More Silly Arguments

Very often while sitting in a Protestant church, I will hear a non-trivial amount of C.S. Lewis references. It's not every day, but over some unspecified interval, there will easily be more references to Lewis during a sermon than church fathers like Augustine or founding theologians like Luther. These social dynamics are quite interesting. While Protestant lament the idea of a saint, they exalt figures like Lewis, Billy Graham, or any other apologist or charismatic preacher who's "defended the faith." Apologists like Lewis provide believers with stock arguments that can be used against "those skeptics seeking to undermine their beliefs." These arguments are not analytical instruments used to interrogate the internal validity and consistency of their belief system, they are often simple platitudes to be memorized without any critical interrogation into their merit. One in particular that I've heard woven into a few sermons, is Lewis's "Argument from Natural Desire." I think this argument exemplifies Lewis's utter failure as an apologist attempting rigorous philosophical argumentation. While he is an amazing author of fiction, I think he was completely inept at anything requiring philosophical rigor. And yet, when the preacher delivered Lewis's argument, people seemed to find it quite convincing. This is what prompted me to write this post. With a little bit of critical thought, it's quite easy to show the vacuousness. So first, I will steelman it like usual, putting it in argumentation schema format. Then we will identify critical questions associated with the schema. Simply reading the questions should be sufficient to cast doubt on the premises. Lastly, we will construct some obvious counterarguments that show how weak it is. 

Here’s a tight, steel-manned version of C. S. Lewis’s “Argument from (Natural) Desire,” expressed as a Walton-style argumentation scheme, with its assumptions made explicit and followed by comprehensive critical questions. 

Argumentation Scheme: Argument from Natural Desire (Lewis-style, defeasible/abductive)

Role of the scheme: An inference to the best explanation aiming to show that theism (or at least a transcendent Good) is the best explanation of a pervasive, apparently innate human longing that finite goods cannot satisfy.

Informal core intuition

Many people experience a recurring longing—often described as Sehnsucht, homesickness for an ultimate good, “Joy”—that is not fully satisfied by any finite, temporal object. Typically, natural desires (as opposed to learned, “artificial” ones) point to real objects that can satisfy them (hunger → food, thirst → water, curiosity → truth). So, if there is a natural desire for an ultimate good that nothing finite can satisfy, the best explanation is that there exists a transcendent reality that does.

Scheme (premises, warrant, backing, qualifier, conclusion)

P1 (Phenomenological Premise): Many (and arguably all) humans exhibit a recurring, robust longing (D) for a maximal, unconditioned fulfillment that finite goods only partially or temporarily satisfy.

P2 (Natural/Artificial Distinction): Desire (D) is natural/innate rather than merely learned or culturally induced; it arises spontaneously across times and cultures and is not exhausted by specific finite targets.

P3 (Inductive Generalization on Natural Desires): Generally, for each natural desire type (N), there exists a real object (O_N) apt to satisfy it (e.g., hunger → food; thirst → water; sleepiness → sleep; sexual desire → sexual union; curiosity → truth).

P4 (Negative Empirical Premise): No finite or temporal object (F) (or collection/sum of such objects) suffices to satisfy (D) non-defectively (i.e., without remainder, disenchantment, or diminishing returns).

W (Warrant/Bridge Principle—Teleological Fit): If a natural desire (N) lacks any adequate object in the domain of finite realities, then (defeasibly) the best explanation is that there exists an object of a different order—call it (T)—proportionate to that desire.

B (Backing):

• Inductive backing: Robust cross-case pattern linking natural desires to real satisfiers.
• Phenomenological backing: Introspective reports and literary/philosophical testimony that (D) exceeds finite satisfiers in principle.
• Explanatory backing: Postulating (T) explains the otherwise puzzling structure of (D)—its object-directedness, inexhaustibility, and persistence.

Q (Qualifier): Ceteris paribus / defeasible / probabilistic (open to undercutting and rebutting defeaters).

C (Conclusion): Therefore, it is reasonable to infer that there exists a transcendent reality (T) (often identified with God, or union/participation in God) that is the proper object of (D).

Explicit assumptions (made salient)

1. Robust natural/artificial distinction: There is a principled way to identify some desires as natural/innate (not just strongly learned).
2. Cross-cultural robustness: (D) is widespread enough to justify the generalization in P1–P2.
3. Desire–object reliability: Natural desires are generally veridical indicators of real satisfiers (no systematic illusions at that level).
4. No finite sufficiency: Even idealized finite goods cannot, in principle, complete (D) (not merely “we haven’t tried enough”).
5. Explanatory virtue preference: Hypotheses with greater explanatory power, scope, and simplicity are to be preferred; positing (T) scores well on these measures relative to rivals.
6. Non-reductive intentionality: The intentional “aboutness” of (D) isn’t fully reducible to purely adaptive or by-product evolutionary accounts without loss of explanans power.
7. Modal step legitimacy: Moving from “generally, natural desires have objects” to “there exists (T)” is a legitimate abductive (not deductive) step.

Exceptions / defeaters the scheme acknowledges

• Undercutters: Show that P2 is mistaken (e.g., (D) is learned/acculturated), or that P3’s generalization is unreliable (counterexamples where natural desires lack real objects in the relevant sense), or that P4 is premature (finite goods could suffice under ideal conditions).
• Rebutters: Provide an alternative explanation (R) (e.g., evolutionary mismatch, signaling, or predictive-processing accounts) that explains (D) at least as well as (T) without positing transcendent realities.

Critical Questions

A. About classification and scope

1. CQ-A1 (Naturalness Test): What principled criteria are used to classify (D) as natural rather than learned? Do these criteria avoid circularity (e.g., calling a desire “natural” because it seems universal)?
2. CQ-A2 (Universality/Prevalence): How prevalent is (D) across cultures, developmental stages, and personality types? Are there populations lacking (D), and if so, what does that imply?
3. CQ-A3 (Phenomenal Uniformity): Is the phenomenology of (D) sufficiently similar across persons to support a single explanandum, or are we conflating heterogeneous longings?

B. About the generalization from desires to objects

4. CQ-B1 (Inductive Base Quality): Is the sample of natural desires supporting P3 broad and theory-neutral, or cherry-picked?
5. CQ-B2 (Counterexamples): Are there putative natural desires that lack real objects (e.g., for perfect justice, complete moral purity, immortality) or whose “objects” are controversial?
6. CQ-B3 (Ambiguity of ‘Object’): Does “object” mean concrete satisfier, ideal limit, or regulative ideal? If the latter, does that still count as a “real object” in the sense needed?

C. About the negative premise (finite goods don’t suffice)

7. CQ-C1 (Epistemic Access): How could we know that no finite arrangement can satisfy (D)? Are we generalizing from limited life experience?
8. CQ-C2 (Psychology of Disenchantment): Could hedonic adaptation, novelty-seeking, or the “wanting/liking” gap explain the felt insufficiency without positing (T)?
9. CQ-C3 (Idealization Challenge): If human life had unbounded time, resources, and relationships, would (D) still remain unsatisfied? What evidence bears on this counterfactual?

D. About abductive strength and rivals

10. CQ-D1 (Best-Explanation Test): Compared to live rivals (e.g., evolutionary by-product, predictive-processing error, social signaling, Platonist/axiological realism, Buddhist diagnosis of craving), does (T) better explain the data (scope, fit, simplicity, predictive fruitfulness)?
11. CQ-D2 (Over-Specification): Even if (T) is the best explanation, does the argument uniquely support classical theism/Christian theism, or only some thin theism/axial transcendence?
12. CQ-D3 (Parsimony vs. Explanatory Power): Does positing (T) add unnecessary entities relative to naturalistic rivals, or does it earn its keep by explaining more with fewer ad hoc moves?
13. CQ-D4 (Predictive/Confirmatory Payoff): What novel predictions or retrodictions does the (T)-hypothesis make (e.g., transformation patterns, moral/affective fruits) that competitors don’t?

E. About phenomenology and intentionality

14. CQ-E1 (Direction of Fit): Is (D) directed to a particular kind of object (personal, infinite, moral) or is it a mood/state that we interpret theistically?
15. CQ-E2 (Conceptual Mediation): To what extent is the experience of (D) concept-laden (shaped by prior metaphysics or tradition)? Could interpretation bias simulate “transcendence”?
16. CQ-E3 (Competing Satisfiers): If some finite experiences (e.g., contemplative states, aesthetic rapture) temporarily feel all-satisfying, why don’t they count as the sought object?

F. About normativity and risk of equivocation

17. CQ-F1 (From ‘is’ to ‘ought’): Does the argument smuggle in a normative principle (“our deepest desires ought to be satisfiable”) not warranted by the inductive base?
18. CQ-F2 (Equivocation on ‘Satisfaction’): Is “satisfaction” used univocally across examples (e.g., hunger vs. ultimate beatitude), or is the argument illicitly shifting senses?
19. CQ-F3 (Modal Strength): Is the conclusion merely plausible, probable, or necessary? What would appropriately calibrate the modal strength?

G. About data and methodology

20. CQ-G1 (Empirical Grounding): What cross-cultural, developmental, and clinical data support P1–P2? Are there robust measures for (D) distinct from depression, anxiety, or trait openness?
21. CQ-G2 (Pathology Confound): Could conditions like anhedonia or existential depression mimic (D)? How does the argument distinguish the target phenomenon?
22. CQ-G3 (Selection Effects): Are testimonials over-represented from literary/religious subcultures that valorize (D)?

H. Meta-level and practical upshot

23. CQ-H1 (Underdetermination): Even granting all premises, do multiple metaphysical frameworks equally accommodate (D) (e.g., non-theistic Platonism, panentheism)?
24. CQ-H2 (Hiddenness Tension): If (T) exists and is the proper object of (D), why is (T) epistemically hidden or ambiguous for many sincere seekers?
25. CQ-H3 (Transformational Criterion): Does orienting to (T) reliably produce distinct, testable transformations (moral/spiritual flourishing) beyond placebo or community effects?

I think any one of these critical questions is enough to cast serious doubt on the strength and force of the argument. As far as I can tell, there hasn't been any sort of substantiation because the argument simply does not serve that purpose. It acts someone as a tool of social reinforcement, designed to alleviate doubt; like so many other apologetic arguments. But we don't even need to press on any of these sources of critique. The structure of the argument can easily be repackaged to undermine the basic tenets of theism or core Christian claims; rejecting this repackaged argument would mean rejecting the original argument. 

Reductio 1 — The Argument from Natural Autonomy (against an omnipotent, omniscient God)

Target: Undercut classical theism by showing the scheme equally supports a world with no ultimate sovereign.

Scheme (mirroring the Lewis structure)

P1 (Phenomenology): Many humans exhibit a robust, recurring longing (A) for unconditioned self-determination: to be answerable to no superior authority, unobserved, and free from any unappealable will over their lives.

P2 (Natural/Innate): Longing (A) appears across cultures and development, often preceding or resisting socialization; even young children resent perceived overreach and intrusive oversight. (A) is not exhausted by finite targets (jobs, political freedoms); it’s a global desire for final say-so.

P3 (Natural-desires Principle): Generally, each natural desire type (N) has a real object (O_N) proportionate to it (hunger→food; thirst→water; curiosity→truth; attachment→community).

P4 (Negative Premise): No finite arrangement (F) can satisfy (A): however much legal liberty, privacy, or power one acquires, there remains the residual vulnerability to higher orders (states, fate, death) and—if theism is true—inescapable dependence upon and surveillance by God.

W (Warrant): If a natural desire (N) has no adequate object among finite realities, the best explanation is that there exists an object of a different order proportionate to (N).

C: Therefore, there exists a reality (U) whose structure ultimately secures unconditioned autonomy—i.e., a world in which there is no omnipotent, omniscient sovereign over persons (no God of classical theism).

Qualifier: Defeasible / abductive.

Why this conflicts with Christian theism

• Christianity affirms a personal, omnipotent, omniscient God to whom all persons are answerable and who “searches hearts and minds.” That is the negation of (U).

Diagnosis: what went wrong?

• Hidden equivocation in “natural desire.” The scheme counts any widespread, strong, pre-reflective longing as “natural,” but then (A) (for total self-ownership) qualifies just as much as Lewis’s (D) (for unconditioned joy/beatitude).
• Over-strong P3/W. If “natural desires generally have real objects” + “best explanation posits an ontic satisfier” are accepted without constraints, the same machinery equally supports anti-theistic worlds.

Moral: Unless we add principled filters on which “natural desires” count, and what kind of “object” they license (e.g., teleological/desiderative vs. counter-teleological cravings), the scheme proves too much and collapses by parity.

Reductio 2 — The Argument from Natural Non-Intrusion (toward “cosmic privacy,” incompatible with divine omniscience)

Target: Show the scheme yields a reality contradicting God’s omniscience/omnipresence.

Scheme

P1 (Phenomenology): Humans display a persistent, non-pathological longing (P) for absolute mental privacy—to have an inner life not surveilled or readable by any other mind.

P2 (Natural/Innate): (P) emerges early (kids hide diaries; adults guard thoughts), recurs across cultures, and is not reducible to contingent harms; rather, it seems constitutive of personhood-boundary.

P3 (Natural-desires Principle): Natural desires reliably track real satisfiers.

P4 (Negative Premise): No finite institutional fix (laws, encryption, social norms) guarantees absolute mental privacy. If God exists as classically conceived, God infallibly knows all thoughts; hence the object of (P) is impossible in a theistic world.

W (Warrant): When no finite satisfier exists for a natural desire, posit a non-finite reality apt to satisfy it.

C: Therefore there exists a reality whose fundamental structure ensures absolute mental privacy—i.e., no omniscient being exists.

Qualifier: Defeasible / abductive.

Conflict with Christianity

• Christian theism affirms divine omniscience (knowledge of all hearts). The conclusion denies that.

Diagnosis

• Again, P3/W overgenerate: treating privacy as a “natural desire” plus the “real object” rule yields a metaphysical constraint that rules out omniscience.
• The scheme is directionally unstable: depending on which natural desire you spotlight (beatitude vs. privacy/non-intrusion), you get inconsistent metaphysical upshots.

Why these reductios land (and what that tells us about the original)

1. Classifying “natural desires” is under-theorized. Without a principled, non-circular taxonomy (e.g., homeostatic/teleological baselines rooted in flourishing rather than in resistance to flourishing), the scheme lets in incompatible desiderata.
2. The Desire→Object bridge (P3 + W) is too coarse. It ignores (a) regulative ideals (helpful fictions), (b) mismatch/by-product explanations, and (c) normative filtering: some desires are constitutively to-be-disciplined rather than satisfied.
3. Scope creep from “some tendency” to “ontic existence.” Even if natural desires usually have some satisfiers, it doesn’t follow that their unconditioned or absolute forms point to an existent object rather than to an open-ended limit concept or virtue-regulating norm.

Pinpointing what must be false (to block the reductio)

To prevent the scheme from entailing anti-theistic conclusions, at least one of the following must be rejected or significantly restricted:

1. (Reject/qualify P3) “All/most natural desires have real objects” → Replace with a ceteris-paribus law only for homeostatic/organism-maintaining desires (hunger, thirst, sleep), not for axiological/ultimate desires or boundary-protective drives.
2. (Restrict P2) The relevant desire must be shown to be teleologically basic to flourishing rather than a defensive/compensatory craving (e.g., desire for unconditioned autonomy may be a distortion of agency, not a basic orientation).
3. (Weaken W) The best-explanation step must allow non-ontic satisfiers (e.g., ideal limits, conceptual norms, spiritual disciplines) so the move from desire to existence of an entity is blocked.
4. (Narrow C) The conclusion cannot specify robust metaphysics (God, omniscience); at most it licenses a thin, regulative ideal—too weak to ground Christian theism (but also too weak to refute it).

If none of these restrictions are adopted, the same abductive machinery supports mutually inconsistent worlds (with God / without God), so something in the original Lewis-style package must give.

Compact critical questions (aimed specifically at blocking the reductio)

• CQ-Filter: What principled criterion excludes desires like (A) (unconditioned autonomy) or (P) (absolute privacy) from the “natural-desire” base while including Lewis’s (D)?
• CQ-Normativity: By what right do we treat some desires as to-be-satisfied and others as to-be-educated or sublimated?
• CQ-Objecthood: Why think the satisfier must be an existent entity rather than a limit-concept or practice?
• CQ-Consistency: Can the scheme, with your preferred constraints, avoid deriving both theistic and anti-theistic conclusions from equally “natural” longings?

Using the Lewis/Walton structure straight, you can build parallel arguments for cosmic autonomy or absolute mental privacy—conclusions that are absurd or that contradict Christian theism. Therefore, either (i) the generalization from natural desire to real object is false, or (ii) the warrant from “no finite satisfier” to “transcendent entity exists” is illegitimate, or (iii) the notion of “natural desire” must be sharply restricted. Any of these concessions undermines the original argument’s force—or at least reduces it to a much thinner, non-theistic conclusion.