File spoon-archives/bhaskar.archive/bhaskar_2002/bhaskar.0202, message 81


From: "Andrew Brown" <Andrew-AT-lubs.leeds.ac.uk>
Date: Tue, 12 Feb 2002 09:56:43 -0000
Subject: BHA: re: cr and social science


Hi Mervyn,

Sorry - I should have noted the edition of PON I was referring to.

Not really sure why you think I am committing the opposite error. I 
was questioning the 'critical' in RB's 'critical' naturalism. The term 
critical is used by Bhaskar to refer to the *distinguishing* features 
(the aspect of difference) of social science relative to natural 
science (the term 'naturalism' obviously referring to the aspect of 
unity), so I am questioning whether Bhaskar has established any 
relevant methodological differences, and taking the 'unity' as given. 
Bhaskar develops the key methodological distinction from the lack 
of the ability to do scientific experiment social science (however 
you want to qualify exactly what RB means by this). Yet this whole 
discussion of RB's seems to be undermined once one recognises 
that many natural sciences don't perform the sort of experiments 
(artificially induced closure) that Bhaskar has in mind. 

Note: this is a specific problem with Bhaskar's exposition of CR 
method in social science. In many ways the subsequent 
discussion of CR social scientific method, by other critical realists, 
has exposed and overcome the problem. Lawson, for example, I 
think would agree with everything I'm saying. And Lawson explicitly 
suggests a method for social science which he (Lawson) has 
borrowed from outside of social science (from plant breeding). The 
title of his relevant chapter is 'Economic science without 
experimentation' (ch.15, Lawson 1997). But both Doug and 
Marshall have also, in previous discussions on this list, simply 
referred to other realists than Bhaskar to support their views 
regarding critical realist method in social science. Collier is another 
case I have referred to already (but, for him, experiment via 
artificially induced closure is essential to science so he questions 
the very status of social 'science').

Many thanks,

Andy



On 11 Feb 2002, at 17:44, Mervyn Hartwig wrote:

> 
> Dear Andy,
> 
> Thanks for calling this difference between natural scientific and
> social scientific reasoning, according to PON, to my attention (pp.
> 64-5 are the crucial ones in my 1979 edition). I hadn't really
> registered it. What Bhaskar says there is important, too, I think, for
> grasping his understanding of the relation between philosophy and
> science (natural as well as social).
> 
> I still think, though, that you are committing the opposite error and
> emphasising the difference at the expense of commonality. Stressing
> only the difference overlooks that Bhaskar's overall intent is to
> establish the possibility of a non-positivist unity of scientific
> method, and that to say that one argument form is a species of another
> is to stress what they have in common as well as the differences -
> something that is underlined in both PON and DPF by use of the adverb
> 'merely' - 'merely a species of', 'merely types of'. 
> 
> Best,
> 
> Mervyn
> 
> 
> 
> Andrew Brown <Andrew-AT-lubs.leeds.ac.uk> writes
> >Hi Mervyn,
> >
> >Your cite from DPF illustrates preciscely my point. This isn't 
> >surprising because, as you know, RB is usually quite consistent on
> >matters such as these. Recall my point is that transcendental
> >arguements are a *species* of the *genus* which is retroductive
> >arguments. Your mistake is to take transcendental arguments as the
> >genus (you said that they are widespread in natural science, yet they
> >aren't - only retroductive arguments are) . Thereby you miss the
> >essential pont RB is making about social scientific method in PON.
> >This is all quite clear from the section of PON I pointed out in my
> >previous post. I simply do not have time to cite at length from PON,
> >but I assure it is all there.
> >
> >Best wishes.
> >
> >Andy
> >
> >On 11 Feb 2002, at 1:43, Mervyn Hartwig wrote:
> >
> >> Hi Andy,
> >> 
> >> '... transcendental arguments are merely types of the retroductive-
> >> explanatory argument form familiar to science'. - DPF 108 (et
> >> passim).
> >> 
> >> Mervyn
> >> 
> >> Andrew Brown <Andrew-AT-lubs.leeds.ac.uk> writes
> >> >Hi Mervyn,
> >> >
> >> >> >Why develop a 
> >> >> >method in social science, viz. 'transcendental deduction', 
> >> >> 
> >> >> I don't know where you got this from. Transcendental reasoning
> >> >> according to Bhaskar is central to the natural as well as social
> >> >> sciences.
> >> >
> >> >I got it from our friend RB! I think you just need to re-read the
> >> >relevant section of PON (pp.49-54) to remind yourself of this. 
> >> >
> >> >You are wrong re 'transcendental reasoning': this is employed by
> >> >both philosophy and social sciences, but not by natural sciences.
> >> >What I think you must have in mind is 'retroductive reasoning'.
> >> >This latter is the genus of which transcendental deduction
> >> >(argument) is the species. Once again this illustrates that you
> >> >are downplaying what is crucial to Bhaskar's argument on method -
> >> >in this case the specificity of social scientific method - and
> >> >what is apparently undermined by the 'little exaggeration' to
> >> >which you refer.
> >> >
> >> >I think I am basically restating Collier's position here, though
> >> >he stresses all that is left in critical naturalism, viz. it's
> >> >*philosophical* aspects, whilst recognising the problem I stress,
> >> >viz. that Bhaskar's key discussion of *method* in social science
> >> >has had its apparent foundation (lack of ability to do social
> >> >scientific experiment) removed.
> >> >
> >> >Best wishes,
> >> >
> >> >Andy
> >> >
> >> >
> >> >
> >>
> 
> 
>      --- from list bhaskar-AT-lists.village.virginia.edu ---




     --- from list bhaskar-AT-lists.village.virginia.edu ---

   

Driftline Main Page

 

Display software: ArchTracker © Malgosia Askanas, 2000-2005