94 Informational environments
its sub-clauses in Chapter 4). Variability at the source is what makes information
relevant to the receiver. The set of relevant variable conditions at the source and
the likeliness of signals to occur when certain conditions at the source obtain
determine the reference class for the informational relations. On this account, high
variability but not randomness at the source and little variability in their correlations to proximate, signalling states are what provides a function for cognition.
On the view presented by Godfrey-Smith, an environment may be too dull
to warrant any cognitive effort on the one hand, in terms of the relevant distal
conditions being too stable and too predictable to provide sufficient grounding
for any adaptive function for mechanisms of memory, learning, inference and so
forth. Dretske’s magnetotactic bacteria may serve as an example of organisms
living in such a dull environment. There is only one variable that is informationally relevant to the bacterium, and the correlation between the direction of the
Earth’s magnetic field and the gradient of oxygen concentration in marine waters
is extremely stable. Unless, however, the bacteria were simply being pulled downward by magnetic forces instead of the organisms actively following the lines of
the magnetic field as a proxy of the gradient of oxygen concentration, there will
be information involved in tracking that one relevant variable. Still, the evolution
of cognitive abilities will require more than that.
On the other hand, an environment may also be too chaotic to warrant any
cognitive effort if distal conditions change randomly or if there are no reliable
correlations to proximal affairs that the organism could detect. There would be
no information for him to use in cognitive processes. Here, sheer resilience in
organic functions or the ability to reproduce in large quantities whenever conditions are favourable will be workable strategies. This is as close as it gets for
life to being information free. According to Sterelny (2003, 149), this kind of
case forms a class of its own, besides “tracking” favourable conditions (the constitutive strategy in Godfrey-Smith’s terminology) and “ecological engineering”
(the constructive strategy). This is the strategy of “adapting” to conditions in the
environment, which seems to echo that classical interpretation of adaptation in
evolutionary theory on which environmental pressure works on lineages of otherwise inert organic matter. Here, however, it is only one class of strategies among
others to create an organism-environment fit, not the expression of a natural law
of natural selection.
Conditions will be neither too dull nor too chaotic for any animal that uses natural information in cognition. It would look reasonable to assume, within GodfreySmith’s conceptual framework, that purposefully created signals might be used
in the same fashion. If the function of the mind, on his “Environmental Complexity Thesis” (1996a, 3) explicated in the previous quotation, is to “enable the
agent to deal with environmental complexity”, and if environmental complexity is
understood as the heterogeneity and variability of conditions therein, purposefully
produced signalling events or artefacts may serve as an environmentally rooted
means of dealing with a given situation of complexity in a manner analogous to
natural information, as long as they fulfil the second of the aforementioned conditions, namely that of stability of relations between proximal and distal conditions.
its sub-clauses in Chapter 4). Variability at the source is what makes information
relevant to the receiver. The set of relevant variable conditions at the source and
the likeliness of signals to occur when certain conditions at the source obtain
determine the reference class for the informational relations. On this account, high
variability but not randomness at the source and little variability in their correlations to proximate, signalling states are what provides a function for cognition.
On the view presented by Godfrey-Smith, an environment may be too dull
to warrant any cognitive effort on the one hand, in terms of the relevant distal
conditions being too stable and too predictable to provide sufficient grounding
for any adaptive function for mechanisms of memory, learning, inference and so
forth. Dretske’s magnetotactic bacteria may serve as an example of organisms
living in such a dull environment. There is only one variable that is informationally relevant to the bacterium, and the correlation between the direction of the
Earth’s magnetic field and the gradient of oxygen concentration in marine waters
is extremely stable. Unless, however, the bacteria were simply being pulled downward by magnetic forces instead of the organisms actively following the lines of
the magnetic field as a proxy of the gradient of oxygen concentration, there will
be information involved in tracking that one relevant variable. Still, the evolution
of cognitive abilities will require more than that.
On the other hand, an environment may also be too chaotic to warrant any
cognitive effort if distal conditions change randomly or if there are no reliable
correlations to proximal affairs that the organism could detect. There would be
no information for him to use in cognitive processes. Here, sheer resilience in
organic functions or the ability to reproduce in large quantities whenever conditions are favourable will be workable strategies. This is as close as it gets for
life to being information free. According to Sterelny (2003, 149), this kind of
case forms a class of its own, besides “tracking” favourable conditions (the constitutive strategy in Godfrey-Smith’s terminology) and “ecological engineering”
(the constructive strategy). This is the strategy of “adapting” to conditions in the
environment, which seems to echo that classical interpretation of adaptation in
evolutionary theory on which environmental pressure works on lineages of otherwise inert organic matter. Here, however, it is only one class of strategies among
others to create an organism-environment fit, not the expression of a natural law
of natural selection.
Conditions will be neither too dull nor too chaotic for any animal that uses natural information in cognition. It would look reasonable to assume, within GodfreySmith’s conceptual framework, that purposefully created signals might be used
in the same fashion. If the function of the mind, on his “Environmental Complexity Thesis” (1996a, 3) explicated in the previous quotation, is to “enable the
agent to deal with environmental complexity”, and if environmental complexity is
understood as the heterogeneity and variability of conditions therein, purposefully
produced signalling events or artefacts may serve as an environmentally rooted
means of dealing with a given situation of complexity in a manner analogous to
natural information, as long as they fulfil the second of the aforementioned conditions, namely that of stability of relations between proximal and distal conditions.