Chemistry Reference
In-Depth Information
1. addition and deletion of special laws (e.g. the creation different models of the
theory or the creation of specialisation theory nets);
2. change of scale or metrics as well as the salience of dimensions;
3. change in the separability of dimensions;
4. addition and deletion of dimensions which make up the space
In this context,
'
dimensions
'
refers to the terms in the general structure
h
D
1
,
...
,
D
k
,
x
1
...
x
p
+
q
i
rather than to physical dimensions. In other words, the
belief-revision strategies for theory change may involve limited changes of
the theoretical cores as modelled in the structures M.
In Hettema (
2012a
) I proposed that we view reduction relationships as an
instance of an interstructural
link
, sufficient to establish a global unity of science,
but also capable of dropping local
. Belief revision is a stronger theory in
the sense that it provides additional specification of how such a link might work.
Abstract
links are simply relationships between two sets of potential models
(or the
context
'
'
conceptual spaces
) of two theories. An abstract link is defined as (Balzer
'
'
et al. (
1987
), p. 61):
Definition 1 (Abstract link)
L
is an abstract link from M
p
to M
0
p
iff
L
M
0
p
The leading idea of the abstract link is that it provides a relationship between
two different types of potential models, but does little else. In practice, links
are instantiated as connection pairs between terms of the
M
p
conceptual space
of
'
'
x
i
,
x
j
,
one theory to those of another; e.g.
that may have some
additional restrictions in terms of either values that the quantities can take in the
link, or a (law-like) relation between these concepts. The machinery for links can
become cumbersome, but the concept is not conceptually complex: it expresses that
some terms in one
hh
...i
,
h
x
p
,
x
q
,
...ii
conceptual
space can be connected to (a number of) terms in
'
'
the other conceptual space.
It is possible to define additional properties on links, and in this way develop a
concept of interpreting links, reducing links, and so on. Of particular interest is that
in the structuralist approach, the unity of science is formulated in terms of theory
holons
, which are large-scale global structures connected by inter-theoretic links.
In Hettema (
2012a
) I have argued that reduction postulates can fruitfully be
interpreted in terms of links, and the concept of links can be made to fit the three
criteria for reduction postulates that were originally developed by Nagel.
In the remainder I wish to forego many of the details of the structuralist approach
by focusing on links as connections between conceptual spaces. The analysis by
Gardenfors and Zenker adds structural precision to the generative strategies that are
available to establish links. The main import of treating conceptual disconnects in
this way, as G¨rdenfors and Zenker argue, is that the scope of
incommensurability
between a predecessor and successor theory, or between a reduced and reducing
theory, is limited significantly.
From the viewpoint of reduction, this approach is capable of formalising, and
subsequently de-fanging, the discontinuity between concepts in the theories of
chemistry and the theories of physics. In the next section I will discuss this with
the help of a practical example: the theory of absolute reaction rates.
