Taking Religion Seriously
-
Speaking of exercises in steel manning, the notorious Charles Murray, of Bell Curve fame, has written a new book, trying to make a case for the Christian mythos being literally true. Heterodox writer Bo Winegard reviews it:
Written by Bo Winegard.
Concise and engaging, Charles Murray’s Taking Religion Seriously manages to be at once endearingly humble and impressively ambitious. Humble, because it is brief and tempered by the author’s self-aware confessions of limited expertise and spiritual authority. Ambitious, because it attempts an intellectual defense of a literalist Christianity¹, rather than a poetic or moral one. For Murray, Christianity is more than metaphor or ethical inspiration. It is true.
As Shakespeare’s famous Dane said, Aye, there’s the rub.
In my review of Ross Douthat’s Believe, I argued that a literalist Christianity is no longer tenable for intellectuals in a post-Enlightenment world. The faith of Aquinas or Luther, however inspiring, must remain foreign to us. Murray has not changed my view, though he, like Douthat, undertakes a bold (and in my judgement, misguided) defense of several metaphysical claims of Christianity: that the world was created by a transcendent God; that materialism is incomplete; that the moral law is divine and objective; that Jesus has a special relationship to God and may have been resurrected.
I claim this defense is misguided not because I’m a fire-breathing atheist. Quite the contrary. Like Murray, I revere Christianity. Rather, I claim this defense is misguided because Christianity is better understood as a mythopoetic vision than as a literal description of the universe. More Wordsworth² than Aquinas or Calvin. Many of Murray’s arguments are unpersuasive and thus highlight the futility of defending a literalist Christianity. Since I not only find his arguments unconvincing but also counterproductive, much of this review will be critical, but that should not be mistaken for a negative judgment of the book.
Take Religion Seriously is written with admirable clarity, so its arguments are easy to follow and assess. One can disagree with them without disliking the work; and though I found myself disputing Murray throughout, I also found myself eager for each new chapter. What is more, the book is not just a collection of academic arguments. It is also the story of an honest and intelligent man’s journey from agnosticism to faith. Even if one cannot accept the particulars of that faith, one can enjoy reading about it.
Murray’s first argument is a familiar one to Christian apologists. The universe appears designed (fine-tuned) for life. And if it is designed, then it must have a designer:
The universe was created by an unknowable creative force that itself has no explainable source. Aristotle’s unmoved mover. By the late 1990s, that sounded to me like a description of God I could accept … I decided that the entity that I might as well call God had deliberately designed the universe to permit the existence of life.
The philosopher Alvin Plantinga forwarded a lucid account of this argument:
Recently a number of thinkers have proposed a new version of the argument from design, the so-called “Fine-Tuning Argument.” Starting in the late Sixties and early Seventies, astrophysicists and others noted that several of the basic physical constants must fall within very narrow limits if there is to be the development of intelligent life—at any rate in a way anything like the way in which we think it actually happened. For example, if the force of gravity were even slightly stronger, all stars would be blue giants; if even slightly weaker, all would be red dwarfs; in neither case could life have developed. The same goes for the weak and strong nuclear forces; if either had been even slightly different, life, at any rate life of the sort we have, could probably not have developed.
Although the fine-tuning argument is undoubtedly sophisticated and superficially persuasive, it is ultimately unsound. Four objections. First, it fallaciously infers priori probabilities from posterior evidence. Second, the universe does not, in fact, appear fine-tuned for life. Third, positing a fine-tuner only moves the problem one step back, inviting an infinite regress. And fourth, even granting a fine-tuner, there is no reason to assume benevolence or competence.
Objection one. Imagine being asked to draw a number from an enormous bag of numbered slips. You pull out 123,004. How improbable, you might think. So many numbers it could have been instead! But to know the probability, you must first know the sample space, i.e., the set of all possible numbers that could have been drawn, and the prior distribution, i.e., the weights assigned to each. After all, perhaps every number in the bag was the same. The mere fact that you can imagine countless alternatives does not mean that those alternatives were part of the actual sample space, or that the probability was one in half a million.
The same fallacy underlies the fine-tuning argument. We observe one universe, note that it is suitable for life, and infer that its existence must be astronomically improbable because we can imagine trillions of others by tweaking a constant here or a constant there. But this is like marveling at having drawn 123,004 without knowing what other numbers were in the bag. We possess posterior evidence (our universe), but no prior distribution. We do not know whether the physical constants could have varied at all, or even whether it is coherent to speak of alternative universes. And without a prior distribution, we have no basis for judging how likely or unlikely our universe was. Maybe our universe was the only possible universe and the term possibility does not even make sense when applied to the universe.
Furthermore, even if one grants that this universe is exceedingly improbable, there remains another plausible explanation: the multiverse hypothesis. On this view, there are (or have been) millions or perhaps trillions of universes, and we happen to inhabit the one that permits life because otherwise we would not be here to ask the question! The appearance of fine-tuning, in other words, reflects an observer-selection effect (physicists and philosophers call this the anthropic principle) rather than design.
Of this multiverse hypothesis, Murray writes:
Regarding the second option, Roger Penrose’s ratio tells me that millions of universes are required to have a reasonable chance that one of them permits the development of life. I cannot make myself take that option seriously. I am reminded of Samuel Johnson’s response when Boswell asked him how he refuted George Berkeley’s hypothesis that material objects do not exist independently of perception. “I refute it thus!” Johnson said, kicking a rock. When it comes to multiverse theory, “I refute it thus,” by looking into a cloudless night sky.
It is fitting that Murray invokes Johnson’s attempted refutation of Berkeley, for he, like Johnson, is not so much refuting the theory as scoffing at it. Berkeley did not deny that kicking a rock would hurt; he denied that such sensations proved the rock existed independently of perception. Likewise, proponents of the multiverse do not deny that the night sky is beautiful or profound; they deny that its beauty constitutes evidence of divine design.
Objection two: The claim that the universe was fine-tuned for life is implausible for another reason. The vast majority of the universe is, in fact, a deadly vacuum: cold, bleak, and uninhabited by anything save dust, gas, and radiation. Most of the universe’s matter is dark; the rest burns in stars or lies frozen and inert in toxic, unlivable worlds.
Even on Earth, life required billions of years to evolve. Humanity appeared only near the end of that long process and only because trillions upon trillions of other organisms perished along the way. The path from single-celled organisms to homo sapiens is paved with corpses. Natural selection, if it is a tool of creation, must be one of the cruelest imaginable.
Is this what we would expect from a creator who designed the universe for life? Hardly. One can, without much imagination, conceive of countless universes that would more convincingly testify to a life-affirming designer: a cosmos filled with thousands of inhabited worlds; creatures able to sustain themselves on nonliving matter rather than on one another; human beings specially and directly created, without the epochs of waste and suffering that natural selection entails. And so on.
So what is the probability of a life-affirming creator given a universe in which life is nearly entirely absent? I do not know, but I do know that there are many imaginable alternatives more friendly to life. In my view, observing life on one tiny planet and concluding that the universe must have been fine-tuned for life would be like observing a small spot of rust on an exposed joint of a skyscraper and concluding the skyscraper must have been designed for rust.
Objection three: The fine-tuning argument ultimately undermines itself. It begins by claiming that the universe and its physical constants are so improbable that they demand an explanation. Yet it then posits an even more improbable entity: an intelligent agent who set those constants himself. If the fine-tuner is sophisticated enough to fine-tune the universe, one must ask who or what fine-tuned the fine-tuner?
The usual reply is that the fine-tuner is a metaphysically necessary being, something more akin to a logical truth than a contingent intelligence. But this is mere assertion, a kind of verbal hocus-pocus that substitutes definition for explanation. The simpler, more parsimonious view is that the universe is self-organizing, that the laws of nature, left to themselves, unfold in precisely the way we observe, without the need for a cosmic engineer.
Objection four: The Gnostics, a sect of early Christian heretics, held that creation itself was a cosmic catastrophe and that salvation lay in escaping the material world with all its indignities and deformities. Examining the copious suffering on the planet, one must concede that the Gnostics had a point.
Leaving aside the contingent calamities that afflict human existence, consider that most suffering is not accidental but essential to life. Because life feeds on life. Prey animals rarely die from old age. And what years they have on earth are full of peril. One gets a sense of this by watching a bird at the feeder. It is is ceaselessly alert, scanning for threats, always vigilant, always afraid. Things are scarcely better for predators, who must constantly seek food and are always haunted by the specter of starvation.
From this, one might reasonably conclude that even if there were a fine-tuner, he is either malicious or incompetent. As Stendhal is said to have joked, “God’s only excuse is that he doesn’t exist.” Witticism aside, the point is that the fine-tuning argument tells us nothing about the character of the fine-tuner. And from what limited evidence we have, that character, if it exists at all, is just as likely to be cruel or indifferent as benevolent.
Murray continues his case for Christianity by arguing that a “materialist explanation of consciousness is incomplete” and that this incompleteness leaves open the possibility that we “have a soul.” He cites three principal sources of evidence against materialism: psi phenomena, terminal lucidity, and near-death experiences. Taken together, he contends, these “add up to proof” that materialism is either mistaken or at least radically incomplete.
Murray’s assessment of the evidence for these phenomena is quite different from mine. He writes:
The binary yes-no question, “Have psi phenomena been proven to exist?” has been answered “yes” with evidence that meets high standards of scientific proof … A subset of near-death experiences amounts to persuasive evidence that is incompatible with a strict materialist theory of consciousness … A subset of terminal lucidity cases amounts to persuasive evidence that is incompatible with a strict materialist theory of consciousness.
My own reading of the evidence for psi phenomena is that it is notoriously fragile, marked by effect sizes so small they could easily be statistical noise. Moreover, positive results tend to vanish under stricter controls or closer scrutiny. If psi were real, if clairvoyance and other similar capacities genuinely conveyed information beyond the ordinary senses, it would confer an enormous evolutionary advantage. Natural selection would have seized upon such a capacity long ago, spreading it widely through the population of animals and humans. Psi would not be coy or elusive. Rather, it would be as obvious as smell or vision. Yet the alleged effects we have appear only in anecdotes or in laboratories under contrived conditions, typically in experiments conducted by researchers who already believe they exist.
Philosophers in the West long ago eschewed supernaturalism and the traditional notion of the soul, not because they were dogmatic materialists or hostile to religion, but because the concept of an immaterial mind proved impossible to reconcile with what we know about the physical world. If our thoughts and feelings belong to an immaterial soul, how do they interact with the matter of the brain? This, of course, is the infamous interaction problem that vexed Cartesian dualism and continues to vex committed dualists to this day. For we know, beyond any reasonable doubt, that the brain is intimately related to consciousness. Disturb or injure the brain and you disturb or injure consciousness. Alter its chemistry and you alter perception, memory, emotion, and even the sense of the self (or belief in God!). Anesthesia makes the point with clarity: whatever consciousness might be, it depends on the functioning of the brain. When the brain stops or enters deep sleep, the mind disappears.³
None of this is to say that materialism has solved the problem of consciousness. On the contrary, consciousness remains puzzling perhaps inexplicable. But positing an immaterial soul to explain the mystery of consciousness merely compounds that mystery by introducing another enigma. Furthermore, the phenomena Murray cites as evidence that materialism is incomplete, i.e., psi, terminal lucidity, and near-death experiences, are of dubious value. At any rate, they are unlikely to convert a skeptic.
Citing C. S. Lewis, Murray next makes the case that human morality points to the divine. Here is Lewis:
If there was a controlling power outside the universe, it could not show itself to us as one of the facts inside the universe—no more than the architect of a house could actually be a wall or staircase or fireplace in that house. The only way in which we could expect it to show itself would be inside ourselves as an influence or a command trying to get us to behave in a certain way. And that is just what we do find inside ourselves. Surely this ought to arouse our suspicions?
And here is Murray:
But whether the culture reinforces or suppresses it, genuine altruism that is not explained by evolutionary biology seems to be a fundamental component of human psychology.
The claim is straightforward. Humans are more altruistic than natural and cultural selection can explain, and this surplus of altruism points to a divine moral law and a God who works within us to reveal it. Yet this runs into an immediate, and I think fatal, problem that Murray seems not to have noticed. Such altruism, wherever it originates, would be selected against and therefore could not flourish within a population. Supernatural intervention cannot rescue altruism from the ineluctable logic of natural selection.
In evolutionary biology, altruism has a precise meaning: behavior that increases another organism’s fitness while reducing the actor’s own. It must be distinguished from cooperation or prosocial behavior, which often benefits all participants. If I help someone hunt, we may both have more food. That is mutualism, not altruism. If I give away a scarce resource to a sister, that might qualify as altruism, at least before one accounts for inclusive fitness.
The problem for Murray’s claim is that genuine altruism is not an evolutionarily viable strategy. Genes that reduce their bearer’s fitness while increasing the fitness of others will not persist. This poses no difficulty for a creationist who denies the mechanism of natural selection, but it is fatal for Murray, who accepts it. A divine moral law cannot override the mathematics of selection.
Perhaps what Murray means is not biological altruism in the technical sense, but rather prosocial behavior more broadly, whatever its underlying fitness consequences. Put differently, his claim is that humans are more cooperative than natural selection alone would predict, and that this excess of cooperation points to a divine moral law. My somewhat facetious and admittedly cynical response is that the world offers little evidence of such grace. Humans are tribal, selfish, and prone to eruptions of bestial violence, a truth Christianity itself has long acknowledged, which is why it speaks of the corruption of the will through original sin.
My more considered response is that Murray is not wholly wrong. Humans do possess a conscience and a remarkably sophisticated moral sense. They are also vastly more cooperative than most other species, certainly more so than other large mammals. This genuine puzzle leads Murray to a divine explanation, perhaps because he is unaware of the extensive literature on the evolution of morality and cooperation. He mentions only kin selection and reciprocal altruism as possible mechanisms and then writes:
But neither of these discoveries explains what the ancient Greeks called agape: unconditional love, focused on giving rather than receiving, not based on merit or acquaintance with the recipient. Theologically, agape has been used to denote God’s love. Applied to humans, the most unambiguous manifestation of agape is kindness, concern, and help for complete strangers.
Yet this is mistaken. A wide range of evolutionary thinkers have offered naturalistic accounts of kindness, concern, and help for strangers, phenomena sometimes termed ultrasociality. Many have pointed to cultural transmission and group conflict as key drivers of human prosociality. Others, in a related vein, have developed coalitional value theories, according to which cooperative individuals earn prestige within their groups, while groups that promote internal trust and altruistic norms outcompete those that do not. In such contexts, highly cooperative behavior is not inexplicable at all; it is adaptive. Morality, under this view, is not sustained by the divine, but by the processes of natural and cultural selection.
Pressing the point, Murray writes:
I was driving down Massachusetts Avenue in Washington, DC, in the mid-1970s when a car a hundred yards in front of me collided with another car and burst into flames. My instantaneous twin reactions were “You have to stop and pull the driver out of the car” and, much louder in my head, “I DON’T WANT TO,” fearing that the gas tank might explode at any moment. I did stop and, with another man who ran over, pulled the driver to safety. I have been relieved ever since that I did. Not stopping would have haunted me for the rest of my life even if someone else had rescued the driver. And in the moment, I did indeed have a warm sense of having done the right thing. The internal command to stop and help wasn’t the result of any evolutionary mechanism that has yet been identified. All the known laws of evolutionary biology say that such behavior has a large negative effect on reproductive fitness.
But this behavior is explicable and in fact does not on average reduce reproductive fitness. People who help others, especially in dangerous situations, are often rewarded with prestige, which enhances their social standing and ultimately their fitness. The human brain therefore evolved to assist others, even strangers in the appropriate contexts. This admittedly faint biological propensity is then amplified by culture norms and institutions, including religious ones.
We celebrate those who help others precisely because such acts require encouragement. But once the norms are in place, once a culture celebrates and rewards the prosocial and punishes the anti-social, the behavior, pace Murray, does not have a large negative effect on reproductive fitness. If it did, it could not have evolved.
The moral law within that inspires Murray, and that inspired many others before him, deserves to be celebrated, but it does not in fact point to a transcendent creator. And, at any rate, however noble the moral law is, it obviously is fragile. Nero, Borgia, Napoleon, Stalin, Hitler, Bundy. The list could continue. An all-powerful and all-loving God, I think, could have done better. Thus the best explanation for terrestrial morality, admirable but frail, remains naturalistic.
Murray concludes Taking Religion Seriously with a discussion of the Gospels and the historical figure of Jesus. This section is the most scholarly in the book, in that it engages questions of authorship and transmission of ancient texts that require specialized training (and expertise in multiple languages). Murray’s ability to explain debates in New Testament scholarship lucidly is therefore admirable, but his presentation is necessarily limited and one-sided.
He is skeptical of the prevailing scholarly consensus about the authorship and dating of the Gospels, dismissing the evidence for it as “wispy.” That his position defies current scholarly opinion does not make it wrong, of course, but I found his arguments unconvincing. Further, even if one were to grant his conclusions about authorship and chronology, they would not establish that the Gospels offer a historically reliable portrait of Jesus.
Traditionally, the four Gospels were attributed to figures closely connected to Jesus or his immediate circle. Mark was said to be the companion and interpreter of Peter; Matthew, one of the twelve apostles and a former tax collector; Luke, the physician and traveling associate of Paul; and John, the “beloved disciple” of Jesus himself. In this traditional view, the Gospels were written between roughly 45 and 70 CE by eyewitnesses or their direct associates.
Modern critical scholarship, which Murray calls “revisionist,” offers a very different account. The earliest manuscripts we possess are anonymous, and most scholars date their composition between roughly 65 and 100 CE. On this view, the Gospels were not written by eyewitnesses or even by close associates of eyewitnesses, but by educated Greek-speaking Christians⁴ who drew upon oral traditions and possibly earlier written sources to construct theologically motivated narratives of Jesus’s life and teachings. This, in part, explains why the Gospels differ so noticeably from one another. They are not disinterested biographies or neutral histories, but works of faith composed by devoted believers whose primary aim was not historical accuracy.⁵
Murray argues that the conventional dating of Mark, around 70 CE, is driven by skepticism that Jesus could have predicted the destruction of the Temple. (For example, Jesus declares, “Not one stone here will be left on another; every one will be thrown down.”)
The conventional dating depends on the assumption that these were not predictions, but passages inserted after 70 to make it look as if Jesus foresaw the destruction. This supposition is also applied to other references in the Gospels that could possibly (though not definitely) reflect knowledge of events after 70. Together, they make up the slender foundation for dating the Gospels after 70.
This may be true of some scholars, but many who date the Gospels to 70 CE or later do not rely on this reasoning. Bart Ehrman and Mark Goodacre, among others, have argued that Jesus may well have predicted the Temple’s destruction yet still they date Mark later. Their argument is not that the prophecy was fabricated but that it would have carried special resonance after the event. Even if Jesus made such a prediction, the question remains: why did it become so central to Mark and to those who later drew upon him?
There are many other reasons for dating Mark to around 70 CE or later. (1) Mark is not mentioned by Paul, who was traveling and writing in the 50s CE; if such a Gospel had been in circulation, Paul would likely have known and cited it. (2) Mark’s apocalyptic tone is consistent with Jewish disillusionment after the war of 66–73 CE. (3) Mark explains Jewish customs and translates Aramaic expressions, suggesting a Gentile readership outside Palestine likely after Jerusalem’s fall, when Christianity had become increasingly Gentile. (4) Mark seems to treat Jerusalem and its Temple as doomed, which makes sense after their destruction. (5) Some passages seem to presuppose organized persecution of Christians, reminiscent of the reign of Nero in the 60s. (6) Mark reflects theological and literary developments beyond those found in Paul’s letters, implying a later phase of reflection and understanding.
The dating of Mark is crucial because Matthew and Luke (and possibly John) depend on Mark’s Gospel, reproducing much of its material in their own narratives. If Mark was written around 70 CE, then the others must have been composed after that.
This discussion of Gospel chronology may seem arcane. Why does it matter? For Murray, the point is clear: if the Gospels were written by eyewitnesses or by close associates of eyewitnesses within 15 to 25 years of Jesus’ death, they are likely to be reasonably reliable. As he writes:
Jesus had been the teacher; they, his students. In an era when oral transmission was the standard way of conveying information from one generation to the next, the task of the student was to memorize the teacher’s teachings and recite them verbatim …
The disciples probably heard Jesus preach the beatitudes and the parables to his audiences dozens if not hundreds of times. Memorization of his words was not just possible but close to unavoidable. Jesus also sent the disciples out on their own evangelizing missions, instructing them to proclaim the kingdom of God and presumably adding instructions about how to go about it (Luke 9:1–2). These are some of the reasons for thinking that Jesus’s teachings as reported in the Gospels are likely to be attributable to him and in some cases are probably close to verbatim.
There is a problem here, even if one accepts Murray’s revisionist (or reactionary?) arguments against the revisionists. The Gospels themselves present very different portraits of Jesus and his teachings. To take a striking example, in John, Jesus often speaks about himself and his relationship to the Father, declaring, “Truly, truly, I say to you, before Abraham was, I am.” In Mark, by contrast, he rarely speaks about himself, and he certainly does not call himself God. If Jesus had, in fact, made such explicit claims to divinity, provoking some to prepare to stone him for blasphemy, one would expect other eyewitnesses to have noted it. It is not, after all, a trivial detail.
Many similar examples could be given. Mark differs from Matthew, who differs from Luke, who differs from John. I do not mean to suggest that each Gospel presents a Jesus unrecognizable to the others, but I do mean to suggest that Mark and John cannot both be historically accurate portraits of him, at least not without elaborate qualifications. So what, then, are we left with?
Broadly speaking, there are four positions regarding what we can know of the historical Jesus: (1) that we can know almost nothing; (2) that we can reconstruct a minimalist apocalyptic Jewish preacher, but little more; (3) that we can recover a quite reliable historical figure; and (4) that we can know the historical Jesus because the Gospels are divinely inspired. As far as I can tell, Murray holds the third position. I hold the first. I am not even certain that the historical Jesus existed (see Richard Carrier), though I also doubt that it matters for Christianity. The Jesus of faith is necessarily a figure of myth not of verifiable history—a symbol and a literary character rather than a biography.
To my surprise, Murray, at least partially in deference to the view that the Gospels depict an accurate historical Jesus, even entertains the literal truth of the resurrection. He writes:
The implication is that something transformative happened to the apostles and Jesus’s other followers after the crucifixion. Coming up with a description of something that doesn’t look like the resurrection turns out to be harder than you might think.
History is replete with examples of followers who reinterpreted the death of a revered leader and not only preserved his movement but intensified their devotion to it. The pattern is similar. Grief transforms into vision. Despair into conviction. Thus the apostles’ belief that Jesus had risen does not suggest a supernatural event. Instead, it reflects the human capacity to mythologize loss and to transform grief into faith. Reports of similar postmortem visions, from the ancients to moderns, are copious and psychologically understandable.
Furthermore, even if one granted that something exceedingly improbable occurred within the early Christian community, something powerful enough to transform postmortem despair into conviction, many more plausible alternatives remain than the physical resurrection of a corpse. Individual or collective hallucination, though rare, is still far more probable than the suspension of every known biological law.
Put in Bayesian terms, the prior probability of a dead man’s return to life is vanishingly small. Even the believer, after all, would admit that the dead seldom return from the grave in physical form. To overcome such priors would require evidence of overwhelming strength: multiple independent eyewitnesses, contemporaneous documentation, physical verification. We have none of these. What we have are stories, written decades later by believers whose faith preceded their explanation and who were not, in fact, independent of each other.
I have been quite critical of many of the arguments in Taking Religion Seriously, but as I wrote at the outset of this review, I enjoyed the book. Like Murray, I believe that Christianity deserves to be taken seriously. Unlike Murray, I think that requires understanding it not as a set of empirical claims but as a mythopoetic narrative, which enchants but does not literally explain the world. Murray wishes to defend the truth of Christianity; I want to preserve its grandeur.
-
I listened to the first part of his interview with Andrew Sullivan. I was disappointed. One would expect someone with his background making his points to have some familiarity with the centuries-old refutations to these very standard arguments for god and at least address them. But he doesn’t. At least not in a friendly interview.
-
I can't abide Sullivan's paywall style where the conversation is cut off in the middle. Sam Harris does that too. I'll sometimes listen to podcasts where they withhold some juicy personal questions for the paywalled version, but that's a clean breakout session rather than a continuation of the primary conversation.
