Every child born into the world arrives with roughly 70 new mutations that neither parent carried. That number comes not from creationist estimates but from direct genome sequencing of parents and offspring, published in mainstream genetics literature. It is one of the most consistently replicated findings in modern genomics.
The question is not whether mutations happen. Everyone agrees they do. The real question—and the one that cuts to the heart of the creation-evolution debate—is what mutations actually accomplish over time. Can they build the complex biological systems we see in living organisms? Or do they primarily degrade what already exists?
How you answer that question shapes everything about how you understand biological history.
What Mutations Are (and What They Do)
A mutation is simply a change in the DNA sequence. It can be as small as a single nucleotide swap—an A where a G should be—or as large as the duplication, deletion, or rearrangement of entire chromosome segments. Mutations arise from copying errors during cell division, from chemical damage to DNA, or from radiation exposure. Every organism has repair machinery that catches most of these errors, but no repair system is perfect.
Geneticists classify mutations by their effect on the organism. Deleterious mutations reduce fitness—they damage a protein, disrupt a regulatory sequence, or break a metabolic pathway. Beneficial mutations improve fitness in a given environment. Neutral mutations have no detectable effect. There is broad agreement among geneticists of all persuasions that the vast majority of mutations with any functional impact are deleterious. The ratio is not even close.
Michael Lynch, a prominent evolutionary geneticist at Indiana University, estimated in a 2010 PNAS paper that the human genome sustains a deleterious mutation rate of approximately 2.2 per diploid genome per generation. That means each of us carries roughly two new harmful mutations our parents did not. And that is a conservative figure—some estimates place the genomic deleterious rate at three or higher.
This matters enormously for any theory of biological origins that depends on mutations to build new complexity.
The Evolutionary Framework: Mutations as Raw Material
In the standard evolutionary model, mutations provide the raw material on which natural selection acts. The process works something like this: random mutations generate variation, natural selection preserves the rare beneficial ones, and over millions of years, organisms gradually increase in complexity and adapt to new environments.
This is an elegant idea, and it has genuine explanatory power for certain kinds of biological change. Bacteria developing antibiotic resistance is a real, observed phenomenon. Insect populations shifting color patterns in response to environmental changes—also real. Nobody disputes that mutations can produce adaptive changes in populations.
But there is a significant gap between observing that mutations can help an organism survive a specific challenge and claiming that mutations can build fundamentally new biological systems—new organs, new body plans, new biochemical pathways. The former is well-documented. The latter is extrapolated.
Even beneficial mutations, when examined closely, often turn out to involve trade-offs. A bacterium that gains antibiotic resistance frequently does so by losing or degrading an existing system—a transport pump, a ribosomal binding site, an enzymatic function. The organism survives the immediate threat, but at a cost. Kevin Anderson’s analysis of beneficial mutations in bacteria, presented at the International Conference on Creationism, documented this pattern across multiple well-known examples. Resistance was typically gained through loss of pre-existing function, not through the construction of anything new.
Genetic Entropy: The Case That Mutations Degrade
Cornell geneticist John Sanford—a plant geneticist who holds over 30 patents and co-invented the gene gun used in agricultural biotechnology—made a provocative argument in his 2005 book Genetic Entropy and the Mystery of the Genome. His claim: the human genome is not improving over time. It is deteriorating.
Sanford’s argument rests on a well-known problem in population genetics. Most deleterious mutations are only very slightly harmful—so slight that natural selection cannot effectively remove them from the population. Geneticists call these “nearly neutral” mutations, and their existence is not controversial. Even Lynch acknowledges them. The question is what happens when they accumulate generation after generation.
Sanford argues that the answer is genetic entropy—a slow, steady degradation of genomic information that selection is powerless to stop. If the vast majority of mutations are harmful (even slightly), and if selection cannot efficiently remove the mildly harmful ones, then the genome is on a one-way trajectory toward decay. This is not a fringe idea in genetics; it is an extension of well-known theoretical work by Motoo Kimura, the founder of neutral theory, and by Alexey Kondrashov, who famously asked why humans have not “died 100 times over” given the rate of deleterious mutations.
To test this, Sanford and colleagues developed a forward-time population genetics simulator called Mendel’s Accountant. Unlike most population genetics models, which work backward from observed data, Mendel’s Accountant tracks individual mutations forward through generations, modeling realistic parameters like population size, reproduction rates, and selection coefficients. The results consistently showed that realistic populations accumulate harmful mutations faster than selection can remove them.
A Real-World Test Case
Theory is one thing. Data is another. Carter and Sanford found a striking test case in the H1N1 influenza virus.
In a 2012 paper published in Theoretical Biology and Medical Modelling, they analyzed over 4,100 fully sequenced H1N1 genomes spanning from 1918 to the present. What they found was a steady, linear accumulation of mutations over time—more than 1,400 point mutations by 2009, representing over 10% of the genome. The mutations accumulated at a constant rate across the entire genome, and the virus showed progressive erosion of codon optimization.
The virus did not become more fit over time. It became less virulent. It experienced multiple extinction events. The human H1N1 lineage appears to have gone functionally extinct in 2009, replaced by a reassortant strain from swine. Carter and Sanford argued this was a real-time observation of genetic entropy in action—a genome degrading under the weight of accumulating mutations faster than selection could preserve functional sequences.
Critics have pushed back on this interpretation, noting that influenza viruses face strong immune selection pressures and that antigenic drift (mutations that help the virus evade host immunity) may look like degradation from a genomic perspective while serving an adaptive purpose from the virus’s perspective. This is a fair point, and it illustrates why the question of what mutations accomplish depends heavily on how you define “beneficial.”
What Mainstream Science Says
Evolutionary geneticists do not deny the problem of deleterious mutation accumulation. Lynch himself has written extensively about it. The standard response involves several proposed mechanisms.
Synergistic epistasis is one: the idea that mildly harmful mutations become more harmful in combination than they would individually, making selection more efficient at removing them in bunches. This could theoretically prevent mutational meltdown, but the empirical evidence for strong synergistic epistasis in humans remains limited.
Another proposed solution is sexual recombination—the shuffling of chromosomes during reproduction. Recombination can concentrate deleterious mutations onto some chromosomes (which are then selected against) while freeing others. This is a genuine advantage of sexual reproduction and likely does slow the accumulation of harmful mutations. Whether it is sufficient to halt genetic entropy entirely is debated even among evolutionary theorists.
A third response is that slightly deleterious mutations may be effectively neutral in large populations, and that population size dynamics play a more complex role than simple models suggest. This is Lynch’s own area of expertise, and his drift-barrier hypothesis provides a sophisticated framework for thinking about mutation rate evolution. But it does not solve the fundamental problem of accumulation—it reframes it.
Kondrashov’s question remains unresolved in mainstream literature. The deleterious mutation rate in humans is high enough that it poses real theoretical difficulties for any model that requires long-term genomic stability.
How This Fits a Creation Framework
From a creation perspective, genetic entropy is not surprising—it is expected. If the original created genomes were designed to be functional and information-rich, then mutations represent a departure from that designed state. The trajectory is downhill, not uphill.
This does not mean organisms cannot adapt. Creation scientists fully acknowledge that mutations, including beneficial ones, play a role in adaptation within created kinds. Bacteria develop antibiotic resistance. Populations shift allele frequencies in response to environmental pressures. Speciation occurs. These are observed biological realities, and the creation model has room for all of them.
What the creation model questions is whether these adaptive changes represent the same kind of process that could build fundamentally new genetic information—new genes, new regulatory networks, new developmental pathways. The observed pattern is that adaptation through mutation typically involves modification, degradation, or reshuffling of existing information. It does not generate the kind of specified complexity that characterizes the functional architecture of genomes.
Nathaniel Jeanson’s work on Y-chromosome mutation rates has added another dimension to this discussion. By measuring pedigree-based mutation rates directly—counting mutations between fathers and sons—Jeanson has argued that the observed rate of Y-chromosome diversification is consistent with a human population originating thousands, not hundreds of thousands, of years ago. This work has been published in the Answers Research Journal and presented at the International Conference on Creationism, and while it remains contested, it represents an active area of creation genetics research.
Challenges and Open Questions
The genetic entropy argument is not without its difficulties, and honest engagement requires acknowledging them.
First, critics point out that if genetic entropy were as severe as Sanford suggests, we should see obvious fitness declines in human populations over historical time. We do see increasing rates of genetic disease and cancer, but overall human lifespan has increased dramatically (though this is largely due to medicine and nutrition rather than genetic improvement). Distinguishing between environmental masking and genuine genetic trajectory is genuinely hard.
Second, the Mendel’s Accountant simulations, while sophisticated, have been criticized for their parameter choices. Mainstream population geneticists argue that the distribution of mutational fitness effects used in the model may overweight nearly neutral mutations and underweight the efficiency of selection against moderately harmful ones. These are legitimate methodological questions that deserve further investigation.
Third, the relationship between genomic mutation accumulation and actual organismal fitness is complex. Not all genomic degradation translates directly to reduced survival or reproduction, especially when organisms have redundant systems and regulatory buffers. The genome may be more robust to individual mutations than a simple count would suggest—though this robustness itself requires explanation.
Fourth, creation scientists still need to develop more detailed models of how rapid diversification within created kinds occurred after the Flood. If mutations are primarily degrading, what drove the rapid speciation that the creation model requires? Some researchers point to pre-designed genetic variability, epigenetic mechanisms, and transposable element activity, but this area needs significantly more work.
Where the Evidence Points
The mutation question cuts to the foundation of biological origins. If mutations are primarily a creative force—capable of generating novel complexity over time—then the evolutionary narrative has its engine. If mutations are primarily a degrading force—eroding designed information faster than selection can preserve it—then the creation framework better fits the data.
The evidence from direct genome sequencing, from the observed ratio of harmful to beneficial mutations, from the theoretical problem of nearly neutral mutation accumulation, and from real-world case studies like H1N1 all point in the same direction: mutations overwhelmingly damage what already exists. The rare beneficial exceptions, while real and important, tend to involve functional trade-offs rather than genuine innovation.
This is an area where creation science has both genuine strengths and genuine work still to do. The strengths are real: Sanford’s genetic entropy argument draws on mainstream data and well-established theoretical concerns. The work still needed is also real: creation geneticists must continue developing testable predictions, engaging with mainstream critiques, and building models that explain both degradation and diversification within the biblical timeframe.
The conversation is far from over. But the direction of the evidence is worth taking seriously.
Support Creation Research
Understanding what mutations actually do—and what they cannot do—is one of the most important questions in origins science. Research into genetic entropy, mutation rates, and genomic change over time requires funding, computational resources, and trained scientists willing to follow the data wherever it leads.
If you want to support the kind of rigorous, honest creation research that tackles these hard questions, consider giving to one of our active projects.
