Varieties of perception 43
Gibson’s ambiguous characterisation of affordances as properties and as facts
might seem a minor issue, but the implications are significant if facts are taken
to be composed of things, properties and their relations. Context indicates that a
reading of affordances as relationally defined facts rather than properties will be
the most appropriate one. If affordances, their meanings and values were plainly
and entirely properties of the environment and the objects therein, we would not
only have to accept that such normative qualities were constituents of the environment, but also we would end up in a world cluttered with an indefinite array
of such qualities that would have to be embodied or embedded in the objects and
environments in which they are encountered, readily pre-packaged for all possible
perceivers and detectable under all natural conditions of perception. These qualities might even contradict each other in certain cases, for certain perceivers, and
still would have to be granted the same ontological status.
If one, disheartened by this prospect, does not want to settle for the opposite
view that all normative qualities of some perceived object, and all variance therein,
are located inside the organism proper and hence a subjective affair, an alternative
is to locate values, their specificity to the perceiver and their variance between
perceivers in the relations between organism and environment. Affordances, as
Gibson insists, are always related to activities, and are preferably described as
capacities or abilities of the object with respect to the perceiving organism’s constitution and abilities. The chair has the capacity of getting me seated, and the
cliff has the ability to make me fall off. Even if we render these facts as seeming
properties of fall-off-ability or sit-on-ability, these can be defined only in relation
to the animal – with rather different outcomes for ants and primates. These relations will be best captured by a more complex, fact-like description, such as “the
chair is sit-on-able for humans” or “the cliff is fall-off-able for large mammals”.
Hence, it is a relational interpretation of affordances that appears best to make
sense of that other, purposeful vagueness in Gibson’s characterisation of affordances as being objective and subjective at once. This relational, bi-directional
interpretation can be further substantiated by an analysis of the nature of the information involved in affordances as being equally relational and bi-directional. This
is what Gibson himself suggests when says that both affordances and information
point “two ways, to the environment and to the observer” (1979, 141).
On this reconstruction of Gibson’s notion of affordances, it is both realist and
anti-dualist in philosophical spirit. This reconstruction should carry over to an
account of ecological information for perception. There is one seeming ambiguity in Gibson’s underdefined account of information that should be taken to be
systematic.
3
On the one hand, “information [is present] in ambient light to specify
affordances” (Gibson 1979, 143), and as such is present and specific even if and
when an affordance is not perceived. On the other hand, information for perception
is always composed both of relations within the environment and of the perceiving organism’s relations to his environment. Hence, an important part of information for perception is dependent on the perceiving organism, and all information
for perception is relational in kind. Within ecological psychology, there has been a
variety of diverging interpretations of the relation between Gibsonian affordances
Gibson’s ambiguous characterisation of affordances as properties and as facts
might seem a minor issue, but the implications are significant if facts are taken
to be composed of things, properties and their relations. Context indicates that a
reading of affordances as relationally defined facts rather than properties will be
the most appropriate one. If affordances, their meanings and values were plainly
and entirely properties of the environment and the objects therein, we would not
only have to accept that such normative qualities were constituents of the environment, but also we would end up in a world cluttered with an indefinite array
of such qualities that would have to be embodied or embedded in the objects and
environments in which they are encountered, readily pre-packaged for all possible
perceivers and detectable under all natural conditions of perception. These qualities might even contradict each other in certain cases, for certain perceivers, and
still would have to be granted the same ontological status.
If one, disheartened by this prospect, does not want to settle for the opposite
view that all normative qualities of some perceived object, and all variance therein,
are located inside the organism proper and hence a subjective affair, an alternative
is to locate values, their specificity to the perceiver and their variance between
perceivers in the relations between organism and environment. Affordances, as
Gibson insists, are always related to activities, and are preferably described as
capacities or abilities of the object with respect to the perceiving organism’s constitution and abilities. The chair has the capacity of getting me seated, and the
cliff has the ability to make me fall off. Even if we render these facts as seeming
properties of fall-off-ability or sit-on-ability, these can be defined only in relation
to the animal – with rather different outcomes for ants and primates. These relations will be best captured by a more complex, fact-like description, such as “the
chair is sit-on-able for humans” or “the cliff is fall-off-able for large mammals”.
Hence, it is a relational interpretation of affordances that appears best to make
sense of that other, purposeful vagueness in Gibson’s characterisation of affordances as being objective and subjective at once. This relational, bi-directional
interpretation can be further substantiated by an analysis of the nature of the information involved in affordances as being equally relational and bi-directional. This
is what Gibson himself suggests when says that both affordances and information
point “two ways, to the environment and to the observer” (1979, 141).
On this reconstruction of Gibson’s notion of affordances, it is both realist and
anti-dualist in philosophical spirit. This reconstruction should carry over to an
account of ecological information for perception. There is one seeming ambiguity in Gibson’s underdefined account of information that should be taken to be
systematic.
3
On the one hand, “information [is present] in ambient light to specify
affordances” (Gibson 1979, 143), and as such is present and specific even if and
when an affordance is not perceived. On the other hand, information for perception
is always composed both of relations within the environment and of the perceiving organism’s relations to his environment. Hence, an important part of information for perception is dependent on the perceiving organism, and all information
for perception is relational in kind. Within ecological psychology, there has been a
variety of diverging interpretations of the relation between Gibsonian affordances