No scientific support for Bible literalist's Eve

Carl Wieland replied to my critique of his Mitochondrial Eve article: here is my response

Alec MacAndrew

Introduction

In 1998, Dr Carl Wieland published an article on the creationist web site 'Answers in Genesis' claiming that the Miitochondrial Eve concept supported a literal biblical Eve dated at 6,500 years ago: http://www.answersingenesis.org/docs/4055.asp (1).

In 2004, I published a critique of Wieland's and other creationists' claims for Mitochondrial Eve here. (2). The gist of this critique was that creationists had prematurely latched onto scientific data that appeared to support a date for Mitochondrial Eve 6,500 year ago, and had not updated their position with more recent analysis that was consistent with a date for Mitochondrial Eve 175,000 years ago.

I have just noticed (it is October 2006 as I write) that Dr Wieland replied to my article in 2005 in a creationist propaganda publication called 'Journal of Creation' (3). His article was posted on the web in July 2006 and can be found here: http://www.creationontheweb.com/content/view/4472/ (4)

Here is my reply to Wieland. Dr Wieland is one of the more careful proponents of creationism. He tried in his original article, genuinely, I think, to represent the Mitochondrial Eve concept as accurately as he was able. Although he made, as we shall see, at least one major fundamental error, I respect him for trying. Would that he had reciprocated this respect - since he has chosen to characterise me as someone writing for 'effect' rather than to present a serious case, readers will understand that I will not now pull my punches.

His knowledge of the science is inadequate to the task he sets himself. In particular, his ignorance of the basic principle on which the Mitochondrial Eve concept is based leads him into a forest of error, where he loses himself and the argument amidst thickets of misunderstanding and misinterpretation.

His position rests, as he agrees, on a presupposition that what is recorded in an ancient Middle Eastern document of doubtful provenance is literally and entirely true, and he welcomes evidence that supports that view while discounting evidence that does not, regardless of the relative strength of the evidence for and against. In other words he has no interest in following a truly scientific programme in which conclusions follow wherever the evidence leads, because, in his view, his ancient text already determines what is true, and therefore, for him, evidence is useful only insofar as it feeds his prejudice (and he has the gall to accuse me of special pleading!). I utterly deplore this method for getting an understanding of the world, but at least he is honest about it unlike many other creationists who pretend that the oxymoronic 'creation scientists' conduct real science. One problem for Wieland in his approach to 'reasoning' is that as new and inconvenient evidence comes to light, findings that seemed at first to be splendid support for his position become discredited and tattered things, which he must continue to hug to his breast. This is how it is for creationism generally, and it is the case here.

I do not plan to repeat descriptions of mitochondrial genetics or definitions of the Mitochondrial Eve concept, or to reprise the history of the scientific findings and the associated creationist claims. I refer readers to my original article here (2) and to the key scientific papers (5),(6), (7).

Wieland's position

Wieland makes several  claims in his 2005/2006 reply (some of which are echoes of his 1998 article) which I summarise below:

 I intend to address these arguments in turn.

There is absolutely no doubt that Wieland is suffering under the basic misunderstanding that the data obtained from sampling human mitochondrial sequences leads to the conclusion that the existing human population can all trace their ancestry back to one female (the so-called Mitochrondrial Eve whom he attempts to identify with his 'Biblical' Eve). In other words, as he understands the case, Cann, Stoneking and Wilson weren't sure before they analysed the diverse mitochodrial sequences in their famous 1987 paper (5) that the analysis would support the proposition that we can all trace our ancestry in the maternal line back to a single female. See, for example, these quotes from Wieland's two articles:

From the 1998 Wieland article:

Most creationists will have by now heard of the ‘mitochondrial Eve’ hypothesis, the finding that all modern humans can be traced back to one woman
Holders to the ‘Eve’ theory certainly insist that all modern humans are indeed descended from one woman .
In other words, had there been more than one mitochondrial ‘surname’, it would have been a severe challenge to the biblical scenario. And it was not something that was expected by evolutionists.
The biblical creationist would conclude that the one woman suggested by the mitochondrial data is almost certainly the real Eve.
Evolutionists, aware of the way in which the mitochondrial Eve discovery could be seen to have vindicated the Bible...
The mitochondrial Eve findings were, in the first instance, in line with biblically-based expectations; while not proving the biblical Eve, they were consistent with her reality, and were not predicted by evolutionary theory.

And from the recent Wieland reply:

Mitochondrial DNA (mtDNA) indicates that all women have descended from a single woman, called mitochondrial Eve
Creationists have enthusiastically welcomed the ‘mitochondrial Eve’ hypothesis (i.e. that all modern humans can be traced back to one woman) because it clearly supports biblical history and contradicts evolutionary scenarios.
I explained how the mitochondrial Eve findings were in line with biblically based expectations. While not proving the biblical Eve, they were consistent with her reality, and were not predicted by evolutionary theory.

It is quite clear that Wieland believes that the Cann et al sequence data provides evidence for the Mitochondrial Eve concept. But this is completely wrong. I can only assume that Wieland has either never read the Cann et al paper, or if he has, he doesn't understand it. There are no data nor any claims in this paper that either support or undermine the idea that extant humans can trace their ancestry in the mitochondrial line back to a single female. Rather, the concept that different versions of a gene or group of genes must, in the technical language, 'coalesce' to a single ancestral entity is a consequence of following the logic of genetics backwards in time (8), (9) (10), (21), (22). As long as we accept the hypothesis that all humans alive today share a common ancestry, whether that common ancestry is 6000 years ago, 200,000 years ago, or 200 million years ago, then the genetic tree for a gene or group of genes must coalesce to a single common root, and this occurs first, as we look back, in the Most Recent Common Ancestor of the genetic group in question. I recommend the thorough review paper on 'Coalescent Theory' by Magnus Norborg (11) to gain an understanding of the concept and a grounding in its mathematical foundation. It seems that Wieland has been seduced by his own metaphor - ie equating the inheritance of DNA with the inheritance of surnames in a population - into thinking that Cann et al believed that they might find more than one mitochondrial 'surname'. That is simply not true.

So, if Cann et al, and subsequent authors on the subject take the existence of a mitochondrial MRCA as given, which they do, then what is new in their paper?  Why do they publish at all? Well, the differences in the mitochondrial sequences of people from around the world, which arise from mutations in human mitochondria since the MRCA, give us information about the 'phylogenetic tree' of humans. They give us information about when and where various splits in human ancestry occurred based on analysis of probable dates and order of coalescent events in the phylogenetic tree. Cann et al were able to infer from mitochondrial sequence data not just that the human matrilineal MRCA lived in Africa but also that locations outside Africa were colonised by more than one maternal migration.

So the proposition that humans can trace their maternal line to a single female is not a consequence of mitochondrial analysis, nor was this proposition unpredicted or unexpected by scientists (Wieland mistakenly repeats over and over again that the mitochondrial Eve findings were 'unexpected').  On the contrary, according to the coalescent concept, on which this work was based, the existence of a matrilineal MRCA is a given, and the question is not whether the matrilineal MRCA existed but when she lived and where. Wieland's characterisation of the science is fundamentally flawed - Cann et al's findings provided no special support for a 'biblical' Eve. Surely it is incumbent on those like Wieland, whose writing influences many naive people, to understand the science that they analyse. Wieland has failed to do so, and has presented a cheap journalistic simulacrum of the real science - in doing so, he lets down his readers and himself.

Since analysis of mitochondrial data offers no specific support for the MRCA concept and no population geneticist would dispute the proposition that human matrilineal line must coalesce to a single female at some time in the past, the only question at stake is when, according to molecular data, that coalescence arose.  Let us see.

Possible mitochondrial recombination irrelevant?

In my original article, I referred to the fact that at the time of the publication of the two papers which measured a very recent age for Mitochondrial Eve, there was another paper published that reported some evidence of recombination between maternal and paternal mitochondrial DNA (12). Wieland claims that I introduced this second challenge to the Cann et al date for effect and to make him seem uninformed (as we have seen he is ignorant about the basic scientific concepts surrounding Mitochondrial Eve, but that was not the reason for my raising this point). He claims that a) 'a second challenge makes life even harder for the evolutionist, but there is not always a need to strengthen a strong argument even further. One is not obliged to play all one’s aces if one will do..'; and b) the observation of possible recombination is irrelevant to his argument as it would drive the date of the mitochondrial MRCA to be earlier than if mitochondria are inherited strictly in the female line.

His first claim, that the Awadalla et al paper makes life harder for evolutionists, is specious There is no sense in which recombination between paternal and maternal mitochondrial DNA could weaken the case for evolution, since the existence of an MRCA in the female line and the existence of a mitochondrial MRCA are both predicted, as we have seen, straightforwardly and with great confidence by coalescence. Recombination in mitochondria, if it happens frequently, would prevent us from linking the identity of matrilineal and mitochondrial MRCAs and would complicate or prevent detailed inferences about human origins and migration based on mitochondrial data. To that extent it would be inconvenient and we should have to apply a critical re-analysis to conclusions based on analysis of mitochondrial sequences. But since there is no evidence per se for or against evolution in the distribution of mitochondrial haplotypes among the extant human population, if we were to confirm that mitochondrial recombination did occur, then the truth of evolution would be unaffected.

We can dispose of his charge of irrelevance easily - his paper was all about dating the matrilineal MRCA and, in fact, it was titled 'A Shrinking Date for Eve'.  That being so, I don't see how anything that could affect the date that we infer can be irrelevant to his claims . Wieland, of course, would like us to ignore any data that might potentially extend the date back in time, but scientists, unlike creationists, do not build their arguments on selective data mining.

More recent studies are 'special pleading' or 'explaining away contrary evidence'?

Wieland claims that by putting forward the Ingman et al paper which moves beyond the D-loop and RFLP analysis to concentrate on the entire mitochondrial genome outside the D-loop based on whole genome sequencing, and which finds a date for the matrilineal MRCA of 150,000 to 200,000 years, I am indulging in special pleading or irrationally explaining away contrary evidence.

Wieland shows a fundamental misunderstanding of how science works. He wants readers to think that all the possible interpretations of the data are equally rational and probable and that one chooses one or the other interpretation based on one's preconceptions; that one conclusion is as valid as another and subject only to interpretation based on one's philosophical perspective.  It suits Wieland to promote that fallacy because he wants to use biblical 'authority' to choose between incompatible hypotheses. He fails to acknowledge the fact that competing scientific hypotheses are not equal - some are supported by evidence and others are not and, in science, those that have significantly poorer empirical support or that are shown to be irrevocably incompatible with facts die or are withdrawn.

In this case, the conclusion that the matrilineal MRCA could be dated at 6,500 years based on direct measurement of mutational rate on the D-loop was clearly incompatible with large quantities of other evidence, not just about the date of the split of the human/chimpanzee lineage, but also about the date of various human migratory events such as the colonisation of Australia and North America by humans. So, scientifically, there was a need either to confirm the 6,500 year date for Mitochondrial Eve (and coincidentally to explain, in a convincing way, how the mountain of extremely powerful biochemical, molecular and anatomical evidence pointing to a much earlier date can be compatible with a more recent date); or to challenge the validity of the 6,500 year date by uncovering reasons for the apparent anomaly in the experimental methods or data analysis applied by Parsons et al. And this latter is precisely the import of the data-pooling for D-loop data, the Hasegawa et al paper (13) and the Ingman et al paper. Theirs was not an attempt to sweep inconvenient findings under the carpet as Wieland would have it, but rather a process, quite common in science, to understand the reasons for apparently incompatible findings and to reconcile them objectively.

In his reply to me Wieland does not address any of the scientific findings put forward in these studies; in fact, you wouldn't know from his reply that he had even read them. Perhaps he hasn't. His approach is to dismiss them all on the basis that they do not fit with his preconceived beliefs. He fails to acknowledge that pooled pedigree restriction map data on the D-loop point to a matrilineal MRCA at 35,000 - 40,000 years BP which is clearly incompatible with his prejudiced belief. He doesn't begin to address the evidence presented regarding the better suitability of whole genome sequencing and the choice of focusing on  the mitochondrial genome outside the D-loop. He claims that scientists are doing a cover-up but he fails to engage with a single scientific factor (eg that whole genome sequencing is bound to be more accurate than RFLP analysis owing the nature of two approaches), because his purpose is not scientific discussion but propaganda.

Mitochondrial evidence compatible with biblical Eve?

Is the mitochondrial data compatible with biblical Eve, and is there 'no reason at all why the mitochondrial Eve could not be the biblical Eve?

Well, to the first part of that question, if we take the Parson's paper in isolation, the mitochondrial data is compatible with a matrilineal MRCA 6,500 years ago. But that is a premature and naive conclusion as my paper pointed out. Subsequent work with mitochondrial sequences across the human population indicates a date 175,000 years ago (and even pedigree data uncorrected for the differences in short and long term rates and without using more accurate whole genome sequencing gives a date of 35,000 years).

As for there being "no reason at all why the mitochondrial Eve could not be the biblical Eve", Wieland is here indulging in unthinking rhetoric. There is a vast number of palaeontological, anatomical and molecular reasons why it makes no sense at all to date the first human as late as 6,500 BP.

There are other scientific reasons for rejecting the idea that the current human population is descended from a single man and woman The evidence from analysis of extant alleles in highly polymorphic loci suggests a minimum population size or tightest bottleneck in the human lineage of 10,000 individuals, and is totally incompatible with a minimum bottleneck of ancestors of current humans of less than seventy (14), (15), (16).  Exploring this in detail is outside the scope of the current reply except to point out that it contradicts Wieland's claim that the scientific evidence does not preclude identifying  the matrilineal MRCA with biblical Eve. The molecular evidence is incompatible with a sole biological maternal ancestor for extant humans. The 'myth of Eve', as Ayala puts it (14) arises from confusion between gene genealogies and individual genealogies.

Turning to his claim that creationists do not use the Parsons et al paper to attempt to prove the existence of 'biblical' Eve, that might well be true for him, but it is certainly not true for other possibly less cautious creationists, as a review of the creationist web sites I quoted in my original article or that are turned up by googling 'mitochondrial Eve creation' will show. And Wieland explicitly derives comfort from his erroneous belief that mitochondrial sequence data demonstrates that the mitochondrial lineage goes back to one woman. This is a common misconception and it is, as we have seen, plain wrong.

Neanderthals are the same species as us?

This claim of Wieland is something of a peripheral afterthought so we won't spend much time on it.  Suffice to say, comparison of Neanderthal and human mitochondrial DNA in two studies and initial results from nuclear DNA (17), (18), (19) leads to the conclusion not only that the modern human lineage and Neanderthals diverged more than 500,000 years ago, but that there was no interbreeding between modern humans and Neanderthals where their ranges overlapped in Europe. In fact, in both mitochondrial studies, the average difference between modern human and Neanderthal sequences was about four times greater than the average distance between humans; and the difference between Neanderthal and modern human mtDNA was no less for European than for other regional groups. This last finding is very telling, as if there had been interbreeding between Neanderthals and modern humans in Europe one would find less difference between Neanderthal and European mtDNA than between Neanderthal and other regional groups.

[Technical note added 21st November 2006: Note however that two papers, recently published, suggest that there was interbreeding between modern humans and Neanderthals:

Trinkaus et al, Early modern humans from the Petera Muierii, Baia de Fier, Romania, Proc Natl Acad Sci doi:10.1073/pnas.0608443103 find a mosaic of modern human and Neanderthal/archaic features in these remains from Romania securely dated to 30,000 radiocarbon years BP, and suggest that this is evidence of interbreeding between the species.  The conclusion is controversial and not generally accepted.
Green et al, Analysis of one million base pairs of Neanderthal DNA, Nature 444, 330 - 336 (16th November 2006). This is a major paper in which one million base pairs of mitochondrial and nuclear DNA have been sequenced.  The authors find that Neanderthals split off from the lineage leading to modern humans between 461,000 and 825,000 years ago, a finding that is entirely consistent with previous analyses. Neanderthal has the derived allele in 30% of human SNPs, suggesting that these SNPs predate the Neanderthal split - this high percentage suggests some gene flow between modern humans and Neanderthals, and the fact that the X-chromosome is more divergent than the autosomes suggests that gene flow would have occurred predominantly from modern human males into Neanderthals. The authors support the well accepted finding that Neanderthals contributed little or no genetic material to the modern human genome]

Conclusion

There is a vast preponderance of scientific evidence against the notion that the first humans lived as recently as 6,500 years ago. In fact the evidence is that they lived 200,000 years ago (20) Creationists everywhere, including Carl Wieland, have used the Parsons et al work in isolation to support their biblically determined views and have failed to acknowledge more recent work that reconciles the basic discrepancy. The fact is that Wieland is ignorant of the basic science and misrepresents its results.  Other, less scrupulous creationists do so more blatantly and more dishonestly than he. I predicted in my original paper that " both ill-informed creationists and those who should know better will be using this discredited argument 20 years from now". I have had cause to remember this prediction many times since then, as I frequently come across creationists claiming that mitochondrial DNA has proved that we are descended from one woman who lived about 6,000 years ago. Wieland's somewhat more sophisticated argument is no less erroneous and does nothing to discourage this irrational nonsense.

References

1) Wieland, A Shrinking Date for 'Eve', Answers in Genesis, available on line here:

http://www.answersingenesis.org/docs/4055.asp

2) MacAndrew, Misconceptions around Mitochondrial Eve, available on line here

3) Wieland, Mitochondrial Eve and biblical Eve are looking good: criticism of young age is premature, Journal of Creation 19(1), 57–59, 2005

4) Reference 3) is available on-line here:

http://www.creationontheweb.com/content/view/4472/

5) Cann, Stoneking and Wilson, Mitochondrial DNA and Human Evolution, Nature 325 (1987) 31 - 36

6) Parsons TJ et al, A high observed substitution rate in the human mitochondrial control region, Nat Genet 15, 363-368

7) Ingman et al, Mitochondrial genome variation and the origin of modern humans, Nature 408, 708 - 713

8) Ridley, Evolution - Third Edition, ISBN 1-4051-0345-0, Blackwell Science, 2004, p144

9) Futuyama, Evolutionary Biology, ISBN 0-87893-189-9, Sinauer Associates, 1998, p327

10) Introduction to Coalescent Theory, available on-line here:

http://serine.umiacs.umd.edu/teaching/computational_biology/syllabus/files/lecture_01.pdf

11) Norborg, Coalescent Theory, March 2000, available on-line here:

http://walnut.usc.edu/~magnus/papers/wiley.pdf

12) Awadalla, Eyre-Walker, and Maynard Smith, Science 286, 2524 - 252

13) Hasegawa, Cao and Yang, Preponderance of Slightly Deleterious Polymorphisms in Mitochondrial DNA:  Nonsynonymous/Synonymous Rate Ratio is Much Higher within Species than Between Species, Mol Biol Evol 15, 1499 -1505

14) Ayala, The myth of Eve, Molecular biology and human origins, Science 270, 1930 - 1936

15) Xiong et al, No severe bottleneck during human evolution; evidence from two apolipoprotein C II alleles, Am J Hum Genet 48, 383 -389

16) Rogers and Jorde, Genetic evidence on the origin of modern humans, Hum Biol 67, 1 - 36

17) Krings et al, Neandertal DNA sequences and the origin of modern humans, Cell 90, 19 - 30, (1997)

18) Ovchinnikov et al, Molecular analysis of Neanderthal DNA from the northern Caucasus, Nature 404, 490

19) News report by Rex Dalton, Neanderthal DNA yields to genome foray, Nature 441, 260 - 261, (2006)

20) McDougall et al, Stratigraphic placement and age of modern humans from Kibish, Ethiopia, Nature 433, 733 - 736 (2005)

21) Fu and Li, Coalescing into the 21st century: an overview and prospects of coalescent theory, Theoretical Population Biology 56, 1-20

22) Kingman, Origins of the Coalescent, Genetics 156, 1461 - 1463 available on-line here:

http://www.genetics.org/cgi/content/full/156/4/1461