They believe that the parent group, which must also be the oldest group, can be identified by computer programs that sort the population groups into branching tree patterns. Out of the many statistical trees that can be generated, the shortest one, the one with the least number of branchings, is called the “maximum parsimony tree,” and researchers believe it to be identical to the actual historical relationships of the various population groups in the tree. The branch (“clade”) forming the base of the tree (the “basal clade”) is supposed to be the parent group. According to evolutionary theory, it should, in addition to being at the base of the tree, have the most variation (i.e. the most mutations) in its mitochondrial DNA, relative to the other population groups. So in this way, researchers believe they can find where and when the root population existed. But some scientists say that the clock is not very accurate and that the genetic information contained in the mitochondrial DNA in today’s populations is not sufficient to tell us with certainty the geographical location of the first human population.
In one of the original African Eve reports (Cann et al. 1987), researchers analyzed the mitochondrial DNA from groups of modern humans from different regions throughout the world. They analyzed the sequence of nucleotide bases found in a particular section of the mitochondrial DNA in all of the individuals being studied. They then used a computer program to arrange the various kinds of mitochondrial DNA sequences (called haplotypes) into a tree. According to the report, the root (or basal clade) of the maximum parismony tree of haplotypes was the African group. But Templeton (1993, p. 52) pointed out that Maddison (1991) had rerun the data and found ten thousand trees that were shorter (i.e. more parsimonious) than the “maximum parsimony tree” reported by the African Eve researchers. Many of these trees had mixed African/Asian roots. Analyzing another “African Eve” report (Vigilant et al. 1991), Templeton (1992) found 1,000 trees two steps shorter than the one put forward by those researchers, who had claimed it was a “maximum parsimony” tree.All of the thousand more parsimonious trees found by Templeton in his 1992 study had non-African basal clades (Templeton 1993, p. 53). This would be consistent with accounts found in the ancient Sanskrit writings of India, which would place the original human populations on this planet in the region between the Himalayas and the Caspian Sea.
Why such different results? Templeton (1993, p. 52), considering another African Eve report, explained: “Computer programs . . . cannot guarantee that the maximum parsimony tree will be found when dealing with such large data sets as these because the state space is too large to search exhaustively. For example, for the 147 haplotypes in Stoneking, Bhatia, and Wilson (1986), there are 1.68 x 10294 possible trees. Finding the maximum parsimony set among these many possibilities is nontrivial.” The computer programs tend to pick out a tree that is maximally parsimonious only in relation to a subset of the total number of possible trees. Which subset of trees that is selected depends on the order in which data are fed into the computer. To guard against this problem, it is necessary to randomize the sequence in which the data are entered over a series of runs. When one has done this a sufficient number of times, so as to find the maximum parsimony trees for various local subsets of the data, then one can compare these trees and arrive at a conclusion. This was not done in the original African Eve studies (the computer program was run only once), and thus the conclusions are not reliable. Also, even data randomization techniques do not completely solve the problem (Templeton
1993, p. 53). So this means that it really is not possible to conclusively determine the common geographical origin of dispersed human populations from the genetic data available today.
In addition to presenting inaccurate conclusions about maximum parsimony trees with African basal clades, the African Eve researchers (Cann et al. 1987; Vigilant et al. 1991) also made misleading statements about the level of mitochondrial DNA diversity in various populations. The African Eve researchers assumed that mutations occur at some fixed rate, and therefore the population with the most internal diversity, relative to the others, should be the oldest. Because the African populations had a higher level of internal diversity than Asian and European populations, the researchers claimed that the African populations were the oldest. But Templeton (1993, p. 56) noted that “no statistical test is presented in either paper in support of this claim.” He pointed out that when proper statistical methods are applied, there is no significant degree of diversity in the mitochondrial DNA of Africans, Europeans, and Asians (Templeton 1993, p. 57). As Templeton himself put it: “The apparent greater diversity of Africans is an artifact of not using sufficient statistics for making inference about the . . .process that led to the present-day human populations. In summary, the evidence for geographical origin is ambiguous. . . . there is no statistically significant support for an African origin with any mtDNA data” (Templeton 1993, p. 57).
Now let us consider the ages for the antiquity of anatomically modern humans proposed by the original African Eve theorists. They tried to calculate the time it took for the observed mtDNA diversity in today’s human populations to accumulate, based on rates of mutation. This time is called “the time to coalescence,” the time at which all the mtDNA sequence diversity in present human populations coalesces into a single past mtDNA sequence, the source of the present diversity. One group of researchers (Stoneking et al. 1986) got an age of 200,000 years for Eve, within a range of 140,000 to 290,000 years, using intraspecific calculations for the molecular clock. Intraspecific means that they based calculations on rates of mutations in human populations only. Another group (Vigilant et al. 1991), using interspecific calculations, also got an age of 200,000 years for Eve, but with a range of 166,000 to 249,000 years. Interspecific means they based their calculations on assumptions about the time at which the human line separated from the chimpanzee line.
First, let us consider the report from the researchers who relied on interspecific calibration of the rate of mutation (Vigilant et al. 1991). Their calibration of the mutation rate was made using either 4 million or 6 million years as the time since the human line supposedly diverged from the chimpanzee line. These times of divergence, when used in calculations that take into account statistical uncertainty, give times of coalescence for human mtDNA of 170,000 and 256,000 years respectively (Templeton
1993, p. 58). But Gingerich (1985) estimated that the divergence between humans and chimps took place 9.2 million years ago. A rate of change based on this date, would greatly increase the time to coalescence for modern mtDNA diversity, making it as much as 554,000 years (Templeton
1993, pp. 58–59). Furthermore, Lovejoy and his coworkers (1993) pointed out that Vigilant et al. (1991) made a mathematical error (they used the wrong transition-transversion), which when corrected gives an age for Eve of at least 1.3 million years (Frayer et al. 1993, p. 40).
It is easy to see that this whole “molecular clock” business is extremely unreliable, because it is based on speculative evolutionary assumptions. It is not at all certain that humans and chimps had a common ancestor of the kind proposed by Darwinian evolutionists. And, as we have seen, even if we assume that chimps and humans did have a common ancestor, the time at which they diverged from that common ancestor is not known with certainty, thus leading to widely varying calibrations of mutation rates and widely varying age estimates for the time to coalescence of modern mitochondrial DNA diversity.