NORMATIVE SYLLOGISMS By FRITS GÅVERTSSON
I: INTRODUCTION Aristotle appears to have thought that reasoning—both practical and theoretical—can be represented in syllogistic shape. The passages in the Aristotelian Corpus dealing with the matter are fragmentary and not altogether consistent. This raises the question: “How are we to construe the relation between premises and conclusion in the Aristotelian practical syllogism (and the corresponding relation pertaining to the theoretical syllogism)?” To answer this question we must begin by giving an account of what Aristotle means by the phrase ‘practical syllogism’. The perspective taken on Aristotelian logic in what follows is that of a normative meta-theory rather than a system of formal logic. This suggests treating Aristotelian insights as modelling correct reasoning understood as psychological process, or as ‘an argument-like schema for explanations of actions’1. I thus use the phrase ‘practical syllogism’ to denote a structure that can be used to characterize cases of deliberation or “practical reasoning”.2 In Prior Analytics 24b18-20 Aristotle provides a definition of the central concept of deduction (sullogismos): “a deduction [sullogismos] is a discourse [logos] in which, certain things being stated, something other than what is stated follows of necessity from their being so.” Each of the ‘things being stated’ is a premise (protasis) of the argument and what ‘follows of necessity’ is the conclusion (sumperasma). The central working component of this definition is obviously the notion of ‘follows of necessity’ (ex anankês sumbainein), corresponding to the modern notion of logical consequence. We are thus presented with a general definition of ‘valid argument’. Deduction (sullogismos) thus contrasts with induction (epagôgê), the other species of argument recognized by Aristotle which involves an “argument from the particular to the universal”. In De Anima 434a15-20 Aristotle says of the practical syllogism: “[T]he one premiss or judgement is universal and the other deals with the particular (for the first tells us that such and such a kind of man should do such and such a kind of act, and the second that this is an act of the kind meant, and I a person of the type intended).” In the above-quoted passage no description is given of the conclusion of the practical syllogism; and it is a matter of exegetical controversy whether Aristotle regards the conclusion of a practical inference as a belief in some proposition (e.g. ‘I ought to φ’) on the strength of NE 1147a25-30, as an action (e.g. my φ-ing) on the strength of De Motu Animalium 701a28-33, or, as the formation of an intention to act (e.g. my intention to φ). We can illustrate the three interpretations using a familiar example of St. Thomas’ from De malo (On Evil, question 3, article 9, reply to objection 7, trans.: J. Oesterle, Notre Dame 1993) using the syllogism ascribed to the temperate man:
1 2
McDowell, J., Mind, Value & Reality, (Harvard University Press, 1998): 66. In this I follow McDowell, J., Mind, Value & Reality, (Harvard University Press, 1998): 29n15.
MAJOR PREMISE MINOR PREMISE CONCLUSION
PROPOSITIONAL INTERPRETATION
INTENTIONAL INTERPRETATION
ACT INTERPRETATION
“No fornication is to be committed, this act is fornication, therefore, this act is not to be done.”
“No fornication is to be committed, this act is fornication” The formation of an intention not to fornicate
“No fornication is to be committed, this act is fornication” The act of refraining from fornication
Since, given either the act-interpretation or the intentional interpretation, the conclusion of the practical syllogism is an action or an intentional mental state while the conclusion of the theoretical syllogism is a proposition, it would appear to follow that the relation between premises and conclusion in these two types of syllogism is radically different. Yet it would seem that Aristotle makes it clear that he regards this relation in the two types of syllogism as being the same, e.g. De Motu Animalium 701a8-12, NE 1147a25-30, the latter of which, after Aristotle has described the form of the premises in the practical syllogism, reads: ‘[I]n the cases where these two beliefs result in one belief, it is necessary, in one case, for the soul to affirm what has been concluded, but, in the case of beliefs about production, to act at once on what has been concluded’. It should be pointed out that this demand, i.e. that the relation between premises and conclusion be construed as the same in practical as well as theoretical reasoning, is not merely of exegetical interest. It seems a reasonable demand that, if heeded, allows us to explain how practical and theoretical reasoning are both instances of reasoning. On the grounds of the above it might seem as if we have good reason to accept the propositional interpretation of the practical syllogism. However, it seems that this move renders the practical syllogism almost indistinguishable from its theoretical counterpart concerned with practical matters, or as G. E. M. Anscombe puts it: “‘Practical reasoning’, or ‘practical syllogism’, which means the same thing, was one of Aristotle’s best discoveries. But its true character has been obscured. It is commonly supposed to be ordinary reasoning leading to such a conclusion as: ‘I ought to do suchand-such.’ By ‘ordinary reasoning’ I mean the only reasoning ordinarily considered in philosophy: reasoning towards the truth of a proposition, which is supposedly shewn to be true by the premises.”3 If we thus want to retain the act-interpretation or the intentional interpretation of the practical syllogism we are faced with the problem of accounting for the supposed symmetry of the relation between premises and conclusion in the two types of syllogism. For an additional explication of the practical syllogism we can turn to De Motu Animalium 701a9 seq. which, in Foster’s Loeb translation, reads as follows: “The conclusion drawn from the two premises becomes the action. For example, when you conceive that every man ought to walk and you yourself are a man, you immediately walk; or if you conceive that on a particular occasion no man ought to walk, and you yourself are a man, you immediately remain at rest. In both instances action follows unless there is some hindrance or compulsion. Again I ought to create a good, and a house is a good, I immediately create a house. Again, I need a covering, and a cloak is a covering, I need a cloak. What I need I ought to make; I 3
Anscombe, G. E. M., Intention (Blackwell, 2nd ed., 1976 [1957]): §33.
need a cloak, I ought to make a cloak. And the conclusion ‘I ought to make a cloak’ is an action. The action results from the beginning of the train of thought. If there is to be a cloak, such and such a thing is necessary, if this thing then something else; and one immediately acts accordingly. That the action is the conclusion is quite clear; but the premises which lead to the doing of something are of two kinds, through the good and through the possible” David Wiggins offers—in the context of discussing the above De Motu Animalium passage—an account of the general shape of the practical syllogism: “The first or major premiss mentions something of which there could be a desire, orexis, transmittable to some practical conclusion (i.e. a desire convertible via some available minor premiss into an action). The second premiss pertains to the feasibility in the particular situation to which the syllogism is applied of what must be done if the claim of the major premiss is to be heeded.”4 Wiggins’s schema fits straightforwardly when the major premise specifies a determinate goal—in terms of giving the content of an orectic psychological state—that provides the motivational element of the action explained and the minor premise marks out a course of action as means to that goal. When applied to the case of virtuous action the content of the major premise in this schema is very general indeed: it seems to amount to a specification of an answer to the Socratic question ‘How ought I to live?’(cf. Pl. Rep. 347e), i.e. a conception of happiness (eudaimonia). II: SPECIFICATION OF/IS THE PROBLEM The characterization of the problem of the relation between the premises and the conclusion in the practical syllogism has thus far been left somewhat vague. Simply stating that the relation between premises and conclusion seems ‘radically different’ in the practical and theoretical syllogisms given that we not subscribe to a propositional interpretation does not suffice. In actuality the problem is much more serious than such formulations let on since the adaptation of the act- or intentional interpretations not only results in premises and conclusion belonging to different ontological categories (something that has—despite the general characterisation of sullogismos from Prior Analytics 24b18-20 cited above—been taken to mean that practical syllogisms are not syllogisms at all in a strict sense due to their lacking a propositionally structured conclusion) but rather threatens to obscure the distinction between premises and conclusion. To see this we need to consider cases where ‘the end is instrumentally remote from the agent’s immediate behavioural possibilities.’5 In such cases the end posited does not select from a multitude of available means to the perceived end rendering the ‘this is an act of the kind meant’-clause in need of further specification through deliberation (bouleusis; cf. NE 1112a20ff). This can be represented as follows:
4
Wiggins, D., “Deliberation and Practical Reason”, Proceedings of the Aristotelian Society, New Series, Vol. 76 (1975-1976):29-51;40. 5 McDowell, J., Mind, Value & Reality: 24.
MAJOR PREMISE MINOR PREMISE CONCLUSION
PROPOSITIONAL INTERPRETATION
INTENTIONAL INTERPRETATION
ACT INTERPRETATION
orectic psychological state towards φ-ing (p ⊕ q…⊕ n) as means to φ-ing “I ought to φ by means of (p ⊕ q…⊕ n)”
orectic psychological state towards φ-ing (p ⊕ q…⊕ n) as means to φ-ing
orectic psychological state towards φ-ing (p ⊕ q…⊕ n) as means to φing The act of to φ-ing by means of (p ⊕ q…⊕ n)
The formation of an intention to φ by means of (p ⊕ q…⊕ n)
Let us introduce the following example: Elizabeth has spent a long and laborious day at her offices in Somerville College, Oxford, during which she felt especially pleased with herself for coining a new term to signify a family of ethical theories she considers especially abhorrent. She feels justified in taking a little break and ponders a trip to the Orchard Tea Garden in Grantchester outside of Cambridge suggested to her by Ludwig, a personal friend. She now realizes that she can get from her offices to Grantchester in several ways: she could go canoeing on the river, ride her bike, rent an automobile, or take the train. She ultimately decides to take the train since she considers it the best option and makes her way towards the station. This example, up until the last sentence, which will be dealt with in what follows, can be represented as follows: PROPOSITIONAL INTERPRETATION MAJOR PREMISE MINOR PREMISE (pre-deliberation) CONCLUSION
“I desire to go Grantchester, by (canoe ⊕ bike automobile ⊕ train)
INTENTIONAL INTERPRETATION to ⊕
“I ought to go to Grantchester by by (canoe ⊕ bike ⊕ automobile ⊕ train)”
“I desire to go Grantchester, by (canoe ⊕ bike automobile ⊕ train)”
ACT INTERPRETATION to ⊕
The formation of an intention to go to Grantchester by by (canoe ⊕ bike ⊕ automobile ⊕ train)
“I desire to go Grantchester, by (canoe ⊕ bike automobile ⊕ train)”
to ⊕
The act of going to Grantchester by by (canoe ⊕ bike ⊕ automobile ⊕ train)”
In making her way towards the station Elizabeth employs a familiar process of reasoning that, for it to be represented in the form of a practical syllogism, needs deliberation to overcome the instrumental remoteness of the major premise and the conclusion (cf. NE 112b11-16). In the above case “it appears that any of several [possible] means will reach [the end], [Elizabeth] examine[s] which of them will reach it most easily and most finely” (NE 1112b17-18). Presumably Elizabeth does this through a ranking of the available alternatives (i) canoeing on the river (Ludwig’s recommendation, but impractical), (ii) ride her bike (time consuming but economical), (iii) rent an automobile (swift but expensive), or (iv) take the train (moderate expense, fast enough) in a manner that ultimately results in Elisabeth judging the train as being the best alternative. Note that in one sense there is nothing distinctively practical in this deliberation: it seems to amount to the application of theoretical knowledge concerning the efficacy of certain means given in the pursuit of a given end. Nevertheless, such deliberation seems to exclude certain previously available alternative
courses of action in a manner that seems to initiate action: in judging the train to be the best option Elizabeth is already, in some sense, on her way towards the station, or has at least formed an intention of going there. The relation between the minor premise and the conclusion in the practical syllogism thereby becomes muddled. Let us introduce a dotted line between premises and conclusion to signify this in our illustration. We now get the following: PROPOSITIONAL INTERPRETATION MAJOR PREMISE MINOR PREMISE (post-deliberation) CONCLUSION
INTENTIONAL INTERPRETATION
ACT INTERPRETATION
“I desire to go to Grantchester by train,
“I desire to go to Grantchester by train”
“I desire to Grantchester by train”
I ought to Grantchester train”
The formation of an intention to go to Grantchester by train
The act of going Grantchester by train.
go
to by
go
to
to
The above should not be taken to imply that deliberation (bouleusis) and the closely related notion of choice (prohairesis) are concerned exclusively with ‘what is towards the end’ (tōn pros to telos), i.e. means. Rather, the two concepts must have a wider application if we are to account for Aristotle’s claims that (i) choice is closely connected (oikeiotaton) to virtue, and that (ii) the choices of the bad or self-indulgent man (mochtheros; akolastos) reveals true character (i.e. misconception of the end rather than simple deficiency of means-ends reasoning). 6 This means that a corresponding muddling can occur with regards to the major premise’s relation to the conclusion in cases where the specification required concerns the end, such as in cases where the agent asks herself what ‘doing well’ (the perceived yet to be further specified end) in a particular situation really amounts to, given that we understand Aristotelian ethics as involving an element of uncodifiability. We are thus met with two neat possible solutions to our problem: we could either accept the propositional interpretation of the practical syllogism or argue for full codifiability—in the sense of the major premise as being specifiable in terms of a set of rules from where we can proceed deductively—of Aristotelian ethics. I will in what follows proceed under the assumption that these two strategies are undesirable and that an alternative way of maintaining that ‘it is necessary, in one case, for the soul to affirm what has been concluded, but, in the case of beliefs about production, to act at once on what has been concluded’ and preserve the symmetry is called for. III: ACCEPTING NORMATIVITY Our problem is one of symmetry. Given this one option open to us is to amend our understanding of the theoretical syllogism so that it matches the understanding of the practical syllogism arrived at rather than the other way around. One way of achieving a theoretical syllogism where the relation between the premises and the conclusion becomes muddled is by arguing, pace e.g. John Broome7 and Niko Kolodny8, for the normativity of rationality. 6
On this, cf. Wiggins, D., “Deliberation and Practical Reason”, Proceedings of the Aristotelian Society, New Series, Vol. 76 (1975-1976): 29-51; 31. 7 Broome, J., “Is rationality normative?”, Disputatio, 11 (2008): 153–71. 8 Kolodny, N., “Why Be Rational?”, Mind 114:455 (2005): 509-63.
The thesis that rationality is normative comes in different guises. In the rather trivial sense of being a system of requirements or rules and thus embodying a notion of correctness rationality is obviously normative but not in a way that achieves the required muddling and consequently not the sense here intended. What is required is a conception of theoretical reasoning construed in syllogistic form such that the affirmation of the major and minor premises conjointly generates an affirmation of the conclusion. It is customary in the contemporary debate concerning the purposed normativity of rationality to pose the central question in terms of reasons. This approach, which has been called the reasons first approach9, assumes reasons as the basic normative unit in terms of which the rest of the normative sphere is to be accounted for and explained. On this picture it is straightforward enough to discern two levels of normativity: Rationality would be normative in the strong sense iff [TO BE CONTINUED…] IV: A DIFFERENT CONDITIONAL? Another way of achieving a theoretical syllogism where the relation between the premises and the conclusion becomes muddled is through re-construing the very idea of ‘follows by necessity’ so that it resembles a so-called “Ramsey conditional”. In an infamous footnote Ramsey remarks: If two people are arguing ‘If p, then q?’ and are both in doubt as to p, they are adding p hypothetically to their stock of knowledge and arguing on that basis about q; so that in a sense ‘If p, q’ and ‘If p, q’ are contradictories. We can say that they are fixing their degree of belief in q given p. If p turns out false, these degrees of belief are rendered void. If either party believes not p for certain, the question ceases to mean anything to him except as a question about what follows from certain laws or hypotheses.10 According to this view the acceptance/non-acceptance of the conditional is made through a deliberative thought experiment-process that amounts to hypothetically adding the antecedent to one’s stock of beliefs and on the basis of this considers whether the consequent is true with the result that if one accepts the consequent of a conditional under the condition that the antecedent is true, then one is ready to accept the whole conditional as true. Central to our concern here is obviously the fact that the prominent placing of deliberation in this process makes it greatly attractive given an Aristotelian framework. [TO BE CONTINUED…]
9
Way, J. ”Two Accounts of The Normativity of Rationality”, Journal of Ethics & Social Philosophy 2009. Ramsey, F. P. “General Propositions and Causality” (1929), in Ramsey, F., Philosophical Papers, ed. H. A. Mellor, Cambridge: Cambridge University Press, 1990: 154-155n1. 10
