The broad theme of communication seems to present challenges at three levels. Therefore, it seems reasonable to inquire about them in sequence.
LEVEL A: How accurately can symbols of communication be transmitted? (Technical issue.) LEVEL B: How accurately do the transmitted symbols convey the desired meaning? (Semantic issue.)
- The semantic problems involve the interpretation of meaning by the receiver compared to the sender’s intended meaning.
- Cultural differences in understanding meanings. LEVEL C: How does the received meaning effectively influence behavior? Is it in a desirable manner? (Issue of effectiveness.)
- Assumption: The purpose of all communication is to influence the recipient’s behavior.
- The effectiveness problem is closely related to the semantic problem, with some overlap between them and the other suggested problem categories.
- It is important not to confuse information with meaning.
Moving from Level A to Levels B and C, it becomes apparent that considering the statistical characteristics of the receiver is essential. One could imagine inserting an additional box labeled “semantic receiver” between the engineering receiver (device converting signals to messages) and the recipient. This semantic receiver would decode the message secondarily, aligning the decoding request with the statistical semantic characteristics of the message for all receivers or a subset of receivers whose semantic abilities are to be influenced.
Similarly, one could consider inserting another box labeled “semantic noise” between the information source and the transmitter. This box, previously labeled simply as “noise” and now as “engineering noise,” adds unintended distortions and disruptions of meaning to the signal that inevitably affect the recipient. The issue of semantic decoding must consider this semantic noise, adjusting the original message so that the combined meaning of the message and semantic noise matches the intended overall message meaning at the recipient.
Utilizing a powerful theory related to Markov processes seems very promising for semantic research, particularly in addressing one of the most important and challenging aspects - the influence of context on meaning. There is a vague sense of a joint constraint where information and meaning are akin to canonical conjugate variables in quantum theory, where emphasizing one may sacrifice the other.
The idea of expressing language in terms of Markov processes seems to be under consideration.