Maddy on Carnap
Maddy continues her rejection of two-level philosophical accounts (begun in the last chapter on Kant) that posit an empirical level of phenomena and some meta-level account of the framework constraints that structure the various domains of enquiry at the former level.
Maddy explains that Carnap is concerned with a Kantian-style project: that of eradicating metaphysical speculation from our discuorse, or as Carnap puts it, the task of eradicating 'pseudo-sentences' from our language. These sentences have no meaning because they transgress a constitutive feature of any meaningful language: that it be based on a set of governing framework rules of some sort or another that must be adopted in order for the sentences of the language to have meaning (pp67-68).
If sentences can only have meaning relative to such a linguistic framework then we meet with a new way of responding to the sceptic: the demand for external justification of our claims about the world is just wrongheaded, for any such claim only makes sense internal to a linguistic framework. Philosophers who ask ontological questions that are supposed to swing free of any such framework and are intended to then ground them are involved in a similarly wrongheaded enterprise (PP 68-69).
Carnap's linguistic framework approach still leaves us with a legitimate question, just not a philosohpical one. What languages should I adopt, that is, what framework rules should I admit into my repertoire and what ones should I change of those I already employ? This is a pragmatic question for Carnap, for there is no fact of the matter about what framework rules to adopt. Carnap codifies this pragmatism about frameworks in the 'Principle of Tolerance' - one is free to adopt whatever rules one wishes, all that is required for meaningful debate with others is that the framework be made clear so that the sentences of the language be understood in the right light (ppp70-71).
Carnap's position seems to give us a successor understanding of Kant's synthetic a priori: some truths are a priori internal to the framework (require no empirical confirmation within the framework to be accepted as true), but are pragmatic/conventional choices external to the framework. SP finds this odd: questions about what framework rules to adopt are not merely conventional choices outside of any linguistic framework, they are of a piece with the rest of her empirical enquiry.
But Maddy recognises that this (albeit very common) Kantian reading of Carnap's two-level approach is not really fair. He is not trying to supply a successor conception of a priority that would explain what makes sentences in specific scientific languages true without empirical input or verification (that they are 'internal' constitutive rules, but externally viewed mere conventions). Practising scientists do not employ linguistic frameworks in this way, rather, they are a construct of philosophers who impose them on scientific language for the purposes of 'rational reconstruction' (p78) - a philosophical procedure for rationalising scientific practices. The purpose of this procedure is to defuse various philosophical pseudo-debates about what grounds scientific knowledge (realism, idealism, nominalism, etc.). These egregious disagreements can be reduced to mere pragmatic choices via rational reconstruction. The touchstone of such reconstruction is the principle of tolerance which assumes that there is no fact of the matter at stake in various philosophical debates (e.g. realism/nominalism about mathematical objects).
Maddy rightly questions what exactly secures the methodology of rational reconstruction (what makes the principle of tolerance itself true)? Moreover, the pragmatic jusification of the principle seems to lead to the possibility of incoherence - it leaves it open for us to pragmatically adopt a different meta-principle that brings with it substantial ontological commitments. But that looks like letting conventional choices dictate the existence of substantial truths outside of rational reconstruction.
I take it that Maddy's criticisms of Carnap on this score are well taken. But these aside SP just doesn't see the point in the detour of rational reconstruction - what point does this 'peculiarly philosophical project' serve?
I am left wondering what is philosophical about second philosophy? At bottom, any attempt to stray from naturalistic methods of inquiry seems to be met with a 'what's the point?' that sounds like a 'what's the point of philosophy?' To talk very crudely, philosophy is in part an attempt to try to understand how the possibility of very general notions (such as 'thought', 'knowledge', 'meaning') simply assumed at the empirical level can be justified at some other level. Carnap tries to justify a rejection of various such justifications that he thinks cause intractable problems by showing that there's no fact of the matter upon which such debates turn. Of course, this is itself a philosophical position of the type Carnap wants to reject, so there is an incoherence here. But if we don't provide any arguments for why empirical knowledge might or might not be possible, or why this isn't a good question to ask, won't SP just look dogmatic, forced to share the domain of discourse with all sorts of other dogmatic positions that reject its cherished methods of inquiry?
Maddy explains that Carnap is concerned with a Kantian-style project: that of eradicating metaphysical speculation from our discuorse, or as Carnap puts it, the task of eradicating 'pseudo-sentences' from our language. These sentences have no meaning because they transgress a constitutive feature of any meaningful language: that it be based on a set of governing framework rules of some sort or another that must be adopted in order for the sentences of the language to have meaning (pp67-68).
If sentences can only have meaning relative to such a linguistic framework then we meet with a new way of responding to the sceptic: the demand for external justification of our claims about the world is just wrongheaded, for any such claim only makes sense internal to a linguistic framework. Philosophers who ask ontological questions that are supposed to swing free of any such framework and are intended to then ground them are involved in a similarly wrongheaded enterprise (PP 68-69).
Carnap's linguistic framework approach still leaves us with a legitimate question, just not a philosohpical one. What languages should I adopt, that is, what framework rules should I admit into my repertoire and what ones should I change of those I already employ? This is a pragmatic question for Carnap, for there is no fact of the matter about what framework rules to adopt. Carnap codifies this pragmatism about frameworks in the 'Principle of Tolerance' - one is free to adopt whatever rules one wishes, all that is required for meaningful debate with others is that the framework be made clear so that the sentences of the language be understood in the right light (ppp70-71).
Carnap's position seems to give us a successor understanding of Kant's synthetic a priori: some truths are a priori internal to the framework (require no empirical confirmation within the framework to be accepted as true), but are pragmatic/conventional choices external to the framework. SP finds this odd: questions about what framework rules to adopt are not merely conventional choices outside of any linguistic framework, they are of a piece with the rest of her empirical enquiry.
But Maddy recognises that this (albeit very common) Kantian reading of Carnap's two-level approach is not really fair. He is not trying to supply a successor conception of a priority that would explain what makes sentences in specific scientific languages true without empirical input or verification (that they are 'internal' constitutive rules, but externally viewed mere conventions). Practising scientists do not employ linguistic frameworks in this way, rather, they are a construct of philosophers who impose them on scientific language for the purposes of 'rational reconstruction' (p78) - a philosophical procedure for rationalising scientific practices. The purpose of this procedure is to defuse various philosophical pseudo-debates about what grounds scientific knowledge (realism, idealism, nominalism, etc.). These egregious disagreements can be reduced to mere pragmatic choices via rational reconstruction. The touchstone of such reconstruction is the principle of tolerance which assumes that there is no fact of the matter at stake in various philosophical debates (e.g. realism/nominalism about mathematical objects).
Maddy rightly questions what exactly secures the methodology of rational reconstruction (what makes the principle of tolerance itself true)? Moreover, the pragmatic jusification of the principle seems to lead to the possibility of incoherence - it leaves it open for us to pragmatically adopt a different meta-principle that brings with it substantial ontological commitments. But that looks like letting conventional choices dictate the existence of substantial truths outside of rational reconstruction.
I take it that Maddy's criticisms of Carnap on this score are well taken. But these aside SP just doesn't see the point in the detour of rational reconstruction - what point does this 'peculiarly philosophical project' serve?
I am left wondering what is philosophical about second philosophy? At bottom, any attempt to stray from naturalistic methods of inquiry seems to be met with a 'what's the point?' that sounds like a 'what's the point of philosophy?' To talk very crudely, philosophy is in part an attempt to try to understand how the possibility of very general notions (such as 'thought', 'knowledge', 'meaning') simply assumed at the empirical level can be justified at some other level. Carnap tries to justify a rejection of various such justifications that he thinks cause intractable problems by showing that there's no fact of the matter upon which such debates turn. Of course, this is itself a philosophical position of the type Carnap wants to reject, so there is an incoherence here. But if we don't provide any arguments for why empirical knowledge might or might not be possible, or why this isn't a good question to ask, won't SP just look dogmatic, forced to share the domain of discourse with all sorts of other dogmatic positions that reject its cherished methods of inquiry?
Labels: Logical Positivism, Naturalism
posted by matt at 7:35 pm
1 Comments:
In her defence, might not Maddy say that SP is open to philosophy and 'peculiarly philosophical' projects. But that, at a bare minimum, it needs to be explained to her, both in her own terms and in the project's own terms, (i) what the point of the project is, (ii) what the methods of the project are, (iii) what the results of the project are. Her criticism of Carnap seems to involve saying that none of these three are satisfied, or at least certainly not the first two.
Why do these things need to be explained to SP in her own terms (which I take to include both common sense and the terminologies of the various positive scientific disciplines)? Well, if not then it looks as though it is SP's opponent who is dogmatically begging the question - he admits that he can't justify or explain the coherence of his own project in terms that don't aready assume the coherence of the project. Given this, it seems to me to be all important that SP take common sense to be something that her opponents depart from. In the case of Carnap, with his talk of pragmatic decisions to adopt the thing language, this looks rather plausible.
Post a Comment
<< Home