The Kalam God Who Doesn’t Do Kalam Things
A respectful critique of William Lane Craig’s framework
I owe William Lane Craig an enormous intellectual debt, and I want to say that up front before I say anything else. His revival of the Kalam cosmological argument gave an entire generation of Christian thinkers the philosophical tools to stand in any room and defend the rationality of theism. His debates with Hitchens, Krauss, Carroll, and Harris set a standard for rigorous public engagement that most apologists still haven’t matched. The man has two earned doctorates, a publication record that would make most philosophers jealous, and a personal graciousness that anyone who’s met him will confirm.
I’m not here to take shots at Bill Craig. I’m here to press on what I think is a structural incoherence in his position on origins, precisely because his work matters enough to warrant the scrutiny.
The Kalam argument, as Craig presents it, is elegant. Whatever begins to exist has a cause. The universe began to exist. Therefore the universe has a cause. From there, Craig derives that this cause must be transcendent, immaterial, immensely powerful, timeless, and personal. An agent who chose to create. The argument works because Craig insists on the full causal weight of that word “personal.” A mechanistic cause that exists eternally should produce an eternal effect. Only agent causation explains a temporal beginning from an eternal state.
So far, so good. This is first-rate philosophical theology, and I’ve built on it extensively in my own work.
Here’s where the tension starts. Craig holds an old-earth position. He’s sympathetic to progressive creation, skeptical of theistic evolution but comfortable with deep time, and reads the days of Genesis as something other than ordinary sequential days. The specifics vary depending on which interview or debate you catch, but the consistent thread is this: Craig accepts the standard cosmological and geological timescales, and he interprets Genesis in light of them.
That’s a respectable position held by serious scholars. I’m not dismissing it out of hand. But I want to trace where it leads, because I think it leads somewhere Craig doesn’t want to go.
When you accept standard uniformitarian timescales, you’ve made a hermeneutical decision with consequences that cascade through your entire framework. You’ve effectively said that the geological and cosmological record, read through uniformitarian assumptions, provides the controlling context for interpreting Genesis. Scripture gets reinterpreted to fit the timescale rather than the timescale being evaluated in light of Scripture.
Now, Craig would probably push back on that characterization. He’d argue that the text itself is ambiguous on the length of the days, that yom can mean different things, that the literary structure of Genesis 1 permits non-literal readings. Fair enough. Those are real exegetical arguments and they deserve engagement.
But notice what happens to divine fiat when you take this route. “God said, ‘Let there be light,’ and there was light.” “God said, ‘Let the earth bring forth living creatures,’ and it was so.” “God said, ‘Let us make man in our image.’” The repeated pattern in Genesis 1 is divine speech producing immediate creaturely effect. Psalm 33:9 confirms it explicitly: “For he spoke, and it came to be; he commanded, and it stood firm.”
If the days are ages, and the mechanisms are natural processes operating over billions of years, what work is the divine speech actually doing? God “says” something, and then... waits four billion years for chemistry and physics to produce what He said? That’s not fiat. That’s a suggestion with a very long implementation timeline.
And this is where I think the position collapses into what I’d call, with all due respect, a kind of Uniformist Deistic Naturalism wearing Christian formal attire. God fires the starting gun at the Big Bang. Then natural law runs the show for roughly 13.8 billion years. Stars form by gravitational collapse. Planets accrete from dust. Life arises through... well, Craig is wisely cautious here, but the framework pushes toward some form of guided natural process. Then at some point God steps back in for the really important stuff: ensoulment, moral accountability, resurrection.
You end up with a theology that needs three different explanatory frameworks stitched together. Divine causation at the cosmic origin. Functional naturalism for everything between the Bang and humanity. Then divine action again for salvation history. That’s not parsimonious. That’s a Frankenstein architecture with seams showing everywhere.
Compare this with what I’ve been developing as the Literal Programmatic Intervention (LPI) model. One consistent framework: God designs the system, establishes fiat initial conditions at creation, governs through discoverable physical law between interventions, and intervenes at scripturally specified historical points. The fiat conditions aren’t embarrassments to be explained away. They’re the boundary conditions that make everything else coherent. Between those interventions, you model the physics honestly and let the evidence take you where it goes. One architecture. One designer. Consistent methodology throughout.
Here’s the deeper irony. Craig’s philosophical work on the Kalam is superb precisely because it takes divine causation seriously. The whole argument depends on a God who acts. Who brings spacetime into existence from nothing. Who chooses, as a personal agent, to create. The Kalam God is not a spectator. He’s the most causally significant being conceivable.
But when Craig arrives at Genesis, that same God apparently doesn’t act the way Genesis describes. The Kalam God creates the universe ex nihilo, but the Craig God doesn’t create life from dust in the way the text presents. Doesn’t form woman from man. Doesn’t speak kinds into existence in the sequence and manner Genesis records. The philosophical theology demands a God of radical causal power. The hermeneutics then domesticate that power into something compatible with uniformitarian timescales.
The tension isn’t fatal to Craig’s theism. He’s still a theist, and a very sophisticated one. But it creates an internal incoherence between his best philosophical argument and his reading of the text that argument is supposed to support. The God who can create a universe from absolute nothing apparently can’t (or didn’t) create life from dust in a week.
There’s also the progressive revelation problem. Progressive revelation is a crucial hermeneutical principle: God reveals truth progressively, with later revelation deepening and clarifying earlier revelation. Jesus says, “You have heard that it was said... but I say to you,” and He’s not contradicting Moses. He’s fulfilling what Moses pointed toward.
But progressive revelation requires that earlier revelation is true in what it affirms, even when later revelation adds depth. The progression has to progress from something real. If Genesis 1 doesn’t describe anything God actually did in the way described, then there’s nothing for later revelation to deepen. You’re not building on a foundation. You’re building on a metaphor, and metaphors don’t bear load the way foundations do.
When Exodus 20:11 says, “For in six days the LORD made heaven and earth, the sea, and all that is in them, and rested on the seventh day,” it’s grounding the Sabbath command in the creation pattern. If the six days are figurative, the grounding is figurative too. And once you start pulling that thread, the fabric unravels faster than you’d expect.
I want to be clear about what I’m not saying. I’m not saying Craig isn’t a Christian. I’m not saying his work is worthless. I’m not saying old-earth Christians are heretics. The body of Christ has room for this disagreement, and Craig has done more for the public credibility of Christian theism than almost anyone alive.
What I am saying is that on origins specifically, the accommodation to uniformitarian timescales has created a framework that is less coherent, less parsimonious, and less faithful to the text than the alternative. Craig ends up needing to explain why the God of the Kalam argument, who is demonstrably capable of creating a universe from nothing, would choose to work through billions of years of natural process in a way that looks indistinguishable from what naturalism would predict. The answer is always some version of “well, the text allows it.” Maybe. But “the text allows it” is a long way from “the text teaches it,” and the burden should run in the other direction.
The simpler, more parsimonious move is to take the text at face value, accept that the God who creates ex nihilo also creates by fiat within the creation week, and then do the hard scientific work of showing how the physical evidence coheres with that framework. That’s harder work. It means swimming against the current of consensus geology and cosmology. It means developing models like the Global Hydrotectonics framework I’ve been working on, where you actually engage the physics of rapid tectonic reorganization and show the math can work.
But it’s more honest work. And it produces a more unified theology, because the God of your philosophy is the same God who acts in your text is the same God whose fingerprints show up in your science.
Bill Craig gave us the tools to argue that God acts. I just want us to let Him act in Genesis too.
James (JD) Longmire ORCID: 0009-0009-1383-7698 Northrop Grumman Fellow (unaffiliated research)
oddXian.com | Challenge the Consensus: Working out the harmony of God’s Word and God’s World