stabbed Smith? Would this mean that the cost of her acceptance even if mistaken is nil and thus that she need not seek further evidence? Does this mean that in matters over which we have no control, the amount of evidence upon which to accept a premise is a matter of indifference, that we never need seek further evidence? This would seem distinctly counterintuitive.
We reply first that the phrasing of this objection suggests that the pragmatic criterion, which is a necessary condition for acceptability, has been confused with a sufficient condition. We have not said that if the expected cost of acquiring additional evidence for p is higher than the expected cost of acting on the basis of p were p mistaken, p is acceptable for X, but rather only if this condition holds is p acceptable. Additional specifically epistemic factors are required for a sufficient condition for acceptability. Should X be asked to accept p on the basis of some argument, that argument must be cogent. Should p be a basic premise, that the source generating the belief that p vouches for p – whether that source be an interlocutor or one of X’s belief-generating mechanisms – must satisfy certain epistemic conditions including being presumptively reliable. That X’s accepting p would have little expected cost does not mean that X is justified in accepting p. At best it means that if the epistemic conditions are satisfied, X need not seek further evidence.
Suppose however that those conditions are satisfied. Does this mean that X need not seek further evidence if the expected cost of X’s own particular acting on p should p be mistaken be less than the expected cost of X’s own particular seeking further evidence? I believe that intuitions may differ on this question. Clearly, whether or not a statement is acceptable for a given individual depends on the pertinent evidence of which that individual is aware. The same statement p may be acceptable for X but not for Y, given their different bodies of evidence. Likewise, one might want to say, should the expected cost of X’s accepting that p be significant but the expected cost of Y’s accepting that p be minimal, then X has a greater responsibility to seek further evidence. Acceptability then is relevant not only to one’s evidence but also to the expected costs for which one is personally responsible.
A consequence of this view is that X and Y may possess the same or comparably strong bodies of evidence for p, but p may be acceptable only for X and not Y. Juror X’s accepting that Jones stabbed Smith on the basis of only one eyewitness report is not justified, while Y’s acceptance is, where Y is simply attending the court proceedings. But should not the gravity of the expected cost of X’s accepting that p signify the seriousness of this issue for anyone, including Y? That X’s accepting p has significant expected cost means that everyone should accept p on the basis of evidence e available to him or her only if the expected cost of gaining additional evidence outweighs that expected cost of X’s acceptance. Expected cost is not relativized to an individual. That the consequences of X’s mistakenly accepting that p has a certain expect cost is a factor in the expected cost in general or for everyone, not just for X. This position is already reflected in the wording of the pragmatic criterion. Notice that it does not read that X is justified in accepting a proposition p on evidence e only if X’s expected costs of acquiring additional information are greater than X’s expected costs of mistakenly acting on p. Rather, it is the expected cost of acquiring additional information versus the expected cost of mistakenly acting on p. We are talking here about general expected costs, the expected costs of people in general either seeking further evidence or accepting that p. X’s being justified in accepting that p indicates general acceptability. Should the general population include jury members whose vote could convict Jones of a capital crime, the expected cost of their mistakenly voting to convict Jones is part of the expected cost of mistakenly proceeding on accepting that Jones stabbed Smith. In assessing the cost of accepting p on e, one could ask what would happen if everyone else did the same.
We submit, then, that the pragmatic criterion frames a necessary condition for premise acceptability. For a sufficient condition, there must also be what we call a presumption of warrant for a statement p from the challenger’s perspective. What this epistemic conditions entails, however, is the topic of another presentation. (See Freeman, 1995).
[i] This contrasts with Clarke’s subjective definition. See (1989, p. 79).
[ii] Audi points out this connection between intrinsic value and reason for action in (1997). See. p. 248. We develop this point in Chapter Nine of our essay, Warrant, Presumption, Acceptability: An Epistemic Approach to Basic Premise Adequacy (under review).
[iii] This adapts Clarke’s discussion in (1989, p. 75).
Aristotle (1962) Nicomachean ethics (Martin Ostwald, trans.). Indianapolis: The Library of Liberal Arts.
Audi, R. (1997). Moral knowledge and ethical character. New York and Oxford: Oxford University Press.
Blair, J. A. (1995). Premise adequacy. In F. H. van Eemeren, R. Grootendorst, J. A. Blair, and C. A. Willard (Eds.), Analysis and evaluation. Proceedings of the third ISSA conference on argumentation volume II (pp. 191-202). Amsterdam: Sic Sat.
Clarke, D.S., Jr. (1989). Rational acceptance and purpose: an outline of a pragmatist epistemology. Totowa, NJ: Rowman & Littlefield Publishers, Inc.
Freeman, J.B. (1995). Consider the source: One step in assessing premise acceptability. In F. H. van Eemeren, R. Grootendorst, J. A. Blair, and C. A. Willard (Eds.), Analysis and evaluation. Proceedings of the third ISSA conference on argumentation volume II (pp. 203-209). Amsterdam: Sic Sat.
Ross, W.D. (1930). The right and the good. Oxford: Oxford University Press.
Walton, Douglas N. (1989). Informal logic: a handbook for critical argumentation. Cambridge: Cambridge University Press.