260 Ulrich Krohs
A hot topic with respect to biological modularity is how to explain its evolution. I
discuss arguments that have been put forward to explain the evolution of modularity, and
confront them with the fact that modularity is not an all-or-nothing issue, but comes in
degrees (sections 15.3 and 15.4). The evolutionary explanations do not offer any reason
why biological organisms are less than maximally modularized, or why even secondary
integration of modules has occurred, for example, in endosymbiogenesis. A comparison
of the biological arguments with arguments about modularity in technology shows that an
important aspect is missing in the biological considerations: they focus almost exclusively
on the benefi ts of modularity, while in engineering it is acknowledged that a modular
structure may also have disadvantages, such as a larger weight of a modular device as
compared to an integrated one, or less fl exibility in meeting other than standard require-
ments. I show that not only the benefi ts but also the costs must be considered to explain
the evolution of modular organization of biological organisms (section 15.5). This is meant
not as a biological argument about the evolution of modularity but as a contribution to the
question which structure an argument about adaptive processes needs to have in order to
be of explanatory value.
A second aim in this chapter concerns the aspect of functionality in discussions of bio-
logical and technological modularity. Function ascriptions are used to delineate modules
in entities of both realms. However, the results are quite different in both cases. Techno-
logical functional modules largely coincide with structural subunits, while in biological
systems this is often not the case (section 15.6). The mismatch between the results of
structural and functional decomposition gives rise to claims in the fi eld of systems biology
that only the structural approach should count as yielding adequate results. I look into the
divergence of structural and functional modules from a different perspective. I fi rst discuss
the question of whether functional modularity may be of any relevance in explaining the
evolution of organisms, and whether or not cost considerations can help in explaining
evolution in this case as well (section 15.7). My fi nal concern with respect to the mismatch
between functional and structural modules then is to draw some conclusions regarding the
epistemic consequences of accepting the mismatch that occurs in particular systems
(section 15.8).
15.2 Delineating Modules
In many cases modules can be delineated morphologically. The organs of animals, for
example, are discernible structures, which show strong internal interactions and compara-
bly weak and few interactions among each other. They also have distinct functions
(pumping and detoxifying blood, mediating gas exchange, digesting food, etc.).
Morphological criteria, however, are not always applicable when decomposing a
complex system. Think of the network of the metabolism of the cell. Here, the various