The domains of natural information 75
may be integrated by a suitably disposed observer, who is viewed in detachment
from his environment, in inferential processes of knowledge acquisition. On Gibson’s view, informational relations are invariants in the patterns in the ambient
energies that can, in practice, be detected by a perceiver on the concrete grounds
of his constitution and abilities. Some of these invariants, as I will further elaborate in the following chapter, form specific integrated bundles for the organism
that are determined in relation to what he can and needs to do in his environment.
Assessing those invariants in isolation would not do justice to the complexity and
multi-modality of interactions. By virtue of being thus integrated and situated, the
information available in the environment will paradigmatically be sufficient to
directly guide the perceiving organism towards the intended state of affairs, with
no k-condition needing to intervene.
If we want to know how animals and humans, to return to the quote that opened
my discussion of Gibson (see Chapter 3), have come to “communicate with cries,
gestures, speech, pictures, writing, and television” (Gibson 1979, 242), and how
they have thus come to be knowledgeable creatures, we might wish not to include
the knowledge criterion from the start, as most organisms do quite well without
it most of the time. One might also come to more precisely identify the locus of
informational content, namely in the specific relation among, first, what information is present in the environment, second, what of this information is available to
the perceiving organism and third, to what ends he uses it.
Notes
1 The possibility of equivocal symptoms is mobilised against Dretske’s nomological
account of information by Millikan (2001).
2 The different answers that can be given to this question are analogous to the – somewhat
notorious – distinction between K- and r-selection.
3 One example for a more liberal approach in Dretske would be the very passage in Dretske (1988, 56) in which he introduces the quail vs. pheasants example, as Millikan
(2004, 32) notes.
4 In Millikan (2004, 52), she says: “Don’t trust what looks like that needle when you go
up to Mars, but here on earth, the positions of needles on gas gauges all fall in the same
roughly defined locally recurrent sign domain.” By implication, gas gauges belong to a
type of artefact that shares a number of relevant properties with species, namely being
reproductively established as a type of artefact, which is distinguished by its proper
functions, and which populates only a limited set of slices of space-time. Here on earth,
we recognise what a gas gauge is and what it signals when we see one. Similarly, don’t
trust your olfactory organs when you go to the moon and seek to identify objects on the
moon by smell – but we know the shape and function of olfactory organs of different
species, as established in processes of biological evolution, and we know something
about the conditions for performing that function.
5 This exemplary juxtaposition characterises one of the core tenets of Millikan’s theory of
proper functions, as proposed in Millikan (1984) and discussed in more detail in Chapter 7, namely that of (first-order) reproductively established families.
may be integrated by a suitably disposed observer, who is viewed in detachment
from his environment, in inferential processes of knowledge acquisition. On Gibson’s view, informational relations are invariants in the patterns in the ambient
energies that can, in practice, be detected by a perceiver on the concrete grounds
of his constitution and abilities. Some of these invariants, as I will further elaborate in the following chapter, form specific integrated bundles for the organism
that are determined in relation to what he can and needs to do in his environment.
Assessing those invariants in isolation would not do justice to the complexity and
multi-modality of interactions. By virtue of being thus integrated and situated, the
information available in the environment will paradigmatically be sufficient to
directly guide the perceiving organism towards the intended state of affairs, with
no k-condition needing to intervene.
If we want to know how animals and humans, to return to the quote that opened
my discussion of Gibson (see Chapter 3), have come to “communicate with cries,
gestures, speech, pictures, writing, and television” (Gibson 1979, 242), and how
they have thus come to be knowledgeable creatures, we might wish not to include
the knowledge criterion from the start, as most organisms do quite well without
it most of the time. One might also come to more precisely identify the locus of
informational content, namely in the specific relation among, first, what information is present in the environment, second, what of this information is available to
the perceiving organism and third, to what ends he uses it.
Notes
1 The possibility of equivocal symptoms is mobilised against Dretske’s nomological
account of information by Millikan (2001).
2 The different answers that can be given to this question are analogous to the – somewhat
notorious – distinction between K- and r-selection.
3 One example for a more liberal approach in Dretske would be the very passage in Dretske (1988, 56) in which he introduces the quail vs. pheasants example, as Millikan
(2004, 32) notes.
4 In Millikan (2004, 52), she says: “Don’t trust what looks like that needle when you go
up to Mars, but here on earth, the positions of needles on gas gauges all fall in the same
roughly defined locally recurrent sign domain.” By implication, gas gauges belong to a
type of artefact that shares a number of relevant properties with species, namely being
reproductively established as a type of artefact, which is distinguished by its proper
functions, and which populates only a limited set of slices of space-time. Here on earth,
we recognise what a gas gauge is and what it signals when we see one. Similarly, don’t
trust your olfactory organs when you go to the moon and seek to identify objects on the
moon by smell – but we know the shape and function of olfactory organs of different
species, as established in processes of biological evolution, and we know something
about the conditions for performing that function.
5 This exemplary juxtaposition characterises one of the core tenets of Millikan’s theory of
proper functions, as proposed in Millikan (1984) and discussed in more detail in Chapter 7, namely that of (first-order) reproductively established families.