I’m grateful for all the time and effort that our commentators have obviously put into their essays. Faced with such a plethora of detailed comments, I don’t want us—especially those who have not yet read the book—to lose the forest for the trees. So I want to take a step back and look at the big picture and review the comments in light of the case I present.

It’s important to understand my target in this book. My aim is to refute the claim of secularists and revisionists that the existence of Adam and Eve is incompatible with the evidence of contemporary science. Here is an outline of the case I present:

  1. As a result of folklore studies, we are able to identify mythical narratives on the basis of at least ten “family resemblances” shared by myths.
  2. Gen 1–11 exhibits multiple instances of eight out of ten of these family resemblances.
  3. Gen 1–11 may therefore be plausibly classified as mythical.
  4. In view of their metaphoricalness, plasticity, and flexibility, myths need not be interpreted literalistically.
  5. Gen 1–11 exhibits multiple instances of this same metaphoricalness, plasticity, and flexibility.
  6. Gen 1–11 need not therefore be interpreted literalistically.
  7. In view of the genealogies that order the primaeval narratives into a primaeval history that melds seamlessly into the patriarchal narratives, Gen 1–11 also displays a historical interest.
  8. Gen 1–11 may therefore be plausibly classified as mytho-history.
  9. The NT confirms that Adam and Eve were regarded as historical persons, since they had real world effects outside the narratives.
  10. We are therefore biblically committed to Adam and Eve as actual, historical persons from whom all humanity has descended.
  11. Since Gen 1–11 is mytho-historical, it is to contemporary science that we must turn in order to find the approximate date at which Adam and Eve lived.
  12. We may determine the approximate time at which humans originated by looking for “archaeological signatures” of modern cognitive capacity, such as abstract thinking, planning depth, technological innovativeness, and symbolic behavior.
  13. Paleoanthropological evidence from various scientific fields indicates that Neanderthals exhibited modern cognitive capacity.
  14. As the universal progenitors of mankind, Adam and Eve must therefore have been ancestral to both Neanderthals and Homo sapiens.
  15. The last common ancestor of Neanderthals and Homo sapiens is the large-brained human species Homo heidelbergensis, whose members lived somewhere around 750kya.
  16. The existence of a single couple who are the progenitors of the entire human race cannot be ruled out on the basis of population genetics if they lived >500kya.
  17. Adam and Eve may therefore be plausibly identified as members of the species Homo heidelbergensis living sometime prior to 750kya.
  18. Therefore, our biblical commitment to a primordial human pair from whom all humanity has descended is fully compatible with the scientific evidence.

As you can see, my argument for my claim of compatibility has two parts, one hermeneutical (steps 1–10) and one scientific (steps 11–18).

in Quest of the historical Adam

Eerdmans, 2021

In the first part I want to determine, independent of modern science, what our biblical commitments are. By “biblical commitments” I mean what Scripture teaches concerning the relevant question. It’s important that this question be answered first, lest our interpretation of the text be distorted by concordism.

In order to determine what those commitments are, I first undertake a detailed genre analysis of Gen 1–11. Right at the start it became clear to me that the narratives of Adam and Eve must not be interpreted in isolation but within the context of the primaeval history of Gen 1–11, and that primaeval history within the context of the entire book of Genesis, and the book of Genesis within the context of the Pentateuch. The considerable benefit of such an approach is that it yields a unified solution to the problems posed by the primaeval history as a whole, not just by the stories of Adam and Eve.

By contrast the so-called Recent Genealogical Adam hypothesis championed by Michael Murray ignores context, focusing solely on the stories of Adam and Eve, and is not preceded by any genre analysis, which is vital to interpretation of the narratives. I don’t want our discussion today to be derailed by debating this alternative model; but let me simply say that Murray confuses my reasons for thinking that Genesis teaches that Adam is the universal progenitor of mankind with reasons for thinking that Genesis teaches that Adam is very ancient. I affirm the first and deny the second. Contrary to Murray, Scripture does affirm the existence of the natural kind “human” (it’s called adam), and there is no scriptural reason to think that that natural kind extends beyond those who are image bearers.The exploration of the question of genre required me to look first at the nature or character of myth. The set of humans is either identical to the set of image bearers, or else a subset of it, if angels are also image bearers (“Let us create man in our image,” says God). (By the way, if angels are created in the image of God, that puts the Kabash on the functional interpretation of the image, since angels and humans have clearly different functions. The image is then more plausibly an ontological likeness to God.)

The exploration of the question of genre required me to look first at the nature or character of myth. The consensus view in classical studies is the folklorist’s understanding of myth as a traditional, sacred narrative explaining how the world and man came to be in their present form. A myth seeks to explain present realities by anchoring them in the prehistoric past and so to validate a culture’s contemporary institutions and values. In contrast to other forms of folklore such as folktales and legends, myths are taken as authoritative for the culture that embraces them. As sacred narratives, their main characters are not usually just human beings but deities or heroes, whose activities are set in an earlier age or another realm.

Whether one views myth with folklorists as a literary genre of a text or, with Jack Collins, as a social function of a text is, I think, inconsequential; either view will lead to the same conclusions. I decided to go with the consensus view among folklorists rather than with Collins’s view simply in order to maintain continuity with the folklorists’ discussion.

Since the lines between myth, folktale, and legend are apt to be blurry, it is probably impossible to lay down necessary and sufficient conditions for each of these narrative genres. Instead, we ought to be looking for what Wittgenstein called “family resemblances” among stories regarded paradigmatically as myths. This is the same approach that Richard Burridge adopted in his ground-breaking work on the genre of the Gospels. On the basis of various family resemblances exhibited by ancient biographies, Burridge was able to show that the Gospels most resemble the genre of ancient biography. Similarly, I drew up a list of ten family resemblances which folklorists have identified as characteristic of myth and was able to show that Gen 1–11 exhibited eight of the ten family resemblances, most importantly an abundance of etiological motifs grounding realities present to the Pentateuchal author in the primordial past. This provides powerful inductive evidence for classifying Gen 1–11 as mythical. My invitation to those who disagree is simple: provide a more plausible explanation for these family resemblances than that of myth.

Hans Madueme protests this approach to the determination of literary genre: “The mythical understanding of primeval history is an extrabiblical theory that obscures the analogy of faith. Christians should give priority to interpreting Scripture in light of Scripture rather than in light of relatively speculative theories about ANE culture and its putative relationship to the biblical authors.”My invitation to those who disagree is simple: provide a more plausible explanation for these family resemblances than that of myth. This characterization subtly conflates determining the genre of a biblical book with interpreting a biblical book. There is nothing impious or misguided about looking to extra-biblical literature to determine that, say, the Psalms are poetry or that Revelation is apocalyptic or that the Gospels are ancient biography. The question of interpretation comes later.

Similarly, Madueme’s charge that my “analysis of Genesis 1–11 in light of family resemblances among ANE myths presupposes the comparative method” is multiply confused. In the first place these family resemblances compiled by folklorists are not limited to ancient Near Eastern myths but are collected from myths throughout the world from any time. Second, the comparative method involves comparing pieces of literature with a view toward showing parallels and possible lines of dependency between them. That is a project about which I express enormous skepticism in the book.

Madueme also takes umbrage at my claim that Gen 1–11 exhibits fantastic elements, that is, elements which, if taken literally, are so extraordinary as to be palpably false. This is just one of the family resemblances of myth. Madueme doesn’t deny that myths are often fantastic, but he denies that my examples from Genesis 1–11 are fantastic. He alleges that I’m not determining our biblical commitments independent of modern science after all because my “notion of the fantastic doesn’t derive from Scripture . . . but . . . from a mind shaped by modern science.” Again, this claim is subtly confused. First of all, the identification of these fantastic elements does not derive from modern science but has been noticed by people as far back as Origen and Augustine. Second, using Scripture itself to determine what is fantastic assumes that Scripture does not contain any fantastic elements, which is viciously circular, and hence, question-begging. Moreover, Scripture certainly does contain fantastic elements: just read the book of Revelation! Third, and most important, Madueme confuses using modern science to identify fantastic elements in Scripture with using modern science to determine our biblical commitments. I reject the latter as concordist. If, as Madueme seems to think, we are biblically committed to the beliefs that the world is only several thousand years old, that all of the world’s languages originated at a Babylonian ziggurat a few thousand years ago, that there has been within the span of human history a worldwide flood that wiped out all terrestrial life on earth save that aboard the ark, and that every human being who has ever existed is descended from a single human couple who lived just a few thousand years ago, then I want to be aware of that fact! For if that interpretation is, indeed, correct, then the task of reconciling our biblical commitments with the facts of modern science, history, and linguistics becomes hopeless. I don’t think Madueme has any appreciation of just how hopeless such a reconciliation would be. For he retorts, “so much the worse for our modern expectations.” But the truth is that such a situation would plausibly call for the sort of drastic revision of the doctrine of inspiration that I describe as the worst case scenarios in my opening chapter.

Collins thinks that we can reduce people’s resistance to my claim that Gen 1–11 exhibits fantastic elements by characterizing these as pictorial elements. But that’s to get ahead of ourselves. How do we know at this point of the inquiry that they are pictorial language? Maybe Madueme is right that they were intended literally. I myself will ultimately claim in steps (4–6) that these elements are pictorial or figurative, but those premises will require argument. In support of step (4) I adduce three lines of evidence, illustrated by an abundance of examples: the metaphoricalness of myth, the plasticity of myth, and the flexibility of myth. Then in support of (5) I provide examples of each of these phenomena in Gen 1–11. I suggest that in many cases the Pentateuchal author did not intend for his narrative to be taken literally. Here Collins and I are very much on the same page.

In step (7) I proceed to argue that genre of Gen 1–11 is not pure myth, however. Rather the genealogies that guide the narratives transform the primaeval narratives into a primaeval history that melds seamlessly into the patriarchal narratives, which have indisputably a historical interest. Therefore, the best genre analysis of Gen 1–11 involves a blend of myth and history, what has been called mytho-history.

Notice that the plausibility of this genre classification for Gen 1–11 in no way depends upon our ability to tease apart which aspects of the stories are mythical and which historical. That’s not how literary criticism works. I suspect that the blend is more like coffee and cream than like colored marbles in a bag. Madueme protests, “Up to that point in the book Craig has been telling us in great detail which parts of the narratives are mythic and which parts are not—he can’t plead ignorance now!” Again, the charge is subtly confused. Madueme mistakenly takes “mythical” to mean “authoritative but not literal,” which is not the folklorists’ definition of “myth.”Identifying some elements as figurative rather than literal is not the same as separating myth from history. With Collins I think that we can sometimes identify elements in the narratives which are plausibly pictorial or figurative. But identifying some elements as figurative rather than literal is not the same as separating myth from history. The whole narrative is mytho-historical and contains figurative language.

Having examined the OT data concerning Adam and Eve, I then turn in step (9) to the NT data. I agree with Madueme that the greatest challenge to my thesis is that neither Jesus nor the apostles apparently interpreted the early chapters of Genesis mytho-historically. But here I argue that appearances can be deceiving. Madueme’s claim that we should “interpret Scripture in light of Scripture” is unfortunately simplistic. Studies in biblical intertextuality reveal how much more complicated things are. Scholars studying NT authors’ citations of OT texts show that these authors often interpret these texts in ways that are foreign to their original meanings and sometimes rather strange. “Out of Egypt have I called my son.” “Thou wilt not allow thy holy one to see corruption.” Similarly, if we think that a NT author’s citation of a text commits us to the historicity of that text, then we’re going to find ourselves committed to the historicity of persons and events of Jewish folklore and mythology, such as Jannes and Jambres, the Archangel Michael’s dispute with Satan over the body of Moses, and the rolling well that accompanied the Israelites during their wilderness wanderings in the desert.

By distinguishing between a literary figure and a historical figure, between using a text illustratively and using a text assertorically, and between what the author believed and what the author taught we can avoid commitment to such persons and events. Madueme’s claim that we are committed to not only what the NT authors teach but also to what they assume is demonstrably false, on pain of being committed to geocentrism, the moon’s being luminous, the eye’s projecting light, etc. It is plausible that many of the passages about Adam in the NT commit us to no more than truth-in-the-stories of Genesis. But not all of them! Paul’s explanation of the historical consequences of Adam’s fall show that he teaches that this event is historical, since the sin of a fictitious person could not have real world consequences outside the fiction. The NT witness therefore confirms our biblical commitment to the historicity of Adam and Eve (step 10).

With step (11) we turn finally to the scientific evidence in order to determine the approximate time at which Adam and Eve existed. In order to determine this, we look for manifestations of modern cognitive behaviors. Here it’s important to clear up a confusion on the part of both Murray and Hlusko. Hlusko represents me as holding that these behaviors “make us human,” and Murray interprets me to be offering a definition of “human.” In my view, however, these modern cognitive behaviors are neither definitional nor constitutive of humanity but rather serve the paleoanthropologist as evidence of humanity. If you want to see what I think what makes us human and defines “human,” you have to turn to the last chapter of the book, where I defend anthropological dualism. A human being is a rational soul united with a hominin body. But since rational souls cannot be detected empirically, we have to look for evidence of rational behavior in order to determine when human beings first appeared on this planet.

As a result of my study, I am firmly convinced that Neanderthals exhibited modern cognitive behaviors and were therefore human (step 13). Since humanity is unlikely to have evolved twice, it follows that humanity must go back to the most recent common ancestor of Neanderthals and Homo sapiens, namely, Homo heidelbergensis, who also exhibited these same modern cognitive behaviors.

I was really surprised that Michael Murray, who has so inspired my Quest, would call into question the humanity of Neanderthals. The trend over the last several decades, as described in Rebecca Sykes’s new book Kindred, is quite in the opposite direction.Rebecca Wragg Sykes, Kindred: Neanderthal Life, Love, Death, and Art (London: Bloomsbury Sigma, 2020). I think, frankly, that Murray’s position is out of date and is increasingly difficult to defend. To give but one piece of evidence: last year archaeologists recovered a three-ply fiber cord from the Neanderthal site of Abri du Maras in France. Found with implements dating to 40–50kya, the cord has three strings of fibers obtained from the inner bark of a conifer tree and each twisted clockwise and then as group twisted counterclockwise. The excavators led by B. L. HardyI was really surprised that Michael Murray, who has so inspired my Quest, would call into question the humanity of Neanderthals. emphasize that cordage manufacture “requires an understanding of mathematical concepts and general numeracy . . . to create a structure.”B. L. Hardy et al., “Direct evidence of Neanderthal fibre technology and its cognitive and behavioral implications,” Scientific Reports 10, 4889 (2020). As the structure becomes more complex, they state, it “requires a cognitive complexity similar to that required by human language.” They opine that in view of the ongoing revelations of Neanderthal art and technology, “it is difficult to see how we can regard Neanderthals as anything other than the cognitive equals of modern humans.”

The article that Murray cites to call into question Neanderthals’ linguistic abilities supports no such conclusion. The authors push back against the claim that “the substitution at position 114076877 in intron 8 of the FOXP2 gene was involved in the evolution of modern language.”Tomislav Maricic et al., “A Recent Evolutionary Change Affects a Regulatory Element in the Human FOXP2 Gene,” Molecular Biology and Evolution 30/4 (2013): 849. In fact, the authors note that in some African populations up to ten percent of the contemporary population actually carries the ancient allele, not the derived allele, and yet no language impairment has been noticed. A later study published in 2018 found the ancient allele to be present in an even larger fraction of various south African populations, as high as an astonishing forty-three percent in ≠Khomani San, which shows that this mutation is not as important for the evolution of speech and language as may have been imagined.E. G. Atkinson et al., “No evidence for recent selection at FOXP2 among diverse human populations,” Cell 174 (2018): 1424–435.

I fear that evangelicals are in danger here of dehumanizing fellow human beings, just as Charles Darwin believed that Black Africans are closer on the evolutionary scale to gorillas than to White Europeans. This would be a morally unconscionable error that we should be zealous to avoid.

Now if the paleontological and archaeological evidence supports the humanity of Neanderthals, then, as step (14) states, Adam and Eve, as the sole universal progenitors of mankind, must have lived before the divergence of Neanderthals and Homo sapiens. The most recent common ancestor of both is often thought to be Homo heidelbergensis, as stated in step (15). I take to heart Hlusko’s admonition not to identify Adam and Eve too closely with a particular human species. Rather, as I state on the final page of the book, the name Homo heidelbergensis “serves at least as a useful placeholder for that large-brained human species which was ancestral to Homo sapiens and our various sister species of the human family. We can live with uncertainty.”

In step (16) I respond to the challenge posed by population genetics to a single couple origin of the human race. The reason that I did not go into this challenge in greater length is that, having been led by the archaeological evidence to the conclusion that Adam and Eve lived some 750kya, I found that the challenge posed by population genetics simply evaporated. Independent genetic modeling by both Swamidass and Gauger and Hössjer reveal that a single couple origin cannot be ruled out prior to 500kya, and this fact came to be acknowledged by even the most ardent opponents of a single couple origin, such as Dennis Venema.It seemed to me that my hypothesis of an ancient genealogical Adam cannot be ruled out. So it seemed to me that my hypothesis of an ancient genealogical Adam cannot be ruled out.

Hlusko, however, offers significant pushback to this claim. She presses two major objections.

First, she disputes my claim “that most studies have failed to uncover evidence of trans-species variation between humans and nonhuman ancestors involving more than four allele lineages,” which, by implication, could have been carried by a heterozygous Adam and Eve. She rejoins, “this is only true if you limit your investigation to living species,” that is, to Homo sapiens. But if we include the genomes of extinct human species like Neanderthals, that pushes the date of the last common ancestor further back. She says, “It turns out that the vast majority of the human nuclear genome coalesces between one million and five million years ago.”

Let’s not miss the significance of what is acknowledged here! For years we have been told that on the basis of the genetic divergence exhibited by the contemporary population the scientific evidence rules out with heliocentric certainty a single couple origin of the human race. It turns out that that claim was not true.

But what about Hlusko’s revised claim that when we include the genomes of extinct human species, then a single couple origin is ruled out? This revised claim is puzzling for a number of reasons.

First, Hlusko seems to conflate the argument based on incomplete lineage sorting with the substantially different argument from trans-species variation. The former shows only that the ancestral population leading up to the divergence of chimps and humans, for example, is relatively large; but that conclusion is quite consistent with there being a founding pair of humans who were the first to emerge from that ancestral population. In order to rule that out, one needs to prove trans-species variation between that ancestral population and the succeeding human population in excess of four lines of alleles. The claim that the majority of the human nuclear genome coalesces one to five million years ago is, in fact, consistent with the models I mentioned before. Both Swamidass and Gauger and Hössjer calculate a time of the most recent common ancestor (TMRCA) around 2mya. But they both arrive at a time of the most recent four alleles (TMR4A) of around 500kya.

Second, the effect of including genetic evidence from extinct human species on TMR4A will push the date further back, but no one knows by how much. In 2017, geneticist Richard Buggs pointed out that the hypothesis of a bottleneck of two had in fact never been tested scientifically.  Stephen Schaffner in his review of my book observes that since scientists aren’t interested in finding such a founding pair, there is scant formal scientific literature on the subject. Hlusko’s documentation does not support her claim. What the cited passage from David Reich’s book basically says is that most of our genealogical ancestors are genetic ghosts. That’s not relevant to our question. I’m not claiming that Adam and Eve’s DNA has managed to pass down to us. Reich also quotes the Li-Durbin study, whose shortcomings are already explained in my book. I also point out that the HLA complex is unusually susceptible to convergent evolution, which renders the evidence of trans-species variation so uncertain that even Venema would not appeal to it.

I know that Swamidass is working on recalculating his results by including ancient DNA, and I’m very eager to see his results. A complicating factor is that including such genetic information in the models is difficult due to high amounts of degradation in ancient DNAMy model hopes to minimize interbreeding but does not decisively exclude it. sequences as well as to interbreeding between Homo sapiens and extinct species. He anticipates that the date of TMR4A may be pushed back a couple hundred thousand years, which is consonant with my hypothesis.

Finally, all this still leaves open the question of interbreeding between the descendants of Adam and Eve and nonhuman hominins. My model hopes to minimize interbreeding but does not decisively exclude it. Such interbreeding would introduce genetic material into the human race from sources not ultimately traceable to Adam and Eve. So just how much interbreeding does Hlusko think would be required in order for the genetic evidence to permit an ancient genealogical Adam and Eve?

All this goes to undermine the objection to an ancient genealogical Adam and Eve based on the evidence of population genetics. In her conclusion, Hlusko says, “I do not agree with Craig that genetic data support humanity deriving from one man and one woman at any point in evolutionary time.” But so saying misrepresents my claim. I am not claiming that there is “direct genetic evidence” for a single couple origin, but rather that a single couple origin is compatible with the genetic evidence. Hlusko herself concludes, “I also do not think that a lack of direct genetic evidence should make the existence of Adam and Eve any less believable.” Exactly! That’s my point. But I thought that she was arguing that there is direct genetic evidence against the existence of Adam and Eve. That bold conclusion has not, I think, been proved.

Hlusko’s second major objection is that a single couple origin is improbable from the standpoint of genetic health. She explains, “A population limited to just two breeding individuals is not going to survive for long because of the deleterious effects of inbreeding among their descendants.” I take it that she means that a population with a founding pair will not survive for long because of the deleterious effects of inbreeding. This was already an objection voiced by Venema himself. In my book I noted this objection and Buggs’s response:

In response to Venema’s claim that a bottleneck of two would be disastrous for the population’s health, Buggs points out that studies show that ‘even a bottleneck of a single pair would not lead to massive decreases in genetic diversity, if followed by rapid population growth. . . . From a bottleneck of a single ertilized female, if population size doubles every generation, after many generations the population will have over half of the heterozygosity of the population before the bottleneck. If population growth is faster than this, the proportion of heterozygosity maintained will be higher’ (p. 347).See Richard Buggs, “Adam and Eve: A Tested Hypothesis?,” Ecology & Evolution (blog), October 28, 2017.

My proposal has the great benefit of allowing us to maintain the traditional doctrine of biblical inspiration and inerrancy.Perhaps I should have taken this objection more seriously, but I think it would be helpful if Hlusko would take account of Buggs’s response.

In short, I remain unconvinced that an ancient genealogical Adam and Eve has been shown to be incompatible with the scientific evidence.

I resonate strongly with Collins’s metaphor of a cost/benefit analysis of various views of the historical Adam. My proposal has the great benefit of allowing us to maintain the traditional doctrine of biblical inspiration and inerrancy. It preserves the traditional doctrines of the universal progenitorship of Adam and Eve and their historical fall into sin. Its genre analysis treats the primeval history of Gen 1–11 as a unit and so enables us to deal with the problems, not only of the historical Adam and Eve, but also of the Flood, the tower of Babel, the age of the earth, and so on. It is fully compatible with the accumulating evidence of paleoanthropology that humanity on this planet is very ancient. Its principal cost is that it requires us to give up the recent date of the events of the primaeval history, a price that, it seems to me, is well worth the benefits.

In Quest of the Historical Adam: Introducing the Symposium
Ken Keathley
Literality, Incredulity, and Hermeneutical Schizophrenia
Hans Madueme
Exegetical Response: A Cost-Benefit Evaluation
C. John Collins
Humans, Homo sapiens, and the Image of God
Michael J. Murray
Human Genomics and Divine Intervention
Leslea J. Hlusko
Restating the Benefit of My Proposal: A Rejoinder
William Lane Craig