The Information Theory Case Against Evolution

 Mathematician William Dembski, an ardent ID advocate, uses information theory to attempt to prove that life and the universe cannot possibly be the result of either natural processes or chance.  Neither mechanism, he insists, is capable of increasing information. Thus, there must have been an intelligent designer. 

 Dembski is inconsistent in his use of the term "information."  In his words , he implies the common understanding of information as a measure of knowledge (meaningful content) about a system.  But in his mathematics, he defines a quantity of information that is identical to what is known as "Shannon uncertainty", which refers to the number of bits needed to transmit a signal communicating a message, irrespective of the content of the message.  This is important, because, in information theory terms, Shannon uncertainty has nothing to do with the content of the message transmitted.  It has to do with the volume of bits transmitted.  If Dembski's law is meant to be applicable to information theory, as its name implies, it cannot be applied to a text's meaningful content.  He inconsistently switches back and forth between these concepts of information theory. 

 In information theory terms, the conventional definition of information, consistent with the vernacular use of the term mentioned above, is:  the decrease in Shannon uncertainty under the action of some process.  Thus, if fewer bits are needed to describe a system after the process, information about the system has been obtained.  It is not clear the extent to which Dembski understands this.

 More important is Dembski's "law of conservation of information," which states that the number of bits of information cannot change in any natural process such as chance or the operation of some physical law, i.e., chance and laws working in tandem cannot generate "complex specified information" (meaningful content).  Since the universe contains information, that information must have come about by other means (which he labels intelligent design) (Dembski: 1999).  While he insists that this argument does not depend on any specific theological assumptions, his writings promote his interpretation that the design is the work of the Christian god. 

 To provide a background for understanding Dembski's argument and how it is mistaken, a short discussion of information theory is presented.

The Science of Information Theory

 "Information" as understood in information theory has nothing to do with any inherent meaning in a message.  The word is used to mean the degree of order, of non-randomness, that can be measured and treated mathematically much as mass or energy or any other physical quantities are. 

 The science of "information theory" began during World War II.  Information theory is a branch of the mathematical theory of probability and mathematical statistics that studies the communication process, viewed as the transmission of information.  Essentially, information is a measure of a system's randomness, which can be identified, sent, and received only if it is recorded in the structure of a material medium (including electromagnetic waves). 

 Although information is sometimes measured in characters, as when describing the length of an email message, the convention in information theory is to measure information in bits.  A "bit" (the term is a contraction of binary bit) is either 0 or 1.  Because there are 8 possible configurations of three bits (000, 001, 010, 011, 100, 101, 110, and 111), three bits can be used to encode any integer from 1 to 8.  All logarithms are expressed to the base two, so log 8 is 3.  Similarly, log 1000 is slightly less than 10 and log 1,000,000 is slightly less than 20. 

 Suppose a coin is flipped one million times and someone writes down the sequence of results.  If you want to communicate this sequence to another person, assuming it is a fair coin, so that the two possible outcomes, heads and tails, occur with equal probability, and each flip requires 1 bit of information to transmit, transmitting the entire sequence of tosses will require one million bits. 

 Suppose that the coin is biased so that heads occur only 1/4 of the time and tails occur 3/4 of the time.  In this case, the entire sequence can be sent in 811,300 bits, on average.  This implies that each flip of the coin requires just 0.8113 bits to transmit.  But you cannot transmit a coin flip in less than 1 bit, when the only language available is zeros and ones.  However, if the goal is to transmit an entire sequence of flips and the distribution is biased in some way, one's knowledge of the distribution can be used to select a more efficient code, as described below.  A sequence of biased coin flips contains less "information" than a sequence of unbiased flips, so it should take fewer bits to transmit.

<NEXT>

The Information Theory Case Against Evolution: Pages (1), 2, 3, 4, 5, 6, 7

Copyright © 2005 ASUSA