您的当前位置：首页 Justifying practical reason

Justifying practical reason

来源：尚车旅游网

-XVWLI\\LQJ󰀃3UDFWLFDO󰀃5HDVRQ󰀝󰀃:KDW󰀃&KDLP󰀃3HUHOPDQ󰀊V

1HZ󰀃5KHWRULF󰀃&DQ󰀃/HDUQ󰀃IURP󰀃)UHJH󰀊V󰀃$WWDFN󰀃RQ󰀃3V\\FKRORJLVP

Louise Cummings

Philosophy and Rhetoric, Volume 35, Number 1, 2002, pp. 50-76 (Article)

Published by Penn State University PressDOI: 10.1353/par.2002.0001

For additional information about this article

http://muse.jhu.edu/journals/par/summary/v035/35.1cummings.html

Access provided by Beijing International Studies University (13 Jun 2014 08:29 GMT)

Justifying Practical Reason: What Chaïm

Perelman’s New Rhetoric Can Learn from Frege’sAttack on Psychologism1

Louise Cummings

Introduction

Chaïm Perelman’s new rhetoric represents both the foundation of his nor-mative inquiry into the notion of justice and a fascinating exploration ofargumentation that has relevance for philosophers and nonphilosophersalike. Notwithstanding the undoubted merits of the new rhetoric, the pro-cess of theorizing by means of which it is formulated is inherently prob-lematic. The problematic nature of this process derives from its pursuitwithin the unintelligible perspective of a metaphysical standpoint. In orderto demonstrate the unintelligibility of this standpoint and of the theorizingthat issues from it, I establish a parallel between Perelman’s new rhetoricand Frege’s views on logic. Specifically, I claim that while Frege waversbetween a Kantian and an unKantian (scientific) conception of logic,Perelman wavers between a nonscientific and a scientific conception ofpractical reason. A scientific conception of logic and practical reason, Iargue, makes it seem, respectively, that we can subject the laws of logic toa demand for judgment and that we can justify the ideals of practical rea-son. The unintelligibility of such a demand is demonstrated by Frege whenhe challenges the position of psychologism in logic. Indeed, Frege’s chal-lenge in this case effectively determines the form of the challenge that Imount subsequently against the intelligibility of Perelman’s attempt to jus-tify the ideals of practical reason. Both Frege’s and Perelman’s approachesfail, I argue, on account of their institution within philosophical inquiry ofthe methodology of scientific inquiry. Finally, the scientistic nature of theaccounts of these theorists is examined in relation to comments by HilaryPutnam concerning the structure of philosophical controversy in general.

JUSTIFYING PRACTICAL REASON

1. Why Perelman believed we need a new rhetoric

Perelman’s new rhetoric represents the culmination of his study of justice,a study pursued from within the perspective of logical empiricism (logicalpositivism). Central to that study is the concept of formal justice, an un-derstanding of which is dependent on the further notion of action (justice,on the view of Perelman and other writers, is the manifestation of reason inaction). Perelman describes formal justice as follows: “An act is formallyjust if it observes a rule which sets out the obligation to treat all the mem-bers of a given category in a certain way” (1963, 45).

Such a rule faces a demand for justification. This demand issues fromthe fact that the formal character of this rule renders the rule liable to modi-fications, modifications that then validate as formally just, actions of “realinequality”:

We may wonder, not without reason, whether this indefiniteness over the verycontent of the rule may not lead shrewd minds to evade any accusation offormal injustice, while leaving them almost complete liberty of action andallowing them the fullest possible scope for arbitrary behaviour. Indeed, whenwe wish not to treat according to the rule a member of a certain essentialcategory, there is nothing to stop us from modifying the rule by means of asupplementary condition causing two categories to emerge where formerlythere was only one. This subdivision would at once make it possible to differ-entiate between the treatment of persons who would thence-forward belong totwo different categories. . . . What is the upshot of these thoughts? That it ispossible, by means of a modification of the rule, to avoid formal injustice,andthis can be done in all cases where the rule itself is not prescribed. In allthose cases formal justice can coincide with real inequality due to the arbi-trary factor in the rules. The result is that formal justice plays a very minorrole in all cases in which there is no question of established rules, imposed onthe one who has to observe them. (1963, 46–47; emphases added)

Justification consists in deducing the rules of justice “from a broader,more general principle of which they are merely particular cases” (50)(Perelman also describes these principles as theoretical laws). The hierar-chy of ever broader principles arrived at through this process of justifica-tion establishes the rational content of these rules:

The condition imposed by this requirement [i.e., justification] on the ruleswill now apply, not to their form, but to their content. Nevertheless, this con-dition does not exhaust the content of the rule, but imposes on it integration in

LOUISE CUMMINGS

a system. The consequence of this obligation will be to stress the rationalcharacter of the rules of justice. (1963, 47; emphases added)

To the extent that a rule of justice can be logically deduced from ageneral principle, that rule, Perelman argues, is justified. However, the realproblem of justification emerges, according to Perelman, when the valuethat is the most general principle of the normative system of justice mustitself undergo justification. For early Perelman—the Perelman of The Ideaof Justice and the Problem of Argument—this value cannot be validated bythe standards either of logic or of the empirical sciences and is arbitraryfor this reason:

The most general principles of a [normative system], instead of asserting whatis, establish what has value. They lay down a value, the most general value,whence are deduced standards, norms, commandments. Now this value hasno basis either in logic or in reality. Since its affirmation results neither froma logical necessity nor from an experiential universality, value is neither uni-versal nor necessary. It is, logically and experientially, arbitrary. (1963, 52)

It emerges that Perelman is compelled, on account of his adherenceto logical empiricism, to concede the altogether arbitrary nature of his no-tion of justice—the value that constitutes the foundation of this notion is,according to the precepts of logical empiricism, beyond any and all deter-mination of its own rational status:2

We are thus led to distinguish three elements in justice—the value that is itsfoundation, the rule that sets it out, the act that gives it effect. Only the twolatter elements—the less important, incidentally—can be subjected to the re-quirements of reasoning. (1963, 56; emphasis added)

Troubled by the arbitrariness of justice to which his logical empiri-cist analysis had led, Perelman, in conjunction with Olbrechts-Tyteca,embarked upon a study that, it was expected, would provide a much neededmodel of reasoning about values. This study, called The New Rhetoric,sought to mimic the earlier efforts of the logician Frege, who attempted torenew modern logic by means of the analysis of mathematical reasoning.In the same way, Perelman and Olbrechts-Tyteca believed that a “logic ofvalue judgments” was obtainable by examining how people in a range ofdomains actually do reason about values:

JUSTIFYING PRACTICAL REASON

I could see but one way to solve the dilemma to which most currents of con-temporary philosophy had led. Instead of working out a priori possible struc-tures for a logic of value judgments, might we not do better to follow themethod adopted by the German logician Gottlob Frege, who, to cast new lighton logic, decided to analyze the reasoning used by mathematicians? Could wenot undertake, in the same way, an extensive inquiry into the manner in whichthe most diverse authors in all fields do in fact reason about values? By ana-lyzing political discourse, the reasons given by judges, the reasoning of mor-alists, the daily discussions carried on in deliberating about making a choiceor reaching a decision or nominating a person, we might be able to trace theactual logic of value judgments which seems continually to elude the grasp ofspecialists in the theory of knowledge. (1979, 9)

I will argue subsequently that Perelman’s pursuit of Fregean meth-odology within his account of argumentation is a pursuit that ultimatelyinvalidates that account. In the meantime, however, I examine what consti-tutes the central tenets of Perelman’s theory of argumentation.

2. Perelman’s theory of argumentation

Since his 1963 study of justice, Perelman had remained committed to theview that the value at the foundation of this notion is itself not justifiableon the basis of any logical or empirical scientific norms. However,Perelman’s new-rhetoric view differed from the view of this early studyconcerning the issue of the upshot of this particular consideration. Perelmanwas no longer prepared to accept the skeptical conclusion of his earlierwork—that reasoning about foundational values is impossible and that thearbitrariness of values is responsible for the imperfection of justice3—butinstead he proposed an inquiry into value reasoning, an inquiry that in-volves a contradistinction between justification and demonstration. Accord-ing to Perelman, justification constitutes a form of proof in the practicaldomain of behavior:

The object itself of any justification is very different from the object of ademonstration. The latter is developed from statements or propositions of whichwe can ask whether they are true or false, whereas the former is of an entirelypractical nature: We justify an action, a kind of behavior, a disposition to act,a claim, a choice, a decision. (1980, 58)

LOUISE CUMMINGS

In rejecting both the self-evident and the arbitrary nature of the val-ues at the foundation of a normative system, the proponent of justificationmust concede a central role for the rationality of choice within his notionof justification—if, for example, the axiom of a logical system is self-evi-dent, then choice simply does not enter into our acceptance of that axiom;also, an arbitrary axiom, an axiom that lacks a rational justification, cannotbe arrived at through the equally rational notion of choice. In Perelman’sview, the values at the foundation of a normative system require justifica-tion—they can undergo criticism—whereas that which is beyond criticism(here Perelman cites the conduct of God and of a sovereign power) is alsobeyond justification: “Justification . . . can deal only with debatable things,and usually only with things that have been criticized for specific reasons”(1980, 60). It is within this context of justification that practical delibera-tion takes place concerning the choice of values for the foundation of anormative system. Yet this choice, to the extent that it is a rational choice,must itself conform to certain precedents, models, convictions and values,none of which are nontemporal or universally valid within the nonabsolutistperspective4 of justification. It is in order to establish the rationality ofthese precedents, etc., that Perelman turns to a juridical model of criticismand justification.

Jurists, Perelman argues, “have devised institutions and rules of pro-cedure, according to which, certain people have the power to legislate, othershave the power to govern, and still others the competence to judge and toelucidate the law” (1980, 62). However, given that our quest is one of thejustification of the foundational values of a normative system of justice,our concern is not with the mere fact that certain individuals actually dohave the power or authority to legislate, while others have the competenceto judge. Rather, our concern is one of establishing whether or not thisauthority and this competence have any claim to rationality. The authorityof a legislator derives from the confidence that that legislator enjoys amongthe people whose role it is to elect him or her. To enjoy such confidence,legislators must legislate, not by focusing solely on their “personal deci-sions as to what is just,” but by “taking into account the aspirations of thepublic which is the source of their power” (1980, 63). By assimilating thewishes of the community in his lawmaking, the legislator is formulatinglaws that are politically just:

Politically just laws and regulations are those that are not arbitrary, becausethey correspond to the beliefs, aspirations, and values of the political commu-

JUSTIFYING PRACTICAL REASON

nity. If the restraints imposed by a legitimate power are in accordance withthe wishes of the community, the decisions of that power are politically just,because it is the community’s convictions and aspirations which furnish thecriteria of political action. (1980, 63)

However, although these laws are politically just, we require a no-tion of philosophical justice in order to arbitrate between the often incom-patible wishes of a community. Philosophical justice takes us to the heartof Perelman’s new rhetoric, to his idea of the philosopher engaged in argu-mentation with a universal audience.

Perelman extends his juridical model to an analysis of the role of thephilosopher in argumentation. This philosopher, Perelman claims, strivesto attain universal values in much the same way that the common-law judgestrives to attain the laws of a universal legislation. The philosopher beginswith principles and norms that he deems would be acceptable to the mem-bers of the universal audience and that do not require justification for thisreason (the juridical equivalent of these principles and norms is the notionof a precedent).5 From these initial principles and norms, the philosophersets about the task of a universalization of values, a task that involves theacceptance of values by “all reasonable men,” i.e., by the members of theuniversal audience. The objectivity of these values, Perelman argues, de-rives from the fact that in accepting these values, the philosopher is ac-cepting only those values that he believes would be acceptable to theuniversal audience. Here the comparison is with Kant’s categorical im-perative in the Critique of Practical Reason: “Act so that the maxim ofyour will can always at the same time hold good as a principle of universallegislation” (105). However, Perelman was opposed to Kant’s distinctionbetween subjective maxims and objective laws, a distinction that wouldfind the philosopher’s will described as subjective, while the will of theuniversal audience would be described as objective. Perelman was alsoopposed to Kant’s view that a principle is deemed to be appropriate for“universal legislation” depending on its form. According to Perelman, thepropositions that are presented to the universal audience for acceptancepossess a content, a content that exposes them to challenge—not everyone,after all, is in agreement on what should constitute the content of thesepropositions. However, those values that do assume universal status can-not be judged in accordance with the same logical empiricistic criteria thatvitiated earlier attempts to provide a justification of the notion of justice:

LOUISE CUMMINGS

But assertions that represent the systematic formulation of an ideal cannot bejudged the way we judge factual judgments. Their role is not to conform toexperience, but to furnish criteria for evaluating and judging experience and,if necessary, for disqualifying certain aspects of it. . . . Philosophers who refuseto recognize this primacy of practical reason have often exposed themselvesto the criticism of the positivist by presenting ontologies and theories of be-ing as if they were on the same level as scientific theories of reality. (1980,70–71)

On this view, the ideals of practical reason have conceptual priority/primacy over the criteria of logical empiricism and, for this reason, theycannot be subjected to any judgment or criticism that proceeds on the basisof logical empiricistic criteria. This said, the ideals of practical reason arenot beyond any and all assessment of their rational status. Rather, theirrational status derives from the validation of the universal audience withwhom the philosopher imagines himself to argue. Yet even as he imaginedhimself to argue with this audience, with this idealization of rational opin-ion, Perelman’s philosopher is mindful of the need to structure his argu-mentation in a manner that is most likely to secure the conviction of thisaudience. As Kluback and Becker (1979) remark:

Perelman believes . . . that the philosopher who returns and participates insocial and political reality can only do so with Aristotle’s definition of dialec-tic as the search for the plausible and the reasonable. This is a dialectic ofnon-formal argumentation embraced in dialogue and discourse, historicallyconditioned and manifest in opinion and common sense. The argumentationof this non-formal logic presupposes an audience before whom ideas on alllevels must find their acceptance or rejection. How we argue a point of viewdepends on the audience we want to convince and the techniques we must useto make this conviction valid. (39)

So it is that Perelman’s philosopher must argue, for example, fromprinciples that are acceptable to the universal audience6 (the philosopherwho fails to so argue is guilty of petitio principii). And while the philoso-pher conducts his argumentation in accordance with certain norms, he doesso in the knowledge that these norms are not validated by criteria outsideof philosophy itself: “There exist no nonformal criteria transcending allphilosophy to which rational argumentation has to conform” (1980, 74).Rather, these norms “express the convictions and aspirations of a free butreasonable man, engaged in a creative, personal and historically situatedeffort: that of proposing to the universal audience as he sees it, a number of

JUSTIFYING PRACTICAL REASON

acceptable theses” (1980, 74). Of course, the mere fact that this effort ishistorically situated renders these norms susceptible to revision—the phi-losopher cannot predict how future occurrences and the advancement ofknowledge will change the norms according to which he structures the vari-ous theses he proposes to the universal audience. However, irrespective ofthe norm modifications that do occur, these modifications still relate tophilosophical criteria, not to absolute and self-evident criteria:

For lack of impersonal criteria, the philosopher must ultimately rely on philo-sophical constructs to support his beliefs and ideals. For him such constructsare the last resort in matters of rationality and justice. But he remembers thatfor philosophy there is no res judicata .(1980, 75)

Clearly, Perelman engaged in extensive analysis in an attempt to ar-rive at a justification of the notion of justice. And the various observationsand claims that he made as part of this analysis are undoubtedly of value toany theorist with an interest in argumentation in general, and in legal argu-mentation in particular. However, notwithstanding the interest and valueof Perelman’s analysis, I want to argue that, quite apart from serving as ajustification of the values at the foundation of a normative system of jus-tice, this analysis creates the illusion of such a justification. This illusion,I will argue, is indicative of a metaphysical standpoint in theorizing. Inorder to expose this standpoint in the case of Perelman’s analysis, I beginby examining a criticism of Frege pursued by James Conant and by HilaryPutnam.

This criticism consists in the claims that Frege employs two concep-tions of logic, a Kantian conception and an unKantian conception, and thatFrege’s unKantian conception of logic is itself only possible on the (unin-telligible) assumption that we can assume a metaphysical standpoint. Thesediffering conceptions of logic, it is argued, create a tension within Frege’sanalysis of logic. I contend that Perelman’s attempt to follow, in his theoryof justice, the method of analysis pursued by Frege in his investigationsinto logic, results in a tension in Perelman’s account that is similar to thetension in Frege’s analysis. Where for Frege the tension is between a Kantianand an unKantian conception of logic, the tension in Perelman’s thinkingmanifests itself in two conceptions of practical reason, a scientific concep-tion and a nonscientific conception. Perelman’s scientific conception ofpractical reason is motivated, I argue, by the same assumption of a meta-physical standpoint in theorizing that motivates Frege’s analysis. Indeed,

LOUISE CUMMINGS

it is this standpoint, I claim, that makes it seem that a justification of theideals of practical reason is even possible. I describe later how the expo-sure of this standpoint entails the immediate rejection of the demand tojustify practical reason. In the meantime, however, I examine Conant’s andPutnam’s claim that Frege employs two conceptions of logic.

3. Frege’s Kantian and unKantian conceptions of logic

Frege’s Kantian conception of logic, Conant argues, consists essentially inthe view that “accord with the laws of logic is constitutive of the possibil-ity of thought” (1991, 134). These laws, Frege states in the introduction totheGrundgesetze, form “the most general laws of thought . . . [that] pre-scribe universally the way in which one ought to think if one is to think atall” (The Basic Laws of Arithmetic, 12). In emphasizing the general natureof these laws of thought, Frege, like Kant before him, is emphasizing thatthe laws of logic embrace all of our reasoning: these laws, it is argued,constitute rationality itself. In addition to this Kantian conception of logic,Frege pursues the distinctly unKantian view that logic is a branch of posi-tive science. Logic, on this alternative conception, makes no reference tothe properties and relations of sciences such as physics and geometry.Rather, it articulates absolutely general substantial truths, truths that arethe province of no particular science (these truths, it should be noted, pos-sess a content; on Kant’s conception of logic, the laws of logic are purelyformal rules). Frege tries to integrate his Kantian and unKantian concep-tions of logic as follows:

It will be granted by all at the outset that the laws of logic ought to be guidingprinciples for thought in the attainment of truth, yet this is only too easilyforgotten, and here what is fatal is the double meaning of the word “law.” Inone sense a law asserts what is; in the other it prescribes what ought to be.Only in the latter sense can the laws of logic be called “laws of thought”: sofar as they stipulate the way in which one ought to think. Any law assertingwhat is, can be conceived as prescribing that one ought to think in conformitywith it, and is thus in that sense a law of thought. This holds for laws of geom-etry and physics no less than for laws of logic. The latter have a special title tothe name “laws of thought” only if we mean to assert that they are the mostgeneral laws, which prescribe universally the way in which one ought to thinkif one is to think at all. (The Basic Laws of Arithmetic, 12)

JUSTIFYING PRACTICAL REASON

It emerges that on Frege’s view the laws of logic are both descrip-tive, asserting what is, and prescriptive, prescribing what ought to be. Fregetries to integrate these two approaches to logic, claiming that the laws oflogic, conceived of as articulating general substantial truths about the world,prescribe how one ought to think if one wants to think in accordance withthe truth: “Any law asserting what is, can be conceived as prescribing thatone ought to think in conformity with it . . . ” (similarly for physics, lawsstating substantial truths about the physical world prescribe how one oughtto think about physical phenomena if one wishes to think in accordancewith the truth about such phenomena). Frege effectively recasts Kant’sconceptualclaim that the laws of logic form a precondition on the verypossibility of rational thought, as a scientific claim about the world, that inorder to arrive at the truth of a matter, one must think in conformity withthe laws of logic.7 Indeed, it is in relation to the issue of the scientificstatus of logic that the views of Kant and Frege diverge most significantly.Frege’s view that the laws of logic possess a content is in sharp contrast toKant’s view that the laws of logic do not, in themselves, provide us withknowledge of objects and that logic is an infertile science. These very dif-ferent views of the scientific status of the laws of logic are at the center ofa tension in Frege’s thinking, a tension that is identified and discussed byConant and Putnam, and by Wittgenstein before them. However, beforeaddressing this tension, I examine a notion that is related to it, Frege’sconception of judgment.

For Frege, forming a judgment involves moving from the sense of athought to its truth-value: “A propositional question contains the demandthat we should either acknowledge the truth of a thought, or reject it asfalse” (Collected Papers on Mathematics, Logic, and Philosophy, 373).

The satisfaction of the demand for the establishment of the truth orfalsity of a thought—Conant, following Thomas Ricketts, refers to thisdemand as a demand for judgment—proceeds in terms of processes of ra-tional thought, processes that make possible a range of mental activities,such as the understanding of language. However, only one of Frege’s con-ceptions of logic can accommodate, or at least this is how it seems, thesatisfaction of this demand. Frege’s unKantian (scientific) conception oflogic, it will be recalled, takes the laws of logic to possess a content. How-ever, as Conant remarks in his commentary on Frege, “In grasping the con-tent of a thought, we grasp that either it or its negation is true—this is aconstitutive feature of what it is to grasp the content of a thought” (1991,136; emphasis in original). In other words, in grasping the content of a

LOUISE CUMMINGS

thought, we are, according to Frege, effectively satisfying the demand forjudgment. Now if, as Frege is claiming, the laws of logic possess a content,then it must be possible to satisfy the demand for judgment in the case ofthese laws. The satisfaction of such a demand, it was argued above, re-quires the presence of processes of rational thought. However, herein liesthe tension in Frege’s account. For according to Frege’s other conceptionof logic, his Kantian conception, these processes, and the demand for judg-ment that is dependent on them, are unintelligible when the laws of logicthemselves are the subject of judgment. The laws of logic, on Frege’sKantian conception of logic, constitute the rational framework within whichall thinking proceeds. It is this framework that makes thinking and judgingpossible. Outside of this framework there exists only nonsense, just plainnonsense, not philosophically significant nonsense of the sort standardlyattributed to Wittgenstein.8 Outside of this framework we cannot possiblyconceive of what it would be to engage in judgment about a logical law, to“acknowledge” either the truth or the falsity of such a law. Indeed, on thisKantian conception of logic, it is only a certain perspective, a metaphysi-cal perspective, that makes these extralogical processes of rational thoughtand the demand for judgment that proceeds on the basis of these processeseven appear possible. I will have more to say shortly about this metaphysi-cal perspective. In the meantime, I examine Frege’s treatment of an issuein philosophy that is related to logic, the issue of psychologism, as a meansof determining the respective validity of Frege’s two conceptions of logic.

4. Challenging psychologism: Exposing the unintelligibility of

Frege’s unKantian conception of logicThus far, a certain tension in Frege’s view of logic has been established, atension between Frege’s Kantian and unKantian conceptions of logic inrelation to the issue of judgment in the case of the laws of logic. Basically,the tension is this—the one and the same demand for judgment in the caseof the laws of logic is unintelligible within Frege’s Kantian conception oflogic, yet is essential to Frege’s unKantian conception of logic. However,in merely noting this tension, I am saying nothing about the validity of thetwo viewpoints that jointly generate this tension. In order to address thequestion of the validity of these viewpoints, I turn to an examination ofFrege’s discussion of psychologism. The significance of psychologism in

JUSTIFYING PRACTICAL REASON

the present context is that it shares with Frege’s unKantian conception oflogic the assumption that logic is a science. It will become evident from anexamination of this discussion that Frege subordinates his unKantian con-ception of logic to his Kantian conception.

When Frege proposes a thought experiment that involves thepsychologistic philosopher of logic, the purpose of this experiment is todemonstrate that the psychologistic philosopher of logic “conflates the lawsof psychology (the laws of takings-to-be-true) with the laws of logic (thelaws of truth), and . . . , through this conflation, ends by completely blur-ring the distinction between the subjective and the objective” (Conant 1991,142). Through his conflation of the laws of logic with the laws of psychol-ogy, the psychologistic philosopher of logic pursues an account of logic inwhich the fundamental laws of logic are none other than empirically estab-lished generalizations. These generalizations represent inductions basedon the evidence of the actual inferences that individual thinkers make. Theyhave a scientific status that is similar to the scientific status of generaliza-tions in other branches of empirical scientific inquiry—hence, the earlierclaim that psychologism shares with Frege’s unKantian conception of logicthe assumption that logic is a science. However, it is just this scientificstatus of the laws of logic, conceived as empirically established generali-zations, that undermines not only psychologism but also Frege’s unKantianconception of logic. Indeed, it is in an attempt to demonstrate the unintel-ligibility of psychologism that Frege conceives of his thought experimentinvolving logical aliens.

Frege invites the reader to envisage a group of beings whose infer-ential practice differs significantly from our own inferential practice. Theselogical aliens deny the logical law of identity, while we readily subscribeto this law. In subscribing to or denying this particular logical law, thepsychologistic philosopher of logic will argue, we and the logical aliensare doing the only thing that is possible9 for us given certain psychologicalfacts about our minds: we cannot fail to affirm the law of identity given thepsychological facts about our minds; the logical aliens cannot fail to denythe law of identity given the psychological facts about their minds. Each ofour inferential practices is validated by the psychological context in whichit finds expression. Now the relativity (subjectivity) of such a view is clear—according to this view, the laws of logic are inextricably linked to appro-priately circumscribed populations of thinking subjects. Yet it is also clearthat the psychologistic philosopher of logic is making an objective claimabout what kind of a thing a logical law is: notwithstanding the variations

LOUISE CUMMINGS

in logical laws that may occur between subjects, logical laws are alwayslaws that govern the psychological process of reasoning:

When the psychologistic logician first presents his theory, he seems to besuggesting that it represents the truth about certain matters. He is telling uswhat kind of a thing a law of logic is: it is a law which governs the psycho-logical process of reasoning. This account of what kind of a thing a logicallaw is has the appearance of being perfectly general: it is true of beings whoreason as we do, but it will also be true of beings who reason in some otherway (such as our friends, the logical aliens). It appears that we are being of-fered a theory which can encompass our inferential habits and theirs fromsome broader vantage point. (Conant 1991, 145; emphasis in original)

The psychologistic logician wants to make an objective claim aboutthe laws of logic, a claim to the effect that the laws of logic are alwayslaws that govern the psychological process of reasoning. Yet it is also anempirical possibility according to his account that some subjects, for ex-ample, the logical aliens, will not only affirm laws of logic quite differentfrom the laws of logic that we affirm, but will also deny the very laws oflogic that we affirm. It seems that, according to the psychologistic logician’saccount, there is at least the possibility of disagreement between ourselvesand the logical aliens concerning the issue of what constitutes the laws oflogic. However, this possibility is in appearance only, Frege argues, for thereason that the concept of disagreement presupposes laws of logic, laws oflogic that are entirely absent from the metaphysical standpoint that we mustassume in order to acknowledge this possibility. In the absence of a sharedframework of logical laws, no sense can be given to the claim that the lawsof logic affirmed by the logical aliens differ from the laws of logic that weaffirm:

It originally looked as if the psychologistic logician wanted to hold on to theidea that logically alien thought conflicts with ours, but his account deprivesthe notion of one proposition’s conflicting with another of the context in whichit has its life. The underlying claim which fuels Frege’s argument here is thatone can only recognize two judgments as being in conflict with one another ifthe framework of logic is already firmly in place. For the criteria by which weare able to so much as recognize (let alone adjudicate) an instance of dis-agreement presuppose the availability of this shared framework. (Conant 1991,146; emphasis in original)

The psychologistic logician, Frege argues, takes a conflict betweendiffering conceptions of the laws of logic to represent a possibility from

JUSTIFYING PRACTICAL REASON

within his account of logic. However, it now emerges that this possibilityis illusory in nature, for the reason that we cannot even make sense of thenotions of conflict and disagreement in the absence of a prior frameworkof logical laws. The psychologistic logician, like Frege’s scientific logi-cian described earlier, expects to submit the laws of logic to a demand forjudgment in much the same way that one expects to submit a statement ofscience to this demand. This demand in the case of a statement of scienceis entirely rational in nature—in presupposing rational thought, scientificstatements presuppose the very processes from which the satisfaction ofthe demand for judgment proceeds. However, where this same demand isnot rational,10 but is unintelligible, is in the case where it is the laws oflogic themselves that are subject to judgment. Frege wants us to see that itis not possible to judge the laws of logic, because it is these laws that makejudgment itself possible: “Frege’s ultimate aim in the thought experiment .. . is to try to get his interlocutor to see the force of the (Kantian) point thatthere isn’t any sense to be made of the idea of undertaking to disagree witha principle of logic—that it is these principles which make both agreementand disagreement possible” (Conant 1991, 147; emphasis added). Disagree-ment and the demand for judgment, it thus emerges, are the unintelligiblemanifestations of a conflation of science with logic, a conflation that per-meates both the psychologistic logician’s and the scientific logician’s con-ceptions of logic.

The psychologistic logician and Frege, to the extent that Frege iscommitted to a scientific (unKantian) conception of logic, proceed to theo-rize about logic in the same way they proceed to theorize about science.Specifically, they view the most fundamental principles of logic, the lawsof logic, as being empirically established generalizations of the same orderas the generalizations that are pursued by scientific inquiry in its variousforms. Moreover, the empirically established generalizations that are thelaws of logic are subject, they believe, to the same rational appraisal towhich the generalizations of science are subject (the numerous efforts, bothexperimental and argumentative in nature, that are undertaken in order torationally appraise scientific claims, can be seen as attempts to satisfy aFrege-type demand for judgment, in that they require scientists to acknowl-edge the truth of a scientific claim or, alternatively, to reject a scientificclaim as false). However, notwithstanding the aspirations of both Fregeand the psychologistic logician for the attainment of a scientific concep-tion of logic,11 the conflation of science with logic that is undertaken bythese theorists serves only to render such a conception of logic unintelli-

LOUISE CUMMINGS

gible. The manifestations of this unintelligibility are Frege’s demand forjudgment in the case of the laws of logic and the psychologistic logician’snotion of disagreement between competing conceptions of the laws of logic.Both this demand and this notion of disagreement are possible only on theunintelligible assumption that we can occupy a metaphysical standpoint inthe course of our deliberations on logic, a standpoint from which it seemsthat we can judge and disagree about the laws of logic without the laws oflogic being presupposed by such judgment and disagreement. This sameunintelligible assumption, I now want to argue, motivates Perelman’s ac-count of practical reason, particularly Perelman’s attempt to justify practi-cal reason.

By way of demonstration of this claim, I examine how Perelman’sanalysis of practical reason employs two quite different conceptions of thisnotion, one scientific and the other nonscientific. Perelman’s scientificconception of practical reason, I argue, constitutes an attempt to theorizeabout issues of justice in action in much the same way that scientists pro-ceed to theorize about physical phenomena. In attempting to theorize aboutpractical reason, Perelman, I contend, is like the psychologistic logician inattempting to theorize from a metaphysical standpoint. Moreover, the sametheoretical anomalies that were shown above to characterize thepsychologistic logician’s account of the laws of logic from within this stand-point—for example, the psychologistic logician’s attempt to make objec-tive claims in relation to the laws of logic from within the subjectiveconstraints of his own account—can also be shown to characterizePerelman’s analysis of practical reason from within this standpoint. How-ever, before examining these anomalies in Perelman’s analysis and, moreimportant, the implications of these anomalies for this analysis, I examinetwo different conceptions of practical reason at work in Perelman’s ac-count.

5. Perelman’s Scientific and Nonscientific Conceptions of

Practical Reason

InThe Idea of Justice and the Problem of Argument, Perelman pursueswith vigor an analogy between explanation in scientific inquiry and justifi-cation in normative inquiry:

JUSTIFYING PRACTICAL REASON

We have seen the analogy that exists between the explanation of a phenom-enon and the justification of an act, how the just act and the phenomenonexplained both coincide with the conclusion of a syllogism. We shall not be atall surprised to find the same analogy holds between the fact of explaining atheoretical law and the justification of a normative rule. . . . These consider-ations show up once again the relativity both of explanation and of justifica-tion: every explanation is relative to certain more general laws, and everyjustification is relative to more abstract principles. But, in view of their arbi-trary character, we may also wish to explain those laws, and we may also beunder the duty of justifying those principles. Explanation and justificationwill then have recourse to still more general laws, to still more abstract prin-ciples. In the theoretical, as in the practical, field we shall end up by con-structing rational systems. The theoretical system of science will be matchedby a normative system of justice. (1963, 50–51)

Of course, in claiming that an analogy exists between explanation inscience and justification in normative inquiry, Perelman was not makingthe stronger claim that our view of normative justification is effectivelycontrolled by our view of scientific explanation. This stronger claim is es-sential to my contention that Perelman was pursuing a scientific concep-tion of practical reason. This claim can be shown, I believe, to be implicitin Perelman’s account of practical reason. One sign of its implicitness isPerelman’s denial of the Kantian claim that “practical laws” are formal innature:

For Kant, a pure, practical law, established a priori, can only be formal—thatis, its form alone makes it appropriate for universal legislation. My views gobeyond that, for I do not believe that a philosopher should limit himself to theformulation of a purely formal law comparable to the Rule of Justice. (1980,70)

Frege, it will be recalled, pursues a similar denial of the formal char-acter of the laws of logic from within his scientific (unKantian) conceptionof logic. Motivating this denial, it was argued above, is a certain view onthe part of Frege of what is to constitute a logical law. Frege claims, forexample, that such a law possesses a content, that is, is subject to a demandfor judgment. This view, it was further argued, is controlled ultimately byFrege’s desire to pursue an account of logic using the methodology of sci-ence. Now in the same way, Perelman claimed to have established an anal-ogy between the methodologies of science and normative inquiry. However,upon closer examination, what appears to be an analogy between the meth-odologies of these disciplines is more properly the control of the method-

LOUISE CUMMINGS

ology of normative inquiry by the methodology of scientific inquiry. Evi-dence for this control can be found, I am claiming, in Perelman’s denial ofKant’s claim that practical laws are formal in nature. These practical laws,according to Perelman, possess a content and are subject to challenge andcriticism for this reason.12 In attributing content to these practical laws,Perelman is effectively denying the priority of these laws. Where for Kantboth logical and practical laws have a priority over other logical and prac-tical statements that is guaranteed by their form, these same laws occupy aquite different status in the accounts of Frege and Perelman, respectively.For it is the case that these laws are not prior to other logical and practicalstatements, but are themselves subject to judgment (Frege) and challengeand criticism (Perelman) on the basis of these statements. Motivating bothFrege’s and Perelman’s denial of the priority of the laws of logic and ofpractical reason, respectively, is a desire on their part to assimilate themethodologies of logic and of practical reason to the methodology of sci-ence (while Frege openly expresses this desire—see note 11—its presenceis implicit in Perelman’s account). Of the scientific methodological con-ceptions of logic and practical reason to result from this assimilation, logi-cal and practical laws acquire the status of empirical generalizations. Theproblems created by a scientific conception of the laws of logic were ex-amined above in relation to the case of the psychologistic logician. Beforediscussing how essentially similar problems are created by a scientific con-ception of practical reason, I turn to an examination of Perelman’s nonsci-entific conception of this notion.

Perelman’s nonscientific conception of practical reason was describedabove in brief. In that context, it was represented by the following state-ments of Perelman:

But assertions that represent the systematic formulation of an ideal cannot bejudged the way we judge factual judgements. Their role is not to conform toexperience, but to furnish criteria for evaluating and judging experience and,if necessary, for disqualifying certain aspects of it. . . . Philosophers who refuseto recognise this primacy of practical reason have often exposed themselvesto the criticism of the positivist by presenting ontologies and theories of be-ing as if they were on the same level as scientific theories of reality. (1980,70–71)

Practical reason, on this alternative conception, is prior to—has pri-macy over—the theoretical activity of science. It is this conceptual prior-ity of practical reason over science that precludes any judgment of the ideals

JUSTIFYING PRACTICAL REASON

of practical reason on the basis of scientific criteria. In fact, in this view,scientific criteria are not presupposed by these ideals, as would have to bethe case if the ideals of practical reason are to be judged in accordancewith scientific criteria. Rather, scientific criteria are subsumed by theseideals. This nonscientific conception does not strive to assimilate the meth-odology of normative inquiry to that of scientific inquiry. Indeed, the dis-tinctness of these respective methodologies is emphasized. Now, the tensioncreated by the combination of this nonscientific conception of practicalreason and the scientific conception of this notion is clear: practical reasonis both conceptually prior to science (nonscientific conception) and on thesame conceptual level as science (scientific conception).13 In the same waythat a dissolution of this tension in Frege’s view of logic was achievedthrough a demonstration of the unintelligibility of Frege’s scientific(unKantian) conception of logic (which was represented by the position ofpsychologism), a similar dissolution of the tension in Perelman’s accountof practical reason can be achieved through a demonstration of the unintel-ligibility of Perelman’s scientific conception of practical reason. It is tothis latter demonstration that I now turn.

6. Exposing the Unintelligibility of Perelman’s Scientific

Conception of Practical Reason

Frege’s psychologistic logician, it was argued above, proposes a scientificconception of the laws of logic. These laws, the psychologistic logiciancontends, are empirical generalizations that describe psychological factsabout our minds, for example, facts about the inferences that our mindsaccept. Clearly, the emphasis of this scientific conception is on the reason-ing that is actually employed by thinkers. In the same way, I want to arguethat Perelman was committed to a scientific conception of practical reasonon account of his pursuit of an “actual logic of value judgments” (1979, 9).The emphasis of such a logic is on the actual value reasoning engaged inby a range of interested parties. Some of those parties were described ear-lier:

By analyzing political discourse, the reasons given by judges, the reasoningof moralists, the daily discussions carried on in deliberating about making achoice or reaching a decision or nominating a person, we might be able to

LOUISE CUMMINGS

trace the actual logic of value judgments which seems continually to eludethe grasp of specialists in the theory of knowledge .(1979, 9)

However, a scientific parallel with Frege did not end with Perelman’spursuit of an account of the actual value reasoning of subjects. For in thesame way that Frege’s scientific/psychologistic logician believes that hecan judge the laws of logic, in Perelman’s scientific conception of practi-cal reason it seems that what constitutes the ideals of practical reason cancome to be debated. Thus we find Perelman’s philosopher proposing to theuniversal audience for acceptance as ideals of practical reason, objectiveprinciples of action:14

The essential function of the philosopher is, perhaps, to formulate such prac-tical principles, while the scholar exercises a similar function in the realms ofscience or theoretical reason. The specific role of philosophy is, in effect, topropose to humanity objective principles of action that will be valid for thewill of all reasonable men. This objectivity will not consist either in confor-mity to some exterior object or in submission to the commands of any par-ticular authority. It envisages an ideal of universality and constitutes an attemptto formulate norms and values such as could be proposed to every reasonablebeing. (1980, 70)

As this quotation indicates, the objectivity of principles of action isderivative upon the notion of validity for “the will of all reasonable men,”that is, upon the notion of acceptance by a universal audience. However,Perelman ends by describing in essentially subjective terms the very no-tion of acceptance by a universal audience upon which the objectivity ofthese principles is based. So it is, for example, that each member or disci-pline within the universal audience, as this is envisaged by the philoso-pher, applies its own standards and criteria to an assessment of thephilosopher’s argumentation: “Relevance and irrelevance [of arguments]are to be examined according to the rules and criteria recognized by thevarious disciplines and their particular methodologies” (1980, 73).

The member-relative and discipline-relative nature of relevance cri-teria and, ultimately, of the notion of acceptance that is based on thesecriteria, compels the philosopher to construct his argumentation using onlypremises and inferences that he believes will be acceptable to each of themembers and disciplines of the universal audience: “It is this audience,with its convictions and its aspirations, that the philosopher wants to con-

JUSTIFYING PRACTICAL REASON

vince, starting from postulates and using arguments which he thinks willbe acceptable to every one of its members” (1980, 72).

The necessary relativity of the philosopher’s argumentation—neces-sary in the sense that if the philosopher overlooks the criteria subscribed toby particular disciplines or members of the universal audience, his argu-mentation will not be acceptable to that audience—is compounded by therelativity of philosophical argument in general. This latter form of relativ-ity is the relativity of argumentative criteria to the specific interests of dif-ferent branches of philosophical inquiry:

Unfortunately, there exists no methodology common to philosophy, whichwould allow one to decide the value of a philosophical argument. Philoso-phers usually borrow their postulates and their techniques of reasoning eitherfrom the history of philosophy by situating themselves in the continuation ofa system, or else from one of the several disciplines from which they drawinspiration. Until he has constructed his own philosophy, a philosopher willpossess no satisfying criteria for judging a philosophical argument in coher-ent fashion. (1980, 73)

In short, Perelman aimed to construct objective principles of actionwhen his only resources for doing so were the various subjective notionsthat are described above in relation to the argumentation between the phi-losopher and the universal audience. A similar conflation of the objectivewith the subjective occurs in the case of the psychologistic logician. Thepsychologistic logician, it will be recalled, views his own theoretical ac-tivity as that of expressing what kind of a thing the logical laws are: thatthey are always laws that govern the psychological process of reasoning.In this case, the objectivityof the psychologistic logician’s claim derivesfrom its universality, that is, its applicability to all thinking subjects (in thesame way, the objectivity of principles of action derives from the univer-sality of these principles, that is, the acceptance of these principles by allthe members of the universal audience). Yet what initially appears to be anobjective claim about the laws of logic is in fact a subjective claim aboutthose laws: we are precluded by the very content of the objective claimfrom stepping outside of our own psychological process of reasoning andassuming the broader vantage point from which the objective claim is made.In an attempt to overcome the subjective constraints of his account, thepsychologistic logician, it was argued, assumed a metaphysical standpoint.However, in assuming this standpoint, it was further argued that the

LOUISE CUMMINGS

psychologistic logician was not making objective claims; rather, he wasmaking no (intelligible) claims at all.

Now in the same way, the aim of Perelman’s philosopher is to pro-pose to the universal audience objective principles of action, principles ofaction that will form the ideals of practical reason. Yet the only conceptualresources upon which he can construct these principles are what we haveseen to be the essentially subjective notions of argumentation with a uni-versal audience. If argumentation with a universal audience cannot givePerelman’s philosopher access to the type of objective principles that heseeks, then this philosopher concludes that his only recourse is to set asidethe subjective constraints of his position by assuming a metaphysical stand-point. From this standpoint, Perelman’s philosopher takes himself to bearticulating principles of action that are objective, principles of action thatcan undergo assent by the universal audience. In reality, however, theaconceptual nature of such a standpoint precludes any attempt on the partof the philosopher to even state of what such a principle of action wouldconsist. In the absence of prior concepts of rationality, and I am thinkinghere of prior concepts of practical rationality, a principle of action is bothimpossible and unintelligible. In the same way that the laws of logic can-not form the subject of a Fregean demand for judgment—the laws of logicare presupposed by judgment; they cannot form the subject of judgment—so Perelman’s philosopher cannot engage in argumentation with a univer-sal audience about the ideals of practical reason; these ideals arepresupposed by any such argumentation.

7. Explaining Further the Unintelligibility of Perelman’s

and Frege’s Approaches

Perelman and Frege, I contend, are engaged in a scientific reductionistproject when they pursue, respectively, scientific conceptions of practicalreason and logic. However, reductionism in the context of Perelman’s andFrege’s accounts has an effect beyond that of causing the ideals of practi-cal reason and the laws of logic to be treated as if they were empiricalgeneralizations. Perelman and Frege, it will be recalled, waver betweenmaking subjective and objective claims about practical reason and logic.Putnam attributes responsibility in part for this “recoil” effect in philoso-phy—the tendency for perpetual movement between the (equally unsatis-

JUSTIFYING PRACTICAL REASON

factory) sides in a philosophical dispute—to just such a reductionist ap-proach in philosophy:

At the very beginning of these lectures, I spoke of the need to get a deeperunderstanding of the causes of our tendency to “recoil” from one horrendousposition to another in philosophy. In this concluding lecture, I have focusedon what seem to me to be the two principal causes of this tendency. The firstof those causes is a certain kind of reductionism, the kind of reductionismthat makes it impossible to see that when concepts are interlinked, as percep-tion,understanding,representation,verification, and truth are interlinked, thephilosophical task must be to explore the circle rather than to reduce all thepoints on the circle to just one point. (1994, 516; emphases in original)

If a reduction of the ideals of practical reason (of the laws of logic)to empirical generalizations about the values that are acceptable to a num-ber of normative voices (to empirical generalizations about the inferencesthat are acceptable to thinkers) cannot provide us with an objective justifi-cation of practical reason (cannot provide us with an objective account ofwhat kind of a thing a law of logic is), then Perelman and Frege concludethat our only option is to recoil to a metaphysical standpoint. This apparentforced choice between philosophical positions—in the cases of Perelmanand Frege, between reductionism and metaphysical theorizing—constitutes,according to Putnam, a second cause of the “recoil” effect in philosophy:

The second of these causes is the prevalence of the sort of assumption justmentioned—the all too seductive assumption that we know what the philo-sophical options are, and that they amount in each case to a forced choicebetween a funny metaphysical something standing behind our talk (whether itbe talk of “truth” or “reference” or “necessity” or “understanding”) and “toughminded” reductionism (verificationism, or deflationism, or antirealism, orwhatever). (1994, 516–17)

Perelman’s and Frege’s accounts, I have just claimed, are examplesof a recoil effect in philosophy, a recoil from reductionism to metaphysicaltheorizing. Perelman’s account, I now want to argue, is involved in at leastone other type of recoil, a recoil from formal to informal logic. It is notdifficult to find comments by Perelman that suggest that he considers for-mal logic as having failed to achieve one of its theoretical aims through itsneglect of domains of reasoning that are not purely formal in nature. Forexample, in the following remarks by Perelman, formal logic is effectively

LOUISE CUMMINGS

charged with neglecting to study domains of reasoning that were activelystudied by Aristotle, the founder of formal logic:

Formal logic essentially studies proof through calculation, i.e., formally cor-rect demonstrative reasoning. But, the way we reason in a discussion, or in anintimate deliberation, when we give reasons pro or contra, when we criticizeor justify a certain thesis, when we present an argument, e.g., in drawing up apreamble for a legal draft or the justification of a judgment, all the techniquesutilized in these situations have escaped the modern logician’s attention to theextent that he has limited himself to the analysis of purely formal reasoning.It is doubtless that in all these situations we reason, and the nature of thesereasonings did not escape Aristotle, considered by everyone to be the fatherof formal logic. (1979, 56)

Perelman proposes his new rhetoric as an analysis of the reasoningof the many domains not examined by formal logic. It is my claim that inrecoiling from formal logic to informal logic (the new rhetoric is a study ofinformal reasoning), Perelman is simply substituting one metaphysicalposition with another. The formal logician subscribes to a conception ofargument in which the formal logical structure of an argument is presumedto exist independently of any prior concepts of argumentative rationality,15concepts that pertain to the content and the context of an argument, forexample. This view of the formal logical structure of argument is indica-tive of a commitment on the part of the formal logician to undertake thecircumscription of argumentative rationality. The formal logical structureof argument, in this view, represents an “upper limit” on argumentativerationality, a limit beyond which argumentative rationality does not exist.

The privileged status afforded to formal logical structure accordingto this conception of argument (formal logic is, in effect, identified withargumentative rationality) and the concomitant denial of any claim to ra-tionality for those forms of argument that are not formal in nature—if for-mal logical structure represents the full extent of argumentative rationality,then the informal arguments described above by Perelman do not qualifyas being rational—are both flatly rejected by Perelman. For Perelman, ourconception of argumentative rationality needs to be extended beyond for-mal logical arguments to include arguments from a range of other con-texts, contexts that are legal, moral, political, etc. in nature. In an attemptto attain this extended conception of argumentative rationality, Perelmanundertakes an account of actual value reasoning. Within this account, for-mal logic is no longer deemed to be constitutive of the notion of argumen-

JUSTIFYING PRACTICAL REASON

tative rationality itself, but is instead one form of argumentative rational-ity among many. Now, the limitation of a formal logical approach to thestudy of argumentative rationality, as I see it, is that the privileging offormal logical standards in this approach amounts to a metaphysical claimabout the identification of formal logic with argumentative rationality: ifformal logical standards of argumentative correctness, etc., are allowed toexhaust the notion of argumentative rationality, then formal logic is effec-tively identified with argumentative rationality. However, these formal logi-cal standards are themselves only possible and intelligible in the presenceof prior concepts of argumentative rationality.16 Yet there can be no suchprior concepts in the case where formal logic is identified with argumenta-tive rationality.

A similar metaphysical claim, I argue, can be shown to be implicit inPerelman’s analysis of informal reasoning, particularly in his attempt toconstruct an “actual logic of value judgments.” Practical reason is effec-tively identified in Perelman’s account with the actual value reasoning ofsubjects. Moreover, Perelman is attempting to overcome the limitations ofa conception of argumentative rationality that is based exclusively on for-mal logic by extending that conception to include various forms of practi-cal deliberation (practical reason). However, by identifying practical reasonwith the actual value reasoning of a range of subjects (moralists, judges,etc.), and by identifying argumentative rationality with practical reason,Perelman is effectively taking an extended conception of argumentativerationality to be coextensive with value reasoning. Yet Perelman’s extendedconception of rationality is itself not possible or intelligible in the absenceof prior concepts of argumentative rationality, concepts that are not in-cluded as part of Perelman’s extended conception of this notion. Throughfailing to understand the real limitation of a formal logical approach to thestudy of argumentative rationality—that the historical and, consequently,conceptual privileging of formal logical standards of argumentative ratio-nality over other standards of argumentative rationality amounts to themetaphysical claim that formal logic is identified with argumentative ra-tionality—Perelman was destined to employ a similar metaphysical claimin the context of his study of informal reasoning, namely, that argumenta-tive rationality is identified with value reasoning.

School of History, Philosophy and Politics

University of Ulster

74Notes

LOUISE CUMMINGS

1.This article was written while the author was in receipt of the Eila Campbell MemorialAward from the British Federation of Women Graduates. The author acknowledges withgratitude the financial assistance of the federation. The comments of a referee of this journalon an earlier version of this article are gratefully acknowledged.

2.“Indeed, while any value whatever can serve as foundation for a system of justice, thatvalue in itself is not just” (1963, 56; emphasis added). Perelman’s description of the value atthe foundation of justice as being “not just” is an inaccurate characterization of the failure ofjustification in this case. This description presupposes the rational assessment of this value,an assessment that concludes with the determination that this foundational value is not just.However, it was Perelman’s view more generally and, of course, the view of the logicalempiricists that no such rational assessment can proceed within a logical empiricist frame-work, for the reason that the conception of rationality employed by this framework does notextend to the domain of values.

3.“Since . . . the arbitrary element in any normative system serves to sanction naturalinequalities, which are also not susceptible of justification, it follows that, for this doublereason, there is no necessary and perfect justice” (1963, 60).

4.This perspective contrasts with the absolutist view of justification: “The classical ideaof justification is based on an absolutist vision, since it looks for an absolute, irrefutable,and universally valid foundation” (1980, 61).

5.“A mode of behavior that has been adopted without protest creates a precedent, and noone will object to actions that conform to precedents” (Perelman, 1980, 27).

6.“It is to these men, or at least to the universal audience as he imagines it, that thephilosopher speaks. It is this audience, with its convictions and its aspirations that the phi-losopher wants to convince, starting from postulates and using arguments which he thinkswill be acceptable to every one of its members” (1980, 72; emphasis added).

7.“How must I think in order to reach the goal, truth? We expect logic to give us theanswer to this question . . . ” (Posthumous Writings, 128).

8.Wittgenstein, it is standardly argued, makes use of a say/show distinction to claim thatwhile we cannot say with our words what lies beyond the limit of thought, we can use ourwords to show what lies beyond this limit.

9.The possibility at issue here, it will be observed, is quite distinct from the possibilitythat was discussed in the main text in relation to Frege’s Kantian conception of logic: theformer type of possibility (psychological possibility) relates to what is possible given cer-tain psychological facts about us; the latter type of possibility (conceptual possibility) re-lates to Frege’s Kantian claim that it is the laws of logic that make all thinking possible.10.I avoid on purpose the use of the term irrational, for the reason that any description ofthe demand for judgment in the case of the laws of logic as being irrational (not rational) isa description that presupposes a capacity for rational thought on the part of the describer, acapacity that, I am arguing, is absent when that describer is theorizing from a metaphysicalstandpoint.

11.Frege intends his Begriffsscrift (concept-notation) to form a perspicuous representa-tion of his idea that logic is a science. This is clearly evident in the following extract fromtheCollected Papers, where Frege is drawing an analogy between the logician’s definitionof a concept and the scientist’s, specifically the chemist’s, decomposition of a substance:“Kerry contests what he calls my definition of “concept.” I would remark, in the first place,that my explanation is not meant as a proper definition. One cannot require that everythingshall be defined, any more than one can require that a chemist shall decompose every sub-stance. What is simple cannot be decomposed, and what is logically simple cannot have aproper definition. Now something logically simple is no more given to us at the outset thanmost of the chemical elements are; it is reached only by means of scientific work. If some-thing has been discovered that is simple, or at least must count as simple for the time being,we shall have to coin a term for it, since language will not originally contain an expressionthat exactly answers. On the introduction of a name for what is logically simple, a definitionis not possible; there is nothing for it but to lead the reader or hearer, by means of hints, tounderstand the word as it is intended” (182–83; emphasis added). In the latter part of thisextract, Frege is describing a type of inquiry that occurs when the definition that is typical of

JUSTIFYING PRACTICAL REASON

hisBegriffsschrift is not possible. Later in the Collected Papers, Frege describes this inquiryas elucidation and contrasts it with the scientific system of logic that is his Begriffsschrift:“Since definitions are not possible for primitive elements, something else must enter in. Icall it explication [elucidation]. It is this, therefore, that serves the purpose of mutual under-standing among investigators, as well as of the communication of the science to others. Wemay relegate it to a propadeutic. It has no place in the system of a science; . . . ” (300–01;emphasis added).

12.“We know that philosophers who invoke universal values such as truth, goodness, jus-tice, and reality as opposed to appearance, are rarely in agreement as to the criteria and thecontent of those values” (Perelman 1980, 70).

13.The tension in Frege’s account of logic was presented as a tension between the unin-telligibility of the demand for judgment in the case of the laws of logic in Frege’s Kantianconception of logic and the intelligibility, indeed necessity, of this same demand in Frege’sscientific (unKantian) conception of logic. Appearances notwithstanding, this tension is ef-fectively the same tension that is identified in the main text as being inherent in Perelman’saccount of practical reason. The reason why the demand for judgment in the case of the lawsof logic is unintelligible in Frege’s Kantian conception of logic relates to the primacy ofthese laws on this conception: these laws, owing to their primacy/priority, make judgmentpossible; we cannot make sense of the attempt to judge these laws themselves, as such judg-ment is devoid of a logical framework in which to proceed (the notion of judgment can itselfonly exist within such a framework).

14.This quotation demonstrates further the impact of scientific methodology on Perelman’sview of practical reason: “The essential function of the philosopher is, perhaps, to formulatesuch practical principles, while the scholar exercises a similar function in the realms ofscience or theoretical reason.”

15.Of course, I am making here a conceptual claim, a claim about the presumed concep-tual independence of formal logic of prior nonformal concepts of argumentative rationality.Yet this conceptual claim has its source in the various historical developments that logic hasundergone. The historical privileging of formal logic over other areas of logical inquiry(remarked upon by Perelman in his quotation in the main text) has had the effect, I believe,of identifying formal logic with argumentative rationality and, in so doing, of making itseem that formal logic does not presuppose other forms of argumentative rationality. ThatPerelman considered that the various historical developments of logic are responsible forthe current conceptual status of formal logic is evident from the following quotation: “Mod-ern logic has undergone some remarkable developments in the last hundred years. Thesehave contributed to the extraordinary use of formal logic which has become essentially theconcern of mathematicians. This has led to attempts to identify logic with formal logic. Theclaim has even been made that all non-formal reasoning, to the extent that it cannot be for-malized, no longer belongs to logic. This conception leads to a genuine impoverishment oflogic as well as to a narrow conception of reason” (1979, vii).

16.This claim is simply another form of a claim that was used to invalidate the position ofFrege’s psychologistic logician. Disagreement about the laws of logic had only the appear-ance of possibility and intelligibility in the psychologistic logician’s account, for the reasonthat disagreement, as an expression of our rationality, presupposes a rational framework inthe form of logical laws, a rational framework that is absent when it is the logical laws thatform the subject of disagreement. In the same way, I am claiming in the main text that for-mal logic does not represent an “upper limit” on the notion of argumentative rationality, butinstead presupposes argumentative rationality, rationality that confers possibility and intel-ligibility on the criteria of formal logic.

Works Cited

Conant, James. 1991. “The Search for Logically Alien Thought: Descartes, Kant, Frege, and

the Tractatus.” Philosophical Topics 20.1: 115–80.

Frege, Gottlob. 1967. The Basic Laws of Arithmetic. Trans. Montgomery Furth. Berkeley:

University of California Press.———. 1984. Collected Papers on Mathematics, Logic, and Philosophy. Ed. Brian

McGuinness. London: Blackwell.

LOUISE CUMMINGS

———. 1979. Posthumous Writings. Ed. H. Hermes, F. Kambartel, and F. Kaulbach. Chi-cago: University of Chicago Press.

Kant, Immanuel. 1909. Kant’s Critique of Practical Reason and Other Works on the Theory

of Ethics. Trans. Thomas K. Abbott. 6th ed. London: Longmans, Green, and Co.

Kluback, William, and Mortimer Becker. 1979. “The Significance of Chaïm Perelman’s

Philosophy of Rhetoric.” Revue Internationale de Philosophie 33: 33–46.

Perelman, Chaïm. 1963. The Idea of Justice and the Problem of Argument. London: Routledge

and Kegan Paul.———. 1980. Justice, Law, and Argument: Essays on Moral and Legal Reasoning. Dordrecht,

Holland: D. Reidel.———. 1979. The New Rhetoric and the Humanities: Essays on Rhetoric and Its Applica-tions. Dordrecht, Holland: D. Reidel.

Putnam, Hilary. 1994. “The Dewey Lectures 1994: Sense, Nonsense, and the Senses: An

Inquiry into the Powers of the Human Mind.” The Journal of Philosophy 91.9: 445–517.

因篇幅问题不能全部显示，请点此查看更多更全内容

查看全文