104 Informational environments
Notably, IE-1c also implies that an organism’s informational environment may
become enriched, for example by artefacts that are designed to enhance detection of some relevant condition, by providing additional signals or by making
the uptake of information more reliable. In analogy to the case of impoverishment, we should not be tempted to conclude that R’s informational environment
has become more advanced in an absolute sense, nor should we believe that
the regularities connecting signal and source have improved in terms of being
altered from a vaguely probabilistic towards a more law-like state of affairs.
Either way, the regularities underlying the ‘original’, natural signalling relation
remain the same, and they will amount to a specification relation, although the
probabilities of correct mappings between signal and source will have altered
due to intervening factors, which add to or subtract from the quality of what is
received.
Having thus distinguished alterations of the informational relations proper from
alterations of informational environments, what more can be said on the conditions under which an informational environment may count as having changed?
To begin with, whatever change there is and on whatever level it occurs, it must be
persistent with respect to the receiver’s timescale of observation in order to qualify as a genuine change. Momentary aberrations in channel conditions or transient
fluctuations in mapping frequencies that the receiver can correct for or integrate,
or that he might be entitled to ignore as being too insignificant altogether, will not
count. Moreover, the change in question must be consistent with the regularities
in his environment. Inverted spectra or green things turning bleen all of a sudden,
as they do in some philosophical thought experiments, will not count either. If
conditions in an environment change, and even if some of the regularities in that
environment are affected, such change will hardly occur in capricious fashion, nor
is it likely to merely affect isolated variables.
1
In terms of its origins, a change in a receiver’s informational environment may
owe to a change in his ecological environment. It does not necessarily depend
on an ecological change though, and if it indeed does so depend, the ecological
change is what should be the main matter of concern. A change in an informational
environment may, and often will, entirely occur on the informational level, while
having possible repercussions on the ecological level. Such an informationallevel change might be of basically two kinds, as indicated in IE-2b, allowing for
all sorts of shades between them: first, the availability and quality of some of the
information usable to R may vary over time and space, in terms of new channels
or new types of signals being added, removed or beginning to interfere with each
other. Second, however, novel informational relations, with their own mapping
regularities, may emerge that had not been present in the environment before. On
the account of natural information that I have presented in this and the preceding
chapters, an addition of genuinely new informational relations to the environment will occur only if some of the natural regularities that govern the mapping
between signals and sources are altered or new ones are introduced. Supposedly,
this is an exceedingly rare case. Conversely, new kinds of sources of information
may appear to which established regularities apply.
Notably, IE-1c also implies that an organism’s informational environment may
become enriched, for example by artefacts that are designed to enhance detection of some relevant condition, by providing additional signals or by making
the uptake of information more reliable. In analogy to the case of impoverishment, we should not be tempted to conclude that R’s informational environment
has become more advanced in an absolute sense, nor should we believe that
the regularities connecting signal and source have improved in terms of being
altered from a vaguely probabilistic towards a more law-like state of affairs.
Either way, the regularities underlying the ‘original’, natural signalling relation
remain the same, and they will amount to a specification relation, although the
probabilities of correct mappings between signal and source will have altered
due to intervening factors, which add to or subtract from the quality of what is
received.
Having thus distinguished alterations of the informational relations proper from
alterations of informational environments, what more can be said on the conditions under which an informational environment may count as having changed?
To begin with, whatever change there is and on whatever level it occurs, it must be
persistent with respect to the receiver’s timescale of observation in order to qualify as a genuine change. Momentary aberrations in channel conditions or transient
fluctuations in mapping frequencies that the receiver can correct for or integrate,
or that he might be entitled to ignore as being too insignificant altogether, will not
count. Moreover, the change in question must be consistent with the regularities
in his environment. Inverted spectra or green things turning bleen all of a sudden,
as they do in some philosophical thought experiments, will not count either. If
conditions in an environment change, and even if some of the regularities in that
environment are affected, such change will hardly occur in capricious fashion, nor
is it likely to merely affect isolated variables.
1
In terms of its origins, a change in a receiver’s informational environment may
owe to a change in his ecological environment. It does not necessarily depend
on an ecological change though, and if it indeed does so depend, the ecological
change is what should be the main matter of concern. A change in an informational
environment may, and often will, entirely occur on the informational level, while
having possible repercussions on the ecological level. Such an informationallevel change might be of basically two kinds, as indicated in IE-2b, allowing for
all sorts of shades between them: first, the availability and quality of some of the
information usable to R may vary over time and space, in terms of new channels
or new types of signals being added, removed or beginning to interfere with each
other. Second, however, novel informational relations, with their own mapping
regularities, may emerge that had not been present in the environment before. On
the account of natural information that I have presented in this and the preceding
chapters, an addition of genuinely new informational relations to the environment will occur only if some of the natural regularities that govern the mapping
between signals and sources are altered or new ones are introduced. Supposedly,
this is an exceedingly rare case. Conversely, new kinds of sources of information
may appear to which established regularities apply.