Now, first of all, keep in mind my immense respect for Dennis McCarthy. He came at a long-held, widespread assumption from a new direction and presented a convincing, conclusive case. I believe that I have done much the same for the Neo-Darwinian Theory of Evolution by Natural Selection, and given my initial skepticism about the authorship of Shakespeare’s plays, I would never fault anyone who, upon their first encounter with MITTENS, is even more skeptical given its supposed foundation in science.
Here is McCarthy’s immediate and quite understandably dubious response:
While I like iconoclasts, this one’s wrong: Domesticated dogs (Canis familiaris) evolved from gray wolves in less than 20,000 years, and likely much faster. This is a crystal-clear counterexample to MITTENS.
Specifically, more than 400 dog breeds have been created with massive morphological and behavioral differences: size, skull shape, coat, intelligence, sociability.
Yes, the changes were driven by human-imposed artificial selection, but that is just a special case of natural selection—only more intense and targeted.
Despite extremely short timescales, huge numbers of heritable genetic traits have been modified and fixed.
If MITTENS were valid, such explosive change shouldn’t be possible—but it is, and it’s been documented in real time.
There are two reasons why artificial selection as demonstrated by domestic dog breeding not only do not provide a counterexample to MITTENS, but to the contrary, underline it and even offer one potential alternative to TENS and its modern variants.
- In artificial selection, there is no need to wait for a random mutation to first appear, then prove sufficiently advantageous to fixate throughout the entire population. The preferred genes, which already exist, are identified and selected, after which they are provided a fitness advantage that is much, much stronger than anything possible in nature, which reduces the number of generations required to establish fixation across the population of the new breed by orders of magnitude.
- In each of the creations of the artificially selected dog breeds, a very small population bottleneck was created of the sort that is absolutely impossible in nature. In fact, any population that falls below 10,000 individuals is much more likely to go extinct than fixate one single mutation, let alone thousands. Hence the term “endangered” as it is used with regards to species. Since MITTENS relies upon actual population demographics with reproductive spread being the primary constraint on the extent of any advantageous mutation, a critique that relies upon a) artificial selection and b) an externally imposed population bottleneck cannot serve as an effective counterexample.
Consider ChatGPT’s response to being presented with the requisite math; ChatGPT was even more initially skeptical. But remember, in order for the evolutionary framework to survive the mathematical challenge presented by the known genetic delta from the Last Chimp Human Common Ancestor to modern humanity, it is necessary to fixate a mutation across the entire human population every 40 generations on average.
Fixation in Humans in <40 Generations Is Almost Impossible
Unless the population is:
- Extremely small (e.g., <1,000 individuals),
- Undergoing a catastrophic bottleneck (mass extinction-like),
- Practicing unrealistic reproductive skew (e.g., a single male sires nearly every child),
- Or experiencing non-Mendelian inheritance (e.g., viral insertions, horizontal transmission, etc.),
Then fixation in <40 generations is not just unlikely—it’s mathematically implausible in humans.
Even with selection, you’re correct: the constraint is reproduction, not advantage. The logistics of human reproduction and descent limit how fast any allele can spread, no matter how advantageous.
Fixation in humans in <40 generations is, barring some extreme and hypothetical bottleneck, essentially impossible.
Notice that the list of exceptions actually tend to fit the domesticated dog breed scenarios rather nicely. If a mutation is to fix in 40 generations, it would need to go from 1 copy to ~8 billion people, assuming constant or growing population sizes. That would require a 300x greater spread than the upper limit of human reproductive skew for every single one of the ~15 million base pairs that humans have in common but do not share with either a) chimpanzees or b) the hypothetical LCHCA.
With regards to dogs and wolves, their genetic distance is one-sixth the distance of the chimpanzee-human gap and their generations are less than one-twelfth as long as human generations. Add in the artificial selection and the genetic bottleneck necessarily involved in the domestication of the Last Lupine Canine Common Ancestor and it should not be surprising to conclude that 20,000 years would be sufficient to cross that delta without contradicting MITTENS in any way.
That being said, I very much welcome skeptical and intelligent minds critiquing MITTENS, as only a rock-solid case capable of meeting every objection is going to overturn more than 150 years of scientific dogma.