From the ALWS archives: A selection of papers from the International Wittgenstein Symposia in Kirchberg am Wechsel

1. Introduction

In the introduction to his anthology about the linguistic turn Rorty evaluates linguistic philosophy (LP). He concludes that LP's attempts, e.g. of logical positivism (cf. Rorty 1967 6), to turn philosophy into a strict science must fail. Submitting, however, that “no one is able to think of any formulations” of traditional philosophical problems which are immune to its sort of criticism, LP has left philosophy with a “merely critical, essentially dialectical, function”. Reflecting upon the consequences of this “rather pessimistic” conclusion Rorty “envisages at least six possibilities for the future of philosophy, after the dissolution of the traditional problems” (Rorty 1967 33 f.). These possibilities are divided according to the “metaphilosophical struggles of the future [being centred] on the issue of [linguistic] reform versus description [“of the facts” (Rorty 1967 38)], i.e. of philosophy-as-proposal [about how to talk cf. (Rorty 1967 34); cf. Bergmann's view of philosophy as “linguistic recommendation" (Rorty 1967 8], versus philosophy-as-discovery” [of “specifically philosophical truths” (Rorty 1967 36)]. Heidegger, Wittgenstein and Waismann are assigned to the former, Husserl, Austin and Strawson to the latter camp (Rorty 1967 38, 35).

This paper focuses on the rationale for this division being positioned in the horizons of metaphilosophical struggles of the future as Rorty envisages. It is only if the earlier assessment of the current situation of LP, in view of its success in dissolving the traditional problems is correct that this positioning could be justified. Rorty's calling Strawson “the strongest rival” of Austinian style Oxford Analysis (Rorty 1967 37) adds to the relevance of this assessment.

Taking Strawson’s case, I will question this assessment. This impairs the credibility of the suggestion that philosophy should, for its continuation, secure its future role and avert a “'post-philosophical' culture”, the conceivability of which Rorty, incidentally, questions (Rorty 1967 34). This result allows for alternative divisions to the one Rorty provides, not necessarily pointing at a post-philosophical culture. However this does not affect the value of Rorty's division, it does open new battlefields for metaphilosophical struggles in the future.

2. Varieties of linguistic philosophy

Rorty singles out methodological nominalism as a distinctive metaphilosophical assumption of LP. This is the view that if philosophical questions about “concepts, subsistent universals or ‘natures’ … cannot be answered by empirical inquiry concerning the behavior or properties of particulars subsumed under [them] and can be answered in some way, [it is] by answering questions about the use of linguistic expressions, and in no other way" (Rorty 1967 11 orig. emph.). Both branches of LP, ordinary language and ideal language philosophy (OLP, ILP) share this nominalism (Bergmann 1964 177 in Rorty 1967 8). Observing that LP itself generally does not contain a metaphilosophical justification of this assumption, Rorty preliminarily examines the measure of its being presuppositionless and of its having criteria for success which can be rationally agreed upon (Rorty 1967 5). This examination leads Rorty to exonerate LP from the charge, advanced by its critics, of being committed to some “substantive philosophical theses”. This verdict applies to OLP and ILP alike (Rorty 1967 14). It is based, mainly, on his appreciation of the aforementioned assumption as “practical” (Rorty 1967 9 orig. emph.). According to Rorty, this assumption boils down to “a single plausible claim: that we should not ask questions unless we can offer criteria for satisfactory answers to those questions” (Rorty 1967 14).

Thus conceived, Rorty’s verdict seems fair enough, insofar as the assumption is restricted to the nature of the philosophical problems to be put on the agenda. It also captures the conventional division of labour as regards the direction of their (dis)solution, i.e. either “ ... by reforming language, or by understanding more about the language we presently use” (Rorty 1967 3), as professed by ILP an OLP respectively. It seems less appropriate, however, to pass this verdict on Strawson's programme of descriptive metaphysics.

Strawson “resembles the descriptive analyst in that he wishes to make clear the actual behavior of our concepts”, or to describe “the logical behavior of the linguistic expressions of natural languages” (Strawson 1967 318, 316). Strawson also seems committed to methodological nominalism, stating, "to observe our concepts in action is necessarily the only way of finding out what they can and cannot do" (Strawson 1967 319). Strawson's programme of descriptive metaphysics differs from descriptive analysis proper, however, "in scope and generality” (Strawson 1967 318). It is more generally oriented in its attempt to discover (Strawson 1967 320) and exhibit the general structure of our conceptual apparatus (Strawson 1967 318), thus trying to fill "the gap between contingent truths about linguistic behavior and necessary truths about language as such” (Rorty 1967 37). And by submitting that the "actual use of linguistic expressions remains his sole and essential point of contact with the reality which he wishes to understand, conceptual reality" (Strawson 1967 320 my emph.) Strawson indeed seems to extend the scope of linguistic analysis by making a, what one might call 'metaphysical ascent', i.e. a shift from a "description of linguistic usage” (Strawson 1967 313) to descriptive metaphysics (Strawson 1957 318).

Calling his endeavour “descriptive metaphysics” is not just a polemic move (cf. Rossi 1993 77) to mock the parochial muddling through of OLP, but it marks Strawson's departure from OLP as understood by Rorty (cf. Rorty 1967 37). OLP involved a reduction of philosophical problems to instances of linguistic usage. An analysis of those expressions, or Strawson’s meta-analysis (Rossi 1993 77) of their presuppositions might "increase our conceptual understanding” (Strawson 1967 318, 320). Language must not be a "central theme" of such philosophical inquiry. Its "meticulous examin[ation]” of the use of words could be merely regarded as its "central method" (Hacker 2005 14 orig. emph.). By these lights, basic assumptions like "ordinary language is alright” (Wittgenstein 1968 §434) (which is OLP's counterpart of the Misleading Form Thesis (Gamut 1991 214) in ILP), however they presuppose some metaphysical or epistemological preconceptions (Rorty 1967, 1) about the elements or aspects to pick out for analysis ((logical) syntax, (il)locutionary acts, semantics etc.) can be seen as heuristic guidelines about a suitable method of analysis. This inquiry, therefore, remains within the confines of methodological nominalism.

Strawson’s metaphysics, by contrast, involves a philosophical (proto-) theory about language. This theory is based on a unified theory of predication which holds claims about the fundamentality of basic universal-particular combinations in epistemology, logic, ontology (spatio-temporal particulars and universals) and grammar (subject and predicate), about the connection between those combinations, and their links with reality. Succinctly put: "grammar reflects logic, logic reflects reality and predication is the key relation that unifies each of the fields and also provides the links between them” (Moravcsik 1978 329 f.). This theory is the core of descriptive metaphysics (cf. "metaphysical grammar” (Ross 1974)) which "attempts to show the natural foundations of our logical, conceptual apparatus in the way things happen in the world, and in our own natures" (Strawson 1967 317). This phrasing suggests a dogmatic conceptualism that seems to take a naturalistic form (see Katz 1990 250 for this terminology). The theory, moreover, is not uncontroversial. Apart from the unclarity of the argument about the fundamentality of the aforementioned basic combinations, due to the mixture of epistemology and ontology, the argument from "exemplification as the basic mode of judgment” to the "fundamentality of spatiotemporal particulars”, for instance, seems a non sequitur (Moravscik 1978 336, 333 orig. emph.).

Strawson's view on linguistic-conceptual relations resembles "the basic premise of the philosophy of language” Katz advocates, namely "that there is a strong relation between the form and content of language and the form and content of conceptualization” (Katz 1966 4). This similarity adds to the interest of Strawson's controversy with empirical linguistics since it bears upon the demarcation between philosophy and empirical science. Strawson criticizes traditional grammar for its lack of a “general theory” that accounts for the "general principles determining the assignment of [lexical] items to grammatical categories” (Strawson 1971 137). To correct this, Strawson outlines a programme of "Research in Non-Empirical Linguistics" (Strawson 1971 148) that should give an account of the "general semantic types of expression [which] qualify for the basic subject- and predicate-roles in simplest sentences”. This programme should yield a "perspicuous grammar” which aims at "finding explanatory foundations for grammar” (Strawson 1971 140, 145 f.). The vindication of this programme's entitlement to deal with this kind of explanation along with its claim to be a priori, non-historical (Strawson 1967 317) and to not use statistical methods (Strawson 1967a 323) amounts at the least (pace Rorty 1967 26 fn. 48) to a philosophical claim about the proper analysis of language. Conversely, linguistics rivals with philosophy, stating that some persistent philosophical problems can be felicitously represented as questions about the nature of language; and solved by a linguistic theory about logical form as an alternative to a philosophical theory about logical form (Katz 1967 340, 346).

Rorty concedes that his introduction is defective for not "adequately exhibit[ting] the interplay between the adoption of a metaphilosophical outlook and the adoption of substantive philosophical theses” (Rorty 1967 38 f.). By extension, Strawson's exchange with philosophy of language and empirical linguistics should alert us to his adoption of their vocabularies and the changes in the metaphilosophical vocabulary of his descriptive metaphysics that this might cause (cf. 1967 39). Thus, for instance, his notion of 'logical behaviour of expressions' (cf. Rorty 1967 21) could be taken not in its traditional meaning as contrasting with 'grammatical form', nor as referring to a specific level of description like in Chomskyan grammar, but, rather linguistically, in the way it is taken by logical grammar, as the "representation of an expression that determines its meaning” (Gamut 1991 214 ff.). Strawson grants "system-construction ... an ancillary" and "limited” role in "increas[ing] our conceptual understanding" (Strawson 1967 319 f.). Accordingly, his perspicuous grammar could be taken as a version of Montague-grammar, in which "a natural language ... can be represented in its essentials as a formal language”. A "modified version”, that is, due to "the addition of ... pragmatic-functional factors [e.g. the need to have expressions for identifying, describing etc.]” (Moravcsik 1976 342). This addition secures the "natural foundations of our logical, conceptual apparatus” (cf. Strawson 1967 317). Corollarily, Strawson's quest for "fundamental relations ... between ... general concepts” (Strawson 1967a 321) could be viewed as a modified search for the compositionality of meaning by logical grammar (cf. Gamut 1991 215). Modified, for Strawson's principle of compositionality seems a metaphysical rephrasing of an empirical hypothesis, rather than a methodological principle (cf. Gamut 1991 219). It assumes the existence of "assertible non-relational ties between non-linguistic elements ... i.e. copulatives expanded into terms such as 'is a characteristic of'” (cf. Ross 1974 374).

Strawson's advancement of OLP could, along similar lines, be viewed as a development from a philosophical occupation with linguistic performance to a philosophico-linguistic occupation with linguistic competence (cf. for this distinction (Katz 1967 345)). Corollarily, his notion of conceptual apparatus, insofar as it covers the equipment for linguistic behaviour could be compared with linguistic competence and, incidentally, be regarded as LP's rephrasing of the Kantian transcendental apperception.

Rorty’s division of coalitions of philosophers fighting metaphilosophical struggles is suggested by the least common denominator of the philosophers assigned to each group (Rorty 1967 38). Strawson's case shows that it allows on the basis of Rorty's observations for other divisions, cutting across the dichotomy of philosophy-as-proposal and philosophy-as-discovery. They might, perhaps, be more relevant to our times. Thus Strawson could, for his conceptualism, join Husserl and Heidegger, who are separated from their combatants for their repudiation (Rorty 1967 34) of methodological nominalism. Another division would start from Rorty's separation of Waismann and Wittgenstein from Austin and Strawson for their "repudiat[ing] the notion that there are philosophical truths to be discovered and demonstrated by argument (Rorty 1967 36), but would separate Strawson, in turn, from Austin, given the former's interest in linguistics. Yet other divisions would be those according to the measure or aim by which analytical methods are deployed in the construction of an ideal language or in the analysis of ordinary language. Such a division must not match Bergmann's one between OLP and ILP. Thus, for instance, Strawson, for his acceptance of limited constructionalism in the exercise of "philosophical imagination" (Strawson 1967 317 ff.) could accompany Waismann, who favours the construction of ideal languages to create "new, interesting and fruitful ways of thinking about things in general” (Rorty 1967 34), and this constructionalism would for similar reasons even be alluring to the later Heidegger (Rorty 1967 35).

3. Concluding remarks

By calling philosophy-as-proposal the "direct heir” (Rorty 1967 36) of ILP, Rorty's division preserves, in a way, the self-proclaimed bifurcation in ILP and OLP. His evaluation, geared as it seems by this bifurcation, might perhaps be "a little thin and myopic" (Hacker 2005 10). The above considerations purport to show that the metatheoretical association of Strawson's programme of descriptive metaphysics with methodological nominalism underdetermines its character. This, however, should not lead us to overlook Rorty’s highly nuanced and imaginative modulations of the received view about LP. For, as Rorty’s case illustrates, a recognition of the dialectics of substantive philosophical theses and metatheoretical adaptations reveals alternative divisions and speculations about the future of philosophy, marking, perhaps, its "'progress' as a movement toward a contemporary consensus” (Rorty 1967 2).