Reply to Paul D. Part1: info. systems as field of analysis

From: Bruce Robinson (bruce.rob@btinternet.com)
Date: Thu Feb 10 2000 - 04:27:38 PST


Thanks to Judy and Paul for their comments on my paper. I'd like to reply
to Paul's comments at some length (hence the delay) as I think they raise a
number of different issues, ranging from why it takes the form it does,
through painting in some of the background in terms of the Information
Systems discipline through to discussing Ilyenkov's ideas and more
generally the nature of dialectical thought and method. I've divided my
thoughts up into separate posts as they do run to some length.

First, to say that the paper is not supposed to be a complete dialectical
analysis of the field (though that is what I am working towards - and I am
finding this discussion very helpful in doing so). It is a particular
staging post in that development and as such is work in progress. Rather it
aims to draw the attention of IS people to materialist dialectics an
alternative to the dominant naive realism and a growing minority of social
constructionists / interpretivists and to show that recognised problems in
the field can be better understood in this way. I have focussed on
modelling (which as I make clear is not the whole of the process IS is
concerned with) because (a) there is less critical analysis of it than the
embodiment of information systems into forms of activity; (b) modelling is
both central to IS theory and practice and has wider implications as a form
of representation.

With regard to your specific questions about the contradictions within the
IS discipline, by sheer coincidence J, I have another paper which attempts
to apply Vygotsky's concepts and methods in 'The Crisis of Psychology' to
IS. While it focuses on his notion of 'disciplinary crisis' rather than
talking too much about contradiction, I think it would give you an idea as
to the issues that arise at present. (It can be found at:
http://www.mngt.waikato.ac.nz/ejrot/cmsconference/cmsdefault.htm in the
section headed 'Information Technology and Critical Theory'.)

Some points from that paper that are directly relevant here. Firstly, there
are two issues in IS which I see as being central under the heading of
Vygotsky's second point (crisis arising from the specific nature of the
discipline's subject matter): the nature of information itself and the
relationship between the social and technical, which is clearly also
relevant in all disciplines associated with technology development and use.
(I hear the Latourians sharpening their pencils at this point!)

Secondly, one of the problems of IS, a young discipline which has emerged
by a historical process in which it has distinguished itself to varying
degrees from management, sociology and computer science, is precisely that
it lacks a simple focus, a single object which might simply be taken to be
characteristic of the discipline and show all its interconnections and
contradictions. The obvious temptation is to take the information system
itself - consensually defined as some combination of humans, artefact (not
necessarily a computer) and organisation - as that object. In yet another
paper, however, I have shown that this is an inadequate basis for analysis
and explanation of an information system's impact and function. It is too
narrow a 'totality' to make the range of relevant interconnections and
contradictions explicit. There are two possible conclusions to draw from
this. One is that IS isn't a 'real discipline' because it lacks such an
object. (I would certainly accept that that is one aspect of its problems.)
The other is that in this particular case there is no single object that
can act as 'concrete universal', that embodies all contradictions, that is
a 'cell form'. This is not to say that one cannot draw boundaries that show
where the contradictions lie and that what is within them forms in some
sense a totality, but that it is not possible to find an object such as the
commodity or the word-meaning to do it. (In part 2, I'll argue why I think
that dialectical method does not require that there is.) Ironically, this
would match Vygotsky and Ilyenkov's insistence that one examine the precise
subject matter of a discipline as a precondition of seeing how general
dialectical methods apply to it.

However, conversely, examining other forms of knowledge and applying
Vygotsky's framework in 'Crisis' to IS makes it clear that there are
supra-disciplinary aspects to dialectics, which V. refers to as "general
dialectics" or "the general conditions and laws of scientific knowledge"
i.e. general methodological, epistemological and ontological questions. For
example, Vygotsky's critiques of empiricism and subjectivism are highly
relevant to IS without too much alteration. (Similarly, for the lack of
much precise writing by him on method, Marx's on political economy are
applied in other subject areas, though they are perhaps more difficult to
disentangle from their precise subject matter.) Ilyenkov must of necessity
admit this if 'moving from the abstract to the concrete' is to be a general
rule of method.

Thus dialectics itself is a unity of the opposites of subject matter
specific methods and general principles - or to put it another way, exists
at a number of different levels. (I think that comes out in your email
where you say: 'As Ilyenkov presents the matter dialectics works with a
"universal concept underlying the entire system of the categories of
science"', and at the same time 'But there is no method for applying
dialectics to "any object in general"'.) Modelling as a topic lies
precisely on the cusp of the two: it takes specific forms in IS and other
disciplines, yet also can easily serve as a more general model (oops!) of
formalised representation.

So my reply to your saying that my theory could apply 'to any form of
modeling whatsoever' is (a) that I do not think that this (necessarily or
in this case) leads to incorrect conclusions i.e. that may be the right
level at which to analyse these issues; (b) that your 'functional
ignorance' of IS may mean that you are not always where I do address
IS-specific issues, precisely because IS does use general semantic
primitives to model with. Thus, for example, my discussion of 'What is an
entity? What is an attribute?' addresses an widely used IS modelling
technique known as entity (or entity-attribute-relationship) modelling,
which is central in database design and many systems development methods.
Similarly, Artz' discussion of objects and classes relates to another
technique of object-oriented modelling. So general issues of epistemology
and classification lurk just behind IS techniques, which is why those
authors who look at the problems of IS modelling in depth tend to move to
philosophical questions and often towards partial and unconscious
recognition of issues tackled by dialectics (see the examples in the
paper).

I hope that clarifies the issues related to information systems and in
passing raises some other points of broader interest. I'll deal with some
of the more general points about dialectics and issues of my own method in
another post.

Bruce Robinson

=====================================

Bruce Robinson
Information Systems Research Centre
University of Salford
Salford M5 4WT
UK

Phone / fax: 0161 861 7160
Email: bruce.rob@btinternet.com

=====================================



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 17:54:03 PST