What is an informational environment? 95
There might or might not be a competitor claiming that pasture over there as his
territory. His markings will tell me, even if I cannot see him. There might or might
not be a predator around the corner. Another individual’s warning call will tell me,
even if I cannot see that predator myself.
There are two apparent reasons why Godfrey-Smith does not consider the possibility of an analogy of functions between natural information and purposefully
produced signals: first, his concept of cognition is explicitly restricted to a pragmatic interpretation in the Deweyan tradition, where it figures as a problem-solving
capacity possessed by individuals in the first place. Perceptual cues that directly
guide action in a Gibsonian sense are, on Godfrey-Smith’s view, no natural component of this picture of cognition. This, however, might not be the only or the best
way of relating Gibson’s work to the tradition of American pragmatism of which
Dewey was part. Both Anthony Chemero (2009) and Harry Heft (2001) draw a
fairly direct genealogical and systematic line between the “American naturalists”,
particularly and respectively James Dewey and William James, and James Jerome
Gibson’s ecological psychology. Gibson’s conspicuous eschewal of mental representations as explanantia in an empirical psychology is cited as a key testimony
to that heritage. At the end of Chapter 7, I will suggest that Dewey’s notion of
the organism-environment relation offers another analogy to the Gibsonian view.
According to Godfrey-Smith’s view of cognition, Gibsonian perception-action
will be a worthwhile strategy only in an environment that is distally relatively
more dull than, for example, a human environment, but still proximally reliable.
One part of my argument, however, was to maintain in a Gibsonian vein that much
of human activity is just as directly guided by perception as that of cognitively
more humble animals. This does not amount to dismissing the importance of the
additional accomplishment of internal sender-receiver games that constitute cognition, and Godfrey-Smith’s merit lies in describing the conditions under which
cognition, in the traditional narrow sense (no pejorative connotations intended
for either predicate), is valuable. However, maintaining that direct perception is
essential to human action, too, implies that only some, however important, aspects
of human environments provide the right sort of distal-variability/proximal-reliability ratio to ground a set of functions for higher-order, narrow cognition.
Second, if cognition is defined as an internally based problem-solving capacity
on the background of the kind and level of environmental complexity an organism
has to deal with, the possibility will not be acknowledged that the accomplishment of some of the tasks of coping with environmental complexity, and hence, in
Godfrey-Smith’s terms, cognitive tasks, may be delegated to the environment by
means of informational artefacts that still remain constituents of cognition proper.
I will discuss the possibility of constructed entities in the environment that play
this latter kind of role in cognitive processes in more detail in the next two chapters (Chapters 7 and 8). It will suffice to say for now that, even if one sticks to
the view that cognition should be considered as a set of internal, neurally based
processes in the first place, there are good reasons to believe that creating signals
and placing them in the environment does amount to a kind of externalisation of
the accomplishment of cognitive tasks into the environment, especially in view of
There might or might not be a competitor claiming that pasture over there as his
territory. His markings will tell me, even if I cannot see him. There might or might
not be a predator around the corner. Another individual’s warning call will tell me,
even if I cannot see that predator myself.
There are two apparent reasons why Godfrey-Smith does not consider the possibility of an analogy of functions between natural information and purposefully
produced signals: first, his concept of cognition is explicitly restricted to a pragmatic interpretation in the Deweyan tradition, where it figures as a problem-solving
capacity possessed by individuals in the first place. Perceptual cues that directly
guide action in a Gibsonian sense are, on Godfrey-Smith’s view, no natural component of this picture of cognition. This, however, might not be the only or the best
way of relating Gibson’s work to the tradition of American pragmatism of which
Dewey was part. Both Anthony Chemero (2009) and Harry Heft (2001) draw a
fairly direct genealogical and systematic line between the “American naturalists”,
particularly and respectively James Dewey and William James, and James Jerome
Gibson’s ecological psychology. Gibson’s conspicuous eschewal of mental representations as explanantia in an empirical psychology is cited as a key testimony
to that heritage. At the end of Chapter 7, I will suggest that Dewey’s notion of
the organism-environment relation offers another analogy to the Gibsonian view.
According to Godfrey-Smith’s view of cognition, Gibsonian perception-action
will be a worthwhile strategy only in an environment that is distally relatively
more dull than, for example, a human environment, but still proximally reliable.
One part of my argument, however, was to maintain in a Gibsonian vein that much
of human activity is just as directly guided by perception as that of cognitively
more humble animals. This does not amount to dismissing the importance of the
additional accomplishment of internal sender-receiver games that constitute cognition, and Godfrey-Smith’s merit lies in describing the conditions under which
cognition, in the traditional narrow sense (no pejorative connotations intended
for either predicate), is valuable. However, maintaining that direct perception is
essential to human action, too, implies that only some, however important, aspects
of human environments provide the right sort of distal-variability/proximal-reliability ratio to ground a set of functions for higher-order, narrow cognition.
Second, if cognition is defined as an internally based problem-solving capacity
on the background of the kind and level of environmental complexity an organism
has to deal with, the possibility will not be acknowledged that the accomplishment of some of the tasks of coping with environmental complexity, and hence, in
Godfrey-Smith’s terms, cognitive tasks, may be delegated to the environment by
means of informational artefacts that still remain constituents of cognition proper.
I will discuss the possibility of constructed entities in the environment that play
this latter kind of role in cognitive processes in more detail in the next two chapters (Chapters 7 and 8). It will suffice to say for now that, even if one sticks to
the view that cognition should be considered as a set of internal, neurally based
processes in the first place, there are good reasons to believe that creating signals
and placing them in the environment does amount to a kind of externalisation of
the accomplishment of cognitive tasks into the environment, especially in view of