Assumptions, Circular Reasoning, and a Literal Adam and Eve

by Patricia Babin

Many scientists claim that genetic evidence clearly demonstrates humans are descended from chimps and the original population of humans was much larger than a single man and woman (the biblical Adam and Eve). They base these claims on theoretical models of the evolutionary relationship between humans, chimps, gorillas, orangutans, and rhesus monkeys. In this article, I explain the erroneous assumptions and the process used to support this conclusion.

In a recent Today’s New Reason to Believe, I promised to provide a detailed review of the assumptions made during the process of generating phylogenetic trees. Researchers use these trees to derive the evolutionary relationships between humans and the great apes. The trees are constructed using gene sequences similar among the species being considered. For each gene, a separate phylogenetic tree (called a gene tree) can be constructed using computational methods.

If evolution were true, one would expect the vast majority of gene trees to look the same and to reflect the evolutionary history of each species. Based on this reasoning, evolutionists construct a theoretical “species tree” that they expect the majority of gene trees to match. Gene trees also generate theoretical dates at which each species being analyzed is thought to have diverged from the common ancestor. The divergence times in the gene trees should also be consistent with one another and with the theoretical species tree. This is the theory, but the genomic data does not necessarily behave in the way evolutionists expect it to behave.

Scientists must explain why the data is the way it is because it’s often inconsistent with the theory. For example, in “Mapping Human Genetic Ancestry,”¹ the authors acknowledge that the genomic data reveals that the human genome looks more like a mosaic rather than showing the clear, direct descendent from chimps, as expected.

Here we will review the data analysis process used by the authors to explain why the data doesn’t show the expected results and we’ll observe the assumptions and the process used to deal with contrarian data.  (The assumptions are underlined so that you can identify them clearly.) We’ll see that the authors conclude evolution is true despite the fact that the majority of the data doesn’t fit the evolutionary hypothesis.

Step 1. Assume evolution is true. Evolution has to be assumed as the very first step in the process of dealing with the genomic data because the authors will be using phylogenetic trees as their means of analyzing the genomic data. Barry G. Hall, author of Phylogenetic Trees Made Easy,explains that building a phylogenetic tree from genomic data is valid only if the sequence similarity present in the genes under consideration is due to shared evolutionary ancestry, a condition called homology.² If the sequence similarity in the genes is due to common function or common design, then it is invalid to assume that building a phylogenetic tree from the sequences will give you real information.

Starting with the assumption that evolution is true, the authors of the paper identified 26,909 high-quality gene sequences that could be used to build gene trees for humans, chimps, gorillas, orangutans, and rhesus monkeys.

Step 2. Align the sequences of each of the 26,909 genes and generate a gene tree for each of them to yield 26,909 gene trees.

Step 3. Assume that rhesus monkeys were the first of the five species to emerge in the species tree (that is, assume that rhesus monkey is a valid outgroup for building a phylogenetic tree with human and great ape gene sequences) and eliminate any gene trees that don’t show rhesus monkeys as the oldest species. This assumption results in the elimination of 1,409 (five percent) of the gene trees from the analysis. 25,500 gene trees now remain.

Step 4. Assume that genes acquire mutations in a clock-like fashion (molecular clock theory). This assumption is the basis for the belief that one can use the changes in gene sequences from one species to another to calculate the time they diverged from one another. Gene trees that generate divergence times inconsistent with the theoretical species tree are discarded because the authors conclude they aren’t behaving in a clock-like fashion. This process eliminates 2,190 (eight percent) of the gene trees. 23,310 trees now remain in the analysis.

Step 5. Assume that the theoretical species tree is correct and eliminate gene trees that differ drastically from the theoretical tree. This assumption eliminates 11,365 (42 percent) of gene trees, leaving 11,945 trees. The scientists identify the remaining sequences as “phylogenetically informative.” This phrase simply means that the remaining gene trees give them more or less the story of the evolutionary relationship they expected to find. In the end, the researchers have rejected ~56 percent of the initial data because, in one way or another, they don’t match the hypothesized species tree.

Step 6. The researchers can now tackle the paper’s real objective—explaining the inconsistency between so many of the gene trees and the theoretical species tree. Armed with the set of gene trees considered valid and making two additional assumptions—that orangutans emerged ~16 million years ago and that the generation time for the species is 20 years—the scientists calculate the probability that gene trees that do not agree with the species tree would turn up in their analysis.

Fortunately for their research, the probability is exactly what they observed occurring in their data. This calculation serves as a justification for upholding the evolutionary paradigm in spite of so much data being eliminated from consideration because it isn’t consistent with the hypothesis. The authors further conclude that the original population of humans was between 24,000 and 49,000. So, having started with the assumption that evolution is true (see step 1), they conclude that evolution is true and that a literal Adam and Eve cannot be true. Can anyone say “circular reasoning”?