Introduction to Evolutionary Informatics

Robert J Marks II, William A Dembski;Winston Ewert;;

• Publisher: WSPC
• Year: 2017
• eBook ISBN: 9789813142169
• Topic: Computer Science
• Subtopic: Computer Science General
• Index: Computer Science

-->

Science has made great strides in modeling space, time, mass and energy. Yet little attention has been paid to the precise representation of the information ubiquitous in nature.

Introduction to Evolutionary Informatics fuses results from complexity modeling and information theory that allow both meaning and design difficulty in nature to be measured in bits. Built on the foundation of a series of peer-reviewed papers published by the authors, the book is written at a level easily understandable to readers with knowledge of rudimentary high school math. Those seeking a quick first read or those not interested in mathematical detail can skip marked sections in the monograph and still experience the impact of this new and exciting model of nature's information.

This book is written for enthusiasts in science, engineering and mathematics interested in understanding the essential role of information in closely examined evolution theory.

--> -->

1

INTRODUCTION

“The honor of mathematics requires us to come up with a mathematical theory of evolution and either prove that Darwin was wrong or right!”

Gregory Chaitin¹

In order to establish solid credibility, a science should be backed by mathematics and models. Even some soft sciences, such as finance, offer compelling mathematical and computer models that win Nobel prizes. The purpose of evolutionary informatics is to scrutinize the mathematics and models underlying evolution and the science of design.

There is a recognized difference between models and reality. A mantra popular with engineers is: “In theory, theory and reality are the same. In reality they are not.” Models in physics have been shown to display incredible experimental agreement with theory. But what of Darwinian evolution? There have been numerous models proposed for Darwinian evolution. Some are examined in this monograph. Each, however, is intelligently designed and the degree to which they are designed can be measured, in bits, using active information. If these models do indeed capture the Darwinian process, then we must conclude that evolution is guided by an intelligence. Without the application of this intelligence, the evolutionary models simply do not work. The computational resources of our universe and, indeed, the current model of the multiverse proposed by string theory are insufficient to allow the small probabilities of evolution by pure chance. The participation of a designer is mandatory.

Our work was initially motivated by attempts of others to describe Darwinian evolution by computer simulation or mathematical models.² The authors of these papers purport that their work relates to biological evolution. We show repeatedly that the proposed models all require inclusion of significant knowledge about the problem being solved. If a goal of a model is specified in advance, that’s not Darwinian evolution: it’s intelligent design. So ironically, these models of evolution purported to demonstrate Darwinian evolution necessitate an intelligent designer. The programmer’s contribution to success, dubbed active information, is measured in bits.

Mount Rushmore’s carved busts of United States presidents indicate design when compared to, say, Mount Fuji. The Search for Extraterrestrial Intelligence (SETI) assumes that signals received from space containing intelligence can be detected. A model to measure meaningful information from observations is the topic of Chapter 7.

1.1 The Queen of Scientists & Engineers

Engineers don’t brag enough. Scientists did not put man on the moon. Engineers did. Scientists are not responsible for the Internet. Engineers are. The latest medical breakthrough is most likely the work of an engineer, not a scientist or a medical doctor. And from whose viewpoint is it better to address intelligent design? Engineers design things.

The engineer’s job is to understand science and mathematics, apply this understanding to reality, and make things work.

There are fundamental philosophical differences between engineers and scientists. Scientists are generally more interested in simply understanding nature. They formulate models, often beautiful and powerful models, and scrutinize them. Once vetted by the acceptance of most, the models are placed on a throne like a queen where they are worshiped. It often takes a major coup to overthrow a scientist’s ensconced dogma. Engineers, on the other hand, make the queen come down from the throne and scrub the floor. If she works, we use her talents. And if she doesn’t work, we fire her.

The story of the queen describes this monograph. We analyze the computer models of evolution offered by scientists and conclude they work only because the programmers designed them to work. There is no creation of information or spontaneous increase in meaningful complexity. The law of conservation of information precludes it. We are able to examine the proposed computer models, identify the source of active information, and show that the evolutionary process, although successful, is a poor way to use available resources. Since the proposed models do not display characteristics of undirected Darwinian search, the reigning queen of undirected Darwinian evolution must be given a pink slip.

1.2 Science and Models

Science requires explanative models. Darwinian evolution, using the repeated processes of mutation and survival of the fittest, looks on the surface to be a science well positioned for modeling using probabilistically based simulation.

Repeatedly observable laws, such as Newton’s law of motion or the laws of thermodynamics, can be confirmed by repeating experiments again and again. Such laws are said to be formed by the application of inductive inference. Non-repeatable phenomena cannot be modeled this way. The theory of the creation of the universe from the Big Bang is an example. In such cases, abductive inference or inference to the best explanation is used to establish laws. Abductive inference has certainly not been a hindrance in forming a rich theoretical explanation of the Big Bang or the science of geology.

The entirety of Darwinian evolution theory over eons of life on earth cannot be repeated in the laboratory. We have, though, some supportive repeatable science to help. Dogs and horses can be bred, bacteria strains lose their vulnerability to antibiotics and the beaks of finches vary in accordance with food sources on the Galapagos Islands. Cannot we extrapolate a viable model of evolution from these phenomena? Those who support Darwinian evolution say yes. Mathematically, though, extrapolation models of temporal processes can be useless. Small perturbations in observations can result in wildly varying extrapolation results.^a,3 Chapter 6 contains a discussion of published models whose proponents feel they have a successful model of Darwinian evolution. They have not. At best, they have guided the goal-seeking breeding of a thoroughbred horse from available stock.

1.2.1 Computer models

The invention of the computer in the mid-20th century gave rise to expectations in the science of evolution. It was hoped the evolutionary process could, for the first time, be modeled and demonstrated by a computer program. Evolutionary computation was founded on the assumption that, unlike glacially slow biological wetware, the speed of a computer would allow sufficient generations to conclusively demonstrate Darwinian evolution. In 1962, Nils Barricelli wrote⁴

“The Darwinian idea that evolution takes place by random hereditary changes and selection has from the beginning been handicapped by the fact that no proper test has been found to decide whether such evolution was possible and how it would develop under controlled conditions.”

In the mid-1960s J. L. Crosby⁵ looked to the computer of the future as a remedy for this condition.

“In general, it is usually impossible or impracticable to test hypotheses about evolution in a particular species by the deliberate setting up of controlled experiments with living organisms of that species. We can attempt to partially to get around this difficulty by constructing [computer] models representing the evolutionary system we wish to study, and use these to test at least the theoretical validity of our ideas.”

1.2.2 The improbable and the impossible

Contrary to expectation, computer science research has revealed numerous problems for a model of evolution without an intelligent designer. The principle of conservation of information shows that evolutionary processes on average are incapable of generating information. Rather, they are restricted to extracting information from a source of knowledge. The success of any evolutionary process is not due to any magic in the process itself, but rather to the creative knowledge available to that process. Computer simulation of evolution has demonstrated that information sources are created by programmers exploiting their knowledge of problem spaces, a process with no analog in a non-teleological world.

Evolutionary models are stochastic, so one might argue “Sure, it’s not probable. But it’s possible!” This is right in the sense that all probable things are possible but not all possible things are probable or, in the contrapositive sense, everything impossible is improbable but improbable events need not be impossible. But, like many contrasts, there comes a point where the improbable and impossible blur together and, within the resources of our finite universe (or even the hypothesized multiverse), an event can be so improbable as to be accurately labeled as impossible. This proposition is commonly referred to as Borel’s Law.⁶ When I stand, is it possible part of my foot will experience quantum tunneling through the floor? Yes. But the event is so improbable that I can stand and sit every picosecond since the creation of the universe and my toes will never experience quantum tunneling. We argue that this technically possible event is, indeed, impossible in practice. Here’s another example. Suppose I randomly choose a billion atoms in the known universe and, without consulting me, you choose a billion. In the strictest of senses, it is possible that the billion atoms you choose are the same as mine. But the probability of matching atoms is so small we could both choose atoms over and over for trillions of years and there would be no chance our billion atoms would exactly match. A successful matching is impossible with the probability resources available in our universe—or even the largest multiverse predicted by string theorists.

Could the biology we observe today have been created by undirected Darwinian evolution? There may be a minuscule probability but, like the examples of quantum tunneling and atom choosing, the development is impossible. Evolutionary informatics shows the observed universe (or a multiverse) is not large enough nor old enough to allow it.

Notes

1. G.J. Chaitin, Proving Darwin: Making Biology Mathematical (Pantheon, 2012).

2. H.S. Wilf and W.J. Ewens, “There’s plenty of time for evolution.” P Natl Acad Sci, 107, pp. 22454–22456 (2010).

R.E. Lenski, C. Ofria, R.T. Pennock and C. Adami, “The evolutionary origin of complex features.” Nature, 423, pp. 139–144 (2003).

T.D. Schneider, “Evolution of biological information.” Nucleic Acids Res, 28, pp. 2794–2799 (2000).

R. Dawkins, The Blind Watchmaker: Why the Evidence of Evolution Reveals a Universe Without Design (Norton, New York, 1996).

D. Thomas, “War of the Weasels: An evolutionary algorithm beats intelligent design.” Skeptical Inquirer, 43, pp. 42–46 (2010).

G.J. Chaitin, Proving Darwin: Making Biology Mathematical (Pantheon, 2012).

3. R.J. Marks II, Handbook of Fourier Analysis and its Applications (Oxford University Press, 2008).

R.J. Marks II, “Gerchberg’s extrapolation algorithm in two dimensions.” Appl Opt, 20, pp. 1815–1820 (1981).

D.K. Smith and R.J. Marks II, “Closed form bandlimited image extrapolation.” Appl Opt, 20, pp. 2476–2483 (1981).

R.J. Marks II, “Posedness of a bandlimited image extension problem in tomography.” Opt Lett, 7, pp. 376–377 (1982).

D. Kaplan and R.J. Marks II, “Noise sensitivity of interpolation and extrapolation matrices.” Appl Opt, 21, pp. 4489–4492 (1982).

R.J. Marks II, “Restoration of continuously sampled bandlimite...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Contents
• Preface
• About the Authors
• 1. Introduction
• 2. Information: What Is It?
• 3. Design Search in Evolution and the Requirement of Intelligence
• 4. Determinism in Randomness
• 5. Conservation of Information in Computer Search
• 6. Analysis of Some Biologically Motivated Evolutionary Models
• 7. Measuring Meaning: Algorithmic Specified Complexity
• 8. Intelligent Design & Artificial Intelligence
• 9. Appendices
• Index