FOR IMMEDIATE RELEASE                           JON ROLAND 512/374-9585

Texas Textbook Testimony

Austin, Texas, July 9, 2003 — Jon Roland, President of the Constitution Society, testified before the Texas State Board of Education that all of the science textbooks being considered in this round of hearings should be rejected for failure to meet the Texas Essential Knowledge and Skills (TEKS) requirements:

112.43. Biology.

(b)(4) ... certain types of questions can be answered by investigations, and that methods, models, and conclusions built from these investigations change as new observations are made.

(c)(3)(A) analyze, review, and critique scientific explanations, including hypotheses and theories, as to their strengths and weaknesses using scientific evidence and information;

Roland reported that none of the textbooks proposed adequately discuss scientific methods and tools. They persist in using older, imprecise language from before the advent of the more precise language that has emerged from modern physics and computer science. Examples include:

  • The use of terms like "conjecture", "hypothesis", "theory", and "fact" as though they were different things, when they are actually all models in which we may have different levels of confidence.
  • The use of terms like "observation" to mean both the operation and the result of the operation, when the two meanings need to be differentiated, and the concept of well-defined "operation" more fully developed as the way to define observational results.
  • The use of terms like "proof" which has quite different meanings in mathematics, the sciences, and law, without explaining the differences; or "truth", which is a matter of philosophical debate but a distraction for scientific investigation, in which utility is the standard.
  • Failure to adequately discuss how to choose among competing models, on the basis of such utility factors as how well they allow us explain, predict, or control events, how much they cost to use, and how well they support making certain kinds of decisions; and selection criteria such as unity, consistency, refutability, and parsimony.
  • Failure to introduce students to the tools commonly used in many sciences, which they will have to learn in the first couple of years of college, many of which are now available as free or low-cost computer software that they can learn to use competently without first having to master the mathematics involved, with such use motivating them to learn the underlying mathematical techniques.

Roland explained that it is impractical for the textbooks to discuss all the strengths and weaknesses of every scientific model in the body of textbooks, because it would make them too long and cause important material to be omitted, but that a solution would be for the textbooks to provide endnotes that cite links to web pages where the controversies concerning the models and evidence are more extensively discussed.

As an example of how treatment of scientific method is inadequate, he cited the persistent use of the phrase "theory of evolution" as though there were a single model of evolution that could be refuted by a single counterexample, when it is more correct to treat it as an analytic approach to the development of an array of models relating pairs of specimens, which may be arranged as a directed graph, or as a descent tree. He explained that the proposition that there exists only one descent tree for all life on Earth is a doctrine or conjecture, because it overreaches the current state of the evidence, and that it is scientifically appropriate to contemplate contamination models for possible specimens that are not easily fitted into the descent tree.

Roland's written testimony of July 9, 2002, is at . His oral testimony and that of others, and the publishers' responses, will be at

Contents | Text Version