The Unification Argument

“He is before all things, and in him all things hold together.” — Colossians 1:17

The Inference We Already Trust

When James Clerk Maxwell unified electricity and magnetism into a single field theory in the 1860s, no one demanded that he prove the alternative — a patchwork of disjoint laws governing two superficially distinct forces — logically impossible. The patchwork was not impossible. It was simply worse. Maxwell’s equations covered the same phenomena with fewer postulates, exhibited deeper structural unity, and predicted electromagnetic radiation as a consequence rather than as an add-on. That was enough. Science accepted the unified theory and did not look back.

The same move repeats throughout the history of physics. Einstein replaced Lorentz ether theory with special relativity. Both theories agreed on the empirical predictions for the data available at the time. Einstein won because his account was vastly more parsimonious: no luminiferous ether, no contraction conspiracies, no preferred frame conspiring to remain undetectable. The data did not force Einstein over Lorentz. The structure of the explanations did. Heliocentrism over Ptolemaic epicycles was decided on the same grounds, decades before Newton supplied the dynamical reason heliocentrism had to be right. One principle covering many phenomena beats many principles covering the same phenomena. This is not aesthetics. This is the working epistemology of every domain in which empirical adequacy alone fails to single out a winner.

The previous three chapters have laid the groundwork. Chapter 1 argued that reality requires a non-contingent, rationally-structured ground. Chapter 2 argued that this ground is best identified as a Mind, by three converging paths — instantiation, normativity, and parsimony — and committed to the strongest available reading of that Mind’s unity: classical Divine Simplicity. Chapter 3 showed that the world we actually inhabit bears precisely the structural signatures that hypothesis predicts. This chapter does not introduce new evidence. It makes explicit the inference the whole book has been running: the same inference Maxwell ran, that Einstein ran, that Copernicus ran. Inference to the best explanation, applied to the most fundamental question we can ask.

The reader is asked to do nothing they don’t already do when they prefer Maxwell to pre-Maxwell. Hold that posture for the duration of the chapter.

Where the chapter is going. The final comparison will be six, three, and one. Naturalism, fairly counted, leaves six independent primitives at the foundation. Mathematical Platonism, the strongest non-theistic competitor, leaves three. Classical theism, on the strongest available reading of divine unity, leaves one — and on the milder fallback reading, still one. By the same parsimony rule that decided Maxwell, Einstein, and Copernicus, one beats three, and one beats six. That is the destination. The rest of the chapter earns it.

The Standard of Inference

Inference to the best explanation is not a lower bar than deduction. It is a different bar, appropriate to questions where deduction is not on offer. Across physics, cosmology, evolutionary biology, geology, archaeology, and the historical sciences, IBE is the operative inference type. Its criteria are well established: parsimony (fewer independent posits), unification (one principle covering many phenomena), scope (the explanation reaches all the explananda without leaving any as untouched brute facts), and predictive coherence (the world the explanation predicts is the world we actually find).

There is no methodological reason to hold metaphysical questions to a higher epistemic standard than scientific ones. To demand mathematical proof of theism while accepting IBE-grade evidence for general relativity is not neutrality; it is an asymmetric standard, and the asymmetry is itself a substantive philosophical commitment that needs its own defense. Every metaphysical position, including naturalism, is offered as a best-explanation claim. None is deduced from self-evident axioms. The fair contest is to apply the same criteria to all and see which wins.

If you accept this framing, the rest of the chapter is close to forced. If you reject it, the disagreement is methodological, not substantive — and the burden is on you to explain why the inference rules we trust in every empirical domain are suddenly off-limits when the question is what reality fundamentally is.

What Counts as One Primitive?

The comparison turns entirely on a prior question: what does it actually mean to call something one primitive?

The naive answer — one word, one entity, one label — is not usable. By that standard, the naturalist could collapse her position to “one primitive: fundamental reality,” and the theist could collapse his to “one primitive: God,” and the comparison would be empty. Both labels would hide arbitrary complexity inside a single noun, and parsimony would have ceased to do any work.

The serious answer requires a structural criterion. A theory is genuinely committed to one primitive when the features of that primitive are not separable items — when they are aspects of a single nature rather than members of a list, such that you cannot have the primitive at all without having all of those features together. A theory is committed to many primitives, even if it labels them as one, when its alleged single primitive is a bundle of externally related features — features that could vary independently of one another, that coexist, but that no internal necessity binds.

Compare two cases. The fundamental field of quantum electrodynamics has many specifiable features: it transforms under U(1) gauge symmetry, couples to charged particles with a specific strength, propagates massless excitations, obeys relativistic invariance. Physicists do not count this as five primitives. They count it as one, because the features are not free parameters that could each be independently varied — they are internally related descriptions of one structure. Change one and you have changed the others.

The issue is not how many words a theory uses. The issue is whether the named features rise and fall together or merely sit beside one another.

Compare this with a hypothetical “naturalism” that asserts: fundamental reality consists of mathematical structure, plus the laws that instantiate it, plus the specific initial conditions, plus consciousness, plus the rational faculties that track truth, plus the correspondence between math and cosmos. That assertion uses one word, but its features are not internally related. The mathematical structure could in principle be different without changing the initial conditions. The cosmological constant could take a different value without entailing different rules for consciousness. There is no internal necessity binding “the rational faculty’s reliable correspondence with truth” to “the specific coupling of the strong force.” These are separate facts that coexist in our world. The label is one; the commitments are many. Bundling them under a single noun does not earn the parsimony credit.

This is the test. And it cuts both ways. The theist does not get to count “Necessary Rational Mind” as one primitive merely by writing it as one phrase. He gets to count it as one primitive only if rationality, intentionality, causal power, normative authority, freedom, necessary existence, and the capacity to ground finite minds belong to one unified nature in the same way the features of a fundamental field belong to one structure — only if those attributes are aspects of one essence rather than items on a list.

So: are they? The classical theistic tradition’s strongest answer is yes, and the answer is Divine Simplicity, developed by Augustine, the Cappadocian Fathers, Maimonides, Avicenna, and Aquinas. The doctrine is precise: in God there are no parts, no metaphysical components, no separable attributes added to an underlying subject. God’s rationality, agency, necessity, intentionality, normative authority, and being are each identical with God’s essence — and therefore identical with one another. To call the Mind rational, actualizing, normative, and intentional is not to enumerate distinct properties belonging to a single bearer; it is to attach distinct human concepts to one undivided reality, describable under different aspects only because finite cognition cannot grasp it all at once. When Aquinas writes Deus est suum esse — God is His own existing — the move is not poetic. It is the deliberate denial that anything in God is added to anything else in God.

This is unity in the strongest possible sense. Divine Simplicity gives theism a primitive that is not merely internally unified but metaphysically uncompounded. There is nothing to count up. The “list of divine attributes” is a list of names finite minds use; it is not a list of components the Mind possesses. The naturalist’s six commitments are six because the features they name could vary independently; God’s attributes cannot vary independently because they are not separable in the first place. One thing cannot fail to coincide with itself.

The reader who finds strict Divine Simplicity too demanding has a milder fallback. Even short of full identity-of-essence-with-attributes, the doctrine of Internal Relatedness claims that the Mind’s attributes implicate one another such that none can be removed without dissolving the rest. Rationality cannot stand alone without intentionality (rationality is about something — propositional content is constitutive of what rationality is). Intentionality cannot stand alone without normative authority over its objects (an intentional state is correct or incorrect). Normative authority cannot operate without the capacity to bind, which in this sense just is causal-evaluative power exercised by an agent. And rational agency that grounds what is rather than describes what is — is what we mean by necessary existence. Pull on any thread and the others come with it. Strict simplicity wins outright; internal relatedness wins decisively. Either route refutes the bookkeeping move.

The primary case in this chapter rests on Divine Simplicity. It is the historical mainline of classical theism, it delivers the strongest available parsimony claim, and the underlying argument is strong enough that the strongest available framing is the one to defend. The honest concession that has to live in this section is also the chapter’s hinge: if internally-related-Mind is mistaken for bundled-Mind — if the attributes of Mind turn out to be separable facts that could vary independently and happen to coexist — then the theistic primitive is not one, it is many, and the parsimony advantage collapses. The chapter does not assert unity and move on; it earns unity by showing the attributes are identical in essence (strict reading) or mutually implicating (milder reading), and it concedes openly what is at stake. A reader who studies the unity argument and concludes the attributes really are detachable has refuted this chapter at the load-bearing point. The argument depends on getting this right. It gets this right, and the rest of the chapter proceeds on that basis. The concession stays visible.

Naturalism’s Inventory, Properly Counted

Apply the same standard to naturalism. Under naturalism, the integrated structure of reality requires at minimum six commitments that no extant theory has unified:

1. Mathematical structure exists. Numbers, sets, relations, the logical and mathematical truths obtain. Naturalism either reifies them (Platonism-as-supplement, which we will treat in its own section), nominalizes them away (leaving the applicability problem unanswered), or accepts them as inexplicable.

2. Physical law instantiates mathematical structure. That anything is rather than nothing — and that what is happens to conform to mathematical regularity — is a separate fact from the existence of the mathematics itself.

3. This particular structure obtains. From the live space of possible physics, the actual laws are these. Why these laws and not others is a further commitment.

4. The constants land in the life-permitting range. Cosmological constant, strong-force coupling, proton-neutron mass difference. Whether we describe this in the careful language of “discrepancy ratio” (preserving Weinberg’s hedge from Chapter 3) or the careless language of “fine-tuning,” the calibration is a further commitment under naturalism.

5. Consciousness arises from non-consciousness. Phenomenal experience emerges from arrangements of non-experiential matter. The hard problem is not a gap in current science; it is a category problem about what kind of explanation could even apply.

6. Rationality reliably tracks truth despite survival-calibration. Cognitive faculties shaped by selection for survival also yield mathematical and metaphysical insight — including the insight that grounds the theory of evolution itself. Why survival-calibrated faculties should be truth-binding rather than merely useful is a further commitment.

Now apply the test. Are these six features internally related? Is there an internal necessity binding (3) to (5), such that you cannot have this particular cosmological constant without having consciousness arising from non-consciousness? No. The features could in principle vary independently. The mathematical structure could exist without instantiation. Instantiation could occur with different structures. The constants could fail the life-permitting test. Consciousness could fail to arise. Truth-tracking rationality could fail to obtain. Each of these is a separate fact that coexists with the others in our world. They are not aspects of one nature; they are items on a list.

This is the correct application of the Hidden Complexity test to naturalism. Naturalism is not allowed to compress these six into one by labeling them “fundamental reality.” The label hides external relations, not internal ones. The naturalist’s inventory is six.

Working physicists may, in principle, eventually unify (2), (3), and (4) — there may be a deeper physics from which the specific laws and constants follow as consequences. If so, parts of the inventory consolidate. But — and this is essential — even a perfect unification of the physical commitments leaves (1), (5), and (6) untouched. Those are not quantitative gaps a deeper physics will close; they are categorial gaps. No advance in physics will explain why mathematical structure exists in the first place, why non-conscious matter generates conscious experience, or why faculties shaped by survival pressures should be reliably bound to truth in regions of reality their shaping had no access to. These are not problems waiting on the next paper; they are problems waiting on a different kind of explanation altogether.

The naturalist inventory, fairly counted, is six. The burden is on naturalism to actually produce a unification of any of them, not to promise one.

Theism’s Inventory, Properly Counted

Under classical theism, the integrated structure of reality follows from one foundational commitment: a Necessary Rational Mind whose essence is, under Divine Simplicity, identical with each of its attributes and therefore wholly without metaphysical compounding. Each of the six features that naturalism leaves as a disconnected fact follows from what such a Mind is:

Mathematical structure is grounded in the Mind’s rationality. Numbers and necessary truths are not external objects the Mind contemplates; they are intrinsic to what rational mindedness is. To be a rational Mind is, in part, to instantiate the necessary structure that mathematics describes.

Instantiation is grounded in the Mind’s causal-evaluative agency. A Mind can actualize possibilities; abstract structures cannot. That a contingent cosmos exists at all is grounded in the Mind’s exercise of agency, not in a separate fact about why the actualization mechanism happens to operate.

The specific physical laws are the Mind’s selection from the space of rational possibilities. Selection, not coincidence. This does not by itself tell us why these laws and not others — that further specification is a fittingness question the chapter does not need to answer — but it does tell us that the question has the shape “why did the Mind select this?” rather than the shape “why does the brute fact of these particular laws obtain rather than no laws or other laws?”

Fine-tuning is the Mind’s calibration toward the conditions for finite minds and meaningful agency. The narrowness of the life-permitting ranges is what we would expect on a Mind hypothesis; on naturalism it is a coincidence that needs its own brute fact. Calibration, not luck.

Consciousness in finite minds is continuous with the Mind’s own nature. Rational consciousness is the ground state of reality, not an emergent anomaly from non-conscious matter. The hard problem dissolves not because we have explained consciousness mechanically, but because consciousness was never supposed to emerge from non-consciousness in the first place — finite minds participate in a feature of the source.

Rationality reliably tracks truth because finite rational faculties and the truths they track derive from the same Mind. Plantinga’s evolutionary argument against naturalism — the worry that survival-calibrated faculties are not automatically truth-binding — does not arise. The faculty’s reliability is grounded, not lucky.

These are not six separate divine functions bolted together. Under Divine Simplicity they are not even six properties of one bearer; they are one undivided essence described under aspects. The unification is not asserted by relabeling. It is earned by the doctrine of unity argued in the previous section.

Theism’s inventory, fairly counted: one — and under strict Divine Simplicity, uncompounded.

Both Views Have a Brute at the Foundation

This sentence is load-bearing for the chapter’s credibility, and it is preserved exactly as v1 stated it: both views have a brute at the foundation. Theism has not eliminated brute fact; it has reduced six to one. The Mind’s necessity is itself the brute. What changes between the two positions is not whether there is something foundational that has to simply be. There is. The question is how much explanatory work that foundational thing does, and how many further foundational things have to also simply be in order to cover the rest of the data.

A six-primitive theory whose primitives are externally related is, by every standard scientific criterion, weaker than a one-primitive theory whose primitive is uncompounded and from which the relevant phenomena follow as natural consequences. This is the Maxwell-over-pre-Maxwell judgment. It is the Einstein-over-Lorentz judgment. It is the heliocentrism-over-Ptolemy judgment. The judgment here is not categorically different. It is the same judgment.

The IBE Comparison

The choice we face is among three candidates: naturalism, mathematical Platonism (as a supplement to or replacement for bare naturalism), and classical theism. Apply the four IBE criteria to each.

On all four criteria, classical theism dominates — provided the unity claim of the previous section holds. Platonism+ is the strongest non-theistic competitor and still loses on parsimony and unification. Naturalism, without supplementation, loses on every criterion simultaneously.

The Platonism Engagement

Mathematical Platonism deserves direct engagement. It is the most serious non-theistic candidate to survive the cosmological argument of Chapter 1 — a coherent position taken seriously by serious philosophers, not a straw target.

Chapter 2 already gave the three reasons Platonism falls short on its own: Platonic abstracta are causally inert (they don’t instantiate), normatively inert (they describe what must hold but cannot bind a finite reasoner), and require additional primitives to bridge those gaps. Here, in the parsimony comparison, Platonism’s structural cost becomes explicit. To cover the same explanatory ground theism covers with one Mind, Platonism requires: (a) the Platonic realm itself, (b) an actualization mechanism distinct from the realm, since the realm is causally inert and cannot bring about a cosmos by being true of it, and (c) a normativity bridge distinct from both, since abstracta cannot reach into finite minds and bind them. Three primitives.

By the same parsimony criterion that decides Maxwell over pre-Maxwell, theism wins against Platonism by 3-to-1 on the milder reading and decisively more on the strict-simplicity reading, where the theistic primitive is not one entity with related properties but one undivided essence. Platonism is doing real philosophical work and deserves the respect of being beaten on its own terms — by being more economical at the same job.

What This Chapter Does Not Establish

The case has landed at Necessary Rational Mind. It has not landed at loving, just, personally involved, morally perfect Mind. That further specification — whether the Mind has a moral character, whether it has acted in history, whether it stands in any particular relation to finite creatures — is not within reach of parsimony considerations. The Unification Argument ends at the metaphysical category. Chapter 5 takes up what follows if moral perfection is added.

The argument also does not establish necessity-of-conclusion. IBE delivers best-current-explanation, not deductive proof. A future deeper unification of naturalism’s six features, or a successful non-theistic ground-candidate not yet on the table, could in principle shift the inference. The conclusion is held with the same confidence we hold any well-supported scientific theory: high, anchored in the standard criteria, and revisable only by an alternative that does the same work better. That is what taking IBE seriously looks like. It is not a hedge against the argument’s force; it is the discipline of the inference type.

The argument is structural, not phenomenological. It operates on the public data — mathematical intelligibility, fine-tuning, the existence of consciousness as a category, the binding character of rational inference — and reaches its conclusion through inference about explanatory adequacy. It does not require first-person religious experience or introspective certainty about any specific mental state. The argument is available to anyone willing to ask the question.

And the brute remains. Even fully accepted, the Mind is not explained away. What has changed between naturalism and theism is the scope of the brute, not its presence. One brute that does the work of six is the win the Unification Argument claims. Theism wins on parsimony; it does not win by eliminating the foundational question that no metaphysical view can eliminate.

Recognition

By the same inference-to-best-explanation criteria that license our preference for Maxwell over pre-Maxwell, Einstein over Lorentz, and heliocentrism over Ptolemaic epicycles, a Necessary Rational Mind is the best current explanation for the integrated structure of reality. This is not a higher epistemic bar than science. It is the same bar.

The argument lands at this exact claim, no more and no less: reality requires a non-contingent ground; the space of viable candidates is small; naturalism leaves six externally-related primitives; Platonism+ leaves three; classical theism proposes one whose essence is, under Divine Simplicity, uncompounded — and from which the relevant phenomena follow as features of what such a Mind is. By parsimony, unification, scope, and predictive coherence — the four IBE criteria we apply throughout science — classical theism’s Necessary Rational Mind is the dominant model currently available.

The reader is now asked to notice one thing. Permanent suspension of judgment in the face of this comparison is not a neutral stance. To suspend judgment is to choose to live as though the second-best explanation is acceptable. That is itself a substantive position, and it now requires its own justification. Nothing in IBE forces a conclusion. But the inference changes what counts as a default. The default is now to follow the standard scientific criteria where they lead. The burden has shifted.

The chapter does not close the case for the reader; it helps the reader see what the inference has already committed them to. What kind of Mind, how to think about a Mind that could act in history, what fittingness considerations are in play — that is the work of the next chapter. The Unification Argument’s job ends here.