Theistic Evolution: A Critique

Since Adam never existed, he could not have committed a first and original sin, and therefore, sin could not have been passed down from him to the rest of humanity.¹

Logically, no intelligent being—not even God—can direct an undirected process.²

Summary

The question is whether a Christian can accept evolution, including human descent from earlier primates, without giving up anything essential to Christian doctrine.  The answer depends on what counts as essential, and the article answers it at three levels.

1. The creeds survive intact. Nothing in the Apostles' or Nicene Creed mentions Adam, the age of the earth, or how species arose. Gregg Allison, writing against theistic evolution in the 2017 Crossway critique, grants that the Nicene Creed "did not specifically address the issue of evolution in general or theistic evolution in particular."² BioLogos affirms the Trinity, the incarnation, the bodily resurrection, miracles and answered prayer, and says God "continues to actively uphold and sustain creation."³ A theistic evolutionist can recite every line of the creeds and mean it.
2. Paul's Adam does not survive intact in any version. Paul's argument in Romans 5 and 1 Corinthians 15 needs one man, the first, through whose sin death came to all people. That is argued in the sibling article Adam, the Fall, and Death. Every theistic-evolution position gives up at least one part of it. Denis Lamoureux gives up Adam himself: "Adam never existed."¹ Francis Collins, founder of BioLogos, leans toward reading Adam and Eve as "a poetic and powerful allegory."⁴ John Walton keeps a real Adam and gives up the claim that Genesis says how his body was made.⁵ William Lane Craig keeps Adam and Eve as the sole ancestors of all humans and moves them back more than 750,000 years, which the genealogies of Genesis cannot reach.⁶⁶ Joshua Swamidass's hypothesis keeps a recent Adam made without parents, and keeps him only by a miracle in human origins that Collins's own definition of theistic evolution rules out.⁴⁷
3. Human death before the first sin is a cost every version pays or defines away. If humans descend from a population of primates, creatures in our line died before any first sin. Romans 5:12 says death "spread to all people" through one man's sin, and the sibling article Death Before the Fall shows that this verse speaks of human death. BioLogos reads that death as mainly spiritual.⁸ Lamoureux says "there is no sin-death problem" because Adam never existed.¹ Craig and Swamidass avoid the problem only by counting the creatures who died before Adam as not yet human in Scripture's sense. Animal death before the Fall is a weaker charge than the young-earth movement makes it, because no verse states that animals did not die before Adam.
4. The charge that "guided evolution" is a contradiction is true only of one version. Stephen Meyer argues that "no intelligent being—not even God—can direct an undirected process."² That holds against Kenneth Miller and others who call evolution unplanned in the full sense. It does not hold against Collins or BioLogos, and the 2017 volume concedes it: its own definition denies only "empirically detectable" guidance, and Allison writes that Collins does "not specify ... whether it is undirected or directed."²² In the science, "random" means random with respect to a mutation's effects on the organism.⁹ "Purposeless," as George Gaylord Simpson and Jacques Monod used it, is a philosophical addition to the science, as Alvin Plantinga argues.¹⁰¹¹¹²
5. Hidden guidance predicts nothing that unguided evolution does not. Collins says the process "would appear a random and undirected process" from our perspective, while from God's "the outcome would be entirely specified."⁴ A hypothesis built to look exactly like its rival cannot be supported by data, and cannot be refuted by data either. So theistic evolutionists cannot claim the biology as evidence for God's guidance, and their critics cannot claim it as evidence against. Meyer is right that this makes the guidance "vacuous" as science.² He is wrong if he means it makes the guidance false.
6. The real price of hidden guidance is the design argument from biology. Collins accepts that the universe's properties "appear to have been precisely tuned for life" and treats this as pointing to God.⁴ He rejects the same kind of inference in biology as a "God of the gaps" argument.⁴ He gives no reason that fits physics and not biology. This is the most serious weakness in his position, and it is a philosophical weakness, not a doctrinal one.
7. Verdict. A Christian can accept evolution and remain orthodox by the creeds. A Christian cannot accept it and keep the doctrine of Adam as Paul states it and as Wayne Grudem lists it in twelve points.² The theistic evolutionists who say this most plainly are Lamoureux and Peter Enns, and Enns calls it "the burden of anyone who wishes to bring evolution and Christianity together."¹³ How each view reads the text of Genesis is the parent article's question, Competing Readings of Genesis. The four-way comparison of the main Christian views is planned as Young Earth, Old Earth, Design, and Evolutionary Creation.

Part 1: The positions in their own words

BioLogos: "two basic ideas"

BioLogos is the largest organisation promoting theistic evolution among evangelicals.  Francis Collins founded it in 2007.  It prefers the name "evolutionary creation" and defines the view this way:  "EC includes two basic ideas. First, that God created all things, including human beings in his own image. Second, that evolution is the best scientific explanation we currently have for the diversity and similarities of all life on Earth."³

BioLogos gives three reasons for avoiding the older label "theistic evolution."  The third is the one that matters here.  The older label, it says, was linked to deism, the idea of a God who made the world and then stepped back.  Against that, "we affirm the biblical miracles (most centrally the Resurrection), believe God answers prayer, and recognize that God works providentially through natural processes to accomplish his purposes."³  It also says God "continues to actively uphold and sustain creation."³

On Adam, BioLogos takes no side.  "Some ECs believe Adam and Eve were a historical couple. Others see the story as a symbolic retelling of Israel's story, or as a symbolic story about humanity as a whole."³  This article uses "theistic evolution" for the whole family of views and "evolutionary creation" when quoting BioLogos and Lamoureux, who prefer it.

Francis Collins: six premises

Collins led the Human Genome Project and later directed the National Institutes of Health.  His 2006 book The Language of God gave the view its most widely read statement.  He calls theistic evolution "the dominant position of serious biologists who are also serious believers," and lists six premises on which "a typical version rests."⁴⁴  The four that bear on this article are:

2. Despite massive improbabilities, the properties of the universe appear to have been precisely tuned for life. 3. While the precise mechanism of the origin of life on earth remains unknown, once life arose, the process of evolution and natural selection permitted the development of biological diversity and complexity over very long periods of time. 4. Once evolution got under way, no special supernatural intervention was required. 5. Humans are part of this process, sharing a common ancestor with the great apes.⁴

The sixth premise is that humans are "also unique in ways that defy evolutionary explanation and point to our spiritual nature," above all in the Moral Law and the universal search for God.⁴  From these Collins draws his synthesis:  "God chose the elegant mechanism of evolution to create microbes, plants, and animals of all sorts. Most remarkably, God intentionally chose the same mechanism to give rise to special creatures who would have intelligence, a knowledge of right and wrong, free will, and a desire to seek fellowship with Him."⁴

Two of Collins's later statements matter for the tests below.  On guidance, he writes that "from God's perspective the outcome would be entirely specified," while "from our perspective ... this would appear a random and undirected process."⁴  In a 2011 book with the physicist Karl Giberson, he calls this "divinely guided evolution" and says it "requires no 'intrusions from outside' for its account of God's creative process, except for the origins of the natural laws guiding the process. We do not rule these out, of course, nor claim they are not possible."¹⁴  On Adam, Collins sets out "the real dilemma for the believer":  whether Genesis 2 describes "a special act of miraculous creation that applied to a historic couple," or is "a poetic and powerful allegory of God's plan for the entrance of the spiritual nature (the soul) and the Moral Law into humanity."⁴  He advises believers to "be wary of staking out any position too strongly," and two pages later says the story of Adam and Eve does "not carry that same historical ring" as the eyewitness history of the Bible.⁴⁴

Denis Lamoureux: evolution is teleological, and Adam never existed

Lamoureux holds doctorates in theology and in biology and taught science and religion at St Joseph's College in the University of Alberta.  He is the most candid of the theistic evolutionists about what the view costs.  He insists that evolution is purposeful, not blind:  "the God of the Bible created the world through teleological evolution, which is a design-reflecting natural process that He ordained and sustained."¹  In his longer work he states the contrast with other Christian views:  "Evolutionary creationists reject dramatic miracles in origins. They argue that God created everything by ordaining and sustaining a teleological evolutionary process."¹⁵

On Adam he does not hedge.  Paul, he grants, "believed that Adam existed," but Paul's belief rests on "the origins science-of-the-day for Paul and his readers."¹  So:  "To conclude, there is no sin-death problem. Adam never existed, and consequently, sin did not enter the world through him. Nor then did physical death arise as a divine judgment for his transgression, because once again, Adam never existed."¹  The doctrine of original sin goes with him, and Lamoureux says its loss is "ultimately not relevant to Christian faith or to our daily walk with the Lord."¹

Walton, Swamidass and Craig: three ways to keep a real Adam

Three writers who accept common descent, or allow it, keep a historical Adam by different routes.  Their positions are argued in full in Adam, the Fall, and Death;  only their claims are given here.

1. John Walton, an Old Testament professor at Wheaton College, holds that "Adam and Eve should be considered real people in a real past," but that the forming accounts are archetypal, not accounts of how two individuals were made.⁵⁵ So Genesis 2 "does not, however, make definitive claims about the material origins of either Adam or Eve," and if science shows that humans were not created de novo, "we could not necessarily claim that the Bible contested that evidence."⁵
2. Joshua Swamidass, a physician and computational biologist, argues that a couple "could have lived as recently as six thousand years ago in the Middle East. They could have been de novo created, the first 'humans' of Scripture, free of death in a sinless environment," while everyone outside the garden came from "a providentially governed process of common descent."⁷⁷ He presents this as a hypothesis that fits the science, not as his own belief.⁷
3. William Lane Craig, a philosopher, reads Genesis 1 to 11 as "mytho-history" and identifies Adam and Eve with "the last common ancestor of Homo sapiens and Neanderthals ... living more than 750,000 years ago."⁶⁶

Strictly, Swamidass's de novo couple is not theistic evolution as Collins defines it, because a couple made without parents is a "special supernatural intervention" after evolution got under way.⁴  It is a hybrid, and it matters below because it shows what a recent Adam costs.

The 2017 critique: Meyer's three meanings and Grudem's twelve claims

In 2017 Crossway published Theistic Evolution:  A Scientific, Philosophical, and Theological Critique, edited by J. P. Moreland, Stephen Meyer, Christopher Shaw, Ann Gauger and Wayne Grudem.  It runs to more than a thousand pages and has twenty-five contributors.  Its editors wrote their own one-sentence definition of the view they oppose:  "God created matter and after that did not guide or intervene or act directly to cause any empirically detectable change in the natural behavior of matter until all living things had evolved by purely natural processes."²  The words "empirically detectable" matter in Part 4.

Stephen Meyer's opening chapter separates three meanings of "evolution":  change over time, universal common descent, and the creative power of an undirected mechanism.  These give three versions of theistic evolution, "one of which is trivial, one of which is contestable but not incoherent, and one of which appears deeply problematic."²  Only the third is the target.

Wayne Grudem's theological chapter lists twelve claims that, he says, theistic evolution makes and Genesis 1 to 3 denies, if Genesis is read as history.²  The first eight concern Adam and Eve:  that they "were not the first human beings (and perhaps they never even existed)," that they "were born from human parents," that they "were never sinless," and that "Human death did not begin as a result of Adam's sin, for human beings existed long before Adam and Eve and they were always subject to death."²  The last four concern God's direct creation of kinds, his rest, an originally "very good" world, and a curse that "changed the workings of the natural world."²  Grudem's verdict is that "each of these twelve claims contradicts one or more parts of the text in Genesis 1-3, if it is understood as historical narrative."²

Grudem's list is a fair summary of Lamoureux, Enns and the Collins of 2006.  It is less fair to Walton, Craig and Swamidass, each of whom denies some of the twelve and affirms others.  Two items, the direct creation of Eve from a rib and the direct creation of each kind, are Grudem's own inferences from Genesis, not denials any of these writers has to make in those words.  The tests below therefore score each version separately.

Part 2: Test 1, the historical Adam

What the doctrine needs

The sibling article Adam, the Fall, and Death argues the exegesis in full.  Its finding, in one paragraph:  Paul sets "one man" against "one man" in Romans 5 and calls Adam "the first man" in 1 Corinthians 15:45, and his argument is a causal chain, not a comparison.  Sin entered through one man, and death spread to all through that sin.  Joseph Fitzmyer, a Jesuit commentator who reads Adam in Genesis as a symbol, still concludes that the symbolic reading of Romans "does violence to the contrast that Paul uses," so "Paul has historicized the symbolic Adam of Genesis."¹⁶  James Dunn argues that Paul's point would survive if Adam were mythic, and that is the strongest reply on the other side.¹⁷  Even Peter Enns, who rejects a historical Adam, grants that "For Paul's analogy to have any force, it seems that both Adam and Jesus must be actual historical figures."¹³

The traditional doctrine has five parts.  Adam was a real man.  He was the first human.  All humans descend from him and Eve.  He lived within the time the genealogies of Genesis allow.  He was made directly by God, without human parents.  A sixth part, that human death began with his sin, is Test 2.

How each version scores

Lamoureux and Enns give up every part, and both say so.  Enns writes that this means "leaving behind Paul's understanding of the cause of the universal plight of sin and death," and that this "is the burden of anyone who wishes to bring evolution and Christianity together."¹³  Lamoureux treats Paul's belief in Adam as ancient science, like the three-tier universe of Philippians 2.¹

Collins in 2006 wrote that "approximately ten thousand ancestors gave rise to the entire population of 6 billion humans on the planet."⁴  On that figure no single couple is the ancestor of all.  He also argued that Cain's wife shows "other humans present," and that the believer's choice lies between a miraculous historic couple and an allegory.⁴  He leans to the allegory without committing to it.⁴⁴

Walton keeps a real Adam and Eve and says nothing in Genesis 2 settles where their bodies came from.⁵⁵  His position avoids conflict with genetics by holding that the text makes no claim genetics could test.  Its cost is in the reading of Genesis, which is the parent article's question.

The two ways to keep a sole ancestor, and what each costs

Craig keeps the most.  Adam is "a progenitor of the entire human race through whose disobedience moral evil entered the world," and Adam and Eve are the "two sole genetic progenitors of the human race."⁶  To do it he moves them back.  "Adam plausibly lived sometime between around 1 million years ago to 750,000 years ago."⁶  He grants that "the biblical genealogies famously total a scant 1,656 years from Adam until the Flood, with another 367 years from the Flood to the call of Abraham."⁶  His solution is to read Genesis 1 to 11 as "mytho-history, not to be taken literally."⁶  And his Adam is not a product of evolution alone.  "Out of this population, God selected two and furnished them with intellects by renovating their brains and endowing them with rational souls."⁶  That is a special act of God at the origin of humanity, the kind of act Collins's fourth premise says was not needed once evolution got under way.

Swamidass keeps a recent Adam made without parents.  He can do it because of a fact about ancestry.  A person who lived a few thousand years ago and had children can become a genealogical ancestor of everyone alive today, while contributing almost none of their DNA.  The cost is a world of people outside the garden, descended by "a providentially governed process of common descent," whom Genesis never mentions.⁷  Those people are the answer to the question of Cain's wife.  Swamidass also needs a de novo creation of Adam and Eve, which is again a special act of God, not evolution.⁷

The genetic question behind both options is whether a single couple could be the sole genetic ancestors of all humans.  The sibling article Adam, Eve, and Human Genetic Diversity works it through.  BioLogos once published Dennis Venema's claim that "There is in fact no point in our evolutionary history over the last 18 million years or more where our lineage was reduced below 10,000 individuals."  BioLogos deleted that sentence in 2021 and replaced it with "the genetic data in fact rules out a single couple for at least half a million years."¹⁸  Swamidass's own review puts the limit at "about 500,000 years (not 18 million)."¹⁹  Ola Hössjer and Ann Gauger modelled a single couple about 500,000 years ago and fitted the data.²⁰  So the genetics, on mainstream assumptions, allows a sole couple at about half a million years and rules one out at a few thousand.  That is why Craig puts Adam where he does.

That finding is a problem for the young-earth reading as well, and this article does not hide it.  The sibling article finds that no young-earth calculation has yet met the two strongest measurements, the HLA-DRB1 lineages and the median time to four ancestral lineages of 495,000 years.¹⁹

Verdict on Test 1

A theistic evolutionist can keep a real Adam.  Walton, Craig and Swamidass each do it in print, and none of them is incoherent.  What no version keeps is the whole traditional doctrine:  a first man, the sole ancestor of all, made without parents, within the time of the genealogies.  Every version gives up at least one of those, and the two that keep the most, Craig's and Swamidass's, keep them by adding a special act of God at the origin of humans.  That act is not evolution.  It is the one point at which both of them leave Collins's definition of theistic evolution behind.

Part 3: Test 2, death before the Fall

Two different charges

The charge that theistic evolution puts death before the Fall is really two charges, and they have very different strength.  The first is that humans died before the first human sin.  The second is that animals died, suffered and preyed on each other for hundreds of millions of years before any human existed.  The sibling article Death Before the Fall argues the texts on animal death and Romans 8.  Only its findings are used here.

Human death: the cost every version pays or defines away

Romans 5:12 says that "sin entered the world through one man and death through sin, and so death spread to all people because all sinned" (NET).  C. E. B. Cranfield, whose commentary on Romans is the most careful modern treatment of the Greek, reads the verse this way:  "According to this verse, human death is the consequence of human sin."²¹  First Corinthians 15:21 makes the same link and shows that bodily death is meant:  "For since death came through a man, the resurrection of the dead also came through a man."  The resurrection Paul defends in that chapter is bodily, so the death it reverses is bodily too.

If humans descend from earlier primates, creatures in our line died for millions of years before any first sin.  Each version of theistic evolution has to say what that does to Paul's sentence.

1. Lamoureux and Enns say Paul was wrong about the cause. Lamoureux: "Nor then did physical death arise as a divine judgment for his transgression, because once again, Adam never existed."¹ Enns calls this "leaving behind Paul's understanding of the cause of the universal plight of sin and death."¹³
2. Collins's 2006 position has a population of about ten thousand ancestors and no first pair.⁴ On it, humans were dying before any sin that could be called the first. Grudem's seventh claim fits it exactly: "Human death did not begin as a result of Adam's sin, for human beings existed long before Adam and Eve and they were always subject to death."²
3. BioLogos leaves room for a historical Adam inside a population. Its president, Deborah Haarsma, writes in her reply to the 2017 volume that "other evolutionary creationists affirm Adam and Eve as historical figures within a larger population of humans, and the first to commit human sins."²² On that view the other humans in the population were mortal before Adam sinned. So the death that "spread to all people" through Adam's sin has to be something other than bodily death. BioLogos says so on its own page about death before the Fall: "For humans, Genesis 3 and other Bible passages may be speaking primarily of spiritual death, not physical death."⁸ The same page argues that even in 1 Corinthians 15 "death" refers to "much more than mere physical death."⁸ "Much more than" concedes that physical death is included. If it is included, the reading leaves Paul saying that bodily death came through a man, which on this view it did not. The spiritual-death reading is a revision of Romans 5:12, not a reading of it.
4. Craig and Swamidass can say that no human died before Adam, but only by a definition. For Craig, the creatures before Adam and Eve were not yet human; the "radical transition effected in the founding pair" lifted them "to the human level."⁶ For Swamidass, those outside the garden "could be in the image of God, or not."⁷ If they are not human in Scripture's sense, then no human died before Adam. The cost is that on Swamidass's hypothesis the people outside the garden are also our ancestors, so beings from whom every living person descends were dying before the Fall.

No version of theistic evolution reads Romans 5:12 the way Cranfield does and leaves it standing.  Each one either denies Paul's causal claim, narrows the death to spiritual death, or narrows the word "human" so that the deaths before Adam do not count.

Animal death: a weaker charge than the critics make it

Grudem's eleventh and twelfth claims are that God "never created an originally 'very good' natural world in the sense of a world that was a safe environment, free of thorns and thistles and similar harmful things," and that after the Fall God "did not place any curse on the world that changed the workings of the natural world."²  These are fair descriptions of theistic evolution.  The question is whether Scripture teaches what they deny.

The sibling article finds that it does not say so in any verse.  Genesis 1:29-30 gives plants as food, but it neither mentions death nor forbids meat.  Romans 5:12 and 1 Corinthians 15 speak of human death.  Genesis 9:3, Isaiah 11 and 65, and Romans 8 point toward a world without predation, but none of them states that animals did not die before Adam.  Michael Murray, a philosopher of religion, notes that the older theologians who blamed evil on the Fall did so "only to the extent that human creatures are its victims," so that "premodern theists were not much inclined to apply this reasoning to explain pain and suffering in the nonhuman-animal world."²³  Animal death before the Fall is an inference from the most natural reading of several texts.  It is not a stated doctrine, and losing it does not cost a Christian anything the creeds or Paul affirm.

Natural evil: the theodicy cost

There is a real cost here, but it is in theodicy, the defence of God's goodness, not in doctrine.  If God made life by a process that needed "death, pain, predation, and self-assertion" for hundreds of millions of years, then that suffering is part of how God chose to create.  Christopher Southgate, a biochemist turned theologian, accepts this and defends it with what he calls the "only way" argument:  "I hold to the (unprovable) assumption that an evolving creation was the only way in which God could give rise to the sort of beauty, diversity, sentience, and sophistication of creatures that the biosphere now contains."²⁴²⁴  The word "unprovable" is his own.

Garrett DeWeese's chapter in the 2017 volume presses the point.  He argues that the theistic-evolution answer to natural evil "is seriously inadequate since it leaves God directly responsible for natural evil," and that "If natural evil is of necessity a part of evolutionary history, and if evolution is the process instituted by God, then it follows that God is the direct cause of natural evil."²²  That conclusion is DeWeese's inference.  No theistic evolutionist he cites says it in those words.

Denis Lamoureux's reply turns the argument around.  Grudem's curse, he writes, has "one simple and fatal problem":  "If indeed human sin is the reason that God launched the cosmic fall, then humans should appear in the fossil record before the appearance of the many deleterious and deadly events and creatures listed by Grudem above."²⁵  That objection hits any critic of theistic evolution who accepts the mainstream age and order of the fossil record.  It does not hit the young-earth reading, which places the whole fossil record after the Fall.  But that reading pays its own price, set out in the sibling article:  every fossil predator, tumour and venom gland has to date from after Adam.  Murray puts the dilemma sharply from the other side.  If animal suffering is the worry, "Wouldn't God secure all the relevant goods and avoid a massive array of evil simply by creating the universe in much the way the young-universe creationist believes it was created?"²³

Verdict on Test 2

The doctrine that human death came into the world through the first man's sin cannot be held unrevised by any theistic evolutionist.  The doctrine that animals did not die before the Fall is not a stated doctrine, and theistic evolution loses little by denying it.  The problem of natural evil is sharper for theistic evolution than for the young-earth view, and theistic evolutionists answer it with arguments their own authors call unprovable.  It is shared in full by every old-earth critic of theistic evolution.

Part 4: Test 3, guided versus unguided

What "random" means in the science

The scientific claim behind "random mutation" is narrow.  A 2022 paper in Nature states it in one line:  "Since the first half of the twentieth century, evolutionary theory has been dominated by the idea that mutations occur randomly with respect to their consequences."⁹  Random here means that a mutation is no more likely to happen because it would help the organism.  It does not mean uncaused, and it does not mean unplanned.

That claim can be tested, and it has been.  In 1943 Salvador Luria and Max Delbrück grew many separate cultures of bacteria and then exposed them to a virus.  If resistance arose in response to the virus, every culture should have about the same small number of resistant cells.  Instead, "the resistant bacteria appear in similar cultures not as random samples but in groups of varying sizes," which showed that the resistance mutations had arisen at random times before the virus was ever present.²⁶  The 2022 Nature paper, by J. Grey Monroe and colleagues, challenged the claim in the weed Arabidopsis.  They found "mutation frequency is reduced by half inside gene bodies and by two-thirds in essential genes," and concluded that this was "challenging the prevailing paradigm that mutation is a directionless force in evolution."⁹  Other geneticists disputed the finding in the same journal in 2023, arguing that it came from errors in calling mutations.²⁷  Either way, the bias Monroe reported comes from how the plant marks and repairs its DNA.  It lowers harmful mutations in genes that already exist.  It does not anticipate new functions.

What "unguided" means beyond the science

Several leading biologists went further and wrote purpose out of the process.

1. George Gaylord Simpson, 1967: "Man is the result of a purposeless and natural process that did not have him in mind. He was not planned."¹⁰
2. Jacques Monod, 1972: "Pure chance, absolutely free but blind, at the very root of the stupendous edifice of evolution: this central concept of modern biology is no longer one among other possible or even conceivable hypotheses."¹¹
3. Douglas Futuyma's standard textbook, 1998: "By coupling undirected, purposeless variation to the blind, uncaring process of natural selection, Darwin made theological or spiritual explanations of the life processes superfluous."²⁸
4. The National Association of Biology Teachers, which until 1997 described evolution as "an unsupervised, impersonal, unpredictable and natural process."¹²

None of these sentences reports a measurement.  No experiment can show that a process had no purpose behind it.  The philosopher Alvin Plantinga makes the point:  the claim that the process is unguided "is no part of the scientific theory as such; it is instead a metaphysical or theological add-on."¹²  He grants that "a veritable choir of extremely distinguished experts" insists otherwise.¹²  A skeptic can reply that the burden lies with whoever adds a cause the data do not need.  That is a fair reply, and it is a point about parsimony, not a result of biology.

Meyer's three charges, one at a time

Stephen Meyer argues that the third kind of theistic evolution, the kind that accepts the creative power of an undirected mechanism, "generates either (1) logical contradictions, (2) a theologically heterodox view of divine action, or (3) a convoluted and scientifically vacuous explanation."²  Each charge has a real target.  None of them hits every theistic evolutionist.

Charge 1, contradiction.  "Logically, no intelligent being—not even God—can direct an undirected process. As soon as he directs it, the 'undirected' process would no longer be undirected."²  The logic is sound if "undirected" means directed by no one.  It fails if "undirected" means random with respect to fitness, because God could direct a process whose mutations are random in that sense.  BioLogos uses "guide" in the second way.  Jim Stump, its senior editor, writes:  "Yes, we believe God guides evolution, the same as we believe God guides photosynthesis."²⁹  Haarsma writes that "God guided evolution just as much as God guides the formation of a baby from an embryo."²²  Grudem replies that this is "a misleading use of the word 'guide,'" which in ordinary speech means changing the direction of something.³⁰  So the first charge becomes a dispute over a word.  The 2017 volume's own definition denies only guidance that causes an "empirically detectable change."²  Gregg Allison's chapter in the same volume grants that "only some varieties of theistic evolution deny that the process was directed," and that Collins does not say which.²  Stump makes the same point from the other side:  the critics "seem to mean by 'guide' something in the neighborhood of 'causing empirically detectable changes.'"²⁹

Charge 2, a God who does not direct and cannot know.  Meyer argues that if God only sustains natural law, "he could not know, and does not know, what those mechanisms would (or will) produce."²  This charge has a real target in Kenneth Miller, a Brown University biologist and Catholic.  Miller writes that "mankind's appearance on this planet was not preordained, that we are here ... as an afterthought, a minor detail, a happenstance in a history that might just as well have left us out. I agree."³¹  He then argues that a lack of inevitability is not "incompatible with a divine will."³¹  John West's chapter in the 2017 volume reports the astronomer George Coyne, formerly of the Vatican Observatory, saying that "not even God could know ... with certainty" that human life would arise.²  Those positions do limit God's knowledge or control, and Meyer is right that they sit badly with the biblical claim that God foreknew and intended humanity.  But Collins holds the opposite:  "from God's perspective the outcome would be entirely specified."⁴  So does Josh Reeves, writing for BioLogos:  "Chance in natural selection would be real because the outcome was not determined by its preceding causes, but God nevertheless has chosen its result."³²  The second charge hits Miller and Coyne.  It does not hit Collins, Reeves or Lamoureux, who writes that "With unfathomable foresight, the Creator set in motion and upheld over time intelligently designed self-assembling natural processes, including random processes."²⁵

Charge 3, a vacuous explanation.  Of Collins's view Meyer writes that it "adds nothing to our scientific understanding of what caused living organisms to arise," and "has no empirical or scientific content beyond that offered by strictly materialistic evolutionary theories."²  That is an accurate description.  Collins designed the view to be scientifically identical to its rival:  "from our perspective ... this would appear a random and undirected process."⁴  BioLogos accepts the consequence openly:  "We do not propose a special Christian version of scientific facts."³

Does guided evolution predict anything?

This is the question the whole dispute turns on.  There are three ways God might guide evolution, and they give different answers.

1. Guidance through ordinary causes, known and willed from outside time. This is the view of Collins, Reeves, Stump and Haarsma. It predicts nothing that unguided evolution does not, by design. Every mutation and every selective event would look exactly as it would if no one had willed it.
2. Guidance through events that physics leaves open. Robert John Russell, a physicist and theologian, argues that "chance in evolution, at the level of quantum mechanics underlying genetic mutation, is a sign not of epistemic ignorance but of ontological indeterminism," so that nature is "genuinely open to objective special providence without being forced into interventionism."³³ Miller makes a similar case.³¹ On this view God picks outcomes that physics leaves undecided. Because a quantum outcome chosen by God and one left to chance look the same, this also predicts nothing extra, unless God's choices were frequent and lopsided enough to show up as a statistical bias.
3. Guidance that steers particular mutations. The philosopher Robin Collins, in an essay hosted by BioLogos, proposes that "God could guide the evolutionary process by mutating some gamete or even adding new information to the gametes," and says "I also consider it an open question as to whether God's guidance of the evolutionary process is detectable."³⁴ Craig's Adam, whose brain God renovates, perhaps by "a regulatory genetic mutation," is a case of this kind.⁶ Meyer is right that this is "a decidedly non-Darwinian conception" and a form of intelligent design.² It is the only kind that predicts anything: mutations that anticipate future function more often than the known mechanisms allow. The tests that exist, from Luria and Delbrück to Monroe, have found no mutation bias that looks ahead. So this kind of guidance, if it happened, was rare or happened where no one has looked.

The finding of this Part follows.  Guidance of the first two kinds is coherent, orthodox, and empty of scientific content.  So theistic evolutionists cannot cite biology as evidence that God guided evolution, and their critics cannot cite biology as evidence that God did not.  The secular writers who say that biology shows the process had no purpose are making Plantinga's "add-on," not reporting a result.  The design theorists who say that hidden guidance is no guidance are making a different mistake:  they are treating "not detectable by science" as "not real."  Meyer avoids that mistake when he calls the view scientifically "vacuous."  He does not show that it is false.

The real cost: design in biology

The cost of hidden guidance is philosophical, and Collins pays it in his own book.  On the universe he writes that "the Anthropic Principle certainly provides an interesting argument in favor of a Creator."⁴  On biology he writes that intelligent design "is a 'God of the gaps' theory, inserting a supposition of the need for supernatural intervention in places that its proponents claim science cannot explain."⁴  Both arguments have the same form.  Each says a feature of nature is very improbable on chance and law alone, and more probable if a mind intended it.  Collins accepts one and rejects the other.

There is a principled reply open to Collins, and it is worth stating at full strength.  For the physical constants no mechanism is known that could produce their values, whereas for living things natural selection is a known mechanism that does produce adaptation.  A design argument is weaker where a mechanism already explains much of what is seen.  The design theorist's answer is C. John Collins's distinction, in the 2017 volume, between "gaps due to ignorance," which may close, and "gaps due to the nature ... of the things involved," where "the result goes beyond what these natural properties would have brought about."²  Whether the origin of new genes and body plans is a gap of the first kind or the second is a scientific question.  This article does not settle it.  The design arguments are taken up in The Design Inference.

Verdict on Test 3

The charge that "guided evolution" is a contradiction in terms fails against Collins and BioLogos, because they do not mean by "unguided" what Meyer means.  It succeeds against Miller and Coyne, who do limit what God intended or knew.  Hidden guidance is coherent and predicts nothing.  Its price is that a theistic evolutionist gives up biology as evidence for God while keeping cosmology, and Collins has not shown why one is allowed and the other is not.

Part 5: What is kept, what is revised, what is given up

The ledger

Three things stand out.  First, the creedal core is kept by every version.  Allison grants that the Nicene Creed's "maker of heaven and earth" does not address evolution.²  Second, every version gives up or revises at least one part of the doctrine of Adam and the Fall, and the more of Adam a version keeps, the more special divine action it needs at human origins.  Third, the losses are not all equal.  Animal death before the Fall is an inference, not a stated doctrine.  Human death through Adam's sin is stated by Paul.

The strongest case for theistic evolution

The case for theistic evolution is strong, and it deserves to be stated at full strength before the verdict.

1. The evidence for common descent, including human descent, is strong. The sibling article Shared Genetic Errors finds that shared broken genes and mobile-element insertions are the strongest genetic evidence for common ancestry. Cases where the same change arose twice independently are real, but they are measured at 0.05% to 3% of markers, far too rare to explain the pattern. A theistic evolutionist accepts this evidence because, as Giberson and Collins put it, "we hold this position because it is most consistent with what science has discovered about the world."¹⁴
2. God working through natural causes is classical Christian teaching. No Christian thinks that because embryology explains how a baby forms, God did not form the child. Haarsma's analogy is exactly that, and Giberson and Collins argue that "God's creative work can be done through the laws of nature, and not merely by breaking or suspending those laws."¹⁴²²
3. The creeds do not settle the question. The Nicene Creed confesses God as "maker of heaven and earth" and says nothing about how. Allison concedes that "not all early Christians interpreted Genesis 1 literally (Origen, for example, did not)."²
4. Tying faith to gaps is risky. Collins warns that "Advances in science ultimately fill in those gaps, to the dismay of those who had attached their faith to them."⁴ A faith that does not depend on any particular gap cannot be refuted by the next discovery.
5. The critics' own definition concedes the main philosophical point. The 2017 volume opposes only a God who causes no "empirically detectable change."² A God who acts without detectable trace is still a God who acts.

This case is strong on the science, on the doctrine of providence, and on the creeds.  It does not touch the question of Adam and death.  Points 1 to 5 can all be true and Paul's argument in Romans 5 still needs a man through whom death came.

What a theistic evolutionist has to say plainly

An honest statement of the position, built from its own best authors, looks like this.  God made all things and governs every event, including every mutation.  Humans descend from earlier primates.  Either Adam is a figure of ancient thought, as Lamoureux holds, or he is one man chosen or made within or alongside a population, as Haarsma, Craig and Swamidass allow.  Either way, bodies died before the first sin, so the death that came through Adam is not, or not only, bodily death.  Paul believed otherwise, and on this view Paul's belief about the cause of death was part of the science of his day.  Enns and Lamoureux say all of this.  The position is coherent, and many Christians hold it.  What it cannot claim is that nothing Paul taught has been revised.

Objections and responses

1. "The 2017 volume attacks a view nobody at BioLogos holds." Haarsma's reply says of the definition: "No one at BioLogos would describe God's action that way!" and calls it "basically deism."²² Swamidass, reviewing the book in Themelios, writes that "A non-intervention understanding of evolution, however, is a minority position among Christians that affirm evolutionary science."³⁵ Response: partly right. The definition denies only detectable action, and Grudem's reply shows that BioLogos's own wording does not conflict with it: "Haarsma's new definition does not actually conflict with our definition."³⁰ The dispute over "guide" is largely verbal, as Part 4 shows. But on Adam the volume describes BioLogos accurately. Haarsma writes that "evolutionary creationists cannot affirm the traditional de novo view of human origins."²² Swamidass writes: "Grudem is correct; theistic evolutionists have not been sensitive to the theological concerns presented here."³⁵
2. "Paul accommodated the science of his day, as he did with a three-tier universe." This is Lamoureux's view.¹²⁵ Response: it is coherent, but it concedes the point at issue. Paul's claim about Adam is not a passing image. It carries his argument for why all die and why all will be raised (1 Cor 15:21-22). If that claim is ancient science, then Paul's account of the cause of death is wrong, which is what Enns says.¹³
3. "Romans 5 is about spiritual death, so evolution is no threat to it." Response: 1 Corinthians 15:21 pairs the death that came through a man with "the resurrection of the dead," and the resurrection is bodily. BioLogos itself says the death in view is "much more than mere physical death," which includes physical death.⁸ The spiritual reading can be held, but it narrows Paul.
4. "The young-earth view cannot explain the genetics either." Response: true, and this article says so. The sibling article finds that no young-earth calculation has met the HLA-DRB1 lineage count or the 495,000-year median for four ancestral lineages.¹⁹ That is a problem for the young-earth reading, not a defence of theistic evolution's reading of Paul. The two questions are separate.
5. "Hidden guidance is unfalsifiable, so it means nothing." A skeptic will say this. Response: unfalsifiable by science is not the same as meaningless. The claim that the process had no purpose is equally beyond any experiment, and Plantinga calls it "a metaphysical or theological add-on."¹² Both claims are philosophical, and both must be argued on philosophical grounds.
6. "Collins is consistent: fine-tuning has no mechanism, adaptation does." Response: that is the best reply, and Part 4 states it. It turns the question into whether known mechanisms can produce new genes and body plans, which is a scientific question this article leaves to The Design Inference and the biology articles.
7. "Calling creation very good rules out predation, so theistic evolution denies the goodness of creation." Response: the sibling article Death Before the Fall finds that no verse says so, and that church history is divided on it. The real cost is in theodicy, and Southgate's answer rests on what he calls an "(unprovable) assumption."²⁴
8. "Craig and Swamidass show that evolution and a historical Adam fit together." Response: they show that common descent and a historical Adam fit together when God acts specially at human origins. Craig's God "selected two and furnished them with intellects by renovating their brains."⁶ Swamidass's Adam may be "de novo created."⁷ Neither is evolution alone, and both pay for Adam with something else: Craig with the genealogies, Swamidass with a world of people Genesis does not mention.
9. "B. B. Warfield and other conservative theologians accepted evolution." Response: Allison's chapter in the 2017 volume argues that "None of them explicitly embraced theistic evolution as this book defines it," and that none of them denied that "all human beings have descended from Adam and Eve."² Allison is a critic of theistic evolution, so this is his reading. On it, the older conservatives who allowed evolution kept a first pair.
10. "This is a slippery slope to liberal theology." Collins answers that "mature observers are used to living on slippery slopes and deciding where to place a sensible stopping point."⁴ Response: a slope argument is about consequences, not truth, and this article does not use it. But Lamoureux's books show where one stopping point lies: he gives up original sin and calls the loss "ultimately not relevant to Christian faith."¹

Issues and unanswered questions

1. The Miller and Levine textbook sentence. Meyer quotes Miller and Levine's Biology as saying that "evolution works without either plan or purpose."² The archive.org copy of the 1998 textbook is teacher's material and does not contain the sentence, and Meyer's own footnote and bibliography give different editions. The sentence is reported here only as Meyer's quotation. Miller's own book is quoted instead.
2. The NABT statement of 1995. The text of the statement and its 1997 revision were not obtained. The wording is quoted from Plantinga's note.
3. George Coyne's words are quoted through West's chapter in the 2017 volume, not from Coyne's own text.
4. The Monroe dispute. Whether mutation in Arabidopsis is biased toward sparing important genes is contested in Nature: Monroe and colleagues reported it in 2022, Wang and colleagues disputed it in 2023, and Monroe's group replied the same year. The article's argument does not depend on the outcome, because the bias, if real, is a property of DNA repair, not of foresight.
5. Walton on human death. The quoted passages of Walton do not settle whether humans died before Adam on his view, so the ledger leaves the cell open.
6. BioLogos's work on divine action. Haarsma wrote in 2018 that "one sentence cannot fully answer the questions people have about God's action in the evolution of living things," and that "the mirror of this book is accurately showing a place we can improve."²² A later BioLogos page titled "Did God guide evolution?" could not be found in any archive.
7. Whether design can be detected in biology is the question that decides Part 4's last section. It is argued elsewhere and not settled here.
8. The four-way comparison with young-earth creation, old-earth creation and intelligent design is planned as Young Earth, Old Earth, Design, and Evolutionary Creation.

Verdict

Can a Christian accept evolution, including human common descent from earlier primates, without giving up anything essential to Christian doctrine?  If "essential" means the ancient creeds, yes.  God as maker of all things, the Trinity, the incarnation, the bodily resurrection and the need of every person for Christ all survive, and BioLogos affirms each of them.

If "essential" includes the doctrine of Adam that Paul uses to explain why humans die, no.  Every version of theistic evolution gives up or revises some part of it.  Lamoureux and Enns give up Adam and say that Paul was mistaken about the cause of death.  BioLogos leaves Adam open and reads the death of Romans 5 as mainly spiritual.  Walton keeps Adam and leaves his origin to science.  Craig and Swamidass keep the most, a sole ancestor or a recent de novo pair, and pay for it with a special act of God at human origins and either the genealogies or a population Genesis never mentions.  No version keeps, without redefining death or "human," Paul's plain claim that bodily death came into the human world through one man's sin.

The charge that guided evolution is a contradiction holds against Miller and Coyne and fails against Collins and BioLogos.  Guidance hidden inside random mutation is coherent and predicts nothing, so biology can neither confirm nor refute it.  The price of that position is the design argument from biology, which Collins rejects while keeping the parallel argument from cosmology.  That inconsistency is the most serious weakness of his version, but it is philosophical, not doctrinal.

So the honest answer is that theistic evolution is compatible with the creeds and incompatible with the traditional doctrine of Adam and the Fall.  Its most candid defenders agree.  Whether that doctrine is "essential" is the question on which Christians divide, and the article leaves it to the reader with the costs stated.

Sources