The one who states his case first seems right, until the other comes and examines him. — Proverbs 18:17
A Word Before We Begin
The Declared Lens
The invisible lens is the problem. Every researcher, every institution, every author carries assumptions that shape what they see, what they emphasize, and how they interpret ambiguity. The intellectually honest move is not to pretend objectivity — it is to declare the lens so others can account for it.
Here is mine: I am a Christian pastor operating within the just war tradition. I believe the Bible is the authoritative Word of God. I believe the evidence for design in creation is compelling. These convictions inform my perspective, and I declare them openly so you can weigh my analysis accordingly.
This analysis was produced collaboratively with AI research tools (Claude, by Anthropic). The methodology, judgment, and conclusions are mine. The research breadth is AI-assisted. Every claim in this document can be independently verified. I encourage you to do so.
Critically: we apply this framework to our own claims, not only to those we critique. If we demand that secular institutions be honest about what they know and don’t know, we must demonstrate the same honesty.
Proof of Concept: Discarding Our Own Bad Evidence
Young-earth creationists once pointed to the Paluxy River tracks in Texas as evidence that humans walked alongside dinosaurs. The claim circulated for years in creationist literature. Then further investigation revealed the tracks were misidentified — and the Institute for Creation Research itself withdrew the claim.
The Paluxy River tracks are Tier 4. We say so, and we move on. Truth-seekers discard bad evidence. Only narrative-protectors cling to it. If we demand this discipline from secular science, we practice it first.
When Scientific Consensus Was Wrong
The phrase “the science is settled” is used to end debate rather than encourage it. But science, properly understood, is never truly settled. It is a process of continuous inquiry and revision. When any field declares itself immune to questioning, it has ceased to be science and become dogma.
Continental Drift: 50 Years of “Moonshine”
In 1912, Alfred Wegener proposed that continents move. He had compelling evidence: matching coastlines, identical fossils on continents separated by oceans, and matching rock formations across the Atlantic. The scientific establishment rejected him for fifty years. It was not until the 1960s — thirty years after his death — that plate tectonics was finally accepted.
The lesson: Scientific consensus rejected clear evidence for half a century because the theory lacked a mechanism the establishment would accept. The evidence was right. The consensus was wrong. For fifty years.
Thomas Kuhn described this in The Structure of Scientific Revolutions (1962). Science operates within “paradigms” — accepted frameworks. Anomalies that don’t fit are dismissed until they accumulate beyond tolerance. This is not an indictment of science itself. It is an indictment of the human tendency to defend territory rather than follow evidence — a tendency that operates in scientific institutions exactly as it does everywhere else.
The Mathematical Problem Across Disciplines
What Needs to Be Distinguished
Three distinct claims are routinely conflated under “evolution”:
1. Microevolution: Small-scale changes within a population — antibiotic resistance, beak variation, dog breeding. Universally accepted, directly observed, not in dispute.
2. Universal Common Descent: All organisms descended from a single common ancestor. Supported by shared genetic markers, but the mechanism is debated.
3. Unguided Darwinian Evolution: Random mutation and natural selection, without direction, account for all of life’s complexity. This is the claim under examination.
The Steelman: What Evolutionary Biology Actually Has
Intellectual honesty requires presenting the strongest version of positions we may disagree with. Here is what evolutionary biology brings to the table:
Shared genetic markers across species — endogenous retroviruses, pseudogenes, and the human chromosome 2 fusion — provide evidence of common ancestry from multiple independent lines. Evolutionary models have genuine predictive power in medicine (antibiotic resistance prediction, vaccine development) and agriculture. The fossil record, while incomplete, shows documented sequences of change over time in some lineages. Microevolution is not only observed but practically useful.
This evidence is real. We acknowledge it. The question is not whether evolutionary biology has evidence. The question is whether the confidence expressed by institutions matches what the evidence actually supports — and where interpretation has been packaged as fact.
Where Evidence Stops and Narrative Begins
Mainstream evolutionary scientists themselves acknowledge the gaps:
Ernst Mayr (2004): The earliest fossils of Homo are separated from Australopithecus by a large, unbridged gap. Not having any fossils that can serve as missing links, we have to fall back on the construction of a historical narrative. Published by one of the 20th century’s most prominent evolutionary biologists.
Bernard Wood (2023, American Scientist): The existing human fossil record is incomplete in almost all respects, with little chance that any narrative explanation offered today can be the right one. Current peer-reviewed publication.
Note what these scientists say: the evidence requires narrative construction to connect. The fossils exist — that is Tier 1. The stories connecting them are interpretation — Tier 2 at best. Yet textbooks present the narratives with the same confidence as the fossils themselves.
The Probability Calculations: Numbers Anyone Can Check
The observable universe contains approximately 10⁸⁰ atoms. The universe is approximately 13.8 billion years old. The maximum possible random chemical events since the Big Bang: approximately 10⁹⁷. Any event less probable than 1 in 10⁹⁷ exceeds the total number of trials available in cosmic history.
Fred Hoyle (1981): Calculated the probability of obtaining the roughly 2,000 enzymes required for a minimal cell by random chance: one in 10⁴⁰⁶⁰⁰⁰. Hoyle stated: the chance is comparable to a tornado sweeping through a junkyard assembling a Boeing 747.
Douglas Axe (2004, Journal of Molecular Biology): Experimentally measured functional protein sequences at a ratio of approximately 1 in 10⁷⁷. Peer-reviewed laboratory research at the Cambridge Centre for Protein Engineering. Organizational Lens: Axe is now a Senior Fellow at the Discovery Institute, a pro-intelligent-design organization. The 2004 paper was published at Cambridge through standard peer review.
Robert Endres (2025, Imperial College London): Using information theory and algorithmic complexity, found that the entropic and informational barriers to forming a viable protocell are so severe that directed panspermia remains a logically open alternative. A mainstream secular scientist at one of the world’s top research universities, in 2025, finds the math so devastating for unguided abiogenesis that aliens seeding the Earth is a logically open alternative.
The Counter-Arguments — Presented Honestly
Mainstream biologists raise several objections. We present them at their strongest:
Objection 1: Hoyle calculated the wrong thing. Modern abiogenesis theories propose simpler self-replicating molecules that then evolved incrementally.
Objection 2: Many different amino acid sequences can perform the same function, so the probability space is larger than a single-target calculation implies.
Objection 3: Millions of chemical reactions occur simultaneously, dramatically increasing the number of trials.
These are legitimate conceptual objections. They are Tier 2 — reasonable methodological disagreements.
Why the counter-arguments don’t resolve the problem: Not one provides its own probability calculation demonstrating that their proposed mechanism works. The critics say “you calculated the wrong thing” — but they never calculate the right thing and show that the number is feasible. Over 44 years — from Hoyle (1981) to Axe (2004) to Endres (2025) — every time someone does the math, the numbers are devastating. Every time, the response is “you modeled it wrong.” Never: “Here is the math done correctly, and look — it works.”
The claim that “simpler systems solve the problem” lacks a supporting probability calculation.
Fine-Tuning: The Universe’s Settings
The probability problem extends beyond biology. Multiple independent fields of physics and cosmology have identified constants and initial conditions that appear fine-tuned to an extraordinary degree.
Roger Penrose (Oxford, Nobel Prize in Physics 2020): The probability of the universe’s initial low entropy state occurring by chance is 1 in 10^10^123. This number is so large that the digits cannot be written in full even if each of the 10⁸⁰ protons in the observable universe were used to write one digit. Penrose is not a Christian.
Lee Smolin: Estimated the probability of stars forming at 1 in 10²²⁹. The cosmological constant is fine-tuned to one part in 10¹²⁰. Gravity and the weak force must be fine-tuned to one part in 10¹⁰⁰.
Astrophysicist Luke Barnes has documented at least 25 scientists working in cosmology who accept that the universe is fine-tuned for life — including Rees, Susskind, Davies, Penrose, Hawking, and Weinberg.
Every one of these numbers exceeds the total number of events possible in the observable universe since the Big Bang (~10⁹⁷). No single field produced this conclusion. The mathematics from multiple independent disciplines, using different methodologies, all point the same direction. The probability problem is not a gap in one discipline — it is a pattern across all of them.