Information Theory: An Approach to Human Communication
by: Donald K. Darnell
Summarized by: Shannon Curtis

From Samuel F.B. Morse to Claude E. Shannon to Warren Weaver, the idea of Information Theory, according to Donald Darnell, has been in existence for some time. However, we generally attribute the title of "Information Theory" to Claude Shannon. An engineer, Shannon treated communication as a machine that consisted of five essential parts: information source, transmitter, channel, receiver, and destination. With all parts of the machine operating effectively, Shannon's main objective was to reduce noise as much as possible in order to enhance each message. This reduction in noise was treated, by Shannon, in a very mathematical manner. Due to Shannon's technical approach and human's high tolerance for error in everyday communication, Darnell points out that Information Theory has little application in day-to-day exchanges. For example, the theory does not take into account today's little concern for efficiency when communicating.


Darnell later clarifies that the Information Theory is not at all concerned with the substance of a message. Rather, it is concerned with the transmission. Again using a technical approach, Shannon treats effective communication as the transfer of symbols and signals from one palce to another with as little distortation as possible. To enhance effective transmission, uncertainty needs to be eliminated. Darnell then goes into mathematical detail about binary digits (bits). Binary digits are a way of measuring uncertainty. If restricted to the use of bits, one bit per symbol can be transmitted. By applying logarithms to bits, Darnell helps to explain how a mathematician sees Information Theory as reducing transmission distortations through a process of elimination. Darnell tried to clarify the complexity of measuring bits by instructing the reader to compare communication to a game of twenty questions. For example, by asking optimal questions, possible outcomes can be reduced by half. Therefore, trying to guess a number between 1 and 10 would equal log2 10 or 3.322 questions needed to find the unknown number between 1 and 10. This answer (3.322) is refered to as the information value. Clozentropy, a combination of "closed procedures" is a type of test that is discussed throughout this essay as well. Clozentropy simply reiterates the idea of redundency in the English language. For example, if asked to fill in the blank for the sentence...How are _____ today?, there are only so many responses that the general population would give. By summing up the frequencies of each different response, dividing by the total number of responses, and sudstituting p for the relative frequency, you are calculating the entropy or uncertainty. (Entropy is another term for uncertainty.) The calculated answer for entropy is equal to the total number of different possible responses and how often they occur. By subtracting the entropy from the information value, you get a measure that makes a particular response comparible to the rest of the population's responses.


The essay concludes with the notion that by taking on a different perspective when referring to communication, one would most likely come up with a different conclusion to the Information Theory. Darnell chooses to use the legal decision-making system as an example of a different perspective resulting in a different conclusion. Darnell states that a decision-making system is an uncertainty absorber because is estabolishes one outcome when several are possible. By making decisions, possibilities are reduced. Darnell points out that, according to a decision-making system, Information Theory is a theory of choices (decisions). In other words, the outcome would be that a decision-making system capable of rendering justice must be capable of transmitting the maximum amount of information.


In my chosen reading, Darnell states that although he defined the Information Theory and all of its counterparts, he is by no means a complete advocate. Darnell is, however, an advocate of communicating wisely and without manipulation.

INTERPRETATION

Personally, I found the definitions and explanation that were scattered throughout this essay easily understandable. However, the literature that followed such definitions consisted of what I would consider jargon. Tangents consisting of terms such as logarithms, formulas, and percentages cluttered nearly half of the essay. Reading through the jargon several times made it comprehendable, but still difficult. I did enjoy examples such as the one given for clozentropy. However, the unexplained charts that followed the example lost what I had previously understood. In general, Donald K. Darnell educated me on most of the Information Theory, yet lost me on much of it as well. I learned that, in contrast to what Shannon would argue, speech is so much more than symbols and signals. Speech is personal experiences, individual opinions and intelligble communication skills. I would recommend this essay only to those who have a great deal of time and energy to spend rereading. An Approach to Human Communication is by no means light reading. However, if you do have the time and energy, reading and rereading will, without a doubt, educate you on the technical aspects of Information Theory by Claude Shannon.

Darnell, Donald (1972). Information Theory. In Richard W. Budd & Brent D. Ruben (Eds.), Approaches to Human Communication (pp. 156-169). New York: Spartan Books.

Click here for the more personal information on Claude E. Shannon and details about his study on Information Theory

Click here to find out more about uncertainty and other vocabulary words related to Shannon's Information Theory

Return Home