20 Informational environments
hand, the mathematical theory makes no reference at all to the meaning of the
messages whose transmission is the topic of inquiry, whereas an informational
theory of meaning will have to consider the mathematical properties of information.
4
On the other hand, however, the mathematical theory of communication
bases its concept of information on the notion of communicators of linguistic
meanings, whereas informational theorists of meaning, being primarily interested
in the origins of meaning, add qualifications to the role of agents as purposeful communicators. On these latter theories, information is, first and foremost,
natural information, to be encountered and picked up by its receivers in their
environments. It serves to first ground all sorts of intentional relations and hence
cannot presuppose their presence. Purposeful transmission of signals comes later,
historically and systematically.
The content of natural information, and some discontent
Retaining the Shannon-Weaverean presupposition that the presence and the
activities of purposeful communicators of meaning are the basis of an analysis of
information, Bar-Hillel and Carnap (1952), in their semantic theory of information, restricted the domain of their theory to sentences within a language. That
domain includes not only natural languages but also arguably privileges propositions formulated in formal languages. As Bar-Hillel and Carnap argue in some
formal detail, the information transmitted by a sentence simply is the content of
that sentence. This point, however, cannot be readily transferred to cases where
there is information but no sentences or other linguistic forms.
A rather different and influential take on the role of information in meaning
was presented by Dretske (1981) – although it might be argued that the influence of Dretske’s theory on philosophical debates is almost entirely composed of
stimulating arguments against his view. Dretske’s aim was to conceptually root
intentional phenomena in general and the possibility of knowledge in particular in
natural relations that he reconstructed as informational relations. These relations
were couched in probabilistic terms that, although commencing from an interpretation of the mathematical theory of information, depart from it in important
respects.
The obvious starting point for a reconstruction of Dretskean information is his
near-infamous definition of informational content:
Informational content: A signal r carries the information that s is F = The
conditional probability of s’s being F, given r (and k [standing for “knowledge”]), is 1 (but, given k alone, less than 1).
(Dretske 1981, 65)
On this account, information is a certain relation between two world affairs s and
r (things, events, properties or a combination thereof ) in which r is a signal of the
other affair s being or having property F only if and only when the signalled affair
of s being F is the case whenever the signalling affair r occurs. Because there is
hand, the mathematical theory makes no reference at all to the meaning of the
messages whose transmission is the topic of inquiry, whereas an informational
theory of meaning will have to consider the mathematical properties of information.
4
On the other hand, however, the mathematical theory of communication
bases its concept of information on the notion of communicators of linguistic
meanings, whereas informational theorists of meaning, being primarily interested
in the origins of meaning, add qualifications to the role of agents as purposeful communicators. On these latter theories, information is, first and foremost,
natural information, to be encountered and picked up by its receivers in their
environments. It serves to first ground all sorts of intentional relations and hence
cannot presuppose their presence. Purposeful transmission of signals comes later,
historically and systematically.
The content of natural information, and some discontent
Retaining the Shannon-Weaverean presupposition that the presence and the
activities of purposeful communicators of meaning are the basis of an analysis of
information, Bar-Hillel and Carnap (1952), in their semantic theory of information, restricted the domain of their theory to sentences within a language. That
domain includes not only natural languages but also arguably privileges propositions formulated in formal languages. As Bar-Hillel and Carnap argue in some
formal detail, the information transmitted by a sentence simply is the content of
that sentence. This point, however, cannot be readily transferred to cases where
there is information but no sentences or other linguistic forms.
A rather different and influential take on the role of information in meaning
was presented by Dretske (1981) – although it might be argued that the influence of Dretske’s theory on philosophical debates is almost entirely composed of
stimulating arguments against his view. Dretske’s aim was to conceptually root
intentional phenomena in general and the possibility of knowledge in particular in
natural relations that he reconstructed as informational relations. These relations
were couched in probabilistic terms that, although commencing from an interpretation of the mathematical theory of information, depart from it in important
respects.
The obvious starting point for a reconstruction of Dretskean information is his
near-infamous definition of informational content:
Informational content: A signal r carries the information that s is F = The
conditional probability of s’s being F, given r (and k [standing for “knowledge”]), is 1 (but, given k alone, less than 1).
(Dretske 1981, 65)
On this account, information is a certain relation between two world affairs s and
r (things, events, properties or a combination thereof ) in which r is a signal of the
other affair s being or having property F only if and only when the signalled affair
of s being F is the case whenever the signalling affair r occurs. Because there is