Home » Library » Modern Library » Review of Reasonable Faith

Review of Reasonable Faith

Review of Reasonable Faith (2007)

Chris Hallquist

ImageReview: William Lane Craig. 1994. Reasonable Faith: Christian Truth and Apologetics. Wheaton, IL: Crossway Books. 350 pp.



          Chapter 1: Faith and Reason: How Do I Know Christianity is True?

          Chapter 2: Man: The Absurdity of Life Without God

          Chapter 3: God: The Existence of God

          Chapter 4: Creation: The Problem of Miracles

          Chapter 5: Creation: The Problem of Historical Knowledge

          Chapter 6: Sacred Scripture: The Historical Reliability of the New Testament

          Chapter 7: Christ: The Self-Understanding of Jesus

          Chapter 8: Christ: The Resurrection of Jesus



William Lane Craig’s Reasonable Faith is written as an apologetics textbook. Many apologetic works deal with only one facet of the arguments for Christianity; in fact, some apologists write many books without straying from their favorite topic. Craig, in contrast, is considered an authority in two quite distinct areas: arguments for the existence of God and the evidence for Jesus’ resurrection. Reasonable Faith includes concise statements of Craig’s arguments in both of these areas, as well as his views on faith, the meaning of life, miracles, history, and Jesus’ view of himself. It also includes a chapter on the reliability of the New Testament by evangelical scholar Craig Blomberg, author of The Historical Reliability of the New Testament. Blomberg was also called on by Lee Strobel to defend the New Testament’s reliability in The Case for Christ. As such, it presents a broad range of Christian apologetics to critique.

Because it’s a textbook, each chapter has a section on the history of the issue. This is a nice touch, especially if one is interested in further reading, but I’ll mainly be passing over this material except when Craig refers back to it in his own arguments. The sections where Craig presents his arguments are all titled “Assessment”; these are followed by brief sections on “Practical Application.”


Craig opens Reasonable Faith with a definition of apologetics: “Apologetics (from the Greek apologia: a defense) is that branch of Christian theology which seeks to provide a rational justification for the truth claims of the Christian faith.”[1] This is a simple statement of the book’s purpose, but it demands scrutiny.

There are two kinds of beliefs for which it might be reasonable to try to provide rational justification after the fact, but neither is really analogous to the beliefs that apologetics aims to defend. The first is very simple beliefs held since childhood and taken as given, such as “The sun will rise tomorrow” or “2+5=7.” Epistemologists and philosophers of mathematics may try to justify such beliefs as an intellectual exercise, but here the aim is very different from that of apologetics; the intent is not to convert the global skeptic or one who says that 2+5=4, but to uncover the complexity of the issue and perhaps even become doubtful. At best, such exercises may serve to shed light on more difficult matters of belief. Furthermore, unlike religious beliefs, such simple beliefs compel nearly universal consent, and uncovering why that is so is what makes them useful as mental exercises.

The second kind of belief is acceptance of the conclusions of academic disciplines, anything from accepting the existence of atoms to the details of Roman history to the statement “integral from negative infinity to positive infinity of exp(-x2) is the square root of pi.” We typically learn such things before we learn about the evidence or proof supporting them. But such conclusions were initially reached through the use of reason and evidence; the process of “providing rational justification” simply informs students how scientists, historians, or mathematicians arrived at their conclusions, with the possible addition of subsequently discovered empirical support.

By contrast, apologetics evidently aims to do more than simply summarize the evidence that convinced Constantine of the truth of Christianity. Because religious beliefs are rarely embraced after a dispassionate survey of all the relevant evidence, many people implicitly recognize that such beliefs were not reached on entirely rational grounds. In light of this, then, what apologetics evidently aims to do is bolster people’s confidence in such beliefs or help them defend them when debating others.

All of this may seem like overextrapolation from a brief passage that one might be tempted to chalk up to sloppy language on Craig’s part. However, the next chapter provides abundant evidence that his project aims to provide arguments for foreordained conclusions.

Chapter 1: Faith and Reason: How Do I Know Christianity is True?

In the first chapter Craig addresses the relationship between faith and reason, explaining first that, on his view, Christian belief is grounded not in reason, but in the witness of the Holy Spirit. In accusing all unbelievers of willfully rejecting the witness of the Holy Spirit, he denies the very possibility that one could come to unbelief through reasoned argument:

[W]hen a person refuses to come to Christ it is never just because of lack of evidence or because of intellectual difficulties: at root, he refuses to come because he willingly ignores and rejects the drawing of God’s Spirit on his heart. No one in the final analysis really fails to become a Christian because of lack of arguments; he fails to become a Christian because he loves darkness rather than light and wants nothing to do with God.[2]

Let’s set aside the astonishing implication that Jews, Muslims, and members of other non-Christian religions want “nothing to do with God.” What could be more presumptuous than declaring that, whatever they might say to the contrary, all unbelievers have felt the Holy Spirit but willfully rejected it? Consider Paul Doland’s account of how for years he struggled to maintain his Christian belief, but in the end could no longer sustain it:

I seem incapable of “experiencing” God. Many Christians thoughtlessly blame me for this, claiming that I haven’t had enough faith, didn’t try hard enough, or wouldn’t have accepted such experiences even if I had had them. All of these accusations are wide of the mark; they haven’t walked in my shoes. They don’t know how many times I’ve prayed and asked Jesus into my life. Since I don’t go around challenging the validity of Christians’ religious experiences, I would appreciate it if Christians would refrain from passing judgment on my lack thereof.[3]

How can Craig respond to the countless non-Christians throughout the world who claim to sincerely believe that they have never felt the Holy Spirit? It seems that he must maintain that they are lying and therefore culpable for their nonbelief, since if any of them were merely confused about what they have felt, or deceiving themselves, they would not be willingly ignoring the pull of the Holy Spirit. But in any case, his only basis for maintaining that non-Christians are willfully ignorant of the witness of the Holy Spirit is his own religious experience. Does his experience trump the testimony of a billion others who claim to have had no experience, or an experience that leads them to some other religion? Imagine: What would a Christian think of a Hindu who made such assertions about those who report no experience of Ganesh?

In a reply to Michael Martin, Tom Wanchick defended Craig’s line of reasoning by appealing to a couple of analogies: parents who refuse to admit their child’s bad behavior, and those who refused to admit that the evidence showed that O.J. Simpson killed his wife.[4] But a child’s behavior and the evidence against an accused criminal are public knowledge. The only person who has access to the “drawings of the Holy Spirit” is the individual concerned. To maintain that all unbelievers willfully ignore their undeniable experiences of the Holy Spirit is practically a conspiracy theory, analogous to the view that the United States government is suppressing evidence of extraterrestrial visitation, in spite of the silence of countless numbers of people which would be required in order for that to be true, and without any positive evidence in its favor.

Craig’s discussion of Martin Luther’s distinction between different uses of reason illustrates a contradiction at the heart of his apologetics:

I think Martin Luther correctly distinguished between what he called the magisterial and ministerial uses of reason. The magisterial use of reason occurs when reason stands over and above the gospel like a magistrate and judges it on the basis of argument and evidence. The ministerial use of reason occurs when reason submits to and serves the gospel…. Should a conflict arise between the witness of the Holy Spirit to the fundamental truth of the Christian faith and beliefs based on argument and evidence, then it is the former which must take precedence over the latter.[5]

That final sentence reveals the fundamental difference between philosophical inquiry and apologetics. And that difference becomes particularly problematic for Craig because he does not merely aim to show that one can rationally believe in Christianity. He also wants to claim that it is positively irrational not to accept Christianity, and he regularly accuses his opponents of closed-mindedness. In his interview with Lee Strobel in The Case for Faith, for instance, he says: “I think many skeptics act in a closed minded way…. [and] will not allow supernatural explanations even to be in the pool of live options.”[6] But in Reasonable Faith Craig admits that non-Christian hypotheses are not in his pool of live options; if a preponderance of the evidence implied that Christian beliefs were false and he recognized this, he would nevertheless continue to accept those Christian beliefs.

Craig adds: “[I]f the magisterial role of reason were valid, then a person who had been given poor arguments for Christianity would have a just excuse before God for not believing in him.”[7] Thus Craig completes his circular argument: Unbelievers are condemned because they willfully ignore the Holy Spirit, and we know that they willfully ignore the Holy Spirit because they are condemned!

Let us pause for a moment to consider just how entrenched Craig’s view is. If the magisterial view of reason is invalid, as Craig believes, then a skeptic cannot use reason to convince him that his theology–including his view of reason–is wrong, since appealing to arguments and evidence is a de facto application of the magisterial view. Though Craig tries to build on Plantinga’s notion of a “properly basic belief,” Craig’s position is in fact more hardened than Plantinga’s, as Plantinga admits that such beliefs could be defeated by other beliefs.[8]

It is not as if Craig merely regards the witness of the Holy Spirit as an evidentiary trump card which will outweigh all other conceivable evidentiary considerations. For Craig’s witness of Holy Spirit cannot even count as evidence: to use something as (either public or private) evidence is to be ready to reevaluate its worth if given grounds for doing so.

Towards the end of the chapter, Craig is sensitive to–though he does not quite answer–the objection that “anything goes” in theology if the magisterial use of reason is invalid:

It is tremendously liberating to be able to know that our faith is true and to commend it as such to an unbeliever without being dependent upon the vagarities of argument and evidence for the assurance that our faith is true; at the same time we know confidently and without embarrassment that our faith is true and that the unbeliever can know this, too, without falling into relativistic subjectivism.[9]

Presumably, the risk of “falling into relativistic subjectivism” refers to the idea that one faith is as good as any other; but having rejected the so-called “magisterial” use of reason, it is difficult to see how this conclusion is to be avoided. Obviously, if no arguments or evidence could ever–even in principle–persuade a Christian that his religious beliefs are false, then a Muslim or a Hindu could take a similar position vis-à-vis his religious beliefs. To avoid “relativistic subjectivism,” it seems, all that a Craigian Christian can (and would) say is that Christianity is true and all other alternatives to it are false, pure and simple. There is nowhere for genuine dialogue between Christians and non-Christians about religious issues to go, it would seem.

On the purportedly irrational reasons for all unbelief, Craig writes:

At the same time, however, this view reminds us that unbelief is at root a spiritual, not intellectual, problem. Sometimes an unbeliever will throw up an intellectual smoke screen so that he can avoid personal, existential involvement with the gospel. In such a case, further argumentation may be futile and counterproductive, and we need to be sensitive to moments when apologetics is and is not appropriate. If we sense the unbeliever’s arguments and questions are not sincere, we may do better to simply break off the discussion and ask him, “If I answered that objection, would you then really be ready to become a Christian?” Tell him lovingly and forthrightly that you think he’s throwing up an intellectual smoke screen to keep from confronting the real issue: his sin before God.[10]

Reading his use of the term in context, it seems that Craig uses “intellectual smokescreen” to mean “purportedly rational reasons for one’s belief or disbelief in a proposition which are not the real reasons why a person believes or disbelieves that proposition.” Though he warns apologists-in-training to be on the lookout for unbelievers’ ‘insincere’ arguments, he has already made it clear that he would continue believing Christianity even if he saw that all of his arguments for Christianity were unsound and that there were even compelling rational objections to it. Since Craig admits that his arguments for Christianity are not the real reasons that he is a Christian, his own apologetic arguments constitute, by his use of the term, an “intellectual smokescreen.” (If Craig meant something else by “intellectual smokescreen” than what I have inferred from his usage, I am at a loss as to what else he could have meant.) The proper response to Craig’s question is “If I rebutted your answer, would you be ready to become an atheist?” But most targets of evangelism probably won’t be clear-headed enough to see the problem with Craig’s question.

In short, Craig advocates a strange sort of fideism which entails that anyone who disagrees with him about fundamental religious issues is being dishonest about their subjective experiences, an issue for which Craig really has no way of knowing anything. In spite of this, he nevertheless develops arguments intended to convert non-Christians, though he admits that he would not be convinced by the strongest argument against Christianity. Finally, he accuses unbelievers of putting up an intellectual smokescreen to mask their ‘real’ reasons for not accepting Christianity.

Chapter 2: Man: The Absurdity of Life Without God

The second chapter of Reasonable Faith contains few formal arguments to critique. Though I will attempt to spell out formal arguments which seem to underlie Craig’s informal reasoning, note that it is difficult to be sure if my reconstruction is faithful to Craig’s intent given the manner in which he has presented his arguments.

The first subheading reads “No Ultimate Meaning Without Immortality and God.” The immortality part has something like an argument:

If each individual person passes out of existence when he dies, then what ultimate meaning can be given to his life? Does it really matter whether he existed at all? It might be said that his life was important because it influenced others or affected the course of history. But this only shows a relative significance to his life, not an ultimate significance. His life may be important relative to certain other events, but what is the ultimate significance of any of those events? If all the events are meaningless, then what can be the ultimate meaning of influencing any of them. Ultimately it makes no difference.[11]

Craig seems to be doing one of two possible things here. On the one hand, he seems to be equivocating between two different senses of the word “ultimate”–namely between “what eventually happens to a thing” and “what really matters.” Alternatively, he seems to be making an argument containing the hidden premise that the value of a thing depends entirely upon what eventually happens to that thing. If the former, Craig is committing a classic informal fallacy; if the latter, he is making an unsupported and dubious assumption. The latter implies an infinite regress: If each moment is given meaning only by the next, then the next moment must be given meaning by the moment after it, and so on ad infinitum.

Next we turn to this section’s argument that life would be meaningless if God did not exist:

If man and the universe could exist forever, but if there were no God, their existence would still have no ultimate significance….

Twentieth-century man came to understand this. Read Waiting for Godot by Samuel Beckett. During this entire play two men carry on trivial conversation while waiting for a third man to arrive, who never does. Our lives are like that, Beckett is saying; we just kill time waiting–for what, we don’t know. In a tragic portrayal of man, Beckett wrote another play in which the curtain opens to revealing a stage littered with junk. For thirty long seconds, the audience sits and stares in silence at that junk. Then the curtain closes. That’s all.

One of the most devastating novels I’ve ever read was Steppenwolf, by Hermann Hesse. At the novel’s end, Harry Haller stands looking at himself in a mirror. During the course of his life he had experienced all the world offers. And now he stands looking at himself, and he mutters, “Ah, the bitter taste of life!” He spits at himself in the looking glass, and then he kicks it to pieces. His life has been futile and meaningless.

French existentialists Jean-Paul Sartre and Albert Camus understood this, too. Sartre portrayed life in his play No Exit as hell–[in] the final line of the play are the words of resignation, “Well, let’s get on with it.” Hence, Sartre writes elsewhere of the “nausea” of existence. Camus, too, saw life as absurd. At the end of his brief novel The Stranger, Camus’s hero discovers in a flash of insight that the universe has no meaning and no God to give it one. The French biochemist Jacques Monod seemed to echo those sentiments when he wrote in his work Chance and Necessity, “Man finally knows he is alone in the indifferent immensity of the universe.”

Thus, if there is no God, then life itself becomes meaningless. Man and the universe are without ultimate significance.[12]

As a sustained argument this large passage leaves much to be desired. At best, it seems to be nothing more than a fallacious argument from authority. If you tried to put this ‘argument’ into a logical form, it would look something like this:

  1. Some twentieth-century authors portrayed our lives as pointless.

  2. Therefore, if God does not exist, our lives our pointless.

Robert M. Price has said on a couple of occasions that to explain apologetics is to refute it. In this case, to summarize this non sequitur is to parody it.

The next section, “No Ultimate Value Without Immortality or God,” is more of the same. Here Craig presents an annoying contradiction: “If life ends at the grave, then it makes no difference whether one has lived as a Stalin or as a saint…. On this basis, a writer like Ayn Rand is absolutely correct to praise the virtues of selfishness.” Actually, if it doesn’t make any difference how one chooses to live, then there’s no reason to praise selfishness. Put another way, metaethical subjectivism is incompatible with the existence of the objective moral facts posited by ethical egoism. This is an elementary point of metaethics which many people fail to grasp.

Craig then rhetorically asks: “In a world without God, who is to say which values are right and which are wrong?” But the question is ill-conceived, for “who is to say” would remain a legitimate question even in a world where God exists. For who is God to say which values are right and which are wrong? Unless there are objective moral facts that exist whether or not God exists, nothing prevents God from declaring it good, even obligatory, to massacre a defeated tribe (as he is said to have done in the Bible), crash planes into infidels’ skyscrapers, or send people to Hell for eternity. This point has been well-known for over two millennia, ever since Plato formulated a version of what has come to be called the Euthyphro dilemma, named after the dialogue in which it occurs. Admittedly, some have challenged the apparent force of the dilemma, but philosophers widely regard these challenges as unsuccessful.[13] Consequently, it is frustrating that Craig does not even raise the issue, and is apparently content to rest his case on mere appeals to authority.

For instance, in his last section of this chapter, “No Ultimate Purpose Without Immortality and God,” he appeals to a statement made by H. G. Wells and then offers a nascent argument which assumes that a purposeful life, by definition, requires being created for a purpose. One wonders who created God so that His life might have purpose. This whole section consists of, in effect, argumentum ad nauseum. There are few true arguments here; instead, Craig seems to think that he has strengthened his case when he has done nothing more than restate his primary claim in three different ways. And this is followed up with six more pages lamenting how sad life would be if there were no God and no immortality–as if depressing facts are somehow more open to debate simply in virtue of being depressing.

Notably, Craig adds that atheism is unbearable because without an afterlife, “all the acts of evil men go unpunished and all the sacrifices of good men go unrewarded.”[14] The latter concern almost seems to presume egoism, that we should only do good deeds if we gain a greater reward in return for them. Craig also finds it unbearable that people like Hitler might escape punishment for their misdeeds. But when you think about it, which is really more horrifying: the notion that Hitler no longer exists and thus “got away with” his crimes, or that Hitler was able to “pull off” such large-scale genocide, unencumbered, in the first place? Moreover, wouldn’t these crimes be all the more horrifying if there were a ruler of the universe who had seen fit to let them happen? And the amount and severity of the gratuitous suffering we find in the world is not mitigated in the least by the popular “free will defense.” For it would be no impediment to anyone’s free will if, whenever a person causes the deaths of a million of his fellows, God miraculously intervenes to prevent further killing. On this point, theism is clearly the more troubling view.

Since I could not find a clear exposition of Craig’s moral argument in Reasonable Faith itself, I turned to the argument Craig develops in Does God Exist? The Craig-Flew Debate. Unfortunately, Craig’s opening statement there again simply makes various appeals to authority, citing Bertrand Russell, Michael Ruse, and Friedrich Nietzsche. But in the postdebate response, Craig implicitly addresses the Euthyphro dilemma when he criticizes the view that moral concepts like justice could have an abstract existence on their own. Craig approvingly quotes Richard Taylor as saying, “A duty is something that is owed…. But something can be owed only to some person or persons. There can be no such thing as duty in isolation…” But this opens the door to the possibility that we might have obligations to other human beings even apart from God’s commands. Even here, there is little in the way of a full-blown theory, and theorization is clearly secondary to assertion.[15]

When I first encountered similar material in Craig’s The Son Rises, I took it to be a fallacious appeal to emotion designed to persuade readers in the absence of a sound argument. But now I suspect that the sophomoric character of such ‘arguments’ betrays their sincerity. If he could be convinced that there is no God, Craig really would find life unbearable. I can sympathize with this attitude: as I gradually lost my own faith, the belief that God was necessary to ground morality was one of the last beliefs to go, even in spite of my by then firm belief that the simple divine command theory of morality is monstrous. And when I look back on that period of my life, I recognize that it simply ‘made sense’ that God was somehow necessary for morality, even though it was unclear to me exactly how God was required, and in spite of the fact that I could provide no real arguments for that view. Perhaps Craig, too, simply has a similarly intuitive sense that God is somehow necessary for morality.

Since he may not have read all of the authors he cites in Reasonable Faith until later, it seems that the thoughts of the sixteen year old convert capture his primary reasons for accepting Christianity:

In my own case, the virgin birth was a stumbling block to my coming to faith–I simply could not believe such a thing. But when I reflected on the fact that God had created the entire universe, it occurred to me that it wouldn’t be too difficult for him to create the genetic material necessary for a virgin birth![16]

Craig ends this section with a discussion of the idea that nonbelievers might act as if a “noble lie” were true, even though they don’t really believe it. Craig’s imagined noble lie “tells us the universe is infused with value (which is a great fiction),” “makes a claim to universal truth (when there is none),” and “tells me not to live for self-interest (which is evidently false).”[17] Craig adds, quite sensibly, that it would make no sense for anyone to pretend that such things are true if they are, in fact, false.

But by appealing to “the witness of the Holy Spirit” as the basis for his indubitable Christian beliefs, Craig is able to have his cake and eat it too: he can believe the “lie” without recognizing it as such. Moreover, his “noble lie” is, in fact, an ignoble one, insofar as Craig also believes that the Holy Spirit is witnessing to unbelievers, no matter what they claim, and will “bring full justice in dazzling flame upon those who have refused to recognize God” (as Craig quotes 2 Thessalonians 1:8 in The Son Rises). However, to paraphrase Susan Clancy’s Abducted: How People Come to Believe They Were Kidnapped by Aliens, Craig’s personal religious experience might explain why Craig believes “weird things,” but not why he accepts Christian fundamentalism (and not some other religious tradition) specifically.

Chapter 3: God: The Existence of God

The primary aim of Chapter 3 is to lay out Craig’s version of the kalam cosmological argument for the existence of God, which appeals to, among other things, modern Big Bang cosmology. The “Assessment” begins with a statement that does not bode well for an argument that purports to draw heavily on modern science:

Moreover, the gradualism of classical evolutionary theory based upon the mechanism of minor mutations and natural selection has been radically called into question by the proponents of “punctuated equilibrium,” who argue that the transitional forms are absent from the fossil record because they never existed. Rather, they say evolution occurs by intermittent bursts from one form to another. Insofar as this new theory fails to account for these bursts and must appeal to “hopeful monsters”–massive mutations that produce new forms without transitional forms–the hypothesis of design becomes more plausible.[18]

This is a major distortion of the punctuated equilibria theory. The theory simply holds that apparently abrupt speciation arises with rapid change from a geological point of view. Transitional forms are still posited and no “hopeful monsters” are required.[19]

This distortion is, of course, incidental to Craig’s cosmological argument itself. In the “Application” section, Craig relates a story about a man who objected to his argument with the simple question, “If God created the universe, then where did God come from?” Craig comments that the man “thought he was very clever.” In truth, though, this objection is a valid one for most versions of the cosmological argument. Craig’s version of the argument, however, aims to bypass the objection by framing the argument in terms of causality rather than, say, how something came to be.

First Premise: Whatever Begins to Exist Has a Cause

Craig offers little support for his first premise, opting instead to attack those who do not accept it. He supplements this with a consideration of different conceptions of God’s existence ‘within,’ ‘outside of,’ or ‘apart from’ time as we know it:

  1. God began to exist at a point in time.

  2. God existed for an infinite time.

  3. God is timeless.

  4. “[P]rior to creation, God existed in an undifferentiated time in which hours, seconds, days, and so forth simply do not exist.”

Craig concludes that options 1 and 2 are ruled out by his own arguments. On the face of it, option 4 is not really an option since no clear meaning can be given to the phrase “undifferentiated time.” (Or if there is some obscure way to decipher a clearer meaning of the phrase, Craig does not say what it is.)

That leaves us with option 3. What reasons do we have for thinking that God is timeless? Craig says that it is simultaneously intuitively obvious and “may be ‘mysterious’ in the sense of ‘wonderful’ or ‘awe-inspiring,’ but it is not, so far as I can see, unintelligible.”[20] He nevertheless concedes that the idea of a thing simply coming into existence is also conceivable. Thus Craig’s grounds for accepting the existence of a timeless God who created the universe, rather than a universe that simply arose spontaneously (perhaps out of a ‘larger’ timeless multiverse), are simply that he finds the former more intuitive. Similarly, it seems intuitive that everything needs an explanation for its existence. But then an unexplained God fares no better than an unexplained universe (or multiverse). (Perhaps Craig could offer an ontological argument to ‘explain’ God’s existence, but such an argument would make his kalam cosmological argument unnecessary, and could even be constructed to ‘prove’ the existence of an omnipotent yet malevolent being. Such a being could not exist alongside an omnipotent and all-good God, for the absolute power of either being entails that such power could not be shared.)

Any attempt to inductively establish that whatever begins to exist has a cause also fails. For while we may have no experience with things that begin to exist without cause (if we set aside events on the quantum scale), we certainly have no experience at all with things that exist timelessly without explanation. Thus, although postulating an uncaused universe or multiverse may be counterintuitive, we have no reason to avoid one by postulating God.

Second Premise: The Universe Began to Exist

Craig defends the second premise of his kalam cosmological argument, that the universe began to exist, with four arguments: two a priori arguments and two scientific ones. I should note from the outset that it is not my view that the universe necessarily did not begin to exist. Rather, I’m sympathetic to the view taken by Robert Heinlein’s fictional character Jubal Harshaw, who “made a pact with himself to postulate a Created Universe on even-numbered days, a tail-swallowing eternal-and-uncreated Universe on odd-numbered days–since each hypothesis, while equally paradoxical, neatly avoided the paradoxes of the other–with, of course, a day off each leap year for sheer solipsist debauchery.”[21] I will nevertheless review Craig’s reasons for holding that the universe began to exist, starting with the weakest of his arguments, his second a priori argument:

The Impossibility of Forming an Actually Infinite Collection of Things by Adding One Member After Another

  1. The series of events in time is a collection formed by adding one member after another.

  2. A collection formed by adding one member after another cannot be actually infinite.

  3. Therefore, the series of events in time cannot be actually infinite.[22]

In his defense of the second premise of this subargument, Craig responds to an objection by J. L. Mackie which Craig characterizes as ones which “illicitly assumes an infinitely distant starting point.”[23] In fact, Mackie’s objection is more problematic for the first premise of the subargument. For if the history of the universe is infinitely long, then it was not formed by adding one member after another. No matter how far back one goes, there is always going to be an infinitely long history of events preceding that moment. Thus the first premise only holds on the question-begging assumption that the set of all events that ever happened before once consisted of zero events. But that assumption is simply another way of saying that the universe began to exist; in other words, Craig is assuming the very conclusion he is trying to establish.

The Impossibility of an Actually Infinite Number of Things

When I first read Reasonable Faith, Craig’s argument against an actually infinite number of things struck me as his strongest argument for the premise that the universe began to exist, for the very concept of infinity seems to lead to a myriad of contradictions. For instance, an infinite set can be added to without getting larger, and subtracted from without getting any smaller (it won’t be any more or any less infinite). Half of an infinite set will still be infinite, the fraction of the whole being as large as the whole itself. And one can remove half of an infinite set and still have infinity left over. It would seem that infinity – infinity (just might) = infinity.

However, as an anonymous reviewer of this paper has pointed out, listing the apparent contradictions that arise from considering infinities seems to be a rhetorical device “inviting readers to [improperly] import their intuitions about finite things into a situation dealing with infinite things.” The proper mathematical treatment of infinities is a complex subject for which I am no expert, but readers interested in how it impinges on this debate might want to consult a series of online papers between Graham Oppy and William Lane Craig on the kalam cosmological argument.[24] Whoever is right about the specifics, it is clearly unreasonable to treat infinities as if they were finite quantities. Doing so would rule out far more than Craig ever intended to rule out, including, but not limited to, the coherence of basic arithmetic. For instance, the set of natural numbers and the set of odd numbers are both equally infinite: for any natural number that you treat as highest, you can add one to it, producing a higher natural number; and for any odd number that you treat as highest, you can add two to it, producing a higher odd number. Yet there seem to be two natural numbers for every odd number. Is basic arithmetic, then, incoherent? Of course not: it’s simply a mistake to presume that the infinite sets involved cannot behave like finite quantities.

Since Craig emphasizes the difference between potential infinities and actual infinities in this section, I assume he would appeal to this distinction to explain the difference between an infinity of natural numbers and an infinity of past events. For one thing, many philosophers have insisted that numbers have an independent existence as abstract objects, and Craig seems to have committed himself to this view in his defense (among other places) of Alvin Plantinga’s argument from abstract objects.[25] Perhaps, though, his ontology of mathematics is mistaken. Perhaps, for instance, the correct ontology is a sort of Humean account favored by many empiricists, where numbers exist only in our heads and mathematical truths are matters of definition. But such differences do not appear to have any impact on Craig’s arguments: his arguments could be restated along Humean lines simply by inserting “potentially thought of by human beings” in the appropriate places without changing the structure of those arguments. Since arithmetic is so plainly coherent, we must reject the assumption that infinities will function like finite quantities.

Scientific Confirmation

The section purporting scientific confirmation that the universe began to exist contains a lengthy argument from Big Bang cosmology and a shorter one from thermodynamics (the latter noting that, contrary to observation, we would expect infinite entropy in an infinitely old universe given the observed arrow of time). In a personal communication, University of Wisconsin Madison astronomer Don Cox commented that this section seemed to be a fairly accurate summary of the state of cosmology at the time it was written, but nevertheless noted a few errors:

  1. Craig criticizes postulating cold dark matter, saying that it would throw off the rotation of galaxies, when in fact we need it to understand their rotation. Recent measurements indicate that cold dark matter makes up 26% of the matter in the universe. However, it has the effect of increasing the universe’s rate of expansion, so Craig gets his point in against the oscillating model.

  2. Craig dismisses vacuum fluctuations because they open up the possibility that the universe covered by the Big Bang model is only part of a larger whole. But Cox reports: “It is now almost certain the vacuum fluctuations during an early period of rapid inflation gave the universe the microstructure it needed to eventually form the physical structures that have condensed out of it, eventually leading to galaxies and everything in them.”[26]

Craig’s polemics put him in an awkward position here: instead of permitting a wait-and-see attitude, they force him to reject alternatives to or possible explanations of the Big Bang despite the potential for future cosmological evidence which would render them all but certain. In short, his argument depends upon the contingency that the universe began to exist and did not emerge from some larger ‘timeless’ or infinitely old multiverse, but scientific advancement threatens to falsify that contingency.

There’s a more fundamental criticism of Craig’s project in this section, however. Notably, he criticizes a variety of specific testable hypotheses about the Big Bang, but only offers a vague and untestable God hypothesis in its place. More precisely, Craig offers no way to at least attempt to falsify the God hypothesis, and suggests no astronomical observations entailed by it. Contemporary writers who claim that theism (or at least Christian theism) implies any specific cosmology tend to be atheists who argue, in turn, that such theism predicts features of our world that in fact have been falsified by subsequent scientific findings.[27] But if the God hypothesis predicts no specific astronomical observations, it is hypocritical to dismiss similarly unfalsifiable hypotheses on the grounds that they cannot be tested, such as the possibility that, in one sense or another, the universe (or a multiverse from which it emerged) has a history before the Big Bang. The same could be said of Craig’s argument from increasing entropy against the existence of an infinitely old universe: if one can justifiably postulate an untestable being that can create ex nihilo, another might as well postulate a vague and untestable physical process that reverses entropy from time to time. (Though I would propose that if after much cosmological research, no explanation of the Big Bang proves to be of any use in making predictions, we should abandon all untestable explanations of it, whether impersonal or divine, no matter how much such an approach may annoy our intuitions.)

A typical response to this point, in so many words, is that we ought to evaluate supernatural hypotheses by less stringent standards than those we apply to naturalistic ones. It is not clear to me what possible rationale there could be for taking such a view; at the very least, one would think that supernatural explanations should not receive preferential treatment over naturalistic ones. Let us call the view that it is inappropriate and overreaching to apply the same evidential standards to supernatural hypotheses that are brought to bear on naturalistic ones the preferential treatment thesis. (It is notable that preferential treatment proponents frequently claim that science cannot say anything at all about religious claims, though that claim does not seem to follow from the preferential treatment thesis itself.) Tacitly, the thesis seems to amount to a concession that supernatural hypotheses do not stand a chance of meeting the basic evidential standards required of naturalistic ones. But if that really is the case, why should anyone treat supernatural hypotheses as live options in the first place? In any case, even if we were to grant the preferential treatment thesis, it does not seem to help Craig’s case, as he proposes the God hypothesis as an alternative to various naturalistic ones implicitly on the grounds that God explains certain features of the world better than naturalistic alternatives which seek to explain the same evidence.

Therefore, the Universe has a Cause

Suppose we grant both of the premises of Craig’s kalam cosmological argument. Then the conclusion that the universe has a cause follows deductively. But this conclusion is a rather mild one. What we really need from an argument for the existence of God is some reason to believe that the cause of the universe is a personal one. And this is where Craig’s ultimate argument for a personal God is weakest.

Craig supplements his kalam argument with the following argument that the cause of the universe is personal:

Let’s say that the cause of water’s freezing is sub-zero temperatures. Whenever the temperature falls below zero degrees Centigrade, the water freezes. Once the cause is given, the effect must follow, and if the cause exists from eternity, the effect must also exist from eternity. If the temperature were to remain below zero degrees from eternity, then any water around would be frozen from eternity. But this seems to imply that if the cause of the universe existed eternally, the universe would also have existed eternally. And this we know to be false.

One might say that the cause came to exist or changed in some way just prior to the first event. But then the cause’s beginning or changing would be the first event, and we must ask all over again for its cause. And this cannot go on forever, for we know that a beginningless series of events cannot exist. There must be an absolutely first event, before which there was no change, no previous event. We know that this first event must have been caused. The question is: how can a first event come to exist if the cause of the event exists changelessly and eternally? Why isn’t the effect as co-eternal as the cause?

It seems that there is only one way out of this dilemma, and that is to infer that the cause of the universe is a personal agent who chooses to create a universe in time. Philosophers call this type of causation “agent causation,” and because the agent is free, he can initiate new effects by freely bringing about conditions which were not previously present. For example, a man sitting from eternity could will to stand up; thus, a temporal effect arises from an eternally existing agent. Similarly, a finite time ago a Creator endowed with free will could have willed to bring the world into being at that moment. In this way, God could exist changelessly and eternally but choose to create the world in time. By ‘choose’ one need not mean that the Creator changes his mind about the decision to create, but that He freely and eternally intends to create a world with a beginning. By exercising his causal power, he therefore brings it about that a world with a beginning comes to exist. So the cause is eternal, but the effect is not. In this way, then, it is possible for the temporal universe to have come to exist from an eternal cause: through the free will of a personal creator.[28]

Whatever you think of the value of this argument, note that postulating a personal cause of the universe is no less problematic than Craig imagines an impersonal cause to be, and may bring in even larger conceptual difficulties. For instance, what does it even mean to say that a timeless God decided to create the universe ‘at some point?’ Our very concept of deciding implies a temporal process in which another temporal process, contemplation, precedes a decision. Indeed, it is extraordinarily difficult to conceive of how any ‘timeless’ thing–personal or not–could produce (yet another temporal process) anything. Craig seems to deny the temporal nature of deciding when he says that we needn’t suppose that God changed his mind, but he offers no alternative analysis which construes deciding in nontemporal terms. In an online article[29] Craig adds that “only a libertarian agent could interrupt the static reign of being of the First Cause sans the universe,” but even a genuinely libertarian agent would seem to need time in order to freely choose. The fundamental problem here is completely independent of whether the cause of the universe is personal or not; it is whether the cause of the universe is timeless or not. For if a perfectly static timeless state or being ‘preceded’ the universe, how could that perfect stasis be interrupted? This appears to be one of those perennial questions, like why there is something rather than nothing, for which no conceivable answer–other than that there never was any such timeless state or being–is available whether one assumes that God or exists or that he does not. Craig often appeals to the intuitive assumption ‘from nothing, nothing comes.’ We might say here that ‘from timelessness, only timelessness comes.’ So we are back to square one: Craig offers no convincing reason to prefer a timeless creator to an uncaused universe, and the latter may even face fewer conceptual difficulties than the former.

The Teleological Argument and the Chapter’s Broader Function

The chapter ends on a one-page sketch of a modern teleological argument based on The Anthropic Cosmological Principle.[30] Craig cites the authors’ ten crucial steps in the evolution of humans, each “so improbable that the sun would have … incinerated the Earth before it would occur.”[31] Then they calculate the odds against the assembly of a human genome at between 4-[180(110,000)] and 4-[360(110,000)].

Such large figures sound astonishing, but they are derived from the specificity of the question ‘What are the odds that the human genome would arise on Earth?’ To make a case that the origin of intelligent organisms was intentional, we would need to know the probability that any intelligent organisms would arise somewhere in the universe. The probability that any particular intelligent species would arise on Earth is very low when we consider all of the potential intelligent species that could have arisen, but this has little relevance for whether God exists. Furthermore, the authors calculated the odds for the human genome to appear “spontaneously”–a word Craig omits from his citation. Since no modern biologist would ever claim that the human genome arose through a purely random process, the cited calculations greatly exaggerate the odds against a natural origin of the human genome.[32]

In short, the conclusion that a natural origin of human beings is so improbable that we must have been created by God does not follow from the “ten crucial steps” analysis, at least not without further argument well beyond what Craig presents in Reasonable Faith. The events labeled as crucial steps, such as the appearance of DNA and the appearance of aerobic respiration, are included because they have occurred only once in Earth’s history. The need for such steps may indicate that the origin of human beings isn’t a particularly probable outcome, but it is quite a large leap from this to the conclusion that the origin of human beings is so improbable that it could not have happened without the intervention of a supernatural agent.

There is an important sense in which the final section of this chapter is more important than its primary focus, the kalam cosmological argument itself. As we have already seen, there are some substantial concerns about the soundness of that argument. But had we concluded that the kalam cosmological argument is an unqualified success–and even conceded the supplementary argument that the cause of the universe must be a person with libertarian free will–this would not support the larger project of Reasonable Faith. Nothing in this sole chapter on the existence of God offers us any reason to believe that the deity has any specific moral characteristics, or takes any interest in life on Earth, though Craig’s last-minute addition of an argument from design might be taken to imply that God had some unknown reason for creating organic sentient beings. Chapter 3, then, has offered us a series of arguments for the existence of a metaphysical abstraction, and nothing more. And yet the position adopted in the previous chapter, that God is needed to make our lives meaningful, assumes that God is a loving being who takes a personal interest in the lives of human beings. It is surprising, then, that not a single line of thought in the current chapter gives us any reason to attribute such characteristics to God.

Interestingly, in various debates and publications, Craig has claimed that demonstrating the existence of God raises the prior probability of the resurrection of Jesus, though he does not explicitly claim this in Reasonable Faith itself.[33] But because the kalam cosmological argument tells us nothing of the Creator’s motives, it could not (at least by itself) increase the probability of the resurrection of Jesus. Nor does it give us any reason to believe that the Creator would be able to resurrect Jesus. To assume otherwise would be analogous to assuming that a scientist capable of creating a Petri dish full of bacteria is also able to raise an individual bacterium from the dead. Thus, nothing in this chapter provides us with any reason to believe in the Christian God specifically, though Craig argues for the truth of Christianity later.

Chapter 4: Creation: The Problem of Miracles

The “Assessment” in Chapter 4 is divided into three parts: a rebuttal of the idea that miracles are ruled out on a Newtonian picture of a law-governed universe, a rebuttal of Baruch Spinoza’s arguments against miracles, and a rebuttal of David Hume’s arguments that belief in miracles is not rationally justified.

The Newtonian World-Machine Versus Quantum Mechanics

Craig first addresses the objection that, in essence, natural laws make the occurrence of miracles impossible. He starts by arguing (correctly, I think) that quantum mechanics does not offer a way around the operation of absolute natural laws, in part because some miracles are overtly inconsistent even with quantum mechanics, and in part because nothing within relativity theory is indeterminate. Moreover, even if quantum mechanics makes it impossible to absolutely rule out miracles, they would nevertheless remain extremely improbable.

Craig says that if laws of nature are just descriptions of regular occurrences (regularities), such descriptions would simply have to take the occurrence of miracles into account as well. He adds that even on the alternative “nomic theory of necessity”–in which natural laws tell us what can and cannot happen–the laws of nature only tell us what happens when there is no supernatural intervention, and a miracle can be described as an event that is only naturally (not supernaturally) impossible. Finally, if natural laws are understood as descriptions of the dispositions of things to act in certain ways, miracles can be accommodated as interferences with natural propensities, and are thus not impossible.

There’s no denying that there is a sense in which anything is possible, or in which we cannot know what is and is not possible. Nevertheless, the mere logical possibility of miracles does not entail that there are not strong reasons to be skeptical of miracles. Objections to belief in miracles posed by Spinoza and Hume provide excellent examples of such reasons.

Spinoza’s Objections

Craig paraphrases three objections to miracles that he attributes to Spinoza: that nature is immutable yet a miracle would overthrow natural law, that a purported miracle might be the act of a lesser being than God, and that a presumed miracle might actually be some undiscovered natural effect. Craig responds to the last objection as follows:

[W]hen miracles occur at a momentous time (for example, a man’s leprosy vanishing when Jesus spoke the words “Be made clean”) and do not regularly recur in history, and when the miracles in question are numerous and various, then the chance of their being the result of unknown natural causes is minimal. Since, as we shall see, most critics now acknowledge that Jesus did perform what we may call miracles, this answer to Spinoza and [contemporary philosopher Antony] Flew seems to be a cogent defense of the supernatural origin of the gospel miracles.[34]

The most questionable part of Craig’s response, that “most critics” concede that Jesus worked miracles (“as we shall see”), refers to a somewhat vague statement on p. 250 in which is it unclear that the critics in question actually concede that the supposedly miraculous events under discussion actually occurred as described. For example, Jesus may have told lepers that they would be healed, but any recoveries may have happened over a period of days. Faith healings and exorcisms in general, if not the specific Gospel accounts, do not even require postulating unknown natural causes. Rather, they can be explained by known forces such as confirmation bias and suggestion.[35]

Moreover, during the time period when the Gospels were written, these natural phenomena were not well-understood and could have easily been misinterpreted as miracles. Even today the sources of false extraordinary claims often go unrecognized, such as the power of suggestion over perception and memory. And then there are well-known factors which are frequently underestimated, such as trickery and fraud. Finally, even to this day we do not fully understand why people sometimes claim to witness events that never happened–but they make such claims nonetheless, particularly about marvelous events. All of this should make us wonder, when we are faced with an apparently inexplicable miracle claim, if we simply lack knowledge of its true natural cause.

Next Craig turns the paucity of well-supported miracle claims found today into an argument for the authenticity of biblical miracles. On the Resurrection, Craig argues:

Moreover, if it were the effect purely of natural causes, then its singularity in the history of mankind becomes very difficult to understand–why has it not happened again? In the nearly two thousand years since that event, no natural causes have been discovered that could explain it.[36]

It seems to me that explaining why such events do not happen today is a far greater difficulty for the plausibility of the view that miracles occur. On Craig’s worldview, God worked numerous overt miracles throughout Israelite history, from Moses to Jesus, and Jesus even gave powers to his disciples to work throughout the world; and then for some reason such incidents evaporated. Why would a God with the means, motive, and opportunity to directly reveal himself to vast numbers of human beings across the globe through miracles limit the undeniable display of his powers to one particular region of the world for a small portion of human history? On the other hand, if miracles do not occur, but there are scores of false miracle claims driven by psychological and sociological factors, it is not surprising that (1) the actual miracle claims we find tend to be poorly documented (including those of Jesus himself) and (2) the miracles purported today tend to be found spurious in situations where they can be thoroughly scrutinized. Because the lack of decent evidence for present-day miracles was central to Hume’s objection to rational belief in them, we will examine this point in more depth shortly.

After lampooning Spinoza’s implicit argument that even eyewitnesses should be skeptical of miracles by quoting Ebenezer Scrooge’s skeptical response to his encounter with a ghost in Charles Dickens’ novel A Christmas Carol, Craig declares:

Perhaps Pascal was right in saying that God had given evidence sufficiently clear for those with an open heart, but sufficiently vague so as to not compel those whose hearts are closed.[37]

This statement raises at least two questions:

  1. Is it obviously absurd for a witness to be skeptical of an apparent miracle? I would think not. If a person perceives a ghost, the possibility the he is hallucinating does deserve some small consideration. In many other cases, we must consider the fact that magicians can produce illusions which seem very convincing to witnesses, and the general fact that unusual events often turn out to have rather mundane explanations. The fact that Spinoza could reasonably point to such cautions is a good thing.

  2. Could the evidence for miracles be sketchy because God does not want to compel belief? Perhaps, but the nature of the evidence that we have for miracles seems to point to other explanations for miracle claims, so it is presumptuous to accuse those who opt for alternative explanations of closing their hearts. That it is admirable to embrace a belief on weak evidence seems to be a popular notion in religious discussions, but even the most faithful believers would hesitate to suggest this for other cases (say, when evaluating whether a defendant is guilty of murder).

Finally, how does Craig account for countless cases of apparently sincere unbelief? Consider former Christian Paul Doland’s apt response to Craig’s claim that God will reveal himself to all who ask:

I can only speak for myself, and I seem incapable of “experiencing” God. Many Christians thoughtlessly blame me for this, claiming that I haven’t had enough faith, didn’t try hard enough, or wouldn’t have accepted such experiences even if I had had them. All of these accusations are wide of the mark; they haven’t walked in my shoes. They don’t know how many times I’ve prayed and asked Jesus into my life. Since I don’t go around challenging the validity of Christians’ religious experiences, I would appreciate it if Christians would refrain from passing judgment on my lack thereof.[38]

Given his comments in the first chapter of Reasonable Faith, Craig apparently believes that Doland and others like him are either outright dishonest or simply deluding themselves. But there seems to be little evidence to support this conjecture.

Hume’s Objections

By far the most famous argument against miracles that Craig addresses is the argument laid out by David Hume in Section 10, Part 1 of his Enquiry Concerning Human Understanding. This argument is typically referred to as the “in principle” objection, while the argumentation of Part 2 concerns a series of “in fact” objections. I will follow Craig in focusing on the argument in Part 1.

Central to Hume’s case is the claim that “experience be our only guide concerning matters of fact,” meaning that we must judge the probability of claims based on our own experience. Given how far miracle claims lie outside of normal human experience, this principle, if correct, provides a serious barrier to belief in miracles. In support of this contention, he observes that the only reason we trust human testimony is because we have, in the past, experienced a connection between testimony and facts about the world. Hume makes a similar point earlier in his Enquiry, in Section 4, when he argues that it is only experience which allows us to infer a person from a voice in the dark. I think that the best argument for Hume’s position, though, may be one unwittingly provided by apologists like Craig, who implicitly presume that experience is the only way we can evaluate other hypotheses. Apologists must constantly appeal to experience in attempting to refute the alternative explanations provided by skeptics. To quote just one rather succinct example from the final chapter of this book: “rival theories [to the Resurrection] are disconfirmed by [already established] beliefs about, for example, the instability of conspiracies, the likelihood of a death as a result of crucifixion, the psychological characteristics of hallucinatory experiences, and so forth.”[39] All three statements are empirical generalizations. Suddenly, Craig must argue not that miracles must be treated like other historical hypotheses, but that miracles must be granted a special exception from our normal requirements for a good hypothesis!

It is important to emphasize that Hume never claimed that no evidence could possibly establish a miracle, only that we must require “that the testimony be of such a kind, that its falsehood would be more miraculous, than the fact, which it endeavours to establish.” As an example of what would be a credible miracle claim, Hume sketches a case where all sources, everywhere in the world, agree that the sky went completely dark for eight days on a particular date. However, Hume says, he would not believe a real miracle had happened should all sources report that the Queen of England rose from the dead, presumably on the grounds that such a thing is much harder to verify than a global darkness, and much more open to falsehood from “the knavery and folly of men.”

Craig begins by correctly pointing out that read literally, Hume’s “uniform experience against every miraculous event” is circular, but his can be taken to mean “general experience” or “nearly uniform experience,” a quite defensible and non-question-begging position. Craig then says that by thinking that “general experience” can rule out miracles, Hume is confusing science and history. This objection ignores rather than refutes what Hume said since he did not merely assert that we must use experience as our guide; he argued the point. Indeed, Craig’s objection ignores the approach he himself uses later on when dealing with rival hypotheses to the Resurrection.

When Craig says we cannot weigh the evidence for miracles on the same scale as the evidence for natural laws, he runs the risk of leaving us with no scale at all to weigh historical evidence. However, earlier Craig provided such a scale: “religio-historical context.”[40] This is supposed to be the real problem with Hume’s example of a hypothetical resurrection claim for the Queen of England: such an event would have no religio-historical context.

There are several problems with this view. First, Craig gives no reason to use a religio-historical context criterion, and it can only be established on hypothetical considerations. Second, evidence does show that spurious claims rarely have no religious context. Even psychic spoon-bender Uri Geller claimed his feats were the work of a higher intelligence, and alien abductions have a quasi-religious edge for some. When “religio-historical context” is broadened to mean any kind of context (and Craig gives no reason not to do so) we can find context for anything. For example, if Uri Geller had made no claims of an “Intelligence in the Sky,” one could still say his powers make sense in the context of the belief that some people can do extraordinary things with their minds. If historians had concluded that a resurrection of Queen Elizabeth took place, they probably would have come up with some context, whether legitimizing the Church of England over that of Rome, or letting her do important work for England. Perhaps someone would’ve even “discovered” a document proving that she descended from Jesus himself! Thus, Hume’s argument stands, and religio-historical context fails as an alternative criterion of probability.

Craig affirms the views of William Paley, Gottfried Less, and Thomas Sherlock–sketched earlier in the “history” section of the chapter–who argued that Hume’s general considerations could lead us to reject the strongest testimony for things outside of our experience, even natural events. I think it’s worth looking at Sherlock’s example: a man in hot climates would be forced to reject all evidence for the existence of ice. Hume dealt with such a view in his essay:

The Indian prince, who refused to believe the first relations concerning the effects of frost, reasoned justly; and it naturally required very strong testimony to engage his assent to facts, that arose from a state of nature, with which he was unacquainted, and which bore so little analogy to those events, of which he had had constant and uniform experience. Though they were not contrary to his experience, they were not conformable to it.

Observe that Hume does not say no evidence could ever establish the existence of ice, merely that the prince was right to be skeptical of the first relations about it. Far from being absurd, this position seems obviously correct once one considers that travelers from distant lands often brought extraordinary false stories.

Also, though I do not think that the contrary/comformable distinction holds, the distinction between ice and miracles does. The issue is not that ice is merely “uncomformable” to experience (one could reasonably say the same about miracles), but that it is repeatable, whereas miracles are not. The prince would find that every visitor from northern climates would agree that water freezes in winter, as would every returning ambassador and trader. This would allow him to build up more evidence than is even possible to conceive of for most miracle claims. The amount of evidence that would be generated in Hume’s global darkness example is almost impossible to get for most one-time claims, but is easy to get for a repeatable phenomenon, such as ice. Of course, it is possible to imagine a world where, in all ages, miracles happen often enough that they are not significantly harder to document than ice. Unfortunately, we do not live in such a world.

Next, Craig argues for the need for a religio-historical context in order to label an event a miracle with the following illustration:

There are all sorts of events that make up the stuff of popular books and television shows on unexplained mysteries (such as levitations, disappearing persons, spontaneous human combustion, and so forth) that have not been scientifically explained, but–judging by their pointless nature, sporadic occurrence, and lack of any religious context–are not miracles. It would be folly for the historian to deny the occurrence of such events in the face of good eyewitness evidence to the contrary, simply because they do not fit in with known natural laws. Yet Hume’s principle would require the historian to say that these events never occurred, which is indefensible.[41]

There are, in fact, good historical grounds for denying these events. Most victims of spontaneous combustion turn out, on further investigation, to have been found near some source of flame, and damage to the body is great only when a fuel source beyond clothes is present.[42] I only know of one famous case of levitation, that of D. D. Home. However, there is evidence that Home engaged in trickery on other occasions. Even ignoring this, the alleged event happened during a spiritualist craze which ended with many fraudulent mediums being exposed by magicians like Harry Houdini.[43] Home survived investigation by Sir Walter Crookes, but mostly held private sittings inaccessible to magicians, and his levitation was the sort of thing that a magician could do. Such “general considerations,” which Craig so deplores, plainly cast some doubt on Home’s alleged feat.

Craig opened chapter 2 with a dismissive comment characterizing the Skeptical Inquirer as one part of a modern humanist campaign of “almost evangelical fervor.” But I think Craig would benefit greatly from a subscription. The endless supply of claims critiqued, many documented better than the claims of Christianity, should lead us to considerable skepticism about miracle claims. Home’s levitation is among these better evidenced claims, as we have statements from three eyewitnesses. The only generally accepted written documentation of potential eyewitness testimony to the resurrection appearances is a brief passage from I Corinthians.

Craig then turns a critique of Antony Flew’s revised Humean argument against the rationality of belief in miracles, offering the following obscure comments about historical possibility:

If one wishes to talk about historical possibility or impossibility at all, these terms ought not to be defined in terms of scientific law, but in terms of historical evidence. Thus, for example, it is historically possible that Nietzsche’s insanity resulted from venereal disease; it is historically impossible that Napoleon won the battle of Waterloo. On this basis, only the evidence can tell us whether it is possible that Jesus rose from the dead.[44]

This is an odd approach to history, since evidence is generally used to establish the factuality, not the mere possibility, that a certain event occurred. It seems that Craig is arguing that the Resurrection is historically possible if there is no contrary evidence, but that is a low standard. I wonder what evidence disproves the hypothesis that Nietzsche was driven insane by a visitation of Cthulhu.

Craig proceeds to claim that Flew’s work is simply a restatement of Ernst Troelstch’s principle of analogy–essentially the idea that we should dismiss reports of historical events that have no parallels in our present experience–but in fact it just restates a Humean point that Craig ignored in his critique of Hume. Craig responds that we should apply this principle in “an unclear historical situation,” but not when an event “bursts all analogies to the present.” This statement may be rhetorically impressive, but it appears to contain little of substance and introduces a number of unanswered questions:

  1. Why should the principle of analogy be applied only in unclear cases, and not universally? Why shouldn’t experience be our guide when evaluating whether the most probable explanation for an extraordinary claim is that the claim is true? On the face of it, the principle appears to be universally warranted, so if Craig wants to restrict its use, he needs to provide good reasons for doing so.

  2. What does Craig mean by “unclear”? On the most charitable assessment of the New Testament, the only evidence for Jesus’ miracles is a handful of partisan sources written quite some time after the alleged events themselves. Why such evidence should not be regarded as unclear is, well, unclear.

  3. What does Craig mean by “bursts all analogies to the present”? If he means “was actually miraculous,” then he is simply begging the question. If he means “the nature of the claims and the evidence for them are far beyond other things we know of,” then he has taken a more reasonable stand, but also a less helpful one: competing miracle claims are so numerous that there will always be at least some analogy between the miracles of one religion and the presumably false ones of others. Perhaps in line with the latter interpretation, Craig considers showing the resurrection narrative to be analogous to known phenomena, though he seems to think critics bear a very high burden of proof involving far too specific comparisons. Does anybody imagine that when historians come across non-Christian extraordinary claims they give them the benefit of the doubt until some specific parallel is produced? More probably they dismiss the claims so long as they are not matched with extraordinary evidence.

In addition to Hume’s famous “in principle” argument against miracles, he also presented a number of “in fact” or evidential reasons for doubting miracle claims, such as the existence of equally well attested competing claims from many different religions. When Craig finally turns to these, he falls on his own sword. Here he says that if Jesus’ miracles are genuine, “we can forgo the investigation of every single counter-Christian miracle, for most of these pale into insignificance next to the gospel miracles.”[45] Actually, the Gospel miracles pale into insignificance compared to many other claims, and a true investigation is impossible. The witnesses are long dead, and though I have heard that Jesus lives, he’s given no sign of being willing to show up for a controlled test of his (alleged) powers. By Craig’s own argument, then, we have sufficient grounds for simply dismissing the alleged evidence for Christianity.

Chapter 5: Creation: The Problem of Historical Knowledge

In Chapter 5 Craig attacks the notion that historical truth is completely unknowable and thus historical reconstruction is entirely subjective, which, if warranted, would be fatal to Christianity’s claim to historicity. I am largely in agreement with him here, but a more moderate historical skepticism could still ground an argument against Christianity. For instance, Craig may be right that history is not qualitatively different from science, but historical conclusions are considerably less certain (or considerably more tentative) than scientific ones. As the 18th-century theologian G. E. Lessing observed:

We all believe that an Alexander lived who in a short time conquered almost all Asia. But who, on the basis of this belief, would risk anything of great, permanent worth, the loss of which would be irreparable? Who, in consequence of this belief, would forswear forever all knowledge that conflicted with this belief? Certainly not I. Now I have no objection to raise against Alexander and his victory: but it might still be possible that the story was founded on a mere poem of Choerilus just as the ten-year siege of Troy depends on no better authority than Homer’s poetry.[46]

This objection is largely secondary to the weighty reasons we have for being skeptical of miracle claims, but I will return to it at the end of this critique.

Chapter 6: Sacred Scripture: The Historical Reliability of the New Testament

(chapter contributed by Craig Blomberg)

The Christian Church has traditionally claimed that the Gospel accounts of Jesus’ life are eyewitness reports from the apostles and their companions: Matthew and John by those members of the twelve, Mark by Peter’s secretary, and Luke by Paul’s physician. However, modern scholarship has called these assignments into question, with even moderate-conservatives doubting the traditional claims. Craig Blomberg begins Chapter 6 with a defense of these claims.

His defense starts with an unsupported statement that the testimony of the early Church should be taken more seriously than many scholars take it. It is a little hard to see why this should be so. Blomberg’s main source, Irenaeus, wrote 85 or so years after any of the Gospels were written. By comparison, the Gospels were written within 70 years of the events they report. To appeal to Irenaeus to save the Gospels from the charge of hearsay is to appeal to an even weaker piece of hearsay.

Blomberg’s specific defense of Matthew’s authorship just doesn’t make sense. He says that it’s unlikely that the Church would have made up a tradition about Matthew because he would be “suspect” as a tax collector, but we hardly get this impression from the Gospels. In contrast, Jesus preaches forgiveness for tax collectors and the general idea is that once one repents, one is no longer suspect. Blomberg also has to come up with a rather convoluted hypothesis to make Papias (an early Christian writer, c. 125) support the claim that Matthew wrote the Gospel we have under that name. But Papias refers to a collection of sayings in Aramaic, not a narrative in Greek. Blomberg proposes that Matthew originally made a collection of sayings in Aramaic that was later translated into Greek, and then Matthew combined it with Mark and further recollections. But there isn’t much evidence for this view. It’s a best shot at harmonizing two statements separated by a half century, one from Irenaeus and one from Papias, neither of which directly supports the view.

To defend the authorship of John, Blomberg cites B. F. Westcott, who has “never been refuted.”[47] Seeing this phrase immediately brought echoes of Brian Flemming’s interview with Earl Doherty, in which Flemming says virtually the same thing about Doherty’s attack on the idea of a historical Jesus. (In what follows note that I am not a fan of Doherty’s work and have a particularly low opinion of those who would use Jesus mythicism as an easy way to discredit Christianity.) Often, “never been refuted” only means “most people don’t think offering a rebuttal is worth their time.” Blomberg’s main reason for maintaining that John was written by an apostle is “his intimate acquaintance with the actions and thoughts of the Twelve,”[48] but this appearance could also be the result of creative, or perhaps unconscious, filling in.

With Mark, there is a legitimate reference from Papias, but this runs into the same problem as Irenaeus’ testimony, if to a lesser degree. Papias wrote at least 50 years after Mark, whereas Mark was written maybe 40 years after the life of Jesus. Once again, a Gospel is saved from the charge of hearsay with a weaker piece of hearsay. Furthermore, as Raymond Brown’s Introduction to the New Testament points out, there are a few other problems with the idea that Mark was written by John Mark, a companion of Peter: the Gospel of Mark does not seem to be a translation from Aramaic (and John Mark was presumably from Palestine); the author seems confused about Palestinian geography; and the author seems to have been working from tradition.[49] This is enough for Brown, hardly a radical in the field of Biblical scholarship, to reject the traditional authorship.

Lastly, we come to Luke, where mainstream scholars would say that Blomberg has the strongest case. Raymond Brown says that it is “not impossible” that the book is by a minor companion of Paul, while noting discrepancies between Acts and Paul’s letters.[50] Blomberg is quite dismissive of these discrepancies, declaring that they “merely prove that Paul did not write Acts!” and that there are no “insuperable problems.” Forgive me for being unwilling to take the latter point on the word of an evangelical scholar. For evangelicals, a problem can be counted as resolved even when one is forced to assert that the passage was miscopied, and many rationalizations are hardly more plausible. When I consider the miracle claims in the book of Luke, I find I must translate Brown’s “not impossible” to mean “not well enough established to lend them credibility.”

Next comes the issue of when the Gospels were written. Mainstream scholars tend to place them from 70 to 100 A.D., approximately. Blomberg counters this conclusion by arguing that Acts was written before Paul’s death because it does not mention the event. This would entail an even earlier date for Mark, because Luke relied on Mark. This argument loses its force, however, once one notes that the author seems to have been pitching Christianity to Gentiles. If this was his purpose, it makes much sense to end Acts with Paul declaring his mission is now to the Gentiles, and little sense to end with the unfortunate fact of Paul’s death at Gentile hands. Blomberg’s allegation that later dating is based on the assumption that Jesus could not have predicted the Temple’s fall hardly squares with most of the Biblical scholarship I’ve read. Brown, who is quite critical of naturalistic bias, accepts the standard dating of Mark around 70 A.D. and the later dating of the other Gospels based on the developed state of the tradition. Bart Ehrman, while somewhat more skeptical than Brown, disavows the idea that the Temple prophecies were later additions: “Most scholars, though, consider this an extreme view…. [I]n one respect, at least, the earliest form of these sayings appears to pass the criterion of dissimilarity, since Jesus’ claim in Mark that not one stone would be left upon another did not, in fact, come true.”[51]

After dealing with questions of authorship and dating, Blomberg turns to the question of harmonizing the countless apparent discrepancies between the Gospels. Here, he asserts, “Commentators on Plutarch and Arrian disclose the same range of suggestions for solving apparent discrepancies among parallel accounts of episodes in Alexander’s life as evangelical biblical scholars apply to the historical texts of the NT.”[52] This is a little hard to swallow, as classical historians are hardly committed to the total inerrancy of even the best ancient sources. To back up this point, Blomberg cites one of his own essays in another work.[53] I cannot shake the feeling that too often in this chapter, major assertions are supported with nothing more than a superscript. In fact, the cited essay does not provide anything like the straining needed to save the divergent birth narratives, and the most challenging discrepancies between the resurrection narratives are likewise more severe than the discrepancies Blomberg addresses.[54]

Blomberg proceeds by offering a long discussion of various types of New Testament criticism, but this is somewhat secondary to his main concern. Consequently, I will skip ahead to the “Burden of Proof” section, where Blomberg says that the books of the New Testament “must be treated at least as generously as other purportedly historical works of antiquity,” and therefore “the burden of proof that any portion of these works is unhistorical must rest squarely on the skeptic’s shoulders.”[55] I am happy to grant the first statement, that we should treat the Gospels as we would treat other works of antiquity of similar content. By all means, let us treat the Gospels as we treat every other ancient account of miraculous things, from the life of Apollonius of Tyana to Herodotus’ tale of a miraculous defense of the temple at Delphi to Josephus’ tale of a cow giving birth to a lamb. We needn’t reject all healings and exorcisms, since people throughout history have deluded themselves and others into thinking that they can produce such effects. However, we ought to reject the truly extraordinary major stories, such as accounts of walking on water, and once we do one wonders which of the more mundane major events in Jesus’ life are ahistorical.

Blomberg concludes his chapter by criticizing the Jesus Seminar, a group of New Testament scholars who famously concluded that only a fraction of the sayings attributed to Jesus are likely to have been uttered by him or simply reflected what he taught. This is one of the weakest sections in the entire book, as Blomberg essentially just presents a string of accusations backed up only by endnotes. Though I sympathize with the position that the Jesus Seminar erroneously took Jesus out of the context of apocalyptic Judaism, Blomberg cites too little evidence of this to be very convincing. There are many well-grounded criticisms of the Jesus Seminar–such as that they were determined to portray Jesus a wise teacher, damn the apocalypticism and faith healing–but Blomberg himself does not offer any of these. Blomberg may be caught between a rock and a hard place, however, since arguing along these lines might convey the impression that Jesus was simply one of many founders of a typical end-times cult in first-century Jerusalem.

Incidentally, in a footnote Blomberg states that he adopted biblical inerrancy as a consequence of his historical studies, but this seems difficult to reconcile with the account of his conversion at the end of the chapter. Nevertheless, through personal communication I have learned that, to his credit, Blomberg explicitly disavows the approach to scholarship laid out in the opening chapter of Reasonable Faith, and concedes that scholarship could, in principle, provide good reason to reject Christianity.[56]

Chapter 7: Christ: The Self-Understanding of Jesus

Before going into the meat of chapter 7, I want to make a note about the strategy Craig takes in that chapter and the following one. Both chapters attempt to defend the historicity of specific parts of the Gospel accounts of Jesus’ life. Chapter 7 argues that Jesus claimed to be the son of God in a unique sense, while chapter 8 deals with the Resurrection. What seems curious at first glance is that neither section makes an explicit appeal to Craig Blomberg’s arguments in the preceding chapter, nor is there any explicit appeal to the alleged eyewitness authority or general reliability of the Gospels. The whole project is portrayed as a ground up one that need take nothing for granted. The one possible exception is a point in chapter 8 where Craig says that claims that Jesus did not die on the cross are ad hoc in that they require denying the historicity of the wound from the Roman spear, even though Craig had not previously defended the historicity of that detail. That would imply that he does think he’s entitled to assume the reliability of the Gospels after all. However, since the passage is so short and runs so counter to the rest of Craig’s presentation, I’m inclined to read it as an ill-conceived throwaway line.

This raises the question of why Craig takes such a strategy, when building on Blomberg’s chapter would be so much easier. One possible explanation is textual: Reasonable Faith a revised version of Craig’s Apologetics: An Introduction published in 1984, and it may be that Blomberg’s chapter was absent from that edition and Craig saw no reason to work it in thoroughly to the new edition, even though he thought Blomberg’s contribution had some general usefulness.

However, I think the main reason is that Craig is anxious to position himself as mainstream. Throughout his writings, he repeatedly claims that it can be proven only using facts that are accepted as the consensus of modern scholarship. In his debates on the existence of God, the appeal to consensus is often the only argument presented for the resurrection.[57] In his debate with Biblical scholar Gerd Lüdemann, Lüdemann argued that the Gospels were not eyewitness reports, and Craig was very quick to say that he wasn’t claiming otherwise.[58] Similarly, when arguing for the existence of God, Craig avoids depending heavily on attacks on evolution, in spite of his occasional parroting of stock creationist canards (an example of which we saw in chapter 2). In fact, in the “Practical Application” section of chapter 2, Craig recommended to his readers that they actively distance themselves from creationism.

Though Craig’s claims about the consensus of Biblical scholarship have been disputed[59], they are at least not obviously false. In contrast, if he claimed that a majority of Biblical scholars believed that the Gospels are eyewitness reports, anybody and everybody with a basic knowledge of the field would be crying foul. Unwilling to position himself against the majority, he concedes the point for the sake of argument.

Though presenting your position as the consensus obviously makes some strategic sense, the cost Craig ends up paying is very high and considering the problems to be described below, it’s not entirely clear to me why he would think his approach is the one most likely to succeed. I can only suspect that, though he may personally accept traditional Christian claims about the Gospels, he has accepted that they cannot be credibly defended before an outsider.

On the other side, though, I should note that an appeal to what was argued in Chapter 6 wouldn’t have much effect on what follows. I think there are good reasons to deny that Blomberg’s chapter established much of anything, so an appeal to it wouldn’t really help Craig’s last two chapters.

The Chapter 7 “Assessment” begins with a clear issue requiring resolution: the claims of modern scholarship “play havoc” with the old “Lord, liar, or lunatic” apologetic promoted by C.S. Lewis. The “trilemma” ran as follows:

  1. Jesus claimed to be “Lord” (God or at least the Son of God).

  2. He was either right or he wasn’t.

  3. If he was wrong, he either believed it or he didn’t.

  4. If he was wrong and didn’t believe it, he was a liar.

  5. If he was wrong and believed it, he was insane, a “lunatic.”

  6. He was not lying or insane.

  7. It follows from 1-6 that we should accept Jesus as Lord.

Modern scholars question the first premise, opening up a third possibility: that the Jesus who claimed to be God is a legend. Questions about the general reliability of the Gospels aside, scholars have noticed a great gap between the portrayal of Jesus in John and the portrayal in the earlier Gospels. This suggests that John is largely ahistorical, in which case Jesus never made such bold claims as “I and the Father are one” (John 10:30). Craig sets out to show that Jesus’ view of himself was basically the view adopted by orthodox Christianity.

First, he defends the claim that Jesus used the title “Son of Man,” which in the book of Daniel is a divine figure. Next Craig concedes, for the sake of argument, that Jesus did not use the titles attributed to him, as most New Testament scholars believe. He then argues that we can establish an “implicit Christology” based on the sayings of Jesus widely regarded to be genuine. This yields a Christology consisting of four broad points: (1) Jesus thought that he was a Son of God in a unique sense; (2) he claimed to act and speak with divine authority; (3) he believed that he could perform miracles; and (4) he claimed to determine people’s eternal destiny before God.

I find it likely that Jesus thought that he was a very important person sent from God to herald in the Kingdom of Heaven on Earth. This easily accounts for the second and third points of Craig’s implicit Christology. The fourth point is supported by a single verse, Luke 12:8-9: “I tell you, every one who acknowledges me before men, the Son of man will also acknowledge before God; but he who denies me before men will be denied before the angels of God.” But this verse only implies that Jesus thought he was a messenger of God, not God incarnate. The latter interpretation recalls that of modern preachers who are so certain that their version of Christianity is correct that they interpret the misfortunes of their opponents as the wrath of God. Even if Jesus thought that his prophetic mission was predicted by the Old Testament prophets, he would not have necessarily identified himself as the Son of God.

Craig’s first sign that Jesus thought of himself as the unique Son of God is his use of “father” to refer to God. But this is weak support. The fact that early Christians also referred to God as “father” without thinking of themselves individually as the unique Son of God undermines this point. Craig maintains that this usage must have originated from Jesus, but if Jesus intended others to imitate his usage, then it would not follow from his own usage that he thought of himself as God’s son in a unique sense.

Then there is the parable of the vineyard, where Jesus implicitly identifies himself as God’s son. Craig argues for its authenticity in part based on a similar saying in the Gospel of Thomas, which “lacks any climax and point.”[60] This lends some support for his case. Though this is not the broad point Craig likely wants his readers to take away from this fact, I wonder if the Gnostic account in Thomas is earlier and truer to what Jesus actually said.

Finally, Craig mentions sayings where “Father” and “Son” appear alongside each other. This also lends some support to his case.

Nevertheless, there are reasons to doubt that Jesus ever thought of himself as the Son of God. The notion is less likely to have arisen within the Judaism of the time, and more likely to be an addition that entered Christianity as the new sect spread to pagan lands. If Jesus used “Son of Man” but his use of “Son of God” is questionable, we may think that he thought of himself as the former but not the latter.

Lastly, there is the matter of the virgin birth. It would of course be very odd to accept it while rejecting the Resurrection. My reasons for rejecting the Resurrection have been partially made out in my analysis of Chapter 4 and more are to follow, but for now I will take it for granted that the virgin birth did not actually happen.

It is possible that Jesus claimed or believed that he was born of a virgin even though he wasn’t. Some early critics claimed that Jesus was illegitimate, and they have their modern followers. Perhaps Mary told her son that he was conceived through God to hide his true origins from him. However, I consider this unlikely for two reasons. First, the doctrine of the virgin birth appears only in the birth narratives in Matthew and Luke. These can be safely regarded as ahistorical based on contradictions between them and the obvious motive to invent such stories in order to exalt Jesus. Also, it seems that Jesus had younger siblings, an idea that hardly squares with the picture of Mary the unwed mother painted by some critics.

Therefore, I do not think that Jesus believed in the virgin birth, and for this reason he could not have thought of himself as Son of God in the same sense that later Christians did. Perhaps he had some notion of adoptionism. Perhaps, though, he simply thought of himself as an agent of God, akin to Moses or David, but not in any way God’s son.

Salvaging the Trilemma

Whatever he exactly believed, it seems that Jesus thought of himself as some kind of great personage. Was he, then, a lunatic?

Consider the case of William Miller. In the first half of the nineteenth century, he predicted that the world would end in 1843 or 1844. I have never read anything to indicate that he was stark raving mad; yet the world did not end in the 1840s. The application to Jesus is clear: Jesus may very well have predicted the end of the world within his lifetime (see Mark 9:1 and all of Mark 13), but he need not have been insane in order to have believed that the end was nigh. Similarly, modern preachers like Jerry Falwell, Pat Robertson, and Reverend Sun Myung Moon fall short of being complete lunatics, but from this it does not follow that whatever they claim is true.

The notion that Jesus is Lord does not fare much better. Even ignoring his false end-times preaching, the New Testament Jesus hardly comes across as infinitely wise. Though Jesus’ injunction against resisting evildoers may have been wise in the face of an invincible Roman Empire, I am glad that it was not taken too seriously during World War II. At one point in the Gospels, Jesus even curses a tree. It takes pretty heavy blinders to read such passages and think that they reflect an infinitely wise God incarnate.

Craig never directly defends C.S. Lewis’ trilemma; he simply assumes that his readers are familiar with it and then supplements it with what he regards as a key missing component. Near the end of the chapter Craig does write: “The balance and soundness of Jesus’ whole life and teachings make it evident that he was no lunatic.”[61] However, he does not provide a single sentence of further support or explanation.

Lewis did not do much better. The space he devotes to developing the trilemma in Mere Christianity is very short, and the trilemma is laden with phrases like “it seems to me obvious” in the place of actual arguments. Lewis appears to have been working on the assumption that his potential converts thought that Jesus was a good teacher but not God. This reading is supported by such phrases as “the really foolish thing that people often say about Him” and “let us not come up with any patronizing nonsense.”[62]

But if Lewis believed that prominent critics of Christianity would grant that Jesus was a good teacher, he was mistaken. In his “Why I Am Not a Christian” lecture, Bertrand Russell made a special point of arguing against the wisdom of Jesus’ teachings, and Robert G. Ingersoll went so far as to say that “the New Testament is infinitely worse than the Old.”[63]

Nevertheless, many of Lewis’ readers, then and now, undoubtedly did take the view he was attacking. Perhaps, then, we should take his argument as a defense of orthodox theology addressed to liberal Christians who view Jesus as a great teacher but who have rejected the old theology. I do not find the liberal view of Jesus particularly credible, so I am somewhat sympathetic to this attack on it. On the other hand, I see no good reason to think that Jesus was an orthodox Trinitarian, or that he said anything attributed to him in the Gospel of John. Craig never disputes this point, probably because he could not do so successfully while basing his argument on the findings of mainstream New Testament scholarship.

Chapter 8: Christ: The Resurrection of Jesus

Though evangelicals often presume the complete historical accuracy of the Gospels when they defend the historicity of Jesus’ resurrection, Craig does not take that approach. Instead, he defends the historicity of a few points in the Gospel accounts and then argues that those “facts” are best explained on the hypothesis that Jesus really did miraculously resurrect from the dead.

The first alleged fact to which Craig appeals is the discovery that Jesus’ tomb was empty. In I Corinthians 15, Paul very briefly (and without any of the detail found in the Gospels) mentions that Jesus was crucified and appeared to many of his followers, at least a couple of whom Paul had met. While New Testament scholars widely doubt that the Gospels were written by the men they are attributed to, this report is significant because most scholars concede that Paul wrote I Corinthians.

Craig claims that, first, Paul’s account is evidence that Jesus was buried and that both local Jews and Christians knew where his tomb was located, so early Christians would not have proclaimed the Resurrection if Jesus’ body remained in the tomb. Second, Craig gives a rather silly argument that Paul implies that the tomb was empty, even though Paul only mentions Jesus’ burial, and never even uses the word “tomb.” Standard practice at the time was to bury executed criminals in a criminal’s graveyard.

Paul may very well have believed that Jesus’ body was absent from his grave, but this does not entail that anyone attempted to verify any reports of an empty tomb. Assuming that anyone even remembered which burial plot belonged to Jesus, his body would have decayed too quickly to have been worth displaying for anything but a very short time. In The Case for the Resurrection of Jesus, Gary Habermas claimed that Jesus’ body could be identified based on “hair, stature, and distinctive wounds.”[64] This strikes me as a particularly weak response to this point given that we have little reason to suppose that Jesus had a wildly different haircut or stature than his contemporaries, and only the historically suspect Gospel of John portrays Jesus’ wounds as those distinguishable from those of other crucified victims. The phrase “on the third day” supposedly refers to the discovery of the empty tomb, but it could easily refer to post-Resurrection appearances of Jesus (it refers to both in the Gospels), or it could be an inference from Old Testament prophecies.

Based on Paul’s testimony alone, the most we could conclude is that Jesus was buried in a criminal’s graveyard. It seems to me that Craig is reading into Paul what he wishes Paul had said, for Paul is the closest thing we have to a historically sound source for the events surrounding Jesus’ life. But by any unbiased reading, Craig’s conclusions cannot be derived from Paul’s testimony itself.

Next let’s consider the argument that had the story been fabricated, its authors would not have made women witnesses to the empty tomb because the testimony of women was then considered worthless. First, this argument might have more force if the earliest version of the story available presented the women as witnesses, but it does not. According to Mark 16:8, the women “told no one,” leaving the reader wondering how the story reached the author’s ears. A “temporary silence” theory would be credible if it were consistent with the story presented by later Gospel writers, but it is not consistent. According to Luke, the women told their story before the first appearances were reported that same day; John says that Mary Magdalene ran to Peter and the Beloved Disciple; and Matthew 28:8 explicitly excludes a temporary silence: “the women hurried away from the tomb, afraid yet filled with joy, and ran to tell his disciples.”

The fact that later writers deleted the silence entirely suggests that they realized that it didn’t make sense. A silence makes sense on the theory that the discovery story was a recent legend, with the silence explaining why people hadn’t heard it before. It does not make sense on the theory that the discovery story is historical. This greatly undermines our trust in the Gospels on this point, leaving us where we were with Paul.

The other arguments are quite weak. The fact that one of Jesus’ enemies is said to have buried him is supposed to support the reliability of the burial story, but perhaps the early Christians reasoned that only a member of the Sanhedrin would have the authority to secure an honorable burial. Craig also claims that the empty tomb story was included in Mark’s source material, but no such material survives, making this assertion mere conjecture. When such conjectures are used to support the Resurrection, I cannot help but hear echoes of Thomas Paine: “is it more probable that nature should go out of her course,” or that a bit of scholarly guesswork should be wrong?[65] Craig also claims that the simplicity of the resurrection narratives is evidence for their veracity, but I doubt if he would conclude that a Muslim miracle claim is likely historical simply because it is only ‘moderately extraordinary.’ Finally, Craig argues that the earliest Jewish propaganda granted the empty tomb, thus the Resurrection must have been a real event; but we have no copies of early Jewish propaganda, just Christian writers’ accounts. That early Christian writers were unbiased guides to their opponents is an assumption as dubious as the assumption that Lee Strobel’s apologetic works are fair guides to the arguments of atheists.

Even cumulatively, I do not think these arguments overcome the difficulty of reconciling Mark’s ending with the historicity of the empty tomb story. Moreover, we have no reason to put that much trust in the Gospel accounts. Consider Craig’s arguments for the historicity of the Gospel accounts of the postmortem appearances of Jesus. Craig cites two authorities, Julius Müller and A. N. Sherwin-White, who claim that there wasn’t enough time for legendary development in these accounts. According to Craig, Sherwin-White maintained that “for the gospels to be legends, the rate of legendary accumulation would have to be ‘unbelievable.'”[66] But Sherwin-White’s own words paint a different picture:

Mr. P. A. B. Bruce has suggested in private correspondence that the study of the Alexander sources is less encouraging for my thesis. There was a remarkable growth of myth around his person and deeds within the life-time of his contemporaries, and the historical embroidery was often deliberate. But the hard core still remains, and an alternative but neglected source–or pair of sources–survived for the serious inquirer Arrian to utilize in the second century A.D. This seems encouraging rather than the reverse. The point of my argument is not to suggest the literal accuracy of ancient sources, secular or ecclesiastical, but to offset the extreme scepticism with which the New Testament narratives are treated in some quarters.[67]

In arguing that the rate of myth-making can be quite rapid, but that there is still some possibility of recovering the truth, Sherwin-White nearly claims the opposite of what Craig attributes to him. Clearly, the fact that the truth was preserved in the case of Alexander does not mean that this happened in all cases. In fact, the whole point of a legend is that at least a few people believe it.

According to Craig, “Müller challenged the scholars of the mid-nineteenth century to show anywhere in history where within thirty years a great series of legends had accumulated around a historical individual and had become firmly fixed in general belief.”[68] But the stories about Jesus were not fixed in “general belief,” for the Jews and pagans of the time did not believe them, and it is unclear what Craig means by “a great series.” Moreover, there is no need to think that the attribution of miracles to Jesus is purely legendary: some miracle claims might be based on actual, though entirely natural, events.

Craig also says that “the earliest Christians would have passed on the Jesus tradition with the care and respect for that tradition which was typical of Jewish folk transmission of traditions, which renders analogies drawn from folk literature or ‘urban legends’ irrelevant.”[69] This is a tall claim with minimal support. In nigh every other case where extraordinary claims have been made, even respectable news outlets have been careless and credulous. Can we be so confident in the hypothetical chain of transmission that we are compelled to accept everything that comes out of it, no matter how extraordinary?

What’s more, experience with modern urban legends, spread today by the magic of the Internet, actually results in a surprisingly close analogy. Think about it: when you hit “forward” on a story a friend sent you, it is reproduced with 100% accuracy. Online legends, then, are not produced one small change at a time, but though periodic and major lapses in what would otherwise be a reliable transmission method. Even granting perfectly accurate oral transmission most of the time, we may suspect that legend crept in nonetheless, one large chunk at a time.

Once Craig conceded that the Gospels are not eyewitness reports, one of the few arguments left with even the slightest chance of saving his case is that there was not enough time for legends surrounding the Resurrection to form. Unfortunately, the argument rests on appeals to authorities who are much more judicious in their conclusions than Craig lets on.

Absent Paul’s testimony, there would be no reason to think that the post-Resurrection appearances were anything but legends. According to Craig, Paul’s testimony “makes certain that on separate occasions different groups and individuals had experiences of seeing Jesus alive from the dead.”[70] But Paul’s testimony about groups is pure hearsay, as he does not tell us who any of these people are or where he got his information. We only know the identity of the twelve disciples through the Gospels, and these same texts fail to corroborate the 500 supposed witnesses to the post-Resurrection appearances mentioned in I Corinthians 15:3-6.

Nevertheless, Craig cites C. H. Dodd’s argument that Paul mentioned the 500 to note that they were still alive at the time and thus available for questioning. Unfortunately, Paul did not offer his contemporaries specific contact information for a single one of these “witnesses.” And even if he had provided as little as their general places of residence, in order to confirm Paul’s claim the members of the Corinth Church would have had to take a long journey to Palestine that they hardly would’ve been ready to take. Even for the twelve, where Paul had the opportunity to check the facts, there is no record that he did so. In any case, it hardly seems wise to hinge the case for a monumental miracle on the mere assumption that a man who lived 2000 years ago was a good fact-checker.

The only thing that we can be reasonably sure of is individual appearances, which easily could have been hallucinations. In fact, on one or more occasions a group of people may have worked themselves up into thinking that they saw Jesus alive. In a famous and potentially analogous case, a group of 19th-century sailors convinced themselves that a floating wreck was the ghost of their dead cook.[71] That the appearances of Jesus happened in different circumstances is a poor response given that we have no sufficiently reliable record of the circumstances in which the appearances occurred.

Finally, Craig argues that the Resurrection is the best explanation of the emergence of the early Christians’ new ideas. Surely, though, this is explanatory overkill, akin to invoking alien abductions to explain coin-shaped bruises on one’s back.[72] Human beings are very good at coming up with all sorts of odd new beliefs, such as:

  1. The belief that diseases can be cured by substances that cause their symptoms (a tenet of homeopathy).

  2. The belief that the Sun is inhabited.

  3. The belief that the Earth is a hollow sphere with a civilization on the inside, and that that subterranean civilization is the source of UFO sightings.

  4. The belief that the Earth is a hollow sphere with a civilization on the inside, and that that subterranean civilization is us.

How does orthodox Christianity explain the emergence of new beliefs? Supposedly, God spent a great deal of time preparing the Jews for Jesus, but failed to clearly explain his coming. Craig argues that the idea of a dying and rising messiah was new to the Jews; but why didn’t God clearly reveal the notion to Isaiah earlier? Christians are forced to read prophecy into the Old Testament in an after-the-fact manner scarcely distinguishable from the interpretation of Nostradamus. This is puzzling on the hypothesis that God directly willed the emergence of Christianity, but unremarkable on the hypothesis that Christianity is just another human idea.

By conceding the received view of the authorship of the Gospels, Craig avoids contradicting mainstream scholarship at the expense of leaving us with no reason to trust Gospels accounts of miracles. Paul is as close as we get to a good early Christian source, but a few sentences from one writer are far from sufficient to establish a miracle.


The defense of Christianity provided in Chapters 3-5, 7, and 8 of Reasonable Faith fails every step of the way. William Lane Craig’s arguments for the existence of God provide the most food for thought, but even if they were successful, they would only provide grounds for belief in a God which is a far cry from the Christian God, particularly in regard to the problem of miracles. Standard objections to miracles are largely independent of whether some sort of god exists, in part due to the fact that they were first formulated by deists skeptical of miracles but not of a Creator.

Craig does not adequately answer arguments that belief in miracles is not warranted, and his rebuttal to Hume is particularly sloppy. After attributing Hume’s main insight to another writer, Craig attempts to rebut it with an argument that could only succeed if the evidence for Christianity’s miracles were completely unambiguous and without parallel, and he offers no grounds to think that this is the case while ignoring relevant parallels. He seems to think that if Jesus shared orthodox Christianity’s view of himself, this would render the Resurrection credible by giving it a “religious-historical context”; but this view is problematic and inadequately developed by Craig. Finally, Craig’s evidence for the Resurrection leans heavily on the Gospel accounts without rebutting the charge that they are hearsay. In order for his overall strategy to work, Craig would not only have to refute Hume’s “in principle” objection to rational belief in miracles, but argue that individuals should not apply the smallest amount of common-sense skepticism to miracle claims.

When I look back on the totality of the argument, though, I realize that Lessing’s objection is perhaps more fatal than I first thought. Craig does not merely want the chain of reasoning to end with the conclusion that Jesus rose from the dead. As he makes clear in his discussion of Jesus’ self-understanding, especially his discussion of the claim that Jesus would determine each person’s eternal destiny, he wants the chain of reasoning to end in orthodox Christianity. This includes the belief that only Christians go to Heaven. He wants people to forswear forever all the moral knowledge which tells them that no one deserves eternal punishment for picking the wrong religion. Craig didn’t come to this belief on rational grounds, and I don’t think it can be defended on rational grounds. In fact, it is a stance that is fundamentally contrary to rational thought.

Ironically, Craig’s first chapter may have demonstrated this with a force never previously achieved by any critic. The key passage is when he says that the magisterial use of reason cannot be accepted if unbelief is to be punished. To embellish his argument ever so slightly, Christianity says that belief is obligatory in all cases (since unbelief is always punished by damnation), yet this means that belief is obligatory when all rational grounds are against it. The point is undeniable and would be trivial if it were not so rarely noticed. To say that we can be obliged to believe something that all reason is against is to disregard reason entirely in the matter. This is the policy that appears to be required at a very fundamental level by Christianity.

Now the only question remaining is whether such a policy is at all credible. Aside from the intuitive considerations against it, there is the problem that any defense of it will invariably appeal to reason, but if a reasoned defense is to have any weight, the principle would have to be rejected. More generally, every apologetic argument supposes that reason must be taken seriously. Thus we reach the conclusion that any apologetic for Christianity must presuppose that Christianity is false. This is, ultimately, why Craig’s apologetic fails.


[1] William Lane Craig, Reasonable Faith (Wheaton, IL: Crossway Books, 1994), p. xi.

[2] Ibid., pp. 35-36.

[3] Paul Doland, “The Case Against Faith: A Critical Look at Lee Strobel’s The Case for Faith” (4th edn., 2006). The Secular Web. </library/modern/paul_doland/strobel.html>.

[4] Tom Wanchick, “Trouble in Paradise? Michael Martin on Heaven” (2003). The Secular Web. </library/modern/paul_doland/strobel.html>.

[5] Craig 1994, p. 36.

[6] Lee Strobel, The Case for Faith (Grand Rapids, MI: Zondervan, 2000), pp. 66-67.

[7] Craig 1994, p. 37.

[8] Ibid., p. 29.

[9] Ibid., p. 49.

[10] Ibid., pp. 49-50.

[11] Ibid., p. 59.

[12] Ibid., pp. 59-60.

[13] See, for example, the Martin-Copan exchange: Atheism, Christian Theism, and Rape; Can Michael Martin be a Moral Realist? Sic et Non; Copan’s Critique of Atheistic Objective Morality; Atheistic Goodness Revisited: A Reply to Michael Martin; and The Naturalistic Fallacy and Other Mistaken Arguments of Paul Copan.

[14] Craig 1994, p. 67.

[15] Does God Exist? The Craig-Flew Debate, ed. Stan W. Wallace (Burlington, VT: Ashgate, 2003), pp. 21-22, 168-172.

[16] Craig 1994, p. 225.

[17] Ibid., p. 71.

[18] Ibid., p. 91.

[19] Stephen Jay Gould, “Evolution as Fact and Theory.” <dead link>.

[20] Craig 1994, p. 94.

[21] Robert Heinlein, Stranger In a Strange Land (New York: Putnam, 1961), p. 140.

[22] Craig 1994, p. 98.

[23] Ibid., p. 99.

[24] Graham Oppy, “Inverse Operations With Transfinite Numbers and The Kalam Cosmological Argument” and “Reply to Professor Craig“; William Lane Craig, “Graham Oppy on the Kalam Cosmological Argument.”

[25] See the transcript of The Washington-Craig Debate at Craig’s website.

[26] Don Cox, personal communication, March 3, 2006.

[27] Richard Carrier, “Why I Am Not a Christian” (2006), Part 4: “Christianity Predicts a Different Universe.”

[28] Craig 1994, p. 117.

[29] William Lane Craig, “Must the Beginning of the Universe Have a Personal Cause? A Rejoinder.” <http://www.leaderu.com/offices/billcraig/docs/morriston.html>.

[30] John D. Barrow and Frank J. Tippler, The Anthropic Cosmological Principle (Oxford: Oxford University Press, 1986), specifically pp. 562-565.

[31] Ibid., p. 118.

[32] Creationists have long promoted this misunderstanding of evolution; see TalkOrigin’s “Five Major Misconceptions about Evolution: The theory of evolution says that life originated, and evolution proceeds, by random chance.” <http://talkorigins.org/faqs/faq-misconceptions.html#chance>.

[33] See Craig’s response to Gary Habermas at the end of Five Views on Apologetics, ed. Steven B. Cowan, Grand Rapids: Zondervan, 2000. For another example, in his debate with Eddie Tabash, Tabash claimed that the initial probability of the Resurrection was extremely low, and in response Craig said, “The resurrection is not improbable intrinsically. What is improbable is the hypothesis that Jesus rose naturally from the dead. That is incredibly improbable, but the Christian agrees that that is not likely to have happened. The hypothesis is God raised Jesus from the dead. And given the existence of God established by the cosmological, the contingency, and the argument from evil that Eddie didn’t refute, we have good reasons to believe that God exists, so this is not an intrinsically improbable event.”

[34] Craig 1994, p. 148.

[35] Useful modern investigations of these subjects include William Nolen’s Healing: A Doctor In Search of a Miracle (New York: Random House, 1974), James Randi’s The Faith Healers (Buffalo, NY: Prometheus Books, 1987), and Michael Cuneo’s American Exorcism (New York: Doubleday, 2001).

[36] Craig 1994, p. 149.

[37] Ibid., p. 150.

[38] Paul Doland: “The Case Against Faith: A Critical Look at Lee Strobel’s The Case for Faith” (2006), “Objection 2: Since Miracles Contradict Science, They Cannot Be True.”

[39] Craig 1994, p. 297.

[40] Ibid., pp. 147-149.

[41] Ibid., pp. 151-152.

[42] Joe Nickell and John F. Fischer. “Incredible Cremations: Investigating Spontaneous Combustion Deaths.” Skeptical Inquirer, Summer 1987, pp. 352-357.

[43] Harry Houdini. A Magician Among the Spirits (New York: Harper and Brothers, 1924).

[44] Craig 1994, p. 153.

[45] Ibid., p. 154.

[46] Henry Chadwick, trans. Lessing’s Theological Writings (London: Adam & Charles Black, 1956), p. 54.

[47] Craig 1994, p. 205.

[48] Ibid.

[49] Raymond E. Brown, Introduction to the New Testament (New York: Doubleday, 1997), pp. 159-160.

[50] Ibid., pp. 327-328.

[51] Bart Ehrman, Jesus: Apocalyptic Prophet of the New Millennium (Oxford: Oxford University Press, 1999), p. 157.

[52] Craig Bloomberg in Craig 1994, p. 207.

[53] Craig L. Bloomberg, “The Legitimacy and Limits of Harmonization” in Hermeneutics, Authority, and Canon (Grand Rapids, MI: Academic Books, 1986): 135-174.

[54] For the problems with the birth narratives, see Tom Flynn’s “Matthew vs. Luke: Whoever wins, coherence loses.” For the resurrection narratives, see Farrell Till’s essays “Did They Tarry in the City?” and “Did They or Didn’t They?

[55] Craig Bloomberg in Craig 1994, p. 222.

[56] Craig Blomberg, personal communication, May 16, 2007 and May 20, 2007.

[57] Transcripts of many of Craig’s debates are available at his website: . Examples of the cited strategy appear in the debates with Edwin Curley, Massimo Pigliucci, and Corey Washington.

[58] Jesus’ Resurrection: Fact or Figment? A Debate Between William Lane Craig and Gerd Lüdemann (Downers Grove, IL: InterVarsity Press, 2000), p. 46.

[59] See, for example, Dale Allison’s comments on the empty tomb in Resurrecting Jesus (New York: T & T Clark, 2005), p. 300.

[60] Craig 1994, p. 245.

[61] Ibid., p. 252.

[62] C. S. Lewis, Mere Christianity (New York: The MacMillian Company, 1958), pp. 40-42.

[63] Bertrand Russell, “Why I Am Not a Christian” and Robert G. Ingersoll, “Why I Am An Agnostic.”

[64] Gary Habermas and Mike Licona, The Case for the Resurrection of Jesus (Grand Rapids: Kregel Publications, 2004), p. 70.

[65] Thomas Paine, “The Age of Reason, Part First” in The Life and Writings of Thomas Paine (New York: Vincent Parke and Company, 1908), p. 93. The full quote is:

If we are to suppose a miracle to be something so entirely out of the course of what is called nature, that she must go out of that course to accomplish it, and we see an account given of such miracle by the person who said he saw it, it raises a question in the mind very easily decided, which is, is it more probable that nature should go out of her course, or that a man should tell a lie? We have never seen, in our time, nature go out of her course; but we have good reason to believe that millions of lies have been told in the same time; it is therefore, at least millions to one, that the reporter of a miracle tells a lie. (pp. 92-93)

[66] Craig 1994, p. 285.

[67] A. N. Sherwin-White, Roman Society and Roman Law in the New Testament (Oxford: Clarendon Press, 1963), pp. 192-193.

[68] Craig 1994, p. 285.

[69] Craig 1994, p. 284.

[70] Ibid., p. 284.

[71] Samuel Hibbert, Sketches of the Philosophy of Apparitions (Edinburgh: Oliver & Boyd, 1824), pp. 403-404.

[72] Susan Clancy, Abducted: How People Come to Believe they Were Kidnapped by Aliens (Cambridge: Harvard University Press, 2005), p. 33.

Copyright ©2007 Chris Hallquist. The electronic version is copyright ©2007 by Internet Infidels, Inc. with the written permission of Chris Hallquist. All rights reserved.