Abstract: The paper contributes to the debate about arguments by analogy, especially the distinction between ‘deductive’ and ‘inductive’ analogies and the question how such arguments can be ‘deductive’, yet nonetheless defeasible. It claims that ‘deductive’ and ‘inductive’ are structural, not normative categories, and should not be used to designate argument validity. Based on Aristotle’s analysis of enthymemes, examples, and metaphors, it argues that arguments from analogy are complex arguments that involve inductive, abductive, and deductive components.

Keywords: abduction, analogy, comparison, deduction, enthymeme, example, induction, metaphor, similarity.

1. Introduction
Arguments by analogy have been a much-disputed subject recently. The most controversial issues in that discussion have been whether or not there are different types of analogical arguments, whether they are to be regarded as basically inductive or deductive or as a completely distinct category of argument of their own, whether or not they involve any hidden or missing premises, and whether it is possible for analogical arguments to be deductive and yet defeasible.

Since the mid-1980s Trudy Govier has repeatedly argued in favor of a view that arguments by analogy should best be regarded as a distinct type of argument, and not as a species of either induction or deduction (Govier 1985; 1987; 1989; 2002), by denying that any universalist generalizations need to be included as unstated or missing premises in such arguments. In response to her view, while basically agreeing with her distinction between ‘inductive’ and a priori analogies, Bruce N. Waller (2001) has tried to restate the case for a deductivist reconstruction of the latter, whereas Marcello Guarini (2004) attempted to show that Waller’s reconstruction was unsubstantiated. Fábio Perin Shecaira, in turn, has defended Waller’s deductivist analysis by introducing some modifications (2013, p. 429) and by declaring analogy arguments that do not fit Waller’s schema to be “defective or sub-optimal instances of their kind” (pp. 407-408; 421). In response to the dispute between Waller, Govier, and Guarini on the possibility of ‘deductive’ arguments by analogy, Lilian Bermejo-Luque has recently (2012, and forthcoming) proposed a new unitary schema for arguments by analogy as complex second-order speech acts to explain how such arguments can be ‘deductive’, but nonetheless defeasible. Independently from Bermejo-Luque, but in a way in some respects not dissimilar to her approach, James Freeman, in an analysis of Govier’s distinctions (2013), has insisted on the necessity of the insertion of a ceteris paribus clause and of qualifiers in a priori analogies and defended their status as defeasible a priori arguments.

I will propose an alternative solution. I would myself prefer to view arguments by analogy within a greater range of argument types that derive from comparisons and similarities (see also Doury 2009), including examples, or even metaphors, and analyze them as complex compound arguments that involve various different types of inferences. I further hold that Aristotle’s logic and rhetoric already provides the tools needed for such an analysis of arguments by analogy.

In a first section, I will briefly analyze the main points of disagreement between scholars on arguments by analogy. I will then argue that categories such as ‘deductive’ and ‘inductive’ are structural, not normative categories, and should therefore not be used to designate argument validity (‘conclusiveness’). In a next step, I will sketch the main features of Aristotle’s theories on arguments involving similarities and comparisons, and will finally demonstrate how arguments from analogy can be reconstructed as complex arguments that involve inductive, abductive, and deductive components.

2. Types of analogies
Govier, Waller and Guarini all agree that there are two types of arguments by analogy: one that operates from empirical data and yields only probable conclusions, and one that proceeds from analogies invented ad hoc and allegedly leads to conclusive inferences. The disagreement is on whether or not the latter can therefore be regarded as deductive. Govier calls those a priori analogies. Waller also adds as a third kind what he calls “figurative analogies” (2001, p. 200), that is analogies that do not actually argue for a certain claim, but simply illustrate a statement for the sake of better understanding (see also Garssen 2009). These are not to be regarded as arguments at all. Bermejo-Luque calls those “explanatory analogies” (2012, p. 6), and appears to further add also a non-discursive, “cognitive-exploratory” function of analogies, in which they act as cognitive tools in that they offer a kind of cognitive proposals for making new objects and phenomena more familiar to us. But the emphasis is on the two primarily argumentative types.

Govier’s analysis of a typical ‘inductive’ analogy runs as follows (1989, p. 141):

(1)
1. A has features x, y, z.
2. B has features x, y, z.
3. A has feature f.
4*. Most things which have features x, y, z, have feature f.
5. Thus, probably, B has feature f.

In this reconstruction, the fourth premise “is starred because, the way most arguments by analogy are worded, it would not be explicit in the argument. It would be unstated.” (p. 141). One should note the qualifier “probably” in the conclusion! While she agrees that such arguments may require some inductive generalization, what she sees involved her is “a hasty generalization – typically a generalization from a single case.” (p. 142). Her example is that war and slavery have a lot in common, yet slavery was abolished by citizen action; hence it should be possible to abolish war by citizen action. Typically, in an ‘inductive’ analogy, “the reality and empirical detail of the analogue matter”, and the conclusion “predicts a result for the primary subject” (p. 142). This is why Guarini (2004, p. 166), just as William R. Brown (1989, p. 162), prefers to call them ‘predictive’.

Govier’s master example for what she calls a priori analogy is Judith Jarvis Thomson’s famous example of the desperate violinist that is hooked on another human being for life support (Thomson 1971, p. 48-49; Govier 1989, p. 142), an ad hoc example that was meant to support the claim that a woman that had gotten pregnant from rape had no moral obligation to keep the foetus alive. According to Govier, in an a priori analogy, the analogue is “constructed”, it “can be entirely hypothetical and may, in fact, be positively fanciful.” (1989, p. 142).

Her analysis of such an a priori analogy is slightly different from that of an ‘inductive’ one (p. 144):

(2)
1. A has x, y, z.
2. B has x, y, z.
3. A is W.
4’. It is in virtue of x, y, z, that A is W.
5. Therefore, B is W.

There is no qualifier such as “probably” here, as there was in ‘inductive’ analogy. On the contrary, Govier even suggests that from premise 4’ (which seems to be presupposed in some way) it is only a short step to the universal premise:

4*. All things which have x, y, z are W.

This is what Govier calls a “U-claim”, a universal claim. In the case of the desperate violinist, the ‘U-claim’ would be something like “No-one has an obligation to support at his or her own inconvenience the life of another human being to which he or she has been unvoluntarily linked.” This premise would make the argument deductively valid. But it would also make premises 1 and 3 logically redundant and thus eliminate the analogy as superfluous. And, what is more, it is clearly an overstatement unwarranted by premises 1 and 3. Based on these considerations and on Stephen F. Barker’s objections that it is often “not possible to state a suitable universal premise” and that “the universal premise […] is nearly always more dubious than the conclusion” (Govier 1989, p. 144; see Barker 1965, pp. 280-290), she is inclined to reject such a deductive reconstruction, and to accept at best that some ‘U-claim’ may be implied, but not presupposed by the argument as an implicit premise (1989, p. 148). She argues that the cases are epistemologically prior to the generalization, and that hence a priori arguments by analogy work better directly from case to case rather than by way of a detour via what she calls a U-claim. In fact: “The trick about analogies – and their charm as well […] – is that we are often able to see or sense important resemblances between cases without being able to spell them out exhaustively […].” (p. 148). This is why she postulates for those analogies a special a priori, but non-deductive category.

Waller, by contrast, finds no sufficient reason “for denying the deductive status of such arguments by analogy” (2001, p. 204), just because the U-claim is hard to formulate or not immediately recognizable. He holds that analyzing an a priori analogy “is not a matter of finding the fixed and final universal principle that rightly governs the analogy” (p. 207). Rather, the analogy forces the audience to think hard and reflect upon their own principles and their implications, so that the analogy does not establish the principle, but gets the audience to recognize the principle (p. 208). In this way, while preserving the deductive status of such analogies, Waller on the other hand denies them any inductive power. According to him, “there is not a shred of induction about them.” (p. 201).

In her reply to Waller, Govier criticizes this view and argues that, if the U-claim were in fact implicit as an unstated premise, as Waller claims, it would be much less required from the audience to think so hard to arrive at it (Govier 2002, p. 156). This criticism of Govier’s, however, appears to underrate the cognitive capacities of audiences, which Aristotle acknowledged when emphasizing the role of the audience in supplying unstated premises in enthymemes (e.g. Rhetoric I 2, 1357a17-21).

To overcome this controversy, Bermejo-Luque (2012) intends to construct a unitary structural schema for both ‘inductive’ and a priori analogies by analyzing them as complex second-order speech acts to explain how such arguments can be ‘deductive’, but nonetheless defeasible. Based on a Toulminian analysis of arguments and a linguistic-pragmatic model of interpretation, by laying strong emphasis on ontological and epistemic qualifiers that qualify the inference-claim as well as the analogue and also the connecting warrant, she proposes to reduce the difference between ‘inductive’ and ‘deductive’ analogies to a matter of such qualifiers (pp. 16-22).

Likewise reducing arguments from analogy to a Toulminian warrant structure by switching the order of some premises in Govier’s analytic schema, and thus reducing differences between types of arguments from analogy to an assessment of ground adequacy and the epistemic status of the warrant, Freeman (2013) also insists on the necessity of the insertion of a ceteris paribus clause and of qualifiers in a priori analogies, lest they be subject to counterexamples (pp. 180-183), and defends their status as defeasible a priori arguments. With Shecaira (2013) he shares the belief that synthetic a priori warrants are typical of moral arguments (Freeman 2013, pp. 179-180).

3. Deduction, induction, abduction
Some confusion in this controversy derives from the fact that in discussions of arguments from analogy terms such as ‘deductive’ and ‘inductive’ are oftentimes applied in a normative sense, implying that a deductive argument is equivalent to a logically conclusive argument, the conclusion of which follows with necessity, whereas an inductive one yields only a plausible or probable conclusion (see Bermejo-Luque 2012, pp. 2-3; 4; 21; explicitly rejected by herself in forthcoming, note 4 and section 4). This dichotomy, as Hitchcock (1980, p. 9) points out, can only be exhaustive if one is willing “to label ‘inductive’ all arguments which are not deductively valid.” In contrast to this, pace Hitchcock’s defence of induction and deduction as types of argument validity (p. 9), I would adhere to the view that ‘deduction’ and ‘induction’ are essentially structural categories and should not be employed as normative terms. Based on Aristotelian logic, a deduction (Aristotle’s term for which is syllogismós) would be structurally defined as an inference from a universal rule and a statement about a particular case being an instance of that rule to a particular assertion about that case, as in the famous example: “All human beings are mortal; Socrates is a human being; hence Socrates is mortal.” When cast in syllogistic form, in deductive arguments the middle term is always the subject in one premise, but the predicate in the other. It is easy to interpret this category in a normative sense, since, given that the premises are true, deductive arguments in standard form typically yield conclusive results, and in fact Aristotle himself reserves the term syllogismós to conclusive deductive arguments only (see Posterior Analytics I 1, 24b18-26). But by far not all formally deductive arguments are logically conclusive, as soon as negations and quantifiers get involved. Consider the following: “All human beings are mortal; Fido is not a human being; hence Fido is immortal.” (It is assumed that Fido is a dog). From a structural point of view, this argument is formally deductive; but it is clearly fallacious (and would hence not count as a syllogismós for Aristotle).

Inductive arguments, by contrast, infer from the particular to the universal (Aristotle, Topics I 12, 105a13-16; Posterior Analytics I 1, 71a8-9; Rhetoric I 2, 1356b14-15: “a proof from a number of similar cases that such is the rule”). Such an inductive argument, however, can be obtained by simply switching propositions within a standard deductive argument, such as when from “Socrates is a human being” and “Socrates is mortal” it is inferred inductively (and in this case by chance correctly) that human beings in general are mortal. Aristotle lists such arguments in his taxonomy of enthymemes from signs (Rhetoric I 2, 1357b10-13; Prior Analytics II 27, 70a16-20), but explicitly remarks that this type of argument is defeasible, since it is not properly deductive (Rhetoric I 2, 1357b13-14). In syllogistic interpretation, the middle term takes the subject position in both premises, such as in the following example: “Socrates is a philosopher; and Socrates is bearded; hence philosophers are bearded.” It is easy to see that in such an argument the conclusion will need a qualifier to make it acceptable if not even valid. For it may be perfectly reasonable to say that the argument does prove that some philosophers are bearded, or perhaps even that as a rule philosophers are likely to be bearded. Yet one single counterexample (such as Kant or Wittgenstein) will suffice to refute any general conclusion such as “All philosophers are bearded.”

However, in addition to deduction and induction, there is yet a third conceivable structural type of argument, which is generally termed ‘abduction’. In an abductive argument what is inferred it is the subsumption of a case under a general rule. The middle term in this case takes the position of predicate in both premises. Using again the obvious standard example, from “Socrates is mortal” and “human beings are mortal”, it may be inferred that the most reasonable explanation for the observed fact will be that “Socrates is a human being”. Arguments of that type are also acknowledged as enthymemes by Aristotle (Rhetoric I 2, 1357b17-19; Prior Analytics II 27, 70a20-24). Of course, as Aristotle himself remarks, even if the premises are true, this type of inference will at no point be safe (Rhetoric I 2, 1357b19-21). Indeed, the Socrates in question may as well happen to be a dog or some other animal.

4. Aristotle on arguments by similarity
Aristotle, in his Rhetoric and Posterior Analytics, calls these latter two types of inferences enthymemes, since, even if all premises are true, the conclusion will only follow with a certain probability. But, as we have seen, they are at the same time quite appropriate descriptions of the structures of induction and abduction. But Aristotle also says something else, namely that, just as the enthymeme is the rhetorical variant of deduction, the example (paradeigma) is the rhetorical variant of induction. This, I take it, is as good a description of analogy as any. Whereas in scientific induction a maximum number of examples must be accumulated to make the induction persuasive, in rhetoric – for reasons of brevity – this is mostly reduced to one single example (or very few), but this one example has to be a particularly significant one: “[T]he example is understood as a kind of qualitative induction in which the fewer number of particular references is compensated by the fact that they are plausible in connection with the circumstances and the audience.” (Gabrielsen 2003, p. 350; cf. Bermejo-Luque, forthcoming, section 2, on quantitative vs. qualitative analogies).

In almost identical words, in the Rhetoric as well as in the Prior Analytics, Aristotle states that arguing by example works neither from part to whole (as induction does) nor from whole to part (as deduction does), but from part to part or from like to like, “when both come under the same genus, but one of them is better known than the other” (Rhetoric I 2, 1357b27-30; Prior Analytics II 24, 69a13-16; see Gabrielsen 2003, p. 351). This is exactly parallel to John Wisdom’s description of analogy arguments as “case-by-case” reasoning (Wisdom 1957, cited in Govier 1989, p. 141). Aristotle’s example is that Pisistratus, when he asked for a bodyguard, became a tyrant; hence it is inferred that when Dionysius asks for a bodyguard, he is aiming at tyranny (Rhetoric I 2, 1357b19). How does this example work? According to Gabrielsen’s reading, “a ‘part to part’ example must be perceived as an unpronounced combination of an inductive and a deductive inference.” (Gabrielsen 2003, p. 351). In Govier’s terms, this would clearly qualify as an ‘inductive’ analogy, since the case adduced is taken from the experience of real life, and the generalization drawing on it (“people who ask for a bodyguard, usually aim at tyranny) would typically be used to predict another case.

Aristotle further says that examples may either be taken from reality or may simply be invented (Rhetoric II 20, 1393a28-31). In my view, this is the same distinction as Govier’s between ‘inductive’ and a priori analogies. Invented examples, he adds, are subdivided into comparisons and fables; the examples he offers for the comparison type are in fact quite similar to the standard examples for a priori analogies: it is, he says, as if one were to say that magistrates should not be chosen by lot, since that would be similar to choosing an athlete for a sports competition by lot instead of by his strength, or to choosing any of the sailors for helmsman (II 20, 1393b4-8). The examples/comparisons are in this case clearly invented ad hoc, and in quite fanciful manner so as to highlight the paradox. Fables (also clearly a fictional genre) may be interpreted as extended forms of such a priori analogies.

Even Aristotle’s theory on the metaphor in the Poetics can be adduced here, since it is based on similarities, and also in view of its cognitive and explanatory power (as Bermejo-Luque has observed, 2012, p. 8). Moreover, Aristotle notes that metaphors can be constructed from genre to species, or from species to genre, but also directly from species to species. Interestingly, he mentions a fourth kind, which he calls “by analogy”, the structure of which is that A relates to X just as B relates to Y; hence in this case the analogy consists in the relationship between two pairs of terms (Poetics 21, 1457b7-9).

In a later passage of the Rhetoric (II 25, 1402b13-14), Aristotle states that enthymemes can be derived from four sources: probabilities, examples, infallible signs, and signs; again we find the example featuring prominently among sources for argument. And here Aristotle explicitly adds that we argue from examples, “when they are the result of induction from one or more similar cases, and when one assumes the general and then concludes the particular by an example” (1402b16-18). He thus links examples to the general realm of similarities; and he analyzes arguments by example as a two-step process, in which in a first step a general statement is established by way of induction, and then from there a particular case (the target claim) is again deduced. Hence in his view, arguments from example do argue from case to case, but they do so via a general principle.

5. Another unitary scheme for arguments by analogy
Based on what we can learn about arguments by various kinds of similarities from Aristotle, I would myself propose the following unitary analysis of arguments by analogy: I endorse the view that arguments by analogy are complex arguments that encompass at least two separate argumentative stages. In a first stage, from the analogue case, by way of an argument from example, a general statement is inductively inferred. This is very clearly the case in so-called ‘inductive’ analogies, since in those cases one or more empirically observed examples serve as the starting point. In a subsequent stage, from this general rule another particular case (the target claim) is inferred deductively. But these two steps can’t be exhaustive. In fact, before the deduction to the target claim can be executed, it will have to be made sure beforehand that the target case is at all an instance of that general rule. This, however, will have to be done by an abductive reasoning based on some other characteristics of the target case. So we have actually a three-stage argument. But this abductive stage has mostly been overlooked in recent reconstructions.

Things may perhaps be slightly different for a priori analogy. Look at Waller’s reconstruction of the structure of such arguments (2001, p. 201):

(3)
1. We both agree with case a.
2. The most plausible reason for believing a is the acceptance of principle C.
3. C implies b (b is a case that fits under principle C).
4. Therefore, consistency requires the acceptance of b.

Shecaira observes that Waller’s schema “does not represent analogical arguments simply as deductive inferences, but rather as complexes of two inferences only one of which is deductive” (2013, p. 407; cf. also p. 424). On our account, however, his analysis in fact involves no less than three inferences. For anyone acquainted with abductive reasoning, premise 2 unmistakably evokes one of the most common standard descriptions of abduction (an “inference to the best explanation”, see Harman 1965; Lipton 2001; cf. Wellman’s “explanatory reasoning” as “reasoning from a body of data to a hypothesis that will render them intelligible”, 1971, p. 52; see Freeman 2013, p. 190). But so does premise 3 (a “case fitting under a principle”) for the target case. Shecaira comes very close to this insight, when he repeatedly speaks of principle C as the “most plausible (i.e., the best) reason for believing a” (2013, p. 429), or notes that this move “resembles an inference to the best explanation” (pp. 430; 435), but at no point he gets beyond calling it, rather vaguely, “a non-deductive sub-argument” (p. 453; cf. pp. 409; 430). Yet if Waller’s analysis is valid, it seems to suggest that in the case of a priori analogies the inductive stage is replaced by a second abductive reasoning that subordinates the ad hoc invented analogue to some principle that is already in some way part of the commitment store of the audience (cf. Waller 2002, p. 213).

This would account for the differences most analysts have observed between these two basic types of arguments by analogy. But since we learn from Aristotle that both inductive and abductive reasonings are by their very definition defeasible, because they are always open to refutation by counterexample, this means that no argument by analogy can be ultimately conclusive. This seems to be trivial for ‘inductive’ analogies. The general statement attained inductively in those arguments necessarily needs to be constrained by a qualifier such as ‘probably’ or ‘presumably’, which will render the ultimate conclusion only a probable or presumable one as well. Bermejo-Luque is certainly right in emphasizing the role of those qualifiers (2012, pp. 16-22). But contrary to what most analysts assume, this equally holds for a priori analogy.

Both Waller and Guarini invoke a number of arguments that challenge the conclusiveness of Thomson’s violinist analogy (Waller 2001, pp. 208-210; Guarini 2004, p. 159), to the effect that, even if the analogy as such holds, it may be abductively related to some different moral intuition such as that one is in fact morally obliged to support any other human being’s life at whatever cost. Freeman’s insistence on the necessity of the insertion of a ceteris paribus clause in such arguments, lest they be subject to counterexamples, points in the same direction (2013, pp. 180-182). And both Guarini and Bermejo-Luque point out that, since all similarities allow for a more or less, arguments by analogy must also allow for degrees of strength (Guarini 2004, p. 159-160; Bermejo-Luque 2012, pp. 16; 23).

Freeman (2013, p. 192) ultimately argues that the epistemic distinction between arguments based on a priori and a posteriori warrants is more fundamental to a general theory of arguments than structural categories (such as inductive and deductive, which in his view mainly concern “the criteria and methods of assessing connection adequacy”, p. 188), but that another distinction is equally fundamental, namely that between conclusive and defeasible arguments, so that the category of defeasible a priori arguments is not only not impossible, but even one out of four fundamental categories in a fourfold system of basic types of arguments (see Freeman 2014, p. 3).

If they are generally defeasible, what, then, is it that makes a priori analogies appear so compelling? There may be a number of explanations. First, there is most certainly the deductive element that comes as the last stage and makes one easily overlook the defeasible abductive or inductive parts. Second, just because in an a priori analogy the analogue is deliberately constructed ad hoc, it is of course constructed in such a way as to ideally support the claim, which makes its compelling force appear much stronger than in ‘inductive’ analogies from empirical data (cf. Bermejo-Luque, forthcoming, section 2, on qualitative a priori vs. quantitative a posteriori analogies). Furthermore, since in a priori analogies both the analogue and the target claim are subordinated to a common general principle in a similar way, namely by an abductive move, they somehow appear to concur in supporting that general principle, so that it seems to get double support (Govier once – perhaps inadvertently – actually says that it is “from A and B” that we move to the U-claim, 1989, p. 148). And finally, the ontological and epistemic qualifiers that, as Bermejo-Luque and Freeman rightly point out, would be needed in most of the propositions involved, are as a rule suppressed, which is something that frequently happens in rhetorical arguments such as enthymemes.

All this may explain why a priori analogies appear so particularly compelling that they are even sometimes interpreted as essentially deductive (in the sense of conclusive) arguments. Although Govier acknowledges the fact that her hypothetical reconstructions of a priori analogies “produce, in effect, a two-stage argument” consisting of “an inductive argument from one case to a universal statement” and “a deductive argument subsuming the subject case under that universal statement.” (Govier 1989, p. 151), nonetheless, in her accompanying diagrams (p. 150-151) the arrows emblematizing an inference all invariably point the same way downward, as if the entire argument were deductive.

6. Conclusion
In conclusion, then, we may say that a lot was to be learned about arguments by analogy and other arguments from similarities from Aristotle. Based on Aristotelian categories, a reconstruction of arguments by analogy seems possible that explains both the commonalities and the differences of ‘inductive’ and a priori analogies and their respective persuasive force. According to this reconstruction, arguments by analogy can be interpreted as complex compound arguments that involve inductive, abductive, and deductive elements. Since inductions are mostly, and abductions generally defeasible, the final step, although formally deductive, rests on defeasible premises and is hence in itself defeasible. On this view, both ‘inductive’ and a priori analogies have basically the same structure; they are invariably defeasible, but allow for degrees of strength.

Acknowledgements
I would like to express my thanks to Lilian Bermejo-Luque for kindly making accessible to me her unpublished work on arguments from analogy, and for many substantial discussions on that subject.

References
Barker, S. F. (1965). The elements of logic. New York: McGraw-Hill.
Bermejo-Luque, L. (2012). A unitary schema for arguments by analogy. Informal Logic, 32(1), 1-24.
Bermejo-Luque, L. (forthcoming). Deduction without dogmas: The case of moral analogical argumentation. Forthcoming in Informal Logic.
Brown, W. R. (1989). Two traditions of analogy. Informal Logic, 11(3), 161-172.
Doury, M. (2009). Argument schemes typologies in practice: The case of comparative arguments. In F. H. van Eemeren & B. Garssen (Eds.), Pondering on Problems of Argumentation. Twenty Essays on Theoretical Issues (pp. 141-155, Ch. 11). Dordrecht: Springer.
Freeman, J. B. (2013). Govier’s distinguishing a priori from inductive arguments by analogy: Implications for a general theory of ground adequacy. Informal Logic 33(2), 175-194.
Freeman, J. B. (2014). What types of arguments are there? In D. Mohammed & M. Lewiński (Eds.), Virtues of Argumentation: Proceedings of the 10th International Conference of the Ontario Society for the Study of Argumentation (OSSA), 22-26 May 2013 (pp. 1-15). Windsor, ON: OSSA (CD ROM).
Gabrielsen, J. (2003). Is there a topical dimension to the rhetorical example? In F. H. van Eemeren, J. A. Blair, C. A. Willard & A. F. Snoeck Henkemans (Eds.), Proceedings of the Fifth Conference of the International Society for the Study of Argumentation (pp. 349-353). Amsterdam: SicSat.
Garssen, B. (2009). Comparing the incomparable: Figurative analogies in a dialectical testing procedure. In F. H. van Eemeren & B. Garssen (Eds.), Pondering on Problems of Argumentation. Twenty Essays on Theoretical Issues (pp. 133-140, Ch. 10). Dordrecht: Springer.
Govier, T. (1985). Logical analogies. Informal Logic, 7(1), 27-33.
Govier, T. (1987). Problems in argument analysis and evaluation. Dordrecht: Foris.
Govier, T. (1989). Analogies and missing premises. Informal Logic, 11(3), 141-152.
Govier, T. (2002). Should a priori analogies be regarded as deductive arguments? Informal Logic, 22(2), 155-157.
Guarini, M. (2004). A defence of non-deductive reconstructions of analogical arguments. Informal Logic, 24(2), 153-168.
Harman, G. (1965). The inference to the best explanation. The Philosophical Review, 74(1), 88-95.
Hitchcock, D. (1980). Deductive and inductive: Types of validity, not types of arguments. Informal Logic Newsletter 2(3), 9-10.
Lipton, P. (2001). Inference to the Best Explanation, London: Routledge.
Shecaira, F. P. (2013). Analogical arguments in ethics and law: A defence of a deductivist analysis. Informal Logic 33(3), 406-437.
Thomson, J. J. (1971). A defense of abortion. Philosophy and Public Affairs, 1(1), 47-66.
Waller, B. (2001). Classifying and analyzing analogies. Informal Logic, 21(3), 199-218.
Wellman, C. (1971). Challenge and response: Justification in ethics. Carbondale, IL: Southern Illinois University Press.
Wisdom, J. (1957). Explanation and proof. Lectures presented at the University of Virginia (unpublished ms.).

Abstract: Evidence-based practice (EBP) is currently a dominating trend in many professional areas. But what do we want evidence for in EBP? Evidence generally speaks to the trustworthiness of our beliefs, but EBP is practical in nature and truth is not really what is at stake. Rather we are after effectiveness in bringing about changes. What we need evidence for is a prediction to the effect that what has worked in one context will also work here. In this paper I argue that is makes good sense to view this prediction as the conclusion of an argument. To set the evidence in an argument will structure our thinking and help us focus on what kinds of evidence we need to support the likelihood that an intervention here will work.

Keywords: Argument, causal role, EBP, effectiveness, enablers, evidence, external validity, local facts, RCT, stability of context

1. Introduction
There exists a vast literature on EBP, hardly surprising given the status of ‘evidence-based’ as a buzzword in contemporary professional debates, such as education, medicine, psychiatry and social policy. Researchers are responding in many ways to political demands for better research bases to inform and guide both policy and practice; some by producing the kind of evidence it is assumed can serve as a base for practice; others by criticizing or even rejecting the whole enterprise of EBP – the latter frequently, but not exclusively, couched in terms of worries about instrumentalization of practice and restrictions in the freedom of professionals to exercise their judgment.

EBP is practical in nature. It is commonly called the what works agenda and its focus is the use of the best available evidence in the bringing about of desirable goals, both for client and society. This is indeed my preferred minimal definition of EBP; the production of desirable change, or conversely how we intervene to prevent certain undesirable outcomes. It is vital to note that EBP is deeply causal: we intervene into a “system” which already produces an output in order to change that output in a desirable direction. These interventions should be based on evidence that shows what works. To say that something (an intervention of some kind) works, is to say that doing it brings us the effects we want. For short, do X and it will lead to Y.

The very term ‘evidence-based practice’ obviously draws attention to evidence. Generally, epistemologists seem to agree that the term ‘evidence’ denotes that which serves to confirm or disconfirm a theory (claim, belief, hypothesis) (e.g. Achinstein, 2001; Kelly 2008). The basic function of evidence is thus summed up in the word support. Evidence speaks to the truth value and the trustworthiness of a claim, and is therefore relevant to all belief formation processes, whether in research or in daily life, including the ones where we form beliefs about the causal relation between action and result, input and output. This basic function can, I submit, in principle be performed by all sorts of data, facts and personal experiences. Indeed, it is worth pointing out that all people have first-hand experiences of the causal kind that we are talking about here. To act as an agent means to intervene in the world and have an influence on it (Menzies and Price, 1993). At the same time, ‘evidence-based practice’ has led to many misunderstandings about the role of evidence as well as to the crux of the matter being overlooked. What is really at stake is the claim that the evidence is evidence for. Evidence is in a sense a servant; good evidence provides us with good reason to believe that the claim is true.

I shall in this paper argue that setting evidence is an argument makes good sense for the practical enterprise of EBP; it serves to clarify and structure our thinking about what we need to know. But to see that, we first have to look at the basic causal structure of EBP and the EBP orthodoxy concerning admissible forms of evidence as well as assumptions concerning uses of evidence. Thus, this paper is mainly a laying-out of the premises I suggest are needed to bolster the conclusion that EBP will be well served by setting the evidence in an argument.

2. The causal nature of EBP
The short version of the causal nature is that EBP is causal because it is about the bringing about of desirable results. That is to say, we have a causal connection between an action or intervention and its effects; between X and Y. The long version of the causal nature of EBP takes into account the many forms of causation; direct, indirect, necessary, sufficient, probable, generic, actual, etc. and develops a more complex and sophisticated picture. In educational contexts, as, I assume, in political contexts, this causal complexity goes highly unrecognized. However, for my purposes in this paper a simplified X-Y relation will by and large do.

My own field is education; a complex field with many factors that interact and influence each other in many different ways. Interventions also vary in nature, from simple actions to highly complex school-wide projects which may take two or three years to run. It is essential to be aware that, regardless of field, any intervention is inserted into pre-existing conditions. The causal system into which we intervene already produces an output; we just wish to change it because we are not entirely happy with the output – in education, student achievement is a typical output of this sort. The already existing output is termed the default value (Hitchcock, 2007, p.506); the value we would expect a variable such as student achievement to have in the absence of intervening causes. The default assumption is that the system will persist in its state and keep producing the default results unless we do something or something happens. The default, Hitchcock stresses, it not that the state or value in question is this or that, but that it will remain this or that unless something happens to change it. When a set of variables all take on their default value and business is run as usual, they cannot by themselves take on a different value. This is a natural principle of causal reasoning, Hitchcock thinks. We tend to assume that if a variable should take on a deviant (or unexpected) value, there must be some outside variable or event that explains it. That is, to change the value of our target variable, whether student achievement or some other desirable outcome, we have to intervene somehow. This certainly seems to be a tacit presupposition of EBP.

For various reasons, the causal theory that best suits the logic of EBP is the manipulationist theory of causation (e.g. Pearl, 2009; Sloman, 2005; Woodward, 2003, 2008). Let us suppose that X produces Y as its default result. To change the value of Y, we must change the value of X. Thus, if we set the value of X to xi rather than xk, then the value of Y should follow in train and change to yi rather than yk. This is precisely what the manipulationist theory of causation tells us: there is an intimate connection between causation and manipulation such that causal relationships are eminently exploitable for the purpose of change. This is one of the reasons why this theory of causation is so popular in disciplines which are to bring about change and development as well as give recommendations for practice and policy.

The point of intervening is that we set the value of X to xi from outside the system rather than letting X be decided by the other variables in the system. That is to say, we manipulate X in order to further the changes in Y we deem desirable, naturally on the assumption that X actually leads to or brings about Y. As Judea Pearl puts it,

The simplest type of external intervention is one in which a single variable, say Xi, is forced to take on some fixed value xi. Such an intervention, which we call “atomic,” amounts to lifting Xi from the old functional mechanism xi = fi(pai, ui) and placing it under the influence of a new mechanism that sets the value xi while keeping all other mechanisms unperturbed (2009, p. 70).

There is, however, more to intervention than this quote tells us. First, it changes the value of Y, even though this is not explicitly mentioned in Pearl’s definition. Changing Y is the main aim of educational interventions and usually the reason why we intervene in the first place (e.g. to improve student achievement). Second, the intervention changes the entire causal model because it cuts the effect (yk) off from its normal causes (xk). When we have intervened on X, the system no longer continues in its default state. Business is no longer run as usual, but is now running in a different way, one we think (or hope) should bring about the desired result or at least increase its probability. Third, the intervention disrupts the relationship between X and its parents. The value of X is no longer determined by the default running of the system, but by the intervention. All other influences on X have been blocked and/or cut off. As the equation in the quote indicates, Xi is lifted from the influence of P, its parents, and U, an error term representing the impact of omitted and/or unknown variables, and its value is decided by a new mechanism, namely the intervention. I prefer to interpret this in line with the causal agency advocated by Menzies and Price, although Pearl himself states that intervention does not necessarily have to involve human activity. But in education interventions require agency, hence my adoption of Menzies and Price’s view on this point.

This is not the place to discuss manipulationist theory in detail, but a couple of issues deserve mention. First, there is Pearl’s view that causal mechanisms (X-Y relations) are autonomous. He thus argues that our intervention on one causal connection leaves the other connections in the system undisturbed. This presupposition seems deeply problematic to me. Educational practice is best understood as an open system where events, actions and factors are somehow locked together, obviously to varying degrees. If factors hang together, the change in Y will depend more on the total structure and it is a mistake – however tempting it is – to look at only small chunks or individual causal mechanisms. In complex systems we cannot assume that intervention on one mechanism leaves all other mechanisms intact. Second, it seems to be a presupposition of manpulationist theorists that X is already a part of the system. For example, Christopher Hitchcock (2007) argues that X-Y relations are internal to the system and that interventions therefore involve exogenous changes to X. My point here is twofold. Firstly, in education a teacher, as an agent within the system can decide to make changes in input X; this qualifies as an intervention in the broad sense of them term, but it comes from inside the system and is thus not exogenous. Second, there are many EBP cases where X is exogenous and inserted into the system as a new element. I view these two points as unproblematic amendments the manipulationist theory of causation. The main point is that X be manipulable and that the intervention alters the causal system.

It is the ambition of EBP to provide knowledge that works; that is, to provide knowledge about how causal input X can be changed to produce desired changes in output Y. For example how implementation of a reading instruction program can improve the reading skills of slow readers, or how a school-wide behavioural support program can serve to enhance students’ social skills and prevent future problem behaviour. But not only that – we wish to know what works generally. That means not only that the effect (output, result) in question is reproducible in principle, but that we know how to achieve it regularly and can plan for it. This kind of practical causal knowledge is future-oriented, in the sense that we, on the basis of experience or other empirical evidence, form the expectation that the desirable results obtained somewhere can somehow be reproduced.

3. What does the evidence tell us?
As suggested above, the basic function of evidence is to speak to the truth value of beliefs. In the EBP case, both advocates and critics simply assume that the evidence speaks to the truth of the belief that there is a causal connection between X and Y, and that this is all the evidence there is (or all we need).

In a similar manner, both advocates and critics often understand EBP to include a hierarchy of evidence as part of its definition. There are various versions of this hierarchy; what they have in common is that they all rank randomized controlled trials (RCTs) on top, and that professional judgment is ranked at or near the bottom (see e.g. Pawson, 2012). The standard criticism is that such hierarchies unduly privilege certain forms of knowledge and research designs, undervalue the contributions of other research perspectives, and especially that they undervalue professional experience and judgment. The privileging of RCT evidence is evident in e.g. the US Department of Education’s User Friendly Guide. EBP literature, such as the User Friendly Guide, provides evidence-ranking schemes (which tell us that the best evidence comes from RCTs), it provides advice guides (which tell us to choose an educational intervention that is backed by good (RCT) evidence, and it often provides “warehouses” (where we find interventions backed by good evidence). Together these three different functions make up the foundation of what has become known as the EBP orthodoxy (see e.g. Cartwright and Hardie, 2012). There is another element to the orthodoxy that I shall return to below.

There are good reasons to adopt the EBP orthodoxy and even better reasons not to adopt it. The principle behind evidential ranking schemes is trustworthiness – our evidence needs to be trustworthy or reliable in order to do its job, which is to speak to the truth value of claims and beliefs. It is no accident that RCTs have established themselves as the gold standard. Nancy Cartwright (2007) divides all research methods in two; clinchers and vouchers. RCTs are clinchers: methods that are deductive and whose logic is such that if all the specific assumptions of the trial are met, a positive result will logically entail the conclusion. The evidence provided is thus sufficient for the conclusion; one might even say that it guarantees it. The evidence, in turn, is guaranteed by the research design. In RCTs we compare groups that are the same with respect to all relevant (causal) factors except one. Random assignment is supposed to ensure that the groups have the same distribution of causal and other factors. The standard result of an RCT is a “treatment effect” (expressed in terms of effect size): average effect in treatment group minus average effect in control group. We assume that the difference between the two groups needs a causal explanation, and since other factors (supposedly) are equally distributed we infer that the treatment, our intervention, is the cause of the outcome. It works; we might be tempted to conclude.

RCTs are strong on internal validity. If we obtain an average positive result and the conditions of the trial are met, we may safely conclude that the causal claim in question is true, X does indeed bring about Y and the evidence supports it. But internal validity is purchased at the expense of external validity, or generality. As Nancy Cartwright (2007) argues, what RCT evidence shows is strictly speaking that the X-Y relation holds where the trial was conducted, for that particular study group (see Cartwright for a detailed discussion of the limitations of the research design). It by no means shows that the X-Y relation holds generally across differing contexts. This fact is not discussed in the EBP literature. Rather, we seem to take it for granted that RCT evidence shows that the causal X-Y relation holds in general, that something works in general. The fact that it does not, is a major premise in the argument for why it is important to set evidence in an argument.

There are several sides to the limitation of RCT evidence. First, we here come across a problem that is also found in the manipulationist theory of causation; namely that one does not distinguish between finding and using causes. Manipulationist theory and empirical research designs alike focus on finding causes. To investigate whether X causes Y, we see if the two are correlated once we have controlled for other possible causes of Y. We hold various background factors fixed, manipulate the values of X and observe whether the values of Y change in train. Basically we conclude that X causes Y if the probability of Y is higher with X than without it, and the evidence we get supports our view. But using causes to bring about desired changes is another matter altogether. I am tempted to say that both manipulationist theory, RCTs and EBP only tell us half the story. They all think in terms of methodology geared at finding causes. When it comes to using causes, it is not the relation between X and Y that matters the most. When we implement an intervention we either change an X that is already part of the system, or we insert it into the system. Either way the pre-existing system, practice, has to be taken into account when we use causes. Hence, what matters is that the probability of Y given X-in-conjunction-with-systemi is larger than the probability of Y given not-X-in-conjunction-with-systemi. And the RCT tells us nothing about this. As Cartwright (2009) points out, the formula that shows that X is a cause of Y, for example expressed in terms of a treatment effect, need not be the right formula for telling whether X will produce Y when we implement it in some concrete system. When we implement X, we generally also change other factors in the system, not only the ones causally downstream from X. But the RCT evidence does not tell us whether X will also affect A, B, and C, and if so how that will affect Y.

The second ramification of the limitation of RCT evidence is a corollary of the first, and concerns the EBP orthodoxy. This orthodoxy also demands faithfulness in implementation, termed fidelity. If you are to implement in your context an intervention that an RCT tells you has worked somewhere, you should do it exactly as it was done there. Take for example a school-wide behavioural program (Arnesen, Ogden and Sørlie, 2006). Components, principles and guidelines are decided in advance, and so is their order and manner of implementation, although the authors concede that some local adjustment is necessary. But basically implementers must be loyal to the procedures prescribed by the program developers. If actual implementation deviates from prescribed implementation, we no longer know exactly what it is that works, the argument goes, and the program suppliers cannot be held responsible for the results. Variations in the efficacy of X are generally due to deviant or unsystematic implementation, the EBP orthodoxy holds. The orthodoxy presupposes similarity of contexts and generality of X-Y relation. The demand for fidelity in EBP is misguided, as it tacitly assumes that the RCT evidence showing the effect of X on Y is all you need.

But what do practitioners need evidence for? I propose that what practitioners, say teachers, want evidence for, is a prediction that X will work here, in my classroom, were I to implement it. The RCT evidence only speaks indirectly to that question, by telling you that X worked somewhere. But how do you get from somewhere to here? This is where the usefulness of an argument comes in.

4. Setting evidence in an argument
Let me back up a little. It is important that we take on board the fact that contributions to an outcome both can and generally do come from different sources. This sounds commonplace, but is easily forgotten; we tend to look for the cause and if we implement an intervention it is only natural that this intervention is salient for us and we ignore other factors. But the overall effect on Y depends on how all these factors add up; thus, an intervention is part of a team of causes and enabling factors which work together.

What, then, should a practitioner look for when trying to make a decision about whether to implement X or not? Which facts must be collected if I am to hedge my bets that X will work here? When is the fact that X worked there relevant to the prediction that it will also work here? We cannot take it for granted that it will, no matter how large the effect size emanating from the RCT evidence. We cannot simply export a causal connection and insert it into a different context and expect it to work. Causal principles are local, Cartwright argues, and it is easy to agree with her. Educational practitioners love to point out that students are different, teachers are different, curricula are different, headmasters are different, parents are different, and school cultures are different. So how can the RCT evidence be made relevant?

I assume that what practitioners want to know is whether an intervention is worth trying in their own concrete context. Will X work here, that is, make a positive causal contribution here if I implement it? RCT evidence does not tell them that. What is does tell them, is that X made a positive contribution to Y somewhere, and that given this positive contribution, we may infer that certain enabling factors were present which allowed X to do its work and make its way to Y. That is to say, the other factors necessary for producing the outcome must also be in place – it is vital to remember that our intervention is part of a constellation of causes which together bring about Y. An effectiveness prediction that X will work here must take the whole constellation into account, as well as possible. It is this task that is made easier and more systematic by thinking of the effectiveness prediction as the conclusion of an argument and that the job is to gather the premises which lead up to the conclusion.

What works somewhere, as shown by the RCT evidence, can be made relevant to what will work here. But a number of other facts must be collected if we are to say something about X-in-conjunction-with-system, which is what we want:

* In “our” context here we already get an outcome, a default result, concerning the student achievements in question, but we want to improve them. How are these results produced? What factors are present in our context and how do they combine to produce the result?

* This constellation of causes is called the causal principle for the outcome (Cartwright and Hardie, 2012 and it is needed to connect the alleged cause with the desired effect.

* Mapping the local causal principle is not enough. Next we have to look at the proposed intervention X and ask whether it can play a positive causal role for producing the desired effect in our setting. How can it work? There is no substitute, Cartwright and Hardie insist, for thinking thoroughly about how X might work if implemented.

* Next we look at the factors that must be in place if the intervention is to be able to play its causal role. Which are they? Are they present? If not present, can they be easily procured? Do they outweigh any disabling factors that might be in place? It is important to remember that some of these enablers may be absences of hindrances. Arnesen, Ogden and Sørlie (2006) provide examples of such local facts, despite their adherence to the EBP orthodoxy and the principle of fidelity. For example, they argue that there must not be personal conflicts among the staff if the behavioural program is to work positively. That is, a conflict is a contextual disabler which hinders or obstructs the working of the program. Conversely we might say that its absence is an enabler. Another local enabling factor is the fact that staff norms and values at least do not contradict the values inherent in the program to be implemented.

* Not only must the necessary enablers be in place, their organization must also be stable. The stability of the system into which we contemplate inserting X is of vital importance for our chances of success. If the system is shifting and unstable, X may never be able to do its work and produce Y. This fact is well-known to teachers, but perhaps not really recognized by EBP proponents. But teachers seek to stabilize the environment, by structuring it in different ways: creating and enforcing rules of conduct, establishing habits and ways of doing things – in short, creating a stable environment which at least to some degree makes for predictability and thus allows us to expect with some confidence that our plans will work out. Time-honoured educational domains such as curriculum theory and didactics can be viewed in this light: they provide knowledge and advice on how to create the stable conditions necessary for goal achievement in general. But since we are trying to predict whether X will work if we implement it, the stability conditions we assess must be linked to X.

* It should be noticed that this kind of “mapping” is not about listing similarities between somewhere and here. Similarities are not important for this kind of generalization. Rather, what it takes is that we have some idea about what a good constellation of factors surrounding X might be, factors which enable X to make a positive contribution to Y. This constellation need not be the same; it can vary from context to context. The important thing is that we map the enablers, procure them if necessary, and that we avoid or remove the disablers.

In sum: local facts are as necessary as they are overlooked. I by no means claim that the issues listed above comprise an exhaustive list of facts a practitioner needs to map in order to hedge his or her bets that a given intervention will work should it be implemented here. Yet it should be evident from this set of issues that it takes a lot of deliberation to figure out the chances that an intervention might work. Setting all these different kinds of evidences into a (reasonably) clear argument structure helps us sort them out and see what facts we need to ascertain. Inspired by Cartwright and Hardie, here is what I propose:

Premise 1: The intervention in question, X, worked somewhere; that is, it played a positive causal role in achieving Y for at least some of the individuals in the study group. The RCT evidence tells us that, and it also indicates how strong the causal influence of X on Y is, given that all other factors are held fixed (the effect size). We should remember, however, that effect size is a statistical entity and only informs us of the aggregate result. A positive aggregate result is perfectly compatible with negative results for some of the individuals in the study group.

Premise 2: Which factors govern the default production of Y here? The RCT evidence does not tell us that.

Premise 3: The intervention can play the same role here as it did there. The RCT evidence does not tell us that.

Premise 4: The enabling factors necessary for the intervention to play a positive causal role for Y are in place here, or we can get them. The RCT evidence does not tell us that.

Premise 5: The system (context) here is stable enough so that the intervention will have time to unfold and work. We know the main factors influencing this stability and we know how to maintain them. The RCT evidence does not tell us that.

Conclusion: Yes, the intervention will most likely work here. There are always unknown factors that might disable or hinder its workings; despite these we think it is worth implementing it. Or we may conclude that since the vital enablers are missing and they are too expensive to get, chances are that this intervention will not contribute positively to Y in our context.

This tentative argument structure can guide you to what kind of evidence you need to ascertain. As should be plain, the RCT evidence alone will not be enough.

5. Conclusion
I have in this paper addressed one aspect of evidence-based practice, namely the fact that a lot more evidence is required in practice than is normally assumed by proponents and critics of EBP alike. The EBP literature, whether written by critics, adherents or researchers, focuses on RCT evidence as the kind of evidence on which practice should be based. Organizations such as CampbellCollaboration and McREL, which collect and vet evidence and produce meta-analyses, adhere to the EBP orthodoxy and the evidence hierarchy and view RCTs if not as the only admissible kind of evidence, then certainly as the preferable kind of evidence. Critics problematize this view point and argue that other kinds of evidence should count as well.

What none of them do, I have argued, is to address the question of what practitioners really need evidence for. If we assume that what practitioners really want to know is whether a proposed intervention will work for them, in their classroom, then it immediately transpires that RCT evidence is not enough. There are two reasons for this. The first is that RCT evidence only pertains to the first of the five premises I have suggested above. The second is that contrary to popular belief, RCT evidence does not show that a causal relation (X-Y) holds in general, it just shows that is holds for the study group from which the evidence emanates. In order to make a decision about whether we actually should implement the intervention in question here, we need to collect a good many local facts and put all our evidences together in an argument structure which allows us to make a sensible all-things-considered judgment. We must never lose sight of the fact that here denotes an already existing practice, a causal system, and that any output has many antecedent events. Changing a factor in the system or inserting a new one will bring changes to the entire system; changes which may affect our desired outcome in good or bad ways. RCT evidence may be highly trustworthy, but it does not even provide half the story. Putting all the different kinds of evidences in a structure will help us think systematically about what we need to know. Thus, EBP as a practical enterprise is indeed well served by setting all the necessary evidences in an argument.

I would like to end this paper with a remark about EBP itself: EBP is much more complicated that advocates and critics alike tend to think. It is essential to distinguish between finding and using causes, and it seems to me that using them to bring about desired results is much more complicated than finding them in the first place. EBP is thus no magical bullet for improving student achievements, but nor is it impossible. As a minimum it requires practitioners who can think for themselves; the EBP orthodoxy is seriously misguided.

References
Achinstein, P. (2001). The book of evidence. Oxford: Oxford University Press.
Arnesen, A., Ogden, T. & Sørlie, M. (2006). Positiv atferd og støttende læringsmiljø i skolen [Positive behaviour and a supporting school environment]. Oslo: Universitetsforlaget.
Cartwright, N. (2007). Are RCTs the gold standard? BioSocieties, 2, 11-20.
Cartwright, N. (2009). How to do things with causes. Proceedings and Addresses of the American Philosophical Association, 83, 2, 5-22.
Cartwright, N. & Hardie, J. (2012). Evidence-based policy. A practical guide to doing it better. Oxford: Oxford University Press.
Hitchcock, C. (2007). Prevention, pre-emption, and the principle of sufficient reason. Philosophical Review, 116, 4, 495-532.
Kelly, T. (2008). Evidence. In E. Zalta (Ed.), Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/entries/evidence/ Retrieved September 16, 2008.
Menzies, P. & Price, H. (1993). Causation as a Secondary Quality. British Journal for the Philosophy of Science, 44, 187-203.
Pawson, R. (2012). Evidence-based policy. A realist perspective. Los Angeles: Sage.
Pearl, J. (2009). Causality. Models, reasoning and inference. Cambridge: Cambridge University Press
Sloman, S. (2005). Causal models. How people think about the world and its alternatives. Oxford: Oxford University Press.
Woodward, J. (2003). Making things happen. A theory of causal explanation. Oxford: Oxford University Press.
Woodward, J. (2008). Causation and manipulability. In E. Zalta (Ed.), Stanford Encyclopedia of Philosophy, http://plato.stanford.edu/entries/causation–mani/ Retrieved March 20, 2012.
US Department of Education (2003). Identifying and implementing educational practices supported by rigorous evidence: A User Friendly Guide. Washington, DC: Coalition for Evidence-Based Policy. http://www2.ed.gov/rschstat/research/pubs/rigorousevid/rigorousevid.pdf. Retrieved June 12, 2014.

Abstract: Biblical scholars have fundamental differences in defining Paul’s argumentative and rhetorical goal in 1 Corinthians 6:12–20. There is no convincing explanation for why the apostle brings 6:12–20 up in the letter. I conduct a pragma-dialectical analysis to account for the argumentation, rhetoric and their interplay in 6:12–20. It turns out that Paul aims at shaming the audience in order to break their resistance.

Keywords: 1 Corinthians, argumentation, Bible, New Testament, Paul, pragma-dialectics, rhetoric, shame, strategic maneuvering, theology.

1. Introduction
Biblical scholars have had significant difficulties in interpreting the argumentation in 1 Corinthians 6:12–20 (Goulder 1999, p. 341; Rosner 1998, p. 336). Two frequent and general problems are brought up to motivate the upcoming analysis of the section in the letter.

The first problem deals with the goal of the section. What does Paul want to argue in the section? Two alternative standpoint options are common (Rosner 1998, p. 336):
a. The apostle argues that the Corinthians should stop a specific behavior, that of having relations with harlots (Drake Williams III 2008, p. 20; Fee 1987, p. 250; Rosner 1998, pp. 341-342);
b. Paul wishes to smother a broader phenomenon: sexual immorality (Conzelmann 1975, p. 108; Lambrecht 2009, p. 486; Rosner 1998, pp. 337-338). Topically speaking, the two themes are related. The question arises, which of the two notions supports the other. Does Paul employ sexual immorality to support the avoidance of harlots or vice versa?

Furthermore, why does the apostle bring up the issue in the first place? Is the control of the Corinthians’ sexual morality an objective in itself for him or does Paul use it to achieve another goal?

The second problem deals with the placement of the section as a part of 1 Corinthians (Fee 1987, p. 250). Does Paul have a certain strategy in his ordering of the different argumentative sections? Or is his approach random (Murphy-O’Connor 1996, p. 253)? I will argue that he has placed the section strategically with a specific intention. In this endeavor, I will occasionally refer to the previous argumentative sections 4:18‒5:13 and 6:1‒11 to support my claims. My general claim is that Paul has certain long-term dialectical and rhetorical aims that he tries to achieve in the section 6:12‒20 (van Eemeren & Houtlosser 2002, pp. 134-135).

I will conduct a pragma-dialectical analysis of 1 Corinthians 6:12–20 in order to solve the problems discussed above. After the analysis, I will draw some conclusions.

2. Analysis
I will apply only those tools and aspects of pragma-dialectics that I deem necessary for the purposes of this study. I will analyze the argumentative section in two main parts. The first one consists of the construction of the analytic overview (van Eemeren et al. 1993, pp. 60-61; van Eemeren & Grootendorst 1992, pp. 93-94; van Eemeren & Grootendorst 2004, pp. 118-122; van Eemeren & Houtlosser 2002, p. 134), which, in turn, entails establishing the following steps:

1. Standpoint(s);
2. Common starting points;
3. Arguments;
4. Argumentation structure.

The analytic overview belongs to the so-called standard pragma-dialectical analysis (van Eemeren & Grootendorst 1992; van Eemeren & Grootendorst 2004). The first three points listed above correspond with the order of the discussion stages of the ideal model for critical discussion: confrontation stage, opening stage and argumentation stage (van Eemeren et al. 1996, pp. 280-283; van Eemeren & Grootendorst 2004, pp. 57-62; van Eemeren & Houtlosser 2002, pp. 132, 138-139). The fourth and last stage, the concluding stage, is missing in the section 6:12–20 and consequently it is left unconsidered in the analysis. After the identification of the standpoint(s), common starting points and arguments, the argumentation structure is reconstructed.

The second part consists of the analysis of the strategic maneuvering which is dealt with theoretically in the extended pragma-dialectical model (van Eemeren 2010). The analytic overview functions as a basis for its analysis. The strategic maneuvering will be assessed by scrutinizing the three inseparable aspects of it which are analytically distinguished from each other: the topical potential, audience demand and presentational devices (van Eemeren 2010, pp. 93-101, 108-113, 118-122; van Eemeren & Houtlosser 2002, pp. 139-141).

2.1 Analytic overview
In the introduction, two alternatives were presented as possible standpoints: stopping relations with harlots and sexual immorality. To map where and how they occur in the text, in Figure 1 I have divided the section into three subsections based on the occurrences of these two topics:

Figure 1: Section 6:12–20 divided into three parts based on the occurrences of the concepts “sexual immorality” and “harlot”

I.
12. All things are permitted for me, but all things are not beneficial; all things are permitted for me, but I will not allow myself to be brought under the power of any.
13. Foods for the stomach and the stomach for foods, but God will abolish both it and them. But the body is not for sexual immorality but for the Lord, and the Lord for the body.
14. God both raised up the Lord and will also raise us up by His power.

II.
15. Do you not know that your bodies are members of Christ? Would I then take the members of Christ and make them members of a harlot? Let it not be!
16. Or do you not know that he who joins oneself to a harlot is one body with her? “The two” for he says, “shall become one flesh”.
17. But he who joins oneself to the Lord is one spirit with Him.

III.
18. Flee sexual immorality! Whatever sin that a man does is outside the body, but he who commits sexual immorality sins against his own body.
19. Or do you not know that your body is the temple of the Holy Spirit who is in you, whom you have from God, and you are not your own?
20. For you were bought with a price. Therefore glorify God in your body!

Sexual immorality (πορνεία) occurs in subsection I in verse 13b. In general, verses 12‒14 hold non-confrontational speech acts. Paul does not explicitly refer to the Corinthians besides in verse 14, in which he includes himself in the audience (ἡμᾶς ἐξεγερεῖ). The apostle does not bring up any apparent dispute in subsection I. Closest to a confrontation is the notion that sexual immorality is not for the body in verse 13b.

In subsection II, in verses 15‒17, Paul confronts the recipients directly (Lambrecht 2009, p. 482). He asks them in verse 15a whether they do not know that their bodies are members of Christ. Then, he makes clear in verses 15bc that the Corinthians should not become members of a harlot. The phrases of “one body,” “one flesh” and “joins oneself” in verse 16 indicate that Paul refers to a sexual relation (Butting 2000, p. 83). Subsection II deals with why sex with harlots in particular should be avoided.

Subsection III begins with a command to flee sexual immorality in verse 18a (Byrne 1983, p. 612). Paul returns to the broader topic he brought up in verse 13b. The argumentation in verses 18‒20 supports the order to flee sexual immorality.

2.1.1 Standpoint(s)
With regard to the standpoint, fleeing sexual immorality (18a) is a too abstract goal to render it as a plausible main aim. Instead, preventing sex with harlots (15bc) is a more concrete and feasible goal for Paul to attempt to achieve, when addressing a single community. It is not feasible to render avoiding prostitutes as a subordinate argument to fleeing sexual immorality. In that case, the argumentation would appear roughly as follows: ”You should flee sexual immorality, since you should not have sex with harlots.” The reasoning becomes understandable when the arguments are reversed: “You should not have sex with harlots, since you should avoid sexual immorality.”

Furthermore, in the two previous argumentative sections, Paul employs the generalization of a problem to support a more concrete instance of that problem. In 4:18‒5:13, he employed the general teaching to exile sinners (5:9–11) as an argument to drive out the single fornicator (5:2). In 6:1‒11, he argued that lawsuits in front of unbelievers are shameful (6:5a, 6), since lawsuits in general are shameful (6:7a). Because of this pattern, I suppose that the current section functions similarly.

2.1.2 Common starting points
In verse 12, Paul states that everything is free to him (see also 10:23). Several scholars treat this as a Corinthian slogan (Barrett 1968 p. 144; Conzelmann 1975, pp. 108-109; Dodd 1995, p. 40; Fee 1987, p. 251; Murphy-O’Connor 2009, p. 24; Rosner 1998, p. 346; Thiselton 2000, pp. 459-460). However, in my mind, the exaggerative formulation reflects Paul’s own view (Dodd 1995, pp. 39, 54), since the apostle had a habit to put forward hyperbole (Lausberg 1998, pp. 263-264, 410-411; Thurén 2000, pp. 202-203, 212). This is not to say that the apostle does not account for a Corinthian view, too. The recipients likely considered themselves free in many respects. However, the apostle wants to appear the expert of freedom by stating that everything is permitted to him. He wishes to promote his ethos. Paul also wants to begin to argue from a common ground.

The phrase “foods for stomach” (and vice versa) reflects common sense that is in itself obvious. I interpret the following phrase in verse 13a about God abolishing as referring to the eschatological judgment. In the previous argumentative sections, 4:18–5:13 and 6:1–11, the apostle refers to eschatological matters (5:5 and 6:2–3). In 6:13, Paul contrasts the negative destruction in the last judgment with the positive resurrection in verse 14, which also occurs at the end of times.

A distinct presentational device (see, chaps. 2.2 and 2.2.3) in the current argumentative section, and in the two previous sections, is the “do you not know that” – question. Paul uses it to convey common starting points (Wuellner 1986, p. 53). The idea is that the Corinthians should have taken into account the apostle’s particular teaching from a previous visit or letter (see, Hurd 1965, pp. 43-58). Consequently, verse 15a is considered a common starting point.

However, verses 16a and 19 are not common starting points, even though they hold an almost identical formulation of the rhetorical question. Instead, they are arguments, because they are supported by starting points in verses 16b and 20a (Wuellner 1986, p. 67). The word “or” in the “do you not know” –questions indicates that the Corinthians should have deduced the conclusion from the starting points that support the arguments inherent in the rhetorical questions (Lambrecht 2009, p. 483).

In verse 16b, Paul quotes Genesis 2:24, which belongs to the presumably authoritative religious writings to him and to the recipients (Drake Williams III 2008, p. 20; Heil 2005, pp. 103-105, 122). Consequently, I render the quote as a common starting point. In verse 20a, the apostle alludes to the sacrifice of Christ (see also 7:23) (Lambrecht 2009, p. 484). God has bought the Corinthians, and believers in general, to himself with Christ’s blood (Conzelmann 1975, p. 113; Goulder 1999, p. 347; Fee 1987, p. 265). This is a fundamental event that establishes the faith of the recipients and belongs to the common starting points.

2.1.3 Arguments
The rest of the text consists of arguments (verses 13b, 16a, 17–19 and 20b). There is no concluding stage, unless one considers verse 20b as belonging to it. I render 20b as a positive repetition to flee sexual immorality which occurs in verse 18a (Fee 1987, p. 265; Lambrecht 2009, pp. 484-485). To be able to glorify God in the body, one needs to flee sexual immorality. The phrases are immediately connected and should be considered as a single argument.

As stated above, the “or do you not know” –questions in verses 16a and 19 are considered as arguments, since they are supported by common starting points.

To sum up, Figure 2 portrays the stages of the ideal model as they appear in section 6:12–20. The text between the two symbols ‘[S]’ indicates the sole standpoint. It does not occur until the midway of the section. The symbols ‘[O]’ and ‘[A]’ similarly indicate the opening and argumentation stages, respectively. Most of the opening stage appears in the first half of the section. The argumentation stage is prominent in the last half of the text. This ordering of the text indicates that Paul approaches the argumentative situation indirectly, making use of insinuatio (Kennedy 1999, pp. 103-104; Lausberg 1998, pp. 121, 124, 132-133, 684). The reason for this is assessed in the analysis of strategic maneuvering (chap. 2.2).

Figure 2: Stages of the ideal model in 1 Corinthians 6:12–20
12. [O] All things are permitted for me, but all things are not beneficial; all things are permitted for me, but I will not allow myself to be brought under the power of any.
13. Foods for the stomach and the stomach for foods, but God will abolish both it and them. [O] [A] But the body is not for sexual immorality but for the Lord, and the Lord for the body. [A]
14. [O] God both raised up the Lord and will also raise us up by His power.
15. Do you not know that your bodies are members of Christ? [O] [S] Would I then take the members of Christ and make them members of a harlot? Let it not be! [S]
16. [A] Or do you not know that he who joins oneself to a harlot is one body with her? [A] [O] “The two” for he says, “shall become one flesh”. [O]
17. [A] But he who joins oneself to the Lord is one spirit with Him.
18. Flee sexual immorality! Whatever sin that a man does is outside the body, but he who commits sexual immorality sins against his own body.
19. Or do you not know that your body is the temple of the Holy Spirit who is in you, whom you have from God, and you are not your own? [A]
20. [O] For you were bought with a price. [O] [A] Therefore glorify God in your body! [A]

2.1.4 Argumentation structure
The argumentation structure, in Figure 3, is constructed based on the assessment of the standpoint, common starting points and arguments (chaps. 2.1.1, 2.1.2 and 2.1.3) and the division of the text displayed in Figure 1. The three subsections frame three argumentative wholes (see, Lambrecht 2009, p. 480).

In subsection II, the first line of defense for the standpoint occurs. The argumentation deals with why sex with harlots specifically is dangerous. Subsections I and III argue why the more general phenomenon, sexual immorality, should be avoided. The reasons for fleeing sexual immorality constitute the second line of defense.

Figure 3: Argumentation structure of 1 Corinthians 6:12–20
1 You should not have sex with harlots (15bc)
1.1a You should not become one with a harlot (15b)
1.1a.1a You are one with Christ (15a)

(1.1a.1b) [You cannot be one both with a harlot and with Christ (15–17)]
1.1b Having sex with a harlot makes you one with her (16a)
1.1b.1 Gen: “The two shall become one flesh” (16b)

1.2a Sexual immorality should be fled (18a)
(1.2a.1) [Sexual immorality is not like acceptable urges such as eating (13)]
(1.2a.1.1a) [How food affects the stomach will not matter in the end (13a)]
1.2a.1.1a.1 God will abolish both (13a)

(1.2a.1.1b) [How sexual immorality affects the body will matter in the end (13b)]
1.2a.1.1b.1 Your bodies will be resurrected (14)
1.2a.1.1b.1.1 The Lord’s body was resurrected (14)

(1.2a.2) [Sexual immorality is a sin against the Holy Spirit (19)]
1.2a.2.1a Sexual immorality is a sin against the body (18b)
1.2a.2.1b The body is the temple of the Holy Spirit (19)
1.2a.2.1b.1 God owns you (20a)
(1.2b) [Having sex with harlots is a case of sexual immorality]

The arguments beginning with 1.1 (subsection II) represent the spiritual danger that sex with harlots specifically causes. Curiously, the argument 1.1a.1b is left implicit. This argument is crucial because it indicates why unity with a harlot is dangerous in view of the unity with Christ: they are mutually exclusive (see, Fee 1987, pp. 251, 257 and Lambrecht 2009, p. 483). Members of the congregation should not be mixed with representatives of sexual immorality. On a more general level, the point is that the holy and the unholy should not be mixed with each other.

The arguments beginning with 1.2 build a bridge between the specific issue of having sex with harlots and the broader phenomenon of sexual immorality (Goulder 1999, p. 345). The implicit argument 1.2b indicates this connection. Two lines of defense, beginning with 1.2a.1 and 1.2a.2, support the notion that sexual immorality should be fled. These correspond with subsections I and III.

In verses 13‒14 (subsection I; arguments beginning with 1.2a.1), Paul manufactures a counter-argument to a view that he implicitly attributes to the Corinthians (see, Fee 1987, p. 253). He compares sexual immorality to acceptable human urges. Eating is used as an example of an acceptable urge. Eating is alluded to by foods and stomach. The point of verse 13a is the following: how food affects the stomach, which is a part of the body, does not matter in the end. This is because God will abolish them. Sexual immorality, however, affects the body in a way that matters in the end. The body is important because it is meant for resurrection and not for destruction (Conzelmann 1975, p. 111; Lambrecht 2009, pp. 481-482). Paul suggests that the unholy sexual immorality harms the holy resurrection body while acceptable urges do not. Thus, sexual immorality differs from other urges.

From the above reasoning, the alleged Corinthian position may be deduced. The Corinthians may have considered sex with harlots as harmless as, for instance, eating. Paul argues that their view is incorrect. However, the analyst has to be careful in making these assumptions regarding the recipients’ stance. The apostle may attribute a false position to the audience in order to make his own case more persuasive. In this case, Paul would commit the fallacy of the straw man (van Eemeren & Grootendorst 1992, p. 126; van Eemeren & Grootendorst 2004, p. 181; van Eemeren & Grootendorst & Snoeck Henkemans 2002, p. 177).

In subsection III, Paul argues that sexual immorality should be fled because it is spiritually dangerous. In verse 18b, the apostle states that sexual immorality is a sin against the body and in verse 19 that the body is the temple of the Holy Spirit. From these two notions, the implicit argument 1.2a.2 can be deduced: sexual immorality is a sin against the Holy Spirit. Again, Paul tries to prevent the mixing of the holy and the unholy.

In verse 20a, the apostle suggests that God has bought the Corinthians. In other words, they are his slaves and not free. This runs contrary to Paul’s initial position in verse 12: everything is free. During the course of the argumentation the apostle attempts to change the recipients’ attitude regarding them being free to its opposite. Already in verses 12–13, Paul qualifies his radical statements (Lambrecht 2009, p. 480). The Corinthians should act according to the will of their master and not according to their alleged freedom. This tactic of Paul functions as an example of his strategic maneuvering which is scrutinized further.

2.2 Strategic maneuvering
In the analysis of strategic maneuvering, its three inseparable aspects, which can be distinguished analytically, are assessed: topical potential, audience demand and presentational devices (van Eemeren 2010, pp. 93-101, 108-113, 118-122; van Eemeren & Houtlosser 2002, pp. 139-141). The analysis concentrates on finding answers to the questions brought up in the introduction. Consequently, some otherwise interesting issues are left unaddressed.

2.2.1 Topical potential
Paul chooses to defend a standpoint which holds that the Corinthians should not have sex with harlots. The topic of the standpoint is substantially strong, because it can be regarded as a sexual misconduct. The apostle explicates this connection in the argumentation stage by implicitly suggesting that having sex with harlots is a case of sexual immorality (1.2b). Paul uses the topic to shame the audience (Goulder 1999, p. 342; see, Moxnes 1993, pp. 22-24).
Connecting the specific issue, sex with harlots, to a broader topic, sexual immorality, provides Paul a pool of new arguments. In addition, sexual immorality makes the specific issue explicitly a sin and thus more easily condemnable. Moreover, from a linguistic point of view the topics, sex with harlots (πόρνη) and sexual immorality (πορνεία), are easy to connect.

The specific phenomenon described in the standpoint has significant topical potential also because the phenomenon is concrete. The Corinthians, according to Paul, are guilty of a physical sin which also has spiritual implications. It is difficult for the audience to deny the accusation or to interpret their behavior as something else. The concreteness leaves little room for defense: either they are guilty or not.

In the opening and argumentation stages, Paul mentions the three persons of the godhead: God, Christ and the Holy Spirit. That the Corinthians’ behavior affects the three persons emphasizes the spiritual danger of having sex with harlots and, perhaps more importantly, the Corinthians’ failure. They have failed to glorify God (20b), to take into account that their bodies are members of Christ (15a), and that they have sinned against the Holy Spirit (18b–19).

2.2.2 Audience demand
Regarding audience demand, Paul approaches the case indirectly. In verses 12‒14, he puts forward mostly starting points. Paul does not put forward the standpoint until midway the section and it is formulated as a rhetorical question. The apostle also refers literally to himself. Why does he choose the insinuatio-approach? Characteristically this approach is chosen, when the case is considered difficult for the arguer (Lausberg 1998, p. 121).

A plausible reason for Paul to opt for insinuatio is that he contributed to the problems in the congregation with his preaching prior to 1 Corinthians (Rosner 1994, p. 125; Thurén 2009). Consequently, Paul does not accuse the recipients directly of wrong behavior but criticizes, instead, their lack of drawing the correct conclusions from his previous teachings (verse 15a). In addition, in 5:9‒11 the apostle corrects the interpretation of his earlier letter (Hurd 1965, pp. 149-150). Moreover, in 11:2 Paul thanks the audience for taking his message literally. When these features are combined with the apostle’s generally hyperbolical presentation, such as that regarding libertinism in 6:12, radical misinterpretations appear plausible.

Paul addresses the whole congregation even though it is not feasible to render all the recipients guilty of sexual misconduct. However, by accusing the whole community the accusation becomes potentially stronger, since the congregation has to share the blame and the shame that follows. Paul wants to claim an authoritative position over the audience and a shared responsibility of the transgression by them helps in this aim.

2.2.3 Presentational devices
Regarding the presentational devices, the hyperbolical formulation in verse 12 reflects Paul’s own rhetorical position more than that of the Corinthians (Dodd 1995, pp. 39, 54). In order to correct his earlier teaching on libertinism without losing his ethos, the apostle chooses to begin from a position that the audience allegedly accepts. Towards the end of the section, he has changed his stance on freedom almost completely (verses 19–20). The Corinthians are not free but instead God’s slaves. Consequently, they should follow his commands which Paul, as their spiritual father, puts forward. Instead of losing authority, he may have felt it necessary to exaggerate his position to portray himself as the expert of freedom.

Besides establishing common starting points, the “do you not know” –questions are designed to shame the audience (see, Wuellner 1986, pp. 61-62, 72). The Corinthians have failed to take into consideration the apostle’s teachings. Paul suggests that they have not realized what consequences the quote from Genesis 2:24 and the notion of them being bought with a price entail. Instead, they have gotten mixed up with unholiness in having relations with harlots. Paul use of the rhetorical questions to shame the audience corresponds with their function and intent in the previous sections 4:18–5:13 and 6:1–11 (Wuellner 1986, pp. 61-62).

3. Conclusion
In the introduction, it was asked, what function section 6:12‒20 has as part of 1 Corinthians. Is it placed randomly or strategically as a part of the letter? Is there an underlying train of thought that connects it to the other argumentative sections, especially 4:18–5:13 and 6:1–11?

Regarding the dialectical aim, Paul wants to keep the holy and the unholy separate. He wishes to prevent the Corinthians from uniting with harlots and consequently being part of sexual immorality. This goal corresponds with the aim of sections 4:18‒5:13 and 6:1‒11. In the former, Paul argues that the recipients should exile the unholy fornicator from their holy congregation. In the latter, the apostle forbids the community of saints to have their lawsuits in front of the unrighteous judges.

Regarding the rhetorical goal, Paul wants to shame the Corinthians by accusing them of sexual immorality. The apostle chooses the topic of sexual misconduct purposefully to inflict the negative feeling. Prolific presentational devices in this attempt are the “do you not know” – questions, which appear also in sections 4:18‒5:13 and 6:1‒11. In addition, in 4:18‒5:13, he brings up the Corinthians’ failure to exile the fornicator as means to shame them. In 6:1‒11, the apostle explicitly states that he argues to shame them in verse 5. The recipients going to law before unbelievers is shameful, because, for instance, they are allegedly worthy and able to resolve the issues themselves.

Paul approaches the case indirectly, by way of insinuatio. Verses 12–14 consists mainly of common starting points and the standpoint in verses 15bc is formulated as a rhetorical question, which refers to the apostle himself. He cannot blame the Corinthians directly, since he has contributed to the issue at hand with his earlier preaching.

Paul’s overall goal in shaming the audience is to diminish their boasting, for which he blames them in chapters 4 and 5. According to the apostle, the Corinthians think that they do not require his leadership anymore (4:8, 15, 18–19). Paul wants to revive his authority and argue that they still need him, since there are several severe, even shameful, problems in the congregation.

References
Barrett, C. K. (1968). The First epistle to the Corinthians. Black’s New Testament Commentaries. London: Adam & Charles Black.
Butting, K. (2000). Pauline variations on Genesis 2.24: speaking of the body of Christ in the context of the discussion of lifestyles. Journal for the Study of the New Testament 79, 79-90.
Byrne, B. S. J. (1983). Sinning against one’s own body: Paul’s understanding of the sexual relationship in 1 Corinthians 6:18. The Catholic Biblical Quarterly 45, 608-616.
Conzelmann, H. (1975). 1 Corinthians. Hermeneia. Philadelphia: Fortress Press.
Dodd, B. J. (1995). Paul’s paradigmatic ‘I’ and 1 Corinthians 6.12. Journal for the Study of the New Testament 59, 39-58.
Drake Williams III, H. H. (2008). Light giving sources: examining the extent of scriptural citation and allusion influence in 1 Corinthians. In Stanley E. Porter (Ed.), Paul: Jew, Greek, and Roman (pp. 7-37). Pauline Studies 5. Leiden & Boston: Brill.
Eemeren, F. H. van (2010). Strategic maneuvering in argumentative discourse: extending the pragma-dialectical theory of argumentation. Argumentation in Context 2. Amsterdam / Philadelphia: John Benjamins Publishing Company.
Eemeren, F. H. van et al. (1993). Reconstructing argumentative discourse. Studies in Rhetoric and Communication. Tuscaloosa / London: The University of Alabama Press.
Eemeren, F. H. van et al. (1996). Fundamentals of argumentation theory: a handbook of historical backgrounds and contemporary developments. Mahwah / New Jersey: Lawrence Erlbaum Associates, Publishers.
Eemeren, F. H. van & Grootendorst, R. (1992). Argumentation, communication, and fallacies: a pragma-dialectical perspective. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers.
Eemeren, F. H. van & Grootendorst, R. (2004). A systematic theory of argumentation: the pragma-dialectical approach. Cambridge: Cambridge University Press.
Eemeren, F. H. van & Grootendorst, R. & Snoeck Henkemans, F. (2002). Argumentation: analysis, evaluation, presentation. Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers.
Eemeren, F. H. van & Houtlosser, P. (2002). Strategic maneuvering: maintaining a delicate balance. In F. H. van Eemeren & Peter Houtlosser (Eds.), Dialectic and Rhetoric. The Warp and Woof of Argumentation Analysis (pp. 131-159). Argumentation Library 6. Dordrecht / Boston / London: Kluwer Academic Publishers.
Fee, G. D. (1987). The First epistle to the Corinthians. Grand Rapids, Michigan: William B. Eerdmans Publishing Company.
Goulder, M. D. (1999). Libertines? (1 Cor. 5-6). Novum Testamentum 41/4, 334-348.
Heil, J. P. (2005). The rhetorical role of Scripture in 1 Corinthians. Studies in Biblical Literature 15. Atlanta: Society of Biblical Literature.
Hurd, J. C. (1965). The origin of 1 Corinthians. London: S.P.C.K.
Kennedy, G. A. (1999). Classical rhetoric and its Christian and secular tradition from ancient to modern times. 2nd ed. Chapel Hill / London: The University of North Carolina Press.
Lambrecth, J. (2009). Paul’s reasoning in 1 Corinthians 6,12-20. Ephemerides Theologicae Lovaniensis 85/4, 479-486.
Lausberg, H. (1998). Handbook of literary rhetoric: a foundation for literary study. Ed. by David E. Orton & R. Dean Anderson. Transl. by Matthew T. Bliss & Annemiek Jansen & David E. Norton. [Handbuch der literarischen rhetoric.] Leiden / Boston / Köln: Brill.
Moxnes, H. (1993). Honor and shame. In R. Rohrbaugh (Ed.), The Social Sciences and the New Testament Interpretation (pp. 19-40). Peabody: Hendrickson Publishers.
Murphy-O’Connor, J. (1996). Paul: a critical life. New York: Oxford Press.
Murphy-O’Connor, J. (2009). Keys to First Corinthians: revisiting major issues. Oxford: University Press.
Rosner, B. S. (1994). Paul, scripture and ethics: a study of 1 Corinthians 5-7. Leiden / New York / Köln: Brill.
Rosner, B. S. (1998). Temple prostitution in 1 Corinthians 6:12-20. Novum Testamentum 40/4, 336-351.
Thiselton, A. C. (2000). The First epistle to the Corinthians: a commentary on the Greek text. The New International Greek Testament Commentary. Grand Rapids, Michigan / Cambridge, U.K.: William B. Eerdmans Publishing Company / Paternoster Press.
Thurén, L. (2000). EGO MALLON – Paul’s view of himself. In Karl-Johan Illman et al. (Eds.), A Bouquet of Wisdom: Essays in Honour of Karl-Gustav Sandelin (pp. 197-216). Åbo: Åbo Akademi University Press.
Thurén, L. (2009). The Corinthians heresies revisited – a rhetorical perspective to the historical situation. In S. Papadopoulos et al. (Eds.), Saint Paul and Corinth – 1950 Years since the Writing of the Epistles to the Corinthians (pp. 777-788). Athens: Psychogios Publications.
Wuellner, W. (1986). Paul as pastor. The function of rhetorical questions in First Corinthians. In A. Vanhoye (Ed.), L’Apôtre Paul: personnalité, style et conception du ministère (pp. 49-77). Leuven: Leuven University Press.

Abstract: A Philosophy for Children teacher must model a discussion that complies with a critical ideal of reasonableness and use effectively all tools necessary to attract the students’ involvement and participation in a meaningful philosophical dialogue. We distinguish the stages of a Philosophy for Children class where the pragma-dialectical rules and the pedagogical devices instrumental to enhance the students’ participation in a community of inquiry ought to be applied.

Keywords: Community of Inquiry, Philosophical dialogue, Philosophical novel, Philosophy for Children, Pragma-dialectical rules

1. Introduction
The Philosophy for children program, created by Matthew Lipman (Lipman, 1980, 1991), centers around the building of a Community of Inquiry through the practice of philosophical dialogue. The Community of Inquiry is considered as a way to foster critical and cooperative thinking through the balance between competition and cooperation in an atmosphere of mutual respect and understanding, similar to the scientific community in that it pursues similar goals through identical methods (Lipman, 1998, p. 57). The Philosophy for Children teacher is a member of the Community of Inquiry with no special privilege but she must see to it that the logical rules that conduct critical thinking are respected and guide the dialogue among the participants.

There is a deep connection between critical thinking and democratic participation. To participate effectively in democracy it is necessary to be able to argue correctly, to have an informed opinion, to express it clearly and to participate in debates both in small groups and in society at large. We restrict the meaning of ‘critical thinking’ to the fundamental aspect in which most definitions coincide: the ability to participate reasonably in a debate and to solve the controversy reasonably. We consider that the pragma-dialectical rules for a critical discussion (van Eemeren & Grootendorst, 1992, p. 208) provide the fundamental criteria to determine what are a reasonable debate and a reasonable resolution of the controversy. Therefore, the Philosophy for Children teacher will find in them an essential tool for the conduction of philosophical dialogue in the Community of Inquiry.

Through the critical rules for a reasonable discussion, Pragma-dialectics provides the theoretical and the practical tools required to debate reasonably. Our intention in this paper is to show how the Philosophy for Children methodology requires and facilitates the introduction of the critical rules in the classroom. The role of argumentation is crucial both in the building of the Community of Inquiry and in education for citizenship. Because of its cooperative thinking strategies, which facilitate the introduction and practice of the critical rules, the Philosophy for Children methodology seems to us the best tool for the teaching of argumentation.

Certain pedagogical strategies are peculiar to Philosophy for Children. Through them, the process of learning argumentation can be initiated and the critical rules can be mastered. In order to visualize this process, we distinguish the different stages that can occur in a Philosophy for Children class and identify the steps that call for the introduction of the critical rules. This distinction was made in a research project in which we studied the development of democratic attitudes in students and teachers through the implementation of Philosophy for Children (Vicuña & López, 1994). We distinguished five stages:

1. Shared reading of the text,
2. Eliciting questions,
3. Finding relationships between questions,
4. Discussion, and
5. Complementary Activities.

Not all the stages are performed in every class, but usually reading, eliciting questions and discussion are present. We illustrate them by dialogues taken from the program’s novels.

2. Stages in a philosophy for children class
In Philosophy for Children a ‘philosophical novel’ is used as a text from which to start in order to create a common ground for discussion and to connect with the interests of the children. This is a narrative text in which the characters are children who interact with each other and with adults, conversing and wondering about everyday incidents both at home and at school. In them genuine children questions are reflected, which, at the same time, refer to some philosophical problems. This stage prepares the ground for the philosophical discussion that will emerge from the children’s different reactions and questions prompted by the story.

For example, in the philosophical novel Kio & Gus (Lipman, 1992), designed for children in first and second grade of elementary school, Kio narrates the following incident occurred when he went with his grandfather to have lunch in town:

Next to the table where we were eating was a coatrack. It had a sign that said, ‘Watch your hat and coat.’ The coatrack was empty, of course, because it was summertime.
The sign bothered me, so I said, ‘Grandpa, why does it say: ‘watch your hat and coat’?’
He said, ‘Because they might disappear’.
So, I guess there are things in the world that will disappear if you don’t watch them! Isn’t that weird!

By means of this dialogue the story relates to the children’s experience of puzzlement concerning what is real. The students may connect with their own personal experiences of situations that cause them to wonder about the permanence of things beyond our perception.

Another example, taken from Harry Stottlemeier’s Discovery (Lipman, 1982, p. 2), designed for children in fifth and sixth grades of elementary school. Harry, the main character in the novel, after realizing that he has made a mistake in answering to a question from his science teacher, reflects in the following way:

‘So, there are things that revolve around the sun that aren’t planets,’ Harry said to himself. ‘All planets revolve about the sun, but not everything that revolves about the sun is a planet.’ And then Harry had an idea. ‘A sentence can’t be reversed. If you put the last part of a sentence first, it’ll no longer be true.

Harry has discovered a logical law. His discovery will lead him to wonder about the meaning of a sentence that starts with the quantifier ‘all’ and to inquire into the boundaries of logical expressions, such as all, no, some. He will eventually engage some of his classmates’ interest into inquiring further about ‘thinking about thinking’. This gives an idea of how the students reading the story can identify with the thinking processes of the characters and be stimulated to connect with their own ways of thinking.

Matthew Lipman, creator of the Philosophy for Children program, thought of the novels as a means to capture the complexity of children’s experiences and, at the same time, as a way to help them organize them with a sense of unity and wholeness. Each novel contains a story which develops and ends having as background a philosophical theme, such as the knowledge of oneself, thinking rigorously, the discovery of the natural world, the foundation of moral norms, etc. In this way, the students can better understand and make sense of the complexity of their experiences. Besides, every novel refers, from a different perspective, to the philosophical problems discussed in earlier novels.

It could be said that the novels constitute a philosophical knowledge that embraces as in a spiral movement the whole of the children’s experience, which is examined in the different levels of learning. This facilitates the students’ exchange of different perspectives and helps them overcome the frustration produced by a way of teaching that presents knowledge as parceled in diverse areas without connection between them. The children’ need for an integrating experience was among the first things that Lipman realized and he saw that philosophy could provide it.

On the other hand, the reading of the text provides the first stage in the building of a Community of Inquiry. Since the reading is shared by all members, they must take turns, listen attentively, pay attention to the turns, respect each other, avoid correcting or mocking a classmate who makes a mistake, etc. This is their first experience of sharing in the community. The teacher must guide this process in a way that generates an atmosphere of respect and empathy which will help prepare for the respect demanded by the critical rules that will be introduced later.

2.2 Stage 2: Eliciting questions
After finishing the reading of the text the students are invited to formulate questions or to share their impressions about the passage just read. The idea is to connect with the genuine interests of the students, so that the philosophical discussion that would ensue is about these interests and not an ‘adult agenda’ imposed upon the students (Lipman, 1980, pp. 102-128). Their questions and commentaries must relate to the text, not necessarily as an interpretation thereof, but as something that the passage brought to mind. Therefore, it is important to ask the students to explain what the connection is between the reading and their questions and commentaries. In this way, the process of eliciting questions is a search for relevance, but not only in relation to the text, but also in relation to the students’ own thinking. This may put the students in a rather vulnerable position, because their classmates may question their ideas or not understand them and they may be forced to clarify their meaning. This latent process of confrontation gives rise to an analysis and scrutiny which leads them to express what they really think instead of repeating opinions inadvertently introduced in their minds by custom or authority figures. In this search for clarifying the students’ true thinking it is also important for the teacher to question opinions that are presented with the only purpose of impressing the audience or simply to establish a position of power. The questions and problems presented must be those that really matter to them, so that they will be willing to clarify them by discussion and common reflection.

Consequently, the process of eliciting questions calls for a teacher that helps the students clarify their contributions without ‘indoctrination’, that is without taking the advantage of reinterpreting what the students say so as to suit the teacher’s preferred meaning. This role is fundamental in the building of the Community of Inquiry. It requires the ability to balance flexibility and rigor; flexibility to invite and admit all opinions, and rigor to demand that they express clearly their real thinking. Therefore, the teacher must ask the students to reformulate what they want to say until it becomes clear for all. In this way, she ensures a connection with the genuine interests of the students and with the shared interests of the group, in order to achieve both ‘thinking for themselves’ and ‘cooperative thinking’.

From what has been said, it seems clear that the pragma-dialectical rules 1, (Freedom rule) and 4 (Relevance rule) (van Eemeren & Grootendorst, 1992, p. 208), may be introduced at this stage, by reinforcing that all questions and opinions should be allowed to be expressed and, at the same time, that the proponents must be able to show how they relate to the text just read.

2.3 Stage 3: Finding significant relationships between questions
In the next step the students’ questions and contributions are grouped by related themes. The students are invited to find significant relationships between their questions in order to determine the different topics of interest and to decide on the subject of the discussion. This requires a deeper understanding of each contribution and developing sensitivity to relevance. It often occurs that some contributions are too personal and originate a long list of anecdotes which may hinder the coherence and consistency of the discussion. Here again the teacher must balance the student’s eagerness to participate against the weight of their contribution towards the cooperative enterprise. An excess of personal anecdotes may stop reflection and make it impossible to go deeper into the proposed theme. Therefore, the teacher must demand that the students go further than their personal experiences and realize that they are part of a more complex and controversial issue. At the same time as the students are invited to connect with their personal experiences as a basis for reflection, they are also made aware that other members of the class have similar experiences and that all this can be seen from a more general perspective.

Once all the questions and comments have been grouped in this manner, the students decide democratically which of the resulting themes they are going to discuss. The authors of the questions that originated the chosen topic must answer them tentatively and commit to a standpoint. Thus, they take a more critical view of their own opinions and become aware of the help that they can get from other members of the class for clarifying and resolving their doubts. This illustrates cooperative thinking. As an example of this process, we may consider the following dialogue from the novel Harry Stottlemeier’s Discovery (Lipman, 1982, pp. 28-29):

‘What I think Laura’s saying,’ said Jill, ‘is that what we call thinking is something we do, like swimming or walking or riding.’
‘That’s right,’ Laura agreed, ‘that’s just what I mean. When I said before I had a mind, I meant that I mind things. I mind the telephone, or my baby sister, or just my own business. ‘Having a mind’ is nothing but ‘minding.’’
But Fran wasn’t happy with the solution Jill and Laura had arrived at. ‘I agree,’ she said, ‘that maybe the mind isn’t quite the same thing as the brain. I know I said before it was, but I’ve changed my mind.’ Everyone giggled for a while, then Fran went on. ‘What I mean is, you can’t see electricity, but it’s real. So why couldn’t our thoughts be something electrical in the brain?’
This time it was Jill’s mother who told the girls they would have to continue the conversation in the morning. ‘Mom,’ said Jill, ‘what’s a mind?’

Although the conversation narrated does not occur in class but at Jill’s home, where Fran and Laura have been invited to stay overnight, it reflects well the kind of interaction that can take place among the children when they are trying to establish relationships between their questions and to clarify the meanings of their contributions. The girls had been talking about the persistency of some memories, like a musical tune and things like that, and the conversation has turned to whether things outside our minds can make us think about them and finally they have asked themselves what is a mind. An adult is present at the end of the dialogue, Jill’s mother, but she is not presented as an authority figure that would settle the discussion. The girls’ opinions are being refined by their own confrontation and analysis of what they mean by them. They may or may not arrive at a satisfactory opinion about the matter, but even if satisfactory, it would be provisory as long as they are willing to explore and reflect more deeply about it.

2.4 Stage 4: Discussion of the selected themes
Once the discussion themes are selected in the manner explained above, we may say that a genuine interest of the children has been expressed. This stage is previous to the introduction of the pragma-dialectical rule 1, Freedom rule (van Eemeren & Grootendorst, 1992, p. 208), because, in order to identify their genuine interests, the children were invited to compare and establish relationships between their questions or comments and the other children’s questions or comments and not to commit to a standpoint yet.

On the other hand, according to Pragma-dialectics, when a language user expresses a standpoint he commits himself to the truth of his standpoint (van Eemeren & Grootendorst, 1992, p. 31), and is under the obligation of defending it, if questioned (Rule 2, Burden of proof rule). In contrast, in Philosophy for Children a standpoint that conforms to the criteria formulated in the pragma-dialectical rules may take some time to be formed and requires some previous steps.

After the group has decided on which of the proposed subjects is going to be discussed, the person who proposed it must give a preliminary answer. This puts him under the obligation of giving reasons, that is, under the pragma-dialectical critical discussion Rule 2. Demanding reasons is, in fact, one of the basic strategies for conducting a session in Philosophy for Children (Lipman, Sharp & Oscanyan, 1980, pp.121-122). Otherwise it would be very difficult to foster cooperative reflection.

The following dialogue, excerpted from the novel Harry Stottlemeier’s Discovery (Lipman, 1982, pp. 22-24), shows an exchange of argumentation among children about the schools’ quality. (We skip the narrative and give the speakers’ names).

Mark: (….) schools are awful everywhere.
Harry: What makes them so bad?
Mark: Grown-ups. They run the schools to suit themselves. (….)
Maria: Well, but someone has to run the schools, and so it has to be the grownups, because they know more than anyone else. It’s the same with other things. You wouldn’t want to fly on an airplane where the pilot was just a little kid, would you? And you wouldn’t want to go to a hospital for an appendicitis operation where the surgeons and nurses were just little kids, would you? So what else is there to do but let grownup people run the schools because they’re the only ones who can do it right?
Mark: I didn’t think up the idea that kids should run the schools – you did -. (….)
Harry: It isn’t a question of whether the grownups should run the schools, or whether the kids should. (….) The real question is whether the schools should be run by people who know what they’re doing, or by people that don’t know what they’re doing.
Maria: What do you mean, ‘know what they’re doing’?
Harry: Understand, I guess. Whoever runs the schools should understand kids, for instance. I think Mark’s right. Lots of times they don’t. But the most important thing they need to understand is why we’re in school in the first place.
Maria: We’re in school to learn.
Harry: Are we? What are we supposed to learn?
Maria: Answers, I suppose. No, no, I take that back. We’re supposed to be learning how to solve problems.
Mark: Should we be learning how to solve problems, or should we be learning how to ask questions?
Harry: We should be learning how to think.
Mark: We do learn how to think, but we never learn to think for ourselves. These teachers don’t want to admit it, but I have a mind of my own. They’re always trying to fill my mind full with all sorts of junk, but it’s not the town junkyard. It makes me mad.

The children are talking to each other after school. Mark states that “the schools are awful everywhere” and, after being questioned by Harry, adds that the adults are guilty, because “they run the schools to suit themselves.” This shows that he has spontaneously put himself under the obligation of giving reasons (Rule 2). Due to the questionable character of this reason, it is challenged by Maria. She says that the adults must run the schools, because “someone has to run the schools, and so it has to be the grownups, because they know more than anyone else.” Next, she offers a counter argument by analogy. In so doing, she is complying with the critical discussion rules (Rule 8, Validity rule) by using a valid argumentative scheme. On the other hand, Mark complains about Maria’s argumentation: “I didn’t think up the idea that kids should run the schools – you did.” That is, he is accusing Maria of violating the critical discussion Rule 3 (Standpoint rule).

Mark’s proposition has led to an aporetic situation. Harry looks for a new alternative that may help to find a better formulation of what Mark has in mind. He says that the people who run the schools must know what they are doing and this means that they must understand kids. He goes on to state that he agrees with Mark that many times adults don’t understand children, but most importantly they need to understand why the children must go to school.

After Harry’s intervention Mark can formulate more clearly his standpoints: ‘We never learn to think for ourselves.’ ‘The teachers try to fill our minds with junk.’ ‘They don’t accept that we have minds of our own.’

The discussion ends because no one comes back to this point. The controversy is unresolved, but this is not important from the point of view of Philosophy for Children, since the children lack the necessary information to resolve it. It is important, however, to notice Maria’s intervention, when she corrects herself. After she had answered Harry’s question, she thinks for a while and takes it back. The stress is put on the cooperation the children get from each other to formulate and reformulate their thinking, and not in the resolution of the controversy. The critical discussion rules are respected along the process, but the resolution would not be possible at this stage due to the students’ lack of the necessary information.

2.5 Stage 5: Complementary Activities: Discussions of concepts
A frequent type of discussion in philosophy is a discussion about concepts. It is difficult sometimes to find a resolution, due to the fact that definitions are often dependent on many factors, especially on the purposes that the arguer has in mind. However, they constitute an excellent training in searching for assumptions, one of the main characteristics of philosophical dialogue (Lipman, 1980, p.119). In the philosophical novel Pixie (Lipman, 1982 p. 50) we find the following discussion:

Miranda said, ‘Pixie, you know what mother said. We mustn’t let anybody in. Rules are rules!’ ‘But mother didn’t mean that we shouldn’t let in people that we know,’ I insisted. Miranda said, ‘There are many weird people that we know and that mother wouldn’t allow us to let in.’

It is difficult to decide which interpretation is correct. Both seem right. Although we could find some flaws in Miranda’s attitude in trying to impose her authority to Pixie without giving reasons, what she says is true. It adjusts literally to what their parents had said. Pixie’s interpretation, on the other hand, appeals to a more contextual prohibition: “Don’t open the door to anyone!” is not an absolute prohibition; it does not apply to the people they know or are friends with. Without more information about the parents’ intentions, it doesn’t seem possible to resolve the discussion between Pixie and Miranda, but the students’ discussion and analysis of this situation provides an excellent training in searching for assumptions underlying what people say. It is this kind of training what enabled Mark, in the previous example, to realize that Maria was unduly assigning to him a standpoint.

Discussions about concepts open a route to the critical rules that have to do with faulty assumptions (e. g. Rule 5, Unexpressed premise rule, van Eemeren & Grootendorst, 1992, p. 208).

Consider, for instance, the following dialogue, excerpted from El libro de Manuel y Camila [Manuel and Camila’s book’] (Tugendhat, López & Vicuña, 2001, pp. 11-21):

(The children had been discussing about crimes and damages and Sebastián had suggested that killing is not such a serious offense because dead people don’t suffer anymore. This caused much wondering to Camila and she talked about it at home. Her uncle suggested a problem that she could propose to Sebastián. We quote just the dialogue indicating the speakers’ names).

Camila: Suppose that you have committed a very serious crime and are permitted to choose whether you want to be executed or spend the rest of your life in prison. What would you choose?
Sebastián: I would choose to be executed, because the suffering would be rather short in comparison with the interminable suffering of years in jail.
Manuel: I don’t think that you mean it seriously. Death is the worst thing that can happen to you.
Sebastián: Why do you say that?
Manuel: Think about the death penalty. Everybody considers it to be the worst punishment, even though it causes short pain.
Camila: That’s it! In the question of death it’s not a matter of suffering pain.
(A little later in the story Álvaro addresses Sebastián)
Álvaro: Would you really prefer to be killed?
Sebastián: I don’t know, maybe.
Manuel: Only because you think that one doesn’t feel pain?
(Sebastián did not answer, but it was apparent that he felt at a loss).

Camila’s doubts are cleared away when she realizes that Sebastián is not making a distinction between damage and suffering pain. This insight has been possible through the interaction with her friends. From the perspective of Philosophy for Children, this interaction is successful, since an important distinction has been made. Although Sebastián does not want to admit it, the distinction is valid. This means that the critical discussion rule 9 (Closure rule) should apply and Sebastián should retract his original standpoint. But to demand this would mean to violate the spirit of the community of inquiry.

Dialogues in Philosophy for Children are different from the controversies that are the object of Pragma-dialectics. They are a little fragmentary, if compared with the resolution of a controversy. It should be taken into consideration also that children do not satisfy all the conditions of a rational arguer. Nevertheless, by participating actively in these dialogues, children develop certain important reasoning strategies, such as establishing distinctions, detecting underlying suppositions, and making adequate definitions of concepts, which will be fundamental for resolving controversies.

3. Conclusion
Philosophical novels provide models of how thinking and dialogue should be. They differ from the controversies examined by Pragma-dialectics in that they emphasize cooperative discussion. Children learn to listen to their classmates’ opinions and to value them. Although the reasons they may give may have deficiencies and may reflect a very peculiar way of looking at the world, it is essential that they learn to give reasons for their opinions and be aware that they can learn from others. In contrast, the pragma-dialectical objective is the reasonable resolution of a controversy by applying the critical discussion rules.

However, there is a strong connection between the critical discussion rules and the development of a community of inquiry. It wouldn’t be possible without the application of Rule 1. The children learn that all contributions are valid, but they also learn that they must be relevant; they must refer to the pertinent passage of the text. There is complete freedom to formulate questions or comments, as long as they are relevant to the subject under discussion. Cooperation in elaborating a contribution also conducts to tolerance towards the opinions of others and this very tolerance demands that we put ourselves under the obligation of giving reasons.

Some steps are implicit in the applying of Rule 1. To get a speaker to formulate a standpoint and to be prepared to back it up with reasons is a process that has been prepared by the first stages described: reading, formulating questions or comments about the text, and refining this contribution so that it may become a standpoint backed up by reasons. Rule 1 leads to Rule 2.

Rules 3 and 4 were mentioned in connection with the ability to detect underlying assumptions in discussions about concepts. Concepts don’t have definitive borders; they can be applied according to context in a more restricted or a more relaxed way. This kind of debate is referred to in Pixie’s discussion about the meaning of the word ‘anybody’ in the sentence: ‘we mustn’t let anybody in.’ Does this mean ‘absolutely nobody’ or just ‘the people we don’t know’? Also in the passage where Mark complains that he didn’t say what Maria has attributed to him. If the teacher had been present, she could have pointed out that this was a violation of Rule 3. Knowing Pragma-dialectics would grant her fundamental tools for the fostering of critical thinking.

Rule 5 was mentioned in connection with the discussion of concepts, since it relates to the ability for detecting underlying assumptions, but we did not give examples. Anyway the teacher must know well all the rules, so that she can point out the argumentative flaws during the discussion process. The teacher’s corrective role will soon be picked up by the students in what is referred to as ‘the self correcting ability of the community of inquiry.’

Rules 7 and 8 are amply respected in the process of cooperative learning. Although Maria had incurred in an argumentative error by violating rule 3, she is still able to present an argument by analogy: ‘children are not able to run a hospital; therefore, they are not able to run a school’. Learning argumentatively valid forms, albeit in a diffused way, is a fundamental part of learning to think cooperatively. In order to organize this learning, the pragma-dialectical rules and the analytical tools provided are indeed extremely valuable, especially for making explicit unexpressed parts of the argumentation and for evaluating arguments.

Rule 9 is not clearly emphasized in the novels, as was seen in the case of Sebastián. The model of a critical discussion that ends successfully is missing, but this deficiency can be overcome without altering the cooperative spirit of the community of inquiry. On the contrary, a discussion that is successfully resolved emphasizes this cooperative spirit, since it reflects respect for certain rules previously agreed upon.

Rule 10 is amply respected along this learning process. Clarifying the children’s contributions, pointing out to language ambiguities, asking the children to be precise or to explain further the meaning of their expressions is something that the teacher of Philosophy for Children is constantly doing since the very early stages of the program.

References
Eemeren, F. H. van, & Grootendorst, R. (1992). Argumentation, Communication and Fallacies. A Pragma- Dialectical Perspective. Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Lipman, M. (1982). Harry Stottlemeier’s Discovery. New Jersey: First Mountain Foundation for the Institute for the Advancement of Philosophy for Children (IAPC).
Lipman, M. (1982). Kio and Gus. New Jersey: First Mountain Foundation for the Institute for the Advancement of Philosophy for Children (IAPC).
Lipman, M. (1982) Pixie. New Jersey: Institute for the Advancement of Philosophy for Children (IAPC).
Lipman, M., Sharp, A. M. & Oscnyan, F. (1980). Philosophy in the Classroom. Philadelphia: Temple University Press.
Lipman, M. (1998). Pensamiento Complejo y Educación. Madrid: Ediciones de la Torre (Spanish Translation of Lipman, M. (1991). Thinking in Education. Cambridge: Cambridge University Press.)
Tugendhat, E., López, C. & Vicuña, A. M. (2001). El libro de Manuel y Camila. Diálogos sobre
Ética. Barcelona: Gedisa.
Vicuña, A. M. & López, C. (1994). Informe Final del Proyecto Fondecyt 0703-91 [Fondecyt 0703-91 Final Report]. (Unpublished final report)

Abstract: This contribution starts a critical analysis and reconstruction of arguments in classical texts of Islamic theology (of the period AD 900-1100) from the viewpoint of the epistemological theory of argumentation. The main question of the analysis is whether these arguments can be reconstructed as being of one of the universal types of argument identified so far by the epistemological approach. The answer is: yes – though non-deductive arguments are not yet well elaborated.

Keywords: deductive arguments, epistemological theory of argumentation, Islamic theological argument, Koran, universality of argument types

1. Aim and structure of this article
Many classical texts of Islamic theology are heavily argumentative; and much of Islamic theology tries to base faith on valid and sound arguments. Some Islamic theologians even think that Islamic doctrines cannot be defended by revelation alone but have always to be justified by rational arguments. The rational approach in Islamic theology was significantly influenced by the Muʿtazila. But also the Māturīdiyya and the As̲h̲ʿariyya have dealt with kalām (Arabic for speculative theology) and applied rational methods in their theology (see e.g. van Ess 1966, pp. 17-33). This argumentative tradition has nearly not been studied in argumentation theory up to this day. This contribution starts to develop a critical analysis and reconstruction of the arguments in classical texts of Islamic theology from the viewpoint of the epistemological theory of argumentation. The theoretical aims of this study are threefold: First, we want to compile (the beginning of) a list of the most important types of arguments used in these texts, giving particular attention to non-deductive arguments. Second, we analyse them with the help of epistemological criteria in order to establish whether they can be captured in this way, in particular whether all of them are intended (in a broad sense) to be or can be reconstructed as being of one of the universal types of argument identified so far by the epistemological approach (deductive, probabilistic or practical arguments or combinations thereof) or whether there are e.g. specifically Islamic types of argument which should extend the present list of epistemologically valuable argument types or whether, on the other hand, there are (frequently used) argument types in Islamic theology which should be abandoned from an epistemological point of view. Third, we assess the examples with the help of the criteria developed in the epistemological theory of argumentation to gain an impression of the state of the art in classical Islamic theological argumentation. The arguments we will analyse in the following are taken from works by Abū Manṣūr Muḥammad ibn Muḥammad ibn Maḥmūd al-Māturīdī (about AD 870-944), by Abū al-Qāsim al-Ḥakīm al-Samarqandī (about AD 890-950) and by Abū Ḥāmid Muḥammad ibn Muḥammad, known as al-Ghazālī (AD 1058–1111), i.e. texts which were written roughly between AD 900 and 1100, i.e. in European terms at the end of the Early and the beginning of the High Middle Ages, before Thomas Aquinas or William of Ockham in Western Europe.

As just said, the argumentation theory which provides the background and criteria of our analysis is the epistemological approach to argumentation and, more specifically, the Practical Theory of Argumentation developed by one of us, because within the epistemological approach, apart from the profound theoretical justification, it provides the most elaborated and precise criteria for good argumentation, the broadest and deepest systematisation of argument types, and an elaborated theory as well as rules for interpreting arguments.[i]

2. Deductive arguments in medieval islamic theology
Islamic theologians of the period under consideration, of course, also use deductive arguments even of a rather sophisticated type. And since the erudite among them were familiar in particular with Aristotle’s logic they even had a theory of deductive arguments at their disposal.

Nice examples of rather good and sophisticated deductive theological arguments can be found e.g. in Māturīdī’s book Kitāb al-Tawhīd (“The Book of Divine Unity”, AD 944). In this book Māturīdī is arguing, among others, against Christian Christology and the doctrine of Trinity. In a long passage of this book he presents a wealth of independent arguments, which try to show that Christian Christology is self-refuting or contradicting well-known facts. An extract reads as follows:

1. [S1.1] The Christians are divided over Christ, [S1.2] for there are those among them who attribute two spirits to him, [S1.3] one of them temporal, the spirit of humanity which is like the spirits of people, [S1.4] and an eternal divine spirit, [S1.5] a part of God, [S1.6] and this came into the body. […]
[S4] Ibn Shabīb said: I heard one of their associates say that he [Christ] was son by adoption and not son by begetting, just as the wives of Muhammad […] are called mothers, and as a man says to another, ‘My little son’.
2. […] [S5] The Master […] said, Say to them: […] [S9.1] Further, it is well-known that a son is younger than a father, [S9.2] so [S9.3] how can they both be eternal? [S10.1] And if the whole is regarded as being in the body, [S10.2] say to him: [S10.3] Which thing in it is the Son? [S11.1] And if he says: The whole; [S11.2] he has made the whole Son and Father, [S11.3] in this making the Father a son to himself. (al-Māturīdī <944> 2008, pp. 97-99)

This passage does not contain a classical argument indicator; however, sentence S5: “The Master said, Say to them” serves this function. It means: the following are proposals how to argue against assertions of the Christian doctrines summarised in S1[-S4], whose negations, of course, are Māturīdī’s theses. The negation exactly of which thesis entailed in S1 is the thesis sustained by S9 is not made explicit; only the content of S9 allows us to infer that S1.4 is the claim under attack, hence the thesis could be: ‘(The divine spirit in) Christ is not eternal.’ That parts of the formulation of the thesis show up only in the reasons is an indicator for a deductive argument. (In the following reconstructions, “S” indicates a sentence from the argument’s original text; “P” indicates a premise; “A” indicates a hypothetical assumption, which is not used as a premise taken to be true; “L” designates a lemma; “T” is the name for a thesis; “e” as well as “<…>” (angle brackets) indicate insertions included in the spirit of the argument and meant to be acceptable for the arguer; “[…]” (square brackets) indicate insertions, comments etc. made by us, the authors.) The argument can be reconstructed as follows:

<eA1 (= part of S6) assumption (not premise) of a part of the Christian doctrine: The divine spirit in Christ is the son of God (the Father).>
P1 (= S9.1): All sons are younger than their fathers. [This can be formalised as follows: For all x and y holds: if x is son of y, then there is a moment tz, for which holds: y subsists at tz, and all moments tw, at which x subsists, are later than tz.]
<eP2 For all x holds: If there is a moment ty, at which x does not subsist, then x is not eternal.>
P3 (= S9.2): ‚So‘: description of an inferential relation: From S9.1 (= P1) [and eA1 and eP2] follows the implicature of S9.3 (= T1): ‘The divine spirit in Christ and God (the Father) are not both eternal.’ [Māturīdī formulates this assertion as a rhetorical question, which implicates the negation of the main propositional content of S9.3.]
∴ —————————————————————————————————
T1(= S9.3): God the Father and the divine spirit in Christ are not both eternal. [Māturīdī has formulated this conclusion as a rhetorical question, which however implicates a negative answer.]

Here the explicit argument terminates. It suffers from two defects, which, however, can be repaired easily. First, the thesis T1 does not follow. Since one of the premises of the inferential relation described in P3, namely eA1, i.e. ‘The divine spirit in Christ is the son of God (the Father)’, for Māturīdī and Islamic theology is just a hypothetical assumption, only a weaker, conditional thesis follows: eT1*:

<eT1* If (eA1:) the divine spirit in Christ is God’s son, then God (the Father) and the divine spirit in Christ are not both eternal.>

(This implication is logically equivalent to the disjunction:

<eT1.1* (¬eA1): The divine spirit in Christ is not God’s son, or God (the Father) and the divine spirit in Christ are not both eternal.>)

Second, for Māturīdī’s overall aim already T1 is too weak because he does not negate God’s eternity and only wants to attack the assumption of Christ’s divinity. With a further implicit premise the result can easily be strengthened in the spirit of Māturīdī’s argument:

<eT1* If (eA1:) the divine spirit in Christ is God’s son, then God (the Father) and the divine spirit in Christ are not both eternal.>
<eP4 God (the Father) is eternal.>
∴ —————————————————————————————————
<eT2 (¬eA1): The divine spirit in Christ is not God’s son, or (the divine spirit in) Christ is not eternal.>

The second horn of this Christian dilemma is in contrast to S1.4, i.e. to one of the Christian doctrines described in the introduction of Māturīdī’s argument.

The argument is a proof of a contradiction:[ii] from some assumptions of Christian Christology and some trivially true premises follows the opposite of one of these assumptions; hence at least one of these assumptions must be false. The argument in its reconstructed form is deductively valid. Furthermore, the premises P1, eP2 and P3 as such are true; we can leave it open whether eP4 (eternity of God the Father) is true; in particular premise P1, i.e. ‘All sons are younger than their fathers’, is true for natural sons. However, a critical point of the argument is whether ‘son’ in the assumption eA1, i.e. ‘Christ is the son of God (the Father)’, can be interpreted as meaning natural sonship. If in Christian Christology not a natural sonship is meant – which probably is the case – we would have two different meanings of ‘son’ in eA1 and P1; this would make the argument invalid. More precisely this would be a fallacy of missing fit, namely of fallacious ambiguity (Lumer 2000, pp. 415-416). By the way, Māturīdī himself in sentence S4 mentions that in Christian Christology at least sometimes the expression ‘son’ is not interpreted in the usual way, but he ignores this critical point in his argument. He could have restricted his result to Christian doctrines which assume a natural sonship and could have objected to other versions that their use of ‘son’ is more than unclear – which for Māturīdī’s critical purposes would be sufficiently strong. Finally, apart from being true, the argument’s premises are also accepted by Māturīdī’s (Muslim and Christian) addressees – which makes the argument adequate in this respect for convincing rationally. – So, altogether the analysed argument of Māturīdī is quite a good deductive argument though in the end it is fallacious.

This was only one example of a deductive argument in medieval Islamic theology. Of course, there are many more of them. Given this wealth of deductive arguments, the theoretical question is no longer whether there are deductive arguments but whether there are non-deductive arguments in medieval Islamic theology.

A particular important kind is practical arguments, i.e. arguments consisting of listings of advantages and disadvantages of an object which justify a specific evaluation of this object. Practical arguments, though perhaps not of a very explicit form, must have been present in daily life of Muslims of the period under consideration, simply because they reflect the basic way of human decision-making. The search for and analysis of respective examples so far is only the topic of further research.

3. Specifically islamic argument types? – authority arguments from the Koran
One of our theoretically central questions is whether there are specifically Islamic argument types, in particular argument types which could be recognised by the epistemological approach to argumentation as being effective in the sense of leading to true or acceptable (e.g. near to truth) beliefs (i.e. whether they are based on effective epistemological principles (cf. Lumer 2005a, pp. 221-222; 231-234) which have not yet been recognised in epistemological argumentation theory). The most obvious candidates are authority arguments from Holy Scriptures, which are present in Islamic theological texts as well as in theological texts from other revealed religions.

Good and instructive kinds of such authority arguments from Holy Scriptures can be found e.g. in Abū al-Qāsim al-Hakīm al-Samarqandī’s screed against the fatalists who think believers do not need to care for subsistence, since Allah already cares for them. At one point e.g. Samarqandī argues with the help of an authority argument from Holy Scriptures that sometimes believers are obliged to strive for their subsistence – though Allah generally cares for the subsistence of human beings. The translated argument is this:

[S1] At certain times it is a duty to strive for living, [S2.1] because [S2.2] the Koran says: [S2.3] ‘And shake the palm tree’s stem by pulling it towards you! [S2.4] Then it lets plunge juicy and fresh dates on you’ [Koran 19:25], [S3.1] and the Koran says: [S3.2] ‘We have created the day for you in order that you gain your livelihood’ [Koran 78:11]. (al-Samarqandī <950> 1838, p. 40)

The argument indicator in S2.1 tells us that the preceding sentence, S1, is the thesis – whose content is sufficient for refuting the fatalists – and that the following sentences, i.e. S2.2 to S3, are the arguments. Sentence S2.4, i.e. one part of the Koran citation, is not necessary for Samarqandī’s argumentative purposes. The rest is a complex argument with two convergent (i.e. each of them sufficient) reasons for the thesis that sometimes it is a duty to strive for one’s living. Both reasons are – independent and correct – citations of Koran verses about doing something for gaining one’s livelihood.

The explicit argument is rather frugal. The transition from the two explicit reasons to the thesis presupposes two groups of implicit reasons. The first group of implicit premises deals with a general problem to be expected in such arguments from authority of Holy Scriptures, e.g. from the authority of the Koran, and, more specifically, how to get from an invitation expressed in the Koran to an effective obligation. This problem can be resolved by inserting some fairly general premises which can be used in most arguments from the authority of the Koran. These general premises are:

E1 – Principle of revelation: Everything written in the Koran is the word of Allah, i.e. a communication by Allah.
E2 – Principle of divine truth: The propositions of all judgements stated by Allah are true.
E3 – Principle of divine duty: All invitations by Allah constitute a respective divine duty (i.e. a duty enforced by Allah).

The other problems which have to be resolved by a second group of implicit premises regard the transition from what is written explicitly in the Koran to the type of invitation or duty formulated in Samarqandī’s thesis, i.e. a duty to strive for living. The first citation expresses a very concrete invitation, namely to shake the palm tree’s stem, whereas the thesis speaks of an abstract duty to strive for living. The context of the Koran citation makes clear that the addressee, i.e. Mary who is in a desperate situation, by shaking the palm tree will contribute to her livelihood. However, commands and duties are intensional texts; and they do not allow for abstractions. I.e. we can say that by shaking the palm tree etc. she contributes to her livelihood, but this does not imply that if Mary has the duty to shake the palm tree, she necessarily also has a duty to contribute to her livelihood (in this situation). The problem is that from one and the same concrete duty enormously many abstractions could be generated, which in other situations will lead to contradicting duties; and we have no formal principle to choose the normatively correct abstraction. (Of course, in the other direction, from the abstract to the concrete, there are no comparable problems: If we have got an abstract duty we can easily classify more concretely described acts as instances of fulfilling that abstract duty.) Hence such abstractions without further substantial premises are not epistemically justified. Since Samarqandī does not provide such substantial premises we do not see any epistemically and interpretively justified reason to proceed from the concrete to the general in Samarqandī’s first argument; its inference is invalid.

The second argument contains smaller technical problems. In the following reconstruction they are resolved by introducing acceptable premises, which in the end make the argument deductively valid. First, the Koran citation in S3.2 speaks only of Allah’s intention to provide a functional commodity, not of a duty. This gap can be bridged by a general normative teleological principle, i.e. that such natural functions (created by Allah) constitute duties to embrace them. (Such normative teleological thinking is also present e.g. in ancient Greek philosophy.[iii]) Second, the Koran quotation in S3.2 simply speaks of gaining one’s livelihood, whereas the thesis S1 speaks of striving for living. Here a premise is needed which says that the duty to do something implies the duty to strive for doing so.

On the basis of these explanations, Samarqandī’s argument can be reconstructed as follows:

Reconstruction of Samarqandī’s sub-argument 1:

P1 – (= S2.2-S2.3): The Koran says: ‘<Mary,> shake the palm tree’s stem by pulling it towards you’ [Koran 19:25].
<eP2 – (= E1) Principle of revelation: Everything written in the Koran is a communication by Allah.>
<eP3 – (= E3) Principle of divine duty: All invitations by Allah constitute a respective divine duty (i.e. a duty enforced by Allah).>
<eP4 – ‘<Mary,> shake the palm tree’s stem by pulling it towards you’ is an invitation.>
∴ —————————————————————————————————
<eL1 – Mary (in the respective situation) has the divine duty to shake the palm tree’s stem by pulling it towards her.>
<eP5 – Mary’s shaking the palm tree’s stem by pulling it towards her is an instance of striving for her living.>
∴ —————————————————————————————————
T1 – (= S1) At certain times it is a duty to strive for living.

Premises P1, eP4 and eP5 are true; the two principles will be discussed in a moment. The first inference is deductively valid, whereas the second is not because of the intensionality problem.

Reconstruction of Samarqandī’s sub-argument 2:

P6 -(= S3): The Koran says: ‘[reformulated:] Allah has created the day for men in order that they gain their livelihood’ [Koran 78:11].
<eP2 – (= E1) Principle of revelation: Everything written in the Koran is a communication by Allah.>
<eP7 – (= E2) Principle of divine truth: The propositions of all judgements stated by Allah are true.>
<eP8 ‘Allah has created the day for men in order that they gain their livelihood’ is a judgement with the proposition that Allah has created the day for men in order that they gain their livelihood.>
∴—————————————————————————————————
<L2 – Allah has created the day for men in order that they gain their livelihood.>
<eP9 – Normative teleological principle: If Allah creates something in order that a human being can do a certain action (and if He communicates this), then to strive for this action is a divine duty.>
∴—————————————————————————————————
<L3 – Human beings have a divine duty to strive for gaining their livelihood.>
∴—————————————————————————————————
T1- (= S1): At certain times it is a duty to strive for living.

The premises P6 and eP8 are true, and the three inferences are deductively valid; and all this is easily recognisable to be so. (The three inferences, of course, can be contracted to one inference only, thereby omitting the two lemmas.)
These four principles are accepted by Muslims but not e.g. by Christians. We can leave open the question whether the principles are true. In any case they rely on strong metaphysical and empirical presuppositions: that Allah exists; that He communicates with human beings; etc. If (some of) these principles are false, the argument is not argumentatively valid and, according to the epistemological theory of argumentation, a fortiori not adequate for rationally convincing. On the other hand, if these principles are true the argument is argumentatively valid and situationally adequate for rationally convincing Muslims; however, the argument is not adequate for convincing other addressees. This reflects the fact that arguments from the authority of the Koran are, of course, addressed to a specific audience, namely Muslims, who believe in the Koran.

This result leads to the question whether these audience-specific arguments constitute a distinctive, sui generis type of Islamic argument. One could for instance reinterpret the principles – in a Toulminian way – as inference rules. As the reconstruction has shown there is no need to do so; authority arguments from the Koran can as well be reconstructed as deductive arguments with particular premises, namely the principles. Therefore the question is which theoretical conceptualisation is generally more appropriate. From an epistemological viewpoint the reconstruction as a deductive argument with particular premises is better in many respects and worse in none than the alternative systematisation. First, it reveals the epistemological foundations, i.e. logically valid inferences and materially true premises with their different respective procedures of validation, which, in addition, are theoretically well established. Furthermore, core questions of argumentation theory regarding the epistemological effectiveness of argument types are thus separated from argumentation theoretically irrelevant questions about the truth of particular material premises; the potential falsity of a (material) premise then does not affect argumentation theory. The alternative approach has nothing to offer in all these respects. Moreover, the deductive reconstruction is parsimonious in providing only one type of argument with many sub-forms constituted by the deductively valid inferences. The alternative approach instead considers every material principle as the basis of a new argument type without any possibility of systematisation.

4. Hermeneutic arguments in islamic theology
Some Islamic theologians of the period under consideration already use a variety of rather sophisticated hermeneutic arguments.

A good source with a wealth of hermeneutic arguments of various types is Ghazālī’s book Against the divinity of Jesus because Ghazālī accepts the authority of the Bible but attacks its Christian interpretation; in particular he advocates a figurative interpretation of many passages which Christians take literally. One part of his argument is this:

[S1] It is well known that this group [the Christians] uses the word ‘God’ for the Messiah […]. [S2.1] If only I knew whether [S2.2] this is just an honorary title because everything mighty is called ‘God’ [S2.3] or whether they really want to say that he [Christ] is God. [S3] If the latter is intended, then this group is more unreasonable than all the others.
[S4.1] They get into such trouble because they hold to the literal sense, [S4.2] even though certainty is given to the clear understanding that the literal sense is not meant. [S5] However, in every law there is text whose literal sense is contrary to reason. [S6] But then the teachers of the respective law have interpreted the texts.
[S7] A group of significant men has been led to similar things. [S8] One of them said: ‘I am sublime.’
[S9] Another said: ‘How mighty I am!’
[S10] And Hallādj said: ‘I am God. And in this cowl is nothing except God!’ […]
[S13] This is a question of reason, because the literal sense cannot be meant. […] (al-Ghazālī 1966, p. 92)

What is interesting in this text from an argumentation theoretical point of view is that Ghazālī uses a simple version of a hermeneutic Principle of Charity, sketched in S4.2 to S6 and S16.2, by which to seemingly nonsensical texts a reasonable figurative meaning can be attributed:

Principle of Charity (= S4.2–S6; S16.2): If in a [holy text or in the text of an authority or of a significant man or in a] law the literal meaning is contrary to reason (S5), <in particular if it is obviously false,> then the text has to be interpreted (S6): then i. the literal sense is not meant (S4.2); ii. instead, to the <text or> word <that leads to the nonsense> a reasonable meaning has to be attributed (S16.2) <i.e. a meaning which makes the utterance reasonable, in particular one that makes it true>.

This Principle of Charity is formulated in such a way as to make Ghazālī’s argument deductively valid. The principle goes in the direction of present-day principles of charity in rationalising interpretations; it is a big progress for hermeneutics because it provides a methodological way to reveal figurative meaning. Nonetheless, in its present form the Principle of Charity is too simplistic and strict for being true. Taken as an empirical hypothesis about the author’s intention, Ghazālī’s Principle of Charity is false: even authorities believe false propositions and sometimes talk nonsense. The relations expressed in the principle hold only frequently. Weakening the principle in this respect to a statistical truth with high frequency would alter the argument entirely, namely make it a defeasible argument – which, however, probably is an argument structure beyond Ghazālī’s theoretical horizon. Furthermore, the Principle of Charity, apart from presupposing the falsity of the literal meaning, does not use at all further evidences (like the context) which could reveal what the author really meant; thereby it does not take seriously the communicative meaning of utterances. This problem leads to some kind of circularity: the reader of the holy text must already know the truth; he cannot use the text to find out what the truth is, in particular the revealed truth about the divinity of Christ. This makes the holy text worthless as evidence.

In the ensuing part of his argument Ghazālī interprets a passage from Paul’s first letter to the Corinthians with the aim to show that even Paul does not affirm the divinity of Jesus and implicitly even denies it. The passage is hermeneutically rich in using a variety of hermeneutical means: text quotes, references for them, references for assertions about word meanings, a disambiguating argument and the hint to an argument which works out an implicature in the Gricean sense. The disambiguating argument and the argument working out an implicature are very interesting from an argumentation theoretical point of view. Ghazālī has a quite good intuition about the structure of these arguments in bringing together many necessary minor premises. But he does not formulate the major premise, i.e. a principle of disambiguation and a principle for revealing implicatures. In our reconstruction we have formulated such principles on the basis of what is said in the minor premises, adding to this some plausible necessary conditions. However, it would be illusory to strive for an argumentatively useful strict principle; all the viable principles are only frequentist or probabilistic, e.g. the principle of disambiguation:

<eP5 Hermeneutic Principle of Contextual Disambiguation: 1. If a speaker s ascribes a quality called “F” to an object a (cf. P1), 2. where “F” has the meanings ‘F1’ and ‘F2’ (cf. eP2), 3. if, furthermore, s in the respective context attributes the qualities ‘F11’, …, ‘F1n’ to a, which are implied by ‘F1a’ (cf. P3, P4), 4. if in the respective context s does not attribute any quality ‘F21’ to a which is implied by ‘F2a’ (cf. eP6) and 5. if no other (in particular opposite) evidences regarding the meaning of “Fa” are present in s (cf. eP7), then s with “Fa” mostly means ‘F1a’.>

This Principle of Contextual Disambiguation is probably true and makes the inference of the first argument (inductively) valid. However, with such a frequentist premise the argument becomes a defeasible statistical argument with a probabilistically qualified thesis. Such arguments are based on a best-evidence principle, according to which the best evidence has to be included in the argument. All this bursts the structure of deductive arguments. Though Ghazālī is at the edge of defeasible argumentation, probably he could neither formulate such a Disambiguation Principle nor did he see the new quality of this kind of arguing and the technical requirements it brings with it. As a consequence, in his arguments he violates in particular the best-evidence principle. Though he has rather good hermeneutic intuitions these technical gaps are impediments for further formally elaborating his hermeneutical arguments.

5. Conclusion
The preceding analyses have shown that Islamic theological texts of the period between AD 900 and 1100 use a wealth of argument types. Apart from deductive arguments in general, we have identified deductive arguments from the authority of the Koran and a remarkable variety of hermeneutic arguments. None of these argument types requires enlarging the list of good argument types recognised as such by the epistemological approach to argumentation. We have found and analysed deductive arguments, but Ghazālī’s arguments in part can be reconstructed as defeasible, statistical arguments. The latter case is particularly interesting because Ghazālī probably did not know or recognise them on a theoretical level. As a consequence his respective arguments are rather rudimentary and, what is more, he could not avoid several risks of these arguments, in particular that they always need to fulfil the best-evidence condition. This argumentation theoretical limitation probably was one obstacle for further developing theological hermeneutics.

NOTES
i. General overview of the epistemological theory of argumentation: Lumer 2005b. Practical Theory of Argumentation: The general approach is developed and justified in: Lumer 1990; 2005a. A systematisation of existing argument types is developed in: Lumer 2011a. Criteria for particular argument types are developed in: deductive arguments: Lumer 1990, pp. 180-209; probabilistic arguments: Lumer 2011b; 1990, pp. 221-260; practical arguments: Lumer 1990, pp. 319-433; Lumer 2014. For theory and rules of interpreting arguments, see: Lumer 2003; for fallacy theory: Lumer 2000.
ii. The argument evaluation in this paragraph uses the criteria exposed in: Lumer 1990, pp. 187-189; abridged criteria: Lumer 2011, p. 14.
iii. Aristotle, e.g., uses the (empirical) fact that something is a unique function of human beings as a reason for a normative (in a broad sense) claim that fulfilling this function is the supreme good for which human beings should strive (NE 1097b-1098a). Thereby he seems to assume a normative implication of supposed “teleological” facts.

References
Al-Ghazālī, Abū Ḥāmid Muḥammad b. Muḥammad al-Ṭūsī (1966). Al-Ghazalis Schrift wider die Gottheit Jesu. Hg. von Franz-Elmar Wilms. Leiden: Brill (German translation of Al-radd al-jamīl li-ilāhiyyat ʿĪsā bi-ṣarīḥ al-Injīl).
Al-Māturīdī, Abū Manṣūr Muḥammad b. Muḥammad (2008). Kitab al-Tawḥīd (The Book of Divine Unity, ca. 900). In D. Thomas (Ed.), Christian Doctrines in Islamic Theology (pp. 96-117: edition and translation of the section on Christianity). Leiden: Brill.
Al-Samarqandī, Abū al-Qāsim al-Hakīm (1838). Al-Sawād al-Aʿẓam. Būlāq. Online: <http://reader.digitale-sammlungen.de/resolve/display/bsb10249849.html>.
Aristotle (NE). Nicomachean ethics. Translation, introduction, and commentary. Translated and with commentary by Sarah Broadie and Christopher Rowe. Oxford [etc.]: Oxford U.P. 2002.
Lohse, B. (2011). Epochen der Dogmengeschichte. 9th edition. Münster; Berlin: Lit Verlag.
Lumer, Ch. (1990). Praktische Argumentationstheorie. Theoretische Grundlagen, praktische Begründung und Regeln wichtiger Argumentationsarten. Braunschweig: Vieweg.
Lumer, Ch. (2000). Reductionism in fallacy theory. Argumentation, 14, 405-423.
Lumer, Ch. (2003). Interpreting arguments. In F. H. van Eemeren, J. A. Blair, Ch. A. Willard & A. F. Snoeck Henkemans (Eds.), Proceedings of the Fifth International Conference of the International Society for the Study of Argumentation (pp. 715-719). Amsterdam: SIC SAT.
Lumer, Ch. (2005a). The epistemological theory of argument – How and why? Informal Logic, 25, 213-243.
Lumer, Ch. (2005b). The epistemological approach to argumentation – A map. Informal Logic, 25, 189-212.
Lumer, Ch. (2011a). Argument schemes – An epistemological approach. In F. Zenker (Ed.), Argumentation. Cognition and Community. Proceedings of the 9th International Conference of the Ontario Society for the Study of Argumentation (OSSA), May 18-22, 2011 (32 pp.). Windsor, Canada: University of Windsor. CD ROM, ISBN 978-0-920233-66-5. Online: <http://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=1016&context=ossaarchive>.
Lumer, Ch. (2011b). Probabilistic arguments in the epistemological approach to argumentation. In F. H. van Eemeren, B. Garssen, D. Godden & G. Mitchell (Eds.), Proceedings of the 7th Conference of the International Society for the Study of Argumentation, Amsterdam, June 29 to July 2, 2010 (pp. 1141-1154). Amsterdam: Rozenberg; Sic Sat (CD ROM).
Lumer, Ch. (2014). Practical arguments for prudential justifications of actions. In D. Mohammed & M. Lewiński (Eds.), Virtues of Argumentation. Proceedings of the 10th International Conference of the Ontario Society for the Study of Argumentation (OSSA), Windsor, Ontario, May 22-26, 2013 (16 pp.). Windsor, Canada: Ontario Society for the Study of Argumentation (OSSA). CD ROM, ISBN 978-0-920233-66-5. Online: <http://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=2077&context=ossaarchive>.
van Ess, J. (1966). Die Erkenntnislehre des ʿAḍudaddīn Al-Īcī. Übersetzung und Kommentar des ersten Buches seiner Mawāqif. Wiesbaden: Steiner.

Abstract: Application arguments in ethics, from an argumentation theoretic perspective, are rather trivial; however they always rely on moral principles whose justification is a notoriously thorny problem. A critique of several trials of such justifications helps to formulate adequacy conditions for good justifications of moral principles. The main part of the article develops an adequate conception of the justification of moral principles as an argument for a specific thesis about such principles.

Keywords: adequacy of justification conceptions, epistemological approach to argumentation, ethical arguments, ethical constructivism, function of morals, instrumentalist justification of morals, justification of moral principles, moral motivation, practical justification, reduction of argument schemes.
1. The aim of the paper
The abundance of argument types and reasoning approaches to ethics is a real jungle. An impression of the complexity of the various types of reasoning and argumentation of the corresponding theoretical issues is provided in Walton’s “Ethical Argumentation” (Walton 2002). To try to give an overview of this material here is illusory. Rather, I will focus on some, in my opinion, systematically central questions:
1. What types of good central arguments are there in applied ethics?
2. What are the main approaches to the justification of moral principles, and how useful are they?
3. How does the best of these approaches to justification, an instrumentalist, constructivist approach, work in detail and what argument types are used in it?

The brief look at the first question serves only to the discussion, which is thus focused on certain aspects of justification in normative ethics; the article’s main aim is to sketch a systematic conception of justifying moral principles. In dealing with certain questions of how to proceed in normative ethics, the article in itself is metaethical: it provides criteria for good argumentation in normative ethics, but not yet moral principles.

The following analysis of argument types and the criteria for their evaluation are based on the epistemological approach in argumentation theory, according to which the standard function of argumentation is to rationally convince, i.e. to guide an addressee in acquiring knowledge or justified belief.[i] The particular approach to justifying moral principles presented in the following is based on my previous metaethical work, most of which has not yet been published in English.[ii]

2. Argumentation in applied ethics – the recourse to moral principles
According to the most broadly accepted understanding, applied ethics should just apply basic and most general moral principles to groups of more specific typical cases or, in cases of singular decisions of great importance – such as the basic lines of a political or economic system or the determination of climate targets –, even to individual cases. If the moral principles are clear, this application should not be a problem in principle. (“In principle” here means that it is clear how to proceed – which neither rules out the possibility that, for example, very complex or comprehensive empirical information, which is not only expensive to procure and process but may exist only in very uncertain or vague form, is needed, nor precludes that evaluations from the perspective of those affected are very difficult to perform.) The two main types of applied ethical arguments conceived in this way are, first, deontic arguments for deontic judgments (about moral obligations) and, second, consequentialist axiological, in particular welfare ethical, arguments for moral appraisals.

Deontic judgments are judgments with the deontic operators ‘(morally) obligatory’, ‘(morally) forbidden’ and ‘(morally) allowed’. Deontic arguments then are arguments that justify deontic judgments from deontic premises. The default case is that in a deductive argument a more specific deontic claim is derived from, first, a general deontic premise, second, empirical premises and, possibly third, interpretive premises (or lemmata) – whether the empirical situation fulfils the conditions of the deontic premise. In the present context it is decisive that the major premise of deontic arguments be a general deontic norm, ultimately – if one considers the justification of less basic deontic norms on the basis of moral principles (in the strict sense) – a deontic moral principle.

Frequently the final, deductive step of a complex deontic argument is relatively trivial in argumentation theoretical terms. What is often more difficult is the justification of the empirical and especially the interpretive premises, as to whether a certain condition of the norm in question is fulfilled. In legal argumentation theory, there are several basic approaches to this interpretation problem. The two most important are:
1. What counts for the interpretation is the legislator’s intention – this approach can not be applied to moral deontic arguments, because there is no legislator.
2. What counts for the interpretation is the (moral) sense of the norm: Which (morally) desirable state is to be achieved with it? Which (morally) undesirable state it to be prevented? The latter question already regards moral evaluations.

Axiological (moral) arguments are arguments for (moral) value judgments or appraisals. Nowadays, the most broadly accepted understanding (and thus the underlying evaluation criterion) of moral value judgments is consequentialist, in particular welfare ethical (or welfarist). The moral value (or the moral desirability or moral benefit) of an object p is then an aggregation or function of the individual utilities of p for all affected by p. Therefore, in comprehensive welfare ethical axiological arguments, first, it is determined who are the beings affected by p. Second, the expected utilities of p for these various individuals is determined; this is done in practical arguments that ultimately list and evaluate the pros and cons of the assessed object p for the person concerned. The third and final step is really moral: These individual expected utilities must be “aggregated” to the moral desirability of p according to one of the ethical evaluation criteria, e.g. a utilitarian, an egalitarian or a prioritaritarian criterion. This final argumentative step is deductive. In the present context it is again decisive that this applied argument presupposes a moral principle, namely a criterion for moral valuation.

As was just shown, the basic structure of applied ethical arguments is simple and easy to systematise in argumentation theoretic terms. But they always presuppose moral principles, namely basic moral norms or moral evaluation criteria. The real problem of ethical argumentation is the justification of the latter.

3. Arguments for moral principles – some competing approaches and some instructive failures
The currently most important approaches to justifying moral principles are moral realism and value objectivism, methodological intuitionism, the game-theoretic approach and the instrumentalist, constructivist approach.

Moral realism and value objectivism are theories according to which there is a moral reality of norms and values independent of the aspirations, motivations and desires of the moral subjects (e.g. Brink 1989; Dancy 2000; McNaughton 1988; Shafer-Landau 2003). These theories have been criticized in ethics from both an epistemological and ontological standpoints: Moral values and norms conceived in this way are, e.g., ontologically odd entities that also are unknowable; and so far nobody has submitted a valid argumentative justification of realistically conceived norms or values (cf. Mackie 1977, ch. 1). I will not repeat these arguments here. In our context, another criticism is even more important: the type of claim that moral realism and value objectivism try to justify misses the particularity of material ethics: Even if these theories were right, then there would exist just one more sort of layer of reality – in addition to colours, smells, shapes, sounds, etc., and theoretical entities (such as electrons, quarks, etc.), there would be also moral entities such as ‘norms’ and ‘values’. However, this would say nothing about how we should behave with respect to these and other entities. The basic question of material ethics is not: ‘How is the world?’, but: ‘What shall I do (from a moral perspective)?’, ‘How shall I decide (morally)?’ (Hampshire 1949). Value objectivism and moral realism overlook the practical side of ethics, its function of effectively orientating our actions. And this practical side means in particular that the material ethical recognition of some morals must motivate the subject (to some degree) to accept and observe that morality. Ethics that are not designed respectively are pragmatically irrelevant; people do not act on such ethics; and, therefore, ethicists can ignore them too. In positive terms, this means: The statements of material ethics must be designed in such a way that, first, what should be done follows from them and information about the respective situation (informational aspect of orientation), and, second, that knowledge of these statements also mostly motivates to the respective actions (motivational aspect of orientation). I call this condition the “practical” or “motivation requirement.”

“Methodological intuitionism” means here a methodological approach which bases the justification of morals primarily on our moral intuitions. Simple forms of methodological intuitionism accept (unfiltered or, alternatively, well-considered) individual intuitions; more sophisticated forms, such as Rawls’ theory of reflective equilibrium, try to develop an intuitively accepted coherent system from the various intuitions by reconsidering intuitions which lead to incoherence (e.g. Rawls <1971> 1999, §§ 4; 9; Daniels 1996; other intuitionist approaches: Audi 2004; Ewing 1953; Humer 2005; Stratton-Lake 2002). In the most condensed (and therefore only thetic) form the main criticisms of this approach are:
1. Our “intuitions” are not primitive and natural psychological reactions, but the result of a lengthy, culturally, cognitively, emotionally and motivationally influenced ontogenetic development process (Lumer 2002; overview of some theories: Lumer 2014b, 27-29).
2. The recourse to one’s own intuitions is not a justification, but begs the question.
3. Since they dispense with any real justification such intuitions are fickle; in particular, they are in principle vulnerable to the challenge of obtaining new information of all kinds. – One important aspect of these three criticisms can be converted into the positive formal requirement: The justification of moral principles must be stable with respect to new information, i.e. the justification must be such that the practical and motivating acceptance of these principles is not affected by new information.

Game theoretical moral justifications (e.g. Binmore 1994; Gauthier 1986) try to show directly, by means of practical arguments, that a certain kind of moral action is optimal for the agent. In particular, they utilise the fact that the individual benefits for all partners can rise through social cooperation. As opposed to the approaches to justification considered so far, game-theoretic moral justifications are real justifications: They show by practical arguments that certain strategies are optimal. They also meet the two previously established conditions of adequacy for the justification of morals: Game theoretical justifications motivate to comply with morals stably with respect to new information. Problems of a game theoretical-justification of morals lie elsewhere.
1. From the point of view of material ethics, they are very weak, only a minimal or business ethics, which for example do not protect the most vulnerable who have nothing to offer for cooperation (Trapp 1998).
2. Game-theoretically justified ethics of cooperation are structurally flawed in a fundamental way: They do not comprise any moral desirability function and no moral evaluation; thereby they also fail to provide the basis for moral emotions. Accordingly, in such ethics, for example, one cannot say that a collaboration was indeed rational for all parties involved, but was still unjust and morally wrong. (Lumer 2010, pp. 564-568.) – In brief, the flaw of the game theoretic-approach is that it ignores the goal or function of morality. In positive terms, this criticism leads to a further requirement for the argumentative justification of morals: moral instrumentality: The justified morality must meet the objectives or the function of morals.

If one wants to meet the practical requirement and the condition of stability with respect to new information, there seems to be no way to do so without the game-theoretical justification of morals. This seems so because, if it has been shown that a particular strategy is optimal, then there is just no alternative strategy that can be shown to be better and to whose compliance we can be motivated stably with respect to new information. But this reasoning is fallacious. The point of departure of game-theoretical ethics is that it wants to satisfy the practical requirement in a too direct, individualistic situation-bound approach. It is asked directly: ‘What action is optimal in (given) cooperation situations?’ and then the respective action is prescribed (mere individual optimisation). Alternatively, this optimality can also be understood as a necessary and limiting condition which must be fulfilled in the end by a well-constructed morality. So one first constructs a morality whose realisation might also change the action situation of the subject, and also sees to it that, in the end, the observance of this morality is also optimal for the subject – but maybe just because the situation has already been changed (socially prestructured optimisation). In this indirect approach, it is then more likely that the demands of such a morality coincide with our stronger intuitive moral beliefs. This alternative approach is to be pursued below.

Another, fourth approach to justifying moral principles is constructivist and instrumentalist: morality is a good instrument for fulfilling certain social functions (cf. e.g. Mackie 1977, ch. 5). This approach can meet the three previously developed conditions of adequacy. It is further elaborated in the following.

4. Instrumentalist arguments for moral principles – the general idea and adequacy conditions for justification theses
The initial problem for a conception of argumentative justification of moral principles, in particular with an epistemological approach, is the following discrepancy: On the one hand, rational arguments have the standard function of leading to knowledge or cognitions, i.e. justified beliefs, where the objects or contents of these beliefs are propositions, or more precisely: judgments (i.e. propositions with an assertive mode), which make up the argument’s thesis. This is the epistemic side of arguments in general. On the other hand, the objects of moral justifications of moral principles, however, are not judgments but moral principles; apart from moral principles one can also morally justify actions, norms, constitutions, evaluation criteria etc., which are not judgments either. Furthermore, apart from not being the right kind of objects of arguments (viz. judgments), the justifications of such objects should not simply lead to new insights, but also to the practical acceptance of these objects, namely to a particular motivation with respect to these objects. This is the moral and practical side of moral justifications.

The simplest and clearest way to bring the epistemic and the practical requirements together is to design such moral justifications as arguments for a thesis about the object of justification, i.e. about the moral principle, etc. However, this cannot be any thesis; but the justification for this thesis must meet certain conditions; a thesis which fulfils these conditions is the justification thesis for moral principles. In this way the epistemic requirement can be met by the fact that the justification still consists in an argumentatively valid and adequate argument which leads to justified belief, and the practical and moral requirements can be met by selecting a particular thesis about the object to be justified. Now my proposal is that the special conditions for moral justification theses about moral principles are identical to (or a superset of) the adequacy conditions already developed in the criticism of the alternative conceptions of the justification of moral principles. Hence the adequacy conditions for moral justification theses are:

Adequacy Condition 1: Motivation or practical requirement: Moral justification theses about moral principles are motivating in the sense that if a prudent addressee (i.e.: an epistemically and practically rational addressee with certain relevant information) is justifiedly convinced of the justification thesis, he is motivated at least to some extent to adopt and observe the moral principle.

Some reasons for the motivation requirement are:
1. The motivation requirement is the specifically practical component of the conception for justifying moral principles. The development and justification of moral principles are part of practical philosophy and as such should generally have a corresponding influence on the practice, lead to the practical and not only to the theoretical acceptance of the justified object.
2. Fulfilling the motivation requirement ensures the relevance of the insights. One could have infinitely many different insights about moral principles. The vast majority of them would be so arbitrary and irrelevant, that we do not even know why what they say should be a reason for the moral principles. Relevances are constituted, however, – leaving aside our feelings – only by a relation to our motives.
3. A justification which satisfies the motivation requirement has the pragmatic advantage that it can actually make a difference.

Adequacy Condition 2: The motivating effect’s stability with respect to new information: The motivating effect of a justified conviction of a justification thesis is stable with respect to new information, i.e. it is not lost as a consequence of acquiring additional true information.

Some reasons for this condition are:
1. Stability with respect to new information is the rational component of the concept of justifying moral principles. The only thing we can directly rationalise (in the sense of making rational) are beliefs, indirectly also actions and other things. And the two main directions of that rationalisation are: first, to make our beliefs true, i.e. to acquire possibly only true beliefs (or correct false beliefs) by observing epistemological rules and, second, to increase the number of true beliefs. The requirement of the motivation’s stability with respect to new information introduces the practically relevant maximum of epistemic rationality into the conception of practical justification.
2. Stability with respect to new information prevents the justification from being persuasive in a pejorative sense, namely that the addressee practically accepts the object of justification only because he does not have certain information. Stability with respect to new information here introduces an element of wisdom, wisdom in the sense of transcending particular and isolated knowledge toward a comprehensive knowledge about the basic questions of life.
3. Stability with respect to new information contributes to the longevity of the motivating effect.

Adequacy condition 3: Moral instrumentality: Moral principles for which the justification thesis is true, fulfill the function of such principles, they meet the instrumental requirements for such principles and morals in general.

Some reasons for this condition are:
1. Moral instrumentality is the specifically moral component of the conception of justification. If the “justified” moral principles do not fulfill the function of morality we are no longer dealing with a justification of a morality.
2. As a consequence of their moral instrumentality the resulting moral principles correspond more easily to what we intuitively expect from morals.

5. The function of moral valuation: prudential consensualism
The next central question of this conception of the justification of moral principles is, what then is the function of moral principles and of morals altogether? And above all, how can we determine this function and again justify it? I see two approaches for identifying and determining the function of morals. One is idealising-hermeneutic, the other is technical-constructive.

With the idealising-hermeneutic approach, one tries to determine the sense and function of the existing morality. First, one explores the general intentions of the morals of the moral agents, which have to do with the function of morals, in particular the intentions of moral reformers; or one infers from the make-up of the moral institutions themselves which function they might have. In this enterprise not all components of the moral subjects’ intention are interesting, but primarily those components that have to do with the intended purpose or the structure and functioning of morals in general, of general components of morality (norms, evaluations, virtues, etc.) as well as of singular concrete elements, i.e. instruments of this morality. The argumentative means to support such statements about the agents’ intentions are interpretive arguments in which the intentional causes of actions are reconstructed. The collection of such contents of intentions leads only to a series of fragments and often only to superficial ideas or even misconceptions. In the systematically second step of the idealising-hermeneutic analysis, the best must be filtered out from such intention pieces and then synthesised to complete ideals: Which conception of morality composed of such fragments of intentions is the best? Practical arguments for (amoral) value judgments (Lumer 2014a) are used for the argumentative justification of this last step.

Idealising hermeneutical justifications of the function of morals flow smoothly into technical-constructive justifications. The aim of technical-constructive justifications is to create good instruments, thus in this case good conceptions of the function of morality, which are valuable to all moral subjects, and therefore are used by them. The argumentative means for the final technical-constructive justification of a function of morals are practical arguments in which the advantages and disadvantages of these functions for the individuals are presented and the best conception is filtered out.

In order to be able to explain the further course of argumentative justification of moral principles, substantive results about the function of morals are required. There are some formal, structural results on the one hand, and real material results on the other. The most important structural results are the following.
1. The basic principles of morality are, first, the criteria for moral evaluation and, second, moral precepts or norms. The relationship between these components which is technically most fertile and best adapted to the human way of deciding is this: First the criteria for moral evaluation are developed; with their help then in the next step all other objects of morality, i.e. norms, rules, institutions, virtues, etc., are instrumentally justified as being morally good, i.e. producing relatively much moral value.
2. With this setup, the question of the aim or function of morality initially is reduced to the question of the function of moral valuations, evaluation criteria and desirability functions.

With respect to the function of a moral value function, so far I have to offer only a hypothesis about the purpose or sense of a socially binding morality, which – unlike an individual morality – is designed to regulate social relations in an intersubjectively binding way. The sense of a socially binding moral desirability function could be prudential-consensualistic:

1. First, there is the consensualistic requirement: Socially binding moral evaluation criteria constitute a common moral value system that provides the intersubjectively shared standard
(i) for assessing socially relevant measures,
(ii) for planning social projects and
(iii) for consensual arbitration of interpersonal conflicts of interest.
In addition, for the individuals the purpose or sense of such an intersubjectively shared value system could be to procure a benchmark for self-transcendent ego ideals and actions. I call this quality of the desired moral value functions “subject universalism”, i.e. the value of all value objects (or more precisely the value relation of every two value objects p and q (= U(p)/U(q)) of this value function is roughly identical for all (or nearly all [iii]) moral subjects of the moral community. (Expressed somewhat formally: for (nearly) all moral subjects i and j and all value objects p and q holds: Ui(p)/Ui(q) ≈ Uj(p)/Uj(q).) So if e.g. for Adam the present well-being of Clara is better than that of Dora, the same should hold for Bert, i.e. for Bert too the present well-being of Clara is better than that of Dora. Subject universalism has to be distinguished from beneficiary universalism, which is the quality of a moral value function to include all possible beneficiaries of a value function, i.e. the objects to whose fate a non-neutral value in that value function is attributed. Subject universalism does not imply beneficiary universalism analytically, but empirically.

2. Second, there is the prudential requirement: Subject universalism speaks of intersubjectively identical valuations, but what kind of valuations are intended here? The prudential requirement is that the subjective value functions to be compared according to subject universalism be parts or components of the subjects’ prudential desirability functions. Prudential desirability functions express what is good for the respective subject and hence rationally or from a prudential point of view should be the guideline of the subject’s decision; prudential desirability functions are constructed similarly to the utility functions of rational decision theory but with much stricter, philosophically developed standards, which also permit the criticism and correction of the subject’s present instrumental or even intrinsic preferences (cf. e.g. Brandt 1979, part I; Lumer <2000> 2009, 241-428; 521-548). Prudential desirability functions are intersubjectively different – that I have a headache is mainly bad for me and neutral for you, and the reverse holds for your headache –; otherwise they could not express the personal good. Therefore, the subject universalistic requirement is not intended to refer to complete prudential desirability functions but only to parts (considering a certain set of value objects) or components thereof. What is a component of a desirability function? In prudential desirability functions the total desirability of an object p (for the respective subject) is consequentialistically conceived as the desirability (and in the end the intrinsic desirability) of its (p’s) various consequences plus the intrinsic desirability of p itself. The various consequences together with the way they come about are the different aspects of the value object, e.g. the hedonic aspect of bringing about immediate pleasure or pain, the financial aspect of altering the subject’s financial endowment, the empathic aspect of altering the person’s state of compassion etc. A component of a prudential desirability function is then a desirability function constituted of the personal desirability of only one particular aspect of the value objects in question – such as the immediate hedonic, the financial or the empathic component of the desirability function which evaluates the objects only in these respects. – While the consensualist, subject universalistic part of the conception of the socially binding morality expresses more directly the function and instrumentality of morality, the prudentialist part already accommodates the conditions formulated in the first two adequacy conditions for moral justification theses:
(i) To be practically influential and to provide a chance of realisation, the subjective desirability functions the consensus of which makes up subject universalism have to be motivational. Prudential desirability functions are motivational because they rely on subjective (decisional) preferences.
(ii) To be really in the interest of the subject and to be stable with respect to new information, the desirability functions should also be prudential.

6. Arguments for moral principles – the justification theses
After this preparatory work we can now formulate the justification thesis about moral value functions:

‘V is the value function which fulfils the function of moral value functions, and stably with respect to new information, motivates (prudent and informed subjects) proportionally to the V-value.’

More specifically, if we fill in the prudential-consensualistic conception of socially binding morals, the thesis is:

‘The value function V is prudential-consensualistic, i.e. V is proportional to the sum of all subject universalistic parts or components of the prudential desirability functions of (nearly) all moral subjects of the moral community.’

The next step of the justification of morals is to enquire empirically, with the help of empirical decision theory and moral psychology, which desirability function fulfils the condition formulated in the justification thesis. This is beyond the topic of this paper. In other publications (Lumer <2000> 2009, 577-616; 2002), however, I have come to the conclusion that interpersonally (nearly) identical components of our prudential desirability functions arise in particular from our expected compassion and our expected feelings of respect. Adam and Bert may e.g. expect to feel similar compassion for Clara who will have a severe headache as a consequence of an accident, where the compassion in turn is also undesirable for Adam and Bert. If this expectancy and empathic desirability can be generalised, Clara’s headache is morally bad. (Elaboration of a moral value function based on compassion: Lumer <2000> 2009, pp. 616-632.)

So far we have dealt with the meaning, sense or function of moral value criteria. The function of all other instruments of morality, that is of moral norms, rules, institutions, virtues, etc., according to the axiological structural approach followed here, then consists in increasing the moral desirability of the world: they are means to the moral improvement of the world. The conception of their justification is straightforward: They are justified by practical arguments, which show that they have the highest possible moral value among the presently realisable instruments of this kind. The justification thesis about moral norms, rules, institutions, virtues, etc., accordingly is: ‘x is a norm (or rule, institution, virtue, etc.), and x is the morally best (or at least rather relatively good) among the presently realisable norms (respectively rules, institutions, virtues, etc.).’

Again, applying this conception of the justification of moral norms etc. is beyond the scope of this paper. One remark, however, might complete the idea of the conception presented. The moral desirability function always is only one component of an individual’s prudential desirability function such that the motivation to do what is morally good often will be too weak and the respective action will not be executed. The key instrument for resolving this problem and for strengthening the motivation to do the morally good is social norms, i.e. general ways of behaviour that in a certain community are followed almost generally and for which it holds that if they are not followed, punishments will be imposed. If these social norms are morally good then the individual moral motivation plus the fear of punishment together may be sufficiently strong to do the normatively required; i.e. in such a structured situation it will mostly be prudentially optimum to fulfill the moral demands.

NOTES
i. For an overview of the epistemological approach to argumentation see: Lumer 2005b. Some major pieces of my own account within the epistemological approach, i.e. the Practical Theory of Argumentation, are: Lumer 1990; 2005a; 2011a; 2014a.
ii. The most comprehensive exposition is: Lumer <2000> 2009, 30-127. Further elaboration of the instrumentalist aspect: Lumer 1999; 2004; 2010. Motivational basis of morals and ethical justification: Lumer 2002. Preliminary work: Lumer 1995. On the instrumentalist approach in philosophy in general: Lumer 2011b.
iii. The exception that the intersubjective equality of valuation is not fulfilled for some subjects is meant to capture very special cases like psychopaths whose personal value functions simply lack certain components. Of course, such exceptions lead to particular problems. However, no empirically based approach would probably ever work without permitting such exceptions.

References
Audi, R. (2004). The good in the right. A theory of intuition and intrinsic value. Princeton; Oxford: Princeton University Press.
Binmore, K. (1994). Game theory and the social contract. Vol. 1: Playing fair. Cambridge, MA: MIT Press.
Brandt, R. B. (1979). A theory of the good and the right. Oxford: Clarendon.
Brink, D. O. (1989). Moral realism and the foundation of ethics. Cambridge [etc.]: Cambridge University Press.
Dancy, J. (2000). Practical reality. Oxford: Oxford University Press.
Daniels, N. (1996). Justice and justification. Reflective equilibrium in theory and practice. Cambridge: Cambridge University Press.
Ewing, A. C. (1953). Ethics. London: St. Paul’s House & English Universities Press.
Gauthier, D. P. (1986). Morals by agreement. Oxford: Clarendon.
Hampshire, S. (1949). Fallacies in moral philosophy. Mind, 58, 466-482.
Huemer, M. (2005). Ethical intuitionism. Basingstoke, GB; New York: Palgrave Macmillan.
Lumer, Ch. (1990). Praktische Argumentationstheorie. Theoretische Grundlagen, praktische Begründung und Regeln wichtiger Argumentationsarten. Braunschweig: Vieweg.
Lumer, Ch. (1995). Die Lücke zwischen Urteil und Handeln und die Bedeutung von “x ist gut für die Person y”. In Ch. Fehige, G. Meggle (Eds.), Zum moralischen Denken (vol. 1, pp. 254-281). Frankfurt: Suhrkamp.
Lumer, Ch. (1999). Quellen der Moral. Plädoyer für einen prudentiellen Altruismus. Conceptus, 32, 185-216.
Lumer, Ch. (<2000> 2009). Rationaler Altruismus. Eine prudentielle Theorie der Rationalität und des Altruismus. Zweite, durchgesehene und ergänzte Auflage. Paderborn: mentis.
Lumer, Ch. (2002). Motive zu moralischem Handeln. Analyse & Kritik, 24, 163-188.
Lumer, Ch. (2004). Vom Primat der Werte – Wertethik versus Pflicht- und Tugendethik. In R. Mokrosch, E. Franke (Eds.), Wertethik und Werterziehung (pp. 39-61). Göttingen: Vandenhoeck & Ruprecht unipress.
Lumer, Ch. (2005a). The epistemological theory of argument – How and why? Informal Logic, 25, 213-243.
Lumer, Ch. (2005b). The epistemological approach to argumentation – a map. Informal Logic, 25, 189-212.
Lumer, Ch. (2010). Moral desirability and rational decision. Ethical Theory and Moral Practice, 13, 561-584.
Lumer, Ch. (2011a). Argument schemes – an epistemological approach. In F. Zenker (Ed.), Argumentation. cognition and community. Proceedings of the 9th International Conference of the Ontario Society for the Study of Argumentation (OSSA), May 18-22, 2011 (32 pp.). Windsor, Canada: University of Windsor. CD-ROM, ISBN 978-0-920233-66-5. Online: <http://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=1016&context=ossaarchive>.
Lumer, Ch. (2011b). Tre tipi di teorie filosofiche. In R. Davies (Ed.), Analisi. Annuario e Bollettino della Società Italiana di Filosofia Analitica (SIFA) (pp. 45-75). Milano; Udine: Mimesis.
Lumer, Ch. (2014a). Practical arguments for prudential justifications of actions. In D. Mohammed, M. Lewiński (Eds.), Virtues of argumentation. Proceedings of the 10th International Conference of the Ontario Society for the Study of Argumentation (OSSA). Windsor, Ontario, May 22-26, 2013 (16 pp.). Windsor, Canada: Ontario Society for the Study of Argumentation (OSSA). CD-rom, ISBN: 978-0-920233-66-5. Online: <http://scholar.uwindsor.ca/cgi/viewcontent.cgi?article=2077&context=ossaarchive>.
Lumer, Ch. (2014b). Morality in times of naturalising the mind – An overview. In Ch. Lumer (Ed.), Morality in times of naturalising the mind (pp. 3-42). Berlin; Boston: de Gruyter.
Mackie, J. L. (1977). Ethics. Inventing right and wrong. Harmondsworth; New York: Penguin.
McNaughton, D. (1988). Moral vision. An introduction to ethics. Oxford: Blackwell.
Rawls, J. (<1971> 1999). A theory of justice. Revised edition. Oxford [etc.]: Oxford University Press.
Shafer-Landau, R. (2003). Moral realism. A defence. Oxford: Clarendon.
Stratton-Lake, Ph. (ed.) (2002). Ethical intuitionism. Re-evaluations. Oxford: Clarendon.
Trapp, R. W. (1998). The potentialities and limits of a rational justification of ethical norms. Or: “what precisely is minimal morals?”. In Ch. Fehige, U. Wessels (Eds.), Preferences (pp. 327-360). Berlin; New York: de Gruyter.
Walton, D. N. (2002). Ethical argumentation. Lanham, MD: Lexington Books.