Does Dawkins' "non-directed selection" succeed in explaining the emergence of biological complexity, or does it require the assumption of prior information?

This question places us at the heart of one of the most heated debates in contemporary philosophy of biology: Is natural selection "non-directed" as described by Richard Dawkins sufficient to explain the emergence of the enormous biological complexity we see in living organisms? Or does this selection require "prior information" embedded in the laws of nature or the fundamental structure of the universe?

Inadequate responses to avoid

From some defenders of intelligent design:

"Evolution is just an unproven theory." This is a dangerous conceptual confusion. Evolution as a historical fact (that organisms have changed over time) is supported by overwhelming evidence from paleontology and comparative genomics. The philosophical debate here is not about "did evolution occur?" but about "is the mechanism of non-directed natural selection sufficient to explain everything we see?" Confusing these levels weakens the serious critical position.

"Irreducible complexity refutes evolution." Michael Behe's argument about "irreducible complexity" has been subjected to detailed scientific criticism, and many of the examples he proposed (such as the bacterial flagellum) have been found to have plausible evolutionary explanations. Relying on this argument alone ignores developments in molecular biology.

From some naturalists:

"Natural selection explains everything, the debate is over." This excessive confidence ignores serious philosophical and scientific problems. Even among evolutionary biologists themselves, there is heated debate about the adequacy of neo-Darwinian mechanisms to explain all aspects of evolution. The "Extended Evolutionary Synthesis" theory proposes additional mechanisms such as epigenetics and niche construction.

"The laws of physics and chemistry are sufficient to explain the origin of life and complexity." This reductionism ignores the problem of the origin of biological information. DNA carries complex digital information, and the transition from chemistry to information is not as clear or simple as sometimes portrayed.

Why these responses are inadequate

The responses from both sides share oversimplification. The issue is not "evolution or no evolution," but about the nature and limits of evolutionary mechanisms. And the deeper philosophical question: Is "non-direction" an accurate description of what happens, or is there direction embedded in the structure of natural laws themselves?

Dawkins' argument: Cumulative selection as a blind watchmaker

In his book "The Blind Watchmaker" (1986), Dawkins presents an important distinction between "single-step selection" and "cumulative selection." The former (like a monkey randomly typing and producing a Shakespearean sentence) is statistically impossible. But the latter — where small improvements are preserved and built upon — can generate enormous complexity over time.

His famous example: A computer program that generates the sentence "METHINKS IT IS LIKE A WEASEL" through random mutations and selection for the closest to the target. In a few generations, the program reaches the required sentence, even though the probability of generating it randomly in one go approaches zero.

Strength: Shows the power of cumulative selection in generating organized results from random processes.

First problem: Assuming a prior target

The basic criticism of Dawkins' example: The program "knows" the target beforehand (the Shakespearean sentence) and measures improvement based on it. But biological evolution is supposedly "blind" and has no prior target. This is not a technical detail — it undermines the argument from its foundation. Selection in nature works on immediate survival and reproduction, not on a future "goal" of complexity.

Dawkins responds that the "goal" in nature is survival and reproduction, and this is sufficient. But this response faces a problem: Why does survival and reproduction lead to increasing complexity? Bacteria are among the most successful organisms in survival and reproduction, yet they have remained relatively simple for billions of years.

Second problem: Search space and prior information

William Dembski and Robert Marks, in a series of mathematical papers, developed what are called "No Free Lunch theorems for search." The central idea: Any search algorithm (including evolution) cannot be better than pure random search unless it exploits prior information about the search space.

In the context of evolution: Natural selection succeeds only because "fitness landscapes" in biology are not completely random, but organized in a way that allows for gradual improvement. This organization itself represents "prior information" embedded in the laws of physics and chemistry.

Strength: Shows that the success of evolution depends on a prior structure of reality.

Weakness: Does not necessarily prove "design" in the traditional sense. This structure may be merely a brute fact about the universe.

Third problem: The origin of biological information

Stephen Meyer in "Signature in the Cell" (2009) presents a different challenge: Even the simplest living cell contains an enormous amount of specified digital information. DNA is not merely a chemical molecule — it carries digital instructions for building proteins.

The problem: Natural selection works only on organisms capable of reproduction. But to obtain the first organism capable of reproduction, you need a minimum of informational complexity. Where did this initial information come from? This is the "chicken and egg" problem in the origin of life.

Contemporary naturalist responses

Biologist Andreas Wagner in "Arrival of the Fittest" (2014) presents a different vision: Nature's "libraries" of possible proteins are organized in vast networks, where one can transition from one function to another through multiple pathways. This greatly facilitates the work of evolution.

Sean Carroll and others focus on the role of gene regulation: Most evolution is not about new genes, but about reusing existing genes in new ways. This greatly reduces the "search space" that evolution needs to explore.

Criticizing the criticism: Is prior information really a problem?

Some philosophers (such as Massimo Pigliucci) see the requirement of "no prior information" as unrealistic. Every process in the universe operates within certain laws and constraints. These are not "inputted information" as much as they are the nature of physical reality.

But this response opens a deeper philosophical question: Why are the laws of nature organized in a way that allows for the emergence and evolution of complexity? This returns us to discussions of fine-tuning and the anthropic principle.

Where we stand in this debate today

The debate continues to rage on multiple levels:

Scientific level: Are neo-Darwinian mechanisms sufficient, or do we need an "extended synthesis" that includes additional mechanisms?

Philosophical level: Is "non-direction" an accurate description, or is there teleology embedded in the laws?

Metaphysical level: Does the existence of laws that allow evolution require explanation, or is it a brute fact?

The reasonable position — according to the method of rational weighing (rajḥān ʿaqlī) — is to acknowledge that natural selection is a powerful and proven mechanism, but it operates within a framework of laws and constraints that themselves need explanation. Does this point to "design"? This depends on how one weighs the different evidence — which is exactly what the god-database project seeks to clarify.

For advanced reading

─ Advanced level: No Free Lunch theorems and their role in criticizing evolutionary algorithms
─ Advanced level: Teleology in biology: From Aristotle to complex systems theory
─ "Teleological Argument: Biological Design" page
─ Dawkins, The Blind Watchmaker (1986)
─ Meyer, Signature in the Cell (2009)
─ Wagner, Arrival of the Fittest (2014)
─ Dembski & Marks, "Conservation of Information in Search" (2009)