The Free Will Debate

A Synthesis of 75 Findings Across 200+ Sources

Compiled March 2026 from deep research on Harris, Sapolsky, Dennett, Kane, Caruso, List, and the neuroscience evidence. Primary audience: Baron Bridgford II.

Epistemic tags: Direct read the source Paraphrase close summary Secondary via another source Inference drawn from evidence Speculation uncertain

I. The Question Nobody Agrees On

The free will debate looks like a disagreement about an answer. It is, at root, a disagreement about which question matters.[1] Inference

Sam Harris thinks the real question is: "Do I, as a conscious being, author my own thoughts and actions?" His answer is no. Thoughts arise unbidden. You can verify this through meditation. The sense of being a conscious decision-maker is an illusion built on top of unconscious neural processes.[2] Paraphrase

Daniel Dennett thinks the real question is: "What capacities ground moral responsibility?" His answer is reason-responsiveness, self-control, and the ability to adjust behavior in light of new information. Whether you "authored" your brain states is irrelevant, he argues, just as asking whether a hurricane "authored" its wind patterns is irrelevant to whether the hurricane is real.[3] Paraphrase

Robert Kane thinks both of them are ducking the real question: "Are you the ultimate source of your own will?" Not just your actions, but the character from which those actions flow. Did you write yourself, or were you written?[4] Paraphrase

Their 2014–2016 exchange is fascinating precisely because it never resolves. It cannot resolve, because Harris and Dennett are not disagreeing about an answer—they are insisting that different questions are the important one.[5] Inference Harris says Dennett is "changing the subject." Dennett says Harris is attacking a strawman nobody defends. They are both right about the other, and neither sees it.

Key Finding

The Harris–Dennett debate never resolves because it is not a disagreement about an answer. It is a disagreement about which question matters. Harris: "Do I have libertarian free will?" Dennett: "What grounds moral responsibility?" These are genuinely different questions with genuinely different implications.

II. The Cast

ThinkerPositionKey MoveStatus
Sam HarrisHard incompatibilist"You didn't choose your brain." Introspection + Libet experiments.Active
Robert SapolskyHard incompatibilist500-page biological evidence cascade: genes → hormones → brain wiring → no gap for agencyActive
Gregg CarusoHard incompatibilist"Luck swallows everything." Targets basic desert specifically. Quarantine model.Active
Daniel DennettCompatibilistFree will = evolved capacity for rational self-governance. Determinism enables freedom.Died April 19, 2024
Robert KaneLibertarianSelf-Forming Actions: rare moments of genuine indeterminacy build characterDied April 20, 2024
Christian ListEmergent compatibilistFree will emerges at higher levels of description, like temperature from moleculesActive

Dennett and Kane died within 24 hours of each other in April 2024.[6] Secondary The two leading defenders of opposing positions—compatibilism and libertarianism—left complete, final bodies of work. The debate between them is now permanently frozen.

Caruso is the figure many casual observers miss. He is the only free will skeptic whom professional philosophers take fully seriously.[7] Inference Where Harris gave us a 96-page popular book that Dennett called "a museum of mistakes,"[5] Direct Caruso published a 200-page book-length debate with Dennett (Just Deserts, 2021) and won the American Philosophical Association's Gittler Award.[8] Secondary He teaches at a community college.[7]

III. Four Positions on a Map

Hard Incompatibilism: "Free will is impossible"

Harris, Sapolsky, Caruso, Pereboom, G. Strawson • ~10% of professional philosophers (declining)[9]

Free will requires something that escapes the causal chain. Nothing escapes the causal chain. Whether the universe is deterministic or indeterministic doesn't matter—randomness isn't control. Therefore: no free will, no basic desert (the idea you intrinsically deserve punishment or reward). Paraphrase

The three versions: Harris (philosophical intuition + neuroscience), Sapolsky (30 years of biological evidence), Caruso (formal philosophical arguments + criminal justice alternative). They agree on the conclusion but differ dramatically in rigor.[7] Inference

Compatibilism: "Free will is real, properly understood"

Dennett, Fischer, Mele, Wolf, P.F. Strawson • ~62% of professional philosophers (growing)[9]

Free will never required escaping causation. It means: acting from your own reasons, responsive to evidence, capable of self-correction, free from coercion or pathology. This capacity is real, measurable, varies by degree, and is fully compatible with determinism. In fact, determinism enables it—only in a law-governed universe can organisms learn from patterns and avoid dangers.[3] Paraphrase

Libertarianism: "Free will is real and requires indeterminism"

Kane, O'Connor, van Inwagen • ~13% of professional philosophers[9]

Both Harris and Dennett are partly right. Free will is incompatible with determinism (agreeing with Harris against Dennett). But indeterminism can rescue it (disagreeing with Harris). In rare moments of genuine inner conflict, quantum indeterminacy at the neural level means the outcome is genuinely open. You author your own character through these "self-forming actions."[4] Paraphrase

The wound: The luck objection remains unresolved. If you replay the universe and get a different outcome, that looks like luck, not freedom. Kane spent 50 years on this and never fully answered it. Nobody else has either.[10] Secondary

Emergent Compatibilism: "Free will is real at the right level of description"

List, Mitchell, Ismael • Emerging position

Physical determinism at the micro level does not preclude genuine agency at the macro level, just as knowing every molecule's position doesn't make "temperature" or "hurricane" illusory. Free will is a higher-level property that supervenes on but is not reducible to physics. Whether this is a genuine fourth position or sophisticated compatibilism is itself debated.[11] Paraphrase

Kevin Mitchell's Free Agents (2023) surveys identical neuroscience as Sapolsky's Determined (2023) and reaches the opposite conclusion. Same evidence. Different framework. This may be the cleanest proof that the debate is about definitions, not data.[12] Inference

IV. What the Neuroscience Actually Shows

The popular narrative: "Brain scans prove your brain decides before you do. Libet proved it. Free will is an illusion." Here is what the research actually shows. Inference

The Libet Experiments (1983)

Benjamin Libet measured three things: brain activity (readiness potential, or RP), reported conscious intention, and muscle movement. He found that brain activity preceded reported intention by about 350 ms.[13] Secondary

What Libet himself concluded: Free will is real. He proposed a "conscious veto"—the brain proposes, consciousness disposes. You can't initiate actions consciously, but you can cancel them. "Free won't." He maintained this position until his death in 2007.[13] Secondary

The person who ran the most famous anti-free-will experiment didn't think it disproved free will.

The Readiness Potential Dissolves (2012–2025)

Aaron Schurger's 2012 accumulator model reinterpreted the readiness potential as stochastic neural noise, not an unconscious "decision."[14] Secondary The brain isn't deciding before you know it—it's fluctuating randomly, and the fluctuation occasionally crosses a threshold that triggers movement. The temporal gap that was the neuroscience "killer argument" against free will may not reflect a decision process at all.

Schurger's work has been extended through 2025, with a probing study further supporting the noise interpretation. The key paper is open access at PMC.[15] Secondary

The fMRI "Prediction" Studies (Soon et al., 2008)

John-Dylan Haynes' team claimed to predict choices 7–10 seconds before subjects were aware of deciding. This became the headline "your brain decides 10 seconds before you do." Paraphrase

The reality:[16] Secondary

The Field Has Reversed

Three findings that shifted the consensus:[17] Secondary

StudyFindingImplication
Trevena & Miller (2010)RP occurs equally whether you move or don't moveRP reflects anticipation, not motor preparation or "decision"
Matsuhashi & Hallett (2008)Using tone probes instead of Libet's clock, intention appears at −1,420 ms—before the RPLibet's clock method underestimates when intention forms
Maoz et al. (2019)RP is absent for meaningful decisions—only appears for trivial, arbitrary button pressesThe entire Libet/Soon program studied a phenomenon irrelevant to real free will

The scholarly consensus by 2024: "Current neuroscience does not present an actual threat to free will."[18] Secondary Both Lavazza (2016, Frontiers) and Mele (2017, Routledge Companion) conclude this independently. The field has moved from "does neuroscience disprove free will?" to "how do we operationalize and measure free will capacities?"

Harris and Sapolsky are arguing a position the neuroscience community has largely moved past. Inference

V. Where They Agree (The Surprising Convergence)

For all the heat, the four positions share remarkable common ground: Inference

Point of AgreementWho Agrees
Determinism (or near-determinism) is true at the physical levelAll six major thinkers
Libertarian free will (contra-causal, soul-based) is falseAll except Kane
Conscious deliberation is causally efficaciousAll six (even Harris concedes deliberation "matters")
The US criminal justice system is deeply brokenAll six[8]
Pure retribution (making people suffer because they "deserve" it) is hard to justifyAll except Kane, who defends it cautiously
Holding people responsible for consequentialist reasons (deterrence, rehabilitation, protection) is legitimateAll six[5]
People who act from coercion, severe mental illness, or addiction are less free than those who don'tAll six
Childhood environment profoundly shapes adult behaviorAll six
The neuroscience evidence (Libet, Soon) doesn't settle the debate by itselfHarris and Sapolsky are the holdouts; the other four plus the neuro community agree

Notice what this means: on every practical question, Harris and Dennett largely converge. They agree on consequentialist responsibility, criminal justice reform, and the falsity of libertarian free will. Dennett accused Harris of being "a compatibilist in everything but name."[5] Direct Harris never provided a convincing rebuttal to this charge.

VI. Where the Disagreement Is Real

After stripping away semantic disputes, definitional confusions, and talking past each other, here is what they actually disagree about: Inference

1. Does causal history matter for moral responsibility?

Dennett says no—what matters is your present capacities. If you are currently reason-responsive and capable of self-control, you are morally responsible.[3] Paraphrase

Caruso says yes—if those present capacities were produced by factors entirely beyond your control (genes, parents, prenatal environment, childhood luck), then you are not the ultimate source of your own agency, and basic desert is unjustified.[7] Paraphrase

This is the deepest genuine impasse. It is a philosophical disagreement about the conditions for desert, not an empirical disagreement about the brain.

2. Is "basic desert" a coherent concept?

Kane says yes, and it requires genuine metaphysical openness at key moments in your history.[4] Dennett says yes, but reinterprets it as forward-looking and consequentialist.[8] Caruso says no—backward-looking desert is incoherent regardless of the causal structure of the universe.[7] Paraphrase

3. Can determined capacities ground genuine agency?

The compatibilist says yes: a thermostat is "free" to regulate temperature; a human is vastly more free by the same logic. Dennett's Game of Life demonstrations show that deterministic systems can exhibit genuine avoidance behavior that is best described at a higher level of abstraction.[3] Paraphrase

The incompatibilist says no: a determined system, no matter how sophisticated, is executing its program. Kane's devastating line: "A marionette who happens to love the strings."[4] Secondary

4. Does indeterminism help?

Kane says yes—it provides genuine openness at self-forming moments. Harris and Caruso say no—randomness isn't control. This is a genuine metaphysical disagreement that empirical evidence cannot resolve. Inference

VII. The Unanswered Arguments

Some of the most important exchanges in this debate ended with one side simply never responding to the other's strongest point. Inference

Dennett → Harris: The Consciousness Asymmetry (NEVER ANSWERED)

Harris treats consciousness as real—something worth reconceptualizing even after folk intuitions are debunked. But he treats free will as pure illusion—not worth reconceptualizing, only discarding. Dennett points out this is the exact same philosophical move applied inconsistently. Harris never provides a principled criterion for why consciousness survives reconceptualization but free will doesn't.[5] Secondary

Multiple independent analysts flag this as Harris's biggest unforced error.

Dennett → Harris: "Compatibilist in everything but name" (NEVER ANSWERED)

Harris endorses consequentialist responsibility, meaningful coercion/non-coercion distinctions, and holds that deliberation causally matters. These are the functional commitments of compatibilism. He simply refuses the label.[5] Direct Harris never explains what practical difference his illusionism makes.

Harris → Dennett: The "Atlantis–Sicily" Problem (NEVER FULLY ANSWERED)

Told Atlantis doesn't exist, compatibilists point at Sicily. "You rarely acknowledge the ways in which Sicily isn't like Atlantis." When ordinary people say "free will," they mean conscious authorship of their thoughts and actions. Compatibilists redefine the term until it no longer refers to that experience.[2] Paraphrase Surveys suggest most laypeople are not natural compatibilists.[19] Secondary

Caruso → Dennett: The Causal History Problem (IMPASSE)

Dennett concedes we are "not responsible for becoming responsible."[8] Direct Caruso presses: if you didn't choose to be reason-responsive, how can being reason-responsive ground desert? Dennett's response—that this proves too much (by the same logic, nobody discovers or creates anything)—is a reductio, not a solution.[5] Inference

Setiya → Sapolsky: The Self-Undermining Objection (INADEQUATELY ANSWERED)

If the argument against free will works, it eliminates not just moral responsibility but rationality itself. If your belief that free will is an illusion is itself just neurons firing (determined by prior causes), why should anyone treat that belief as rational or true? Sapolsky's response—a determined system can still track truth, like a calculator—smuggles in the normative concepts determinism is supposed to eliminate.[20] Secondary

Fischer → Sapolsky: The Definitional Circularity (NEVER ANSWERED)

Sapolsky defines free will as requiring "a neuron independent of the sum of its biological past." This stipulates incompatibilism. He then spends 500 pages proving determinism—which compatibilists already accept. Fischer: "He offers no arguments in the entire book against compatibilism."[20] Secondary The elaborate scientific case reduces to a definitional move on page 15.

Sapolsky admits in the book: "I will not consider most of [the compatibilist arguments] because, to be frank, I can't understand what they're suggesting."[21] Direct A 528-page book attacking a position the author admits he doesn't comprehend.

VIII. Is It Just About Definitions?

Multiple independent lines of evidence converge on an uncomfortable conclusion: much of the free will debate may be semantic. Inference

The Mitchell–Sapolsky Test

Kevin Mitchell's Free Agents (2023) and Sapolsky's Determined (2023) survey identical neuroscience evidence and reach opposite conclusions. Mitchell: free will is an evolved biological capacity for conscious deliberation. Sapolsky: free will is an illusion. Same data, different definitions of "free will."[12] Secondary

The 40-Year Neuroscience Meta-Lesson

Both sides have been citing the same three experiments (Libet, Soon, Schurger) for decades. The fact that both sides claim victory from the same data is itself proof that the dispute is conceptual, not empirical.[17] Secondary Neuroscience accomplished one real thing: it killed the Cartesian picture. Every serious position now accepts that consciousness is not the uncaused first cause of action. The remaining question—whether that matters for freedom—is philosophical.

Kane's Five Levels of Freedom

Kane identified a hierarchy that clarifies the entire debate:[4] Paraphrase

  1. Freedom of self-realization—doing what you want
  2. Freedom from internal constraints—not compulsive or addicted
  3. Reflective self-control—Frankfurt/Fischer type
  4. Self-determination—setting your own path
  5. Self-formation—authoring your own character

Compatibilists defend levels 1–3. Kane insists 4–5 are what actually matter. Harris says none are real. They are arguing at different altitudes.

The Caruso–Dennett Methodological Divide

The deepest split may not be philosophical at all but methodological. Dennett uses evolutionary, fuzzy-boundary thinking (Darwin dissolved essentialist categories). Caruso uses classical analytic philosophy demanding internal consistency. They are doing different kinds of philosophy at each other.[7] Secondary

The verdict on "is it semantic?": Partially. The debate is semantic where it concerns the definition of free will (Harris and Dennett agree on 90% of practical questions). But it is NOT semantic where it concerns (a) whether basic desert is coherent, (b) whether causal history defeats moral responsibility, and (c) whether indeterminism can constitute rather than undermine agency. These are genuine philosophical disagreements that no amount of definitional housekeeping will dissolve. Inference

IX. What Should We Actually Do?

The criminal justice thread is where the debate acquires teeth. Inference

The Convergence on Reform

Every major thinker in this investigation agrees that the US criminal justice system is broken and that pure retribution is either unjustifiable or deeply questionable. The disagreement is about why it should change and how far the change should go.

Caruso's Quarantine Model

The public health-quarantine model: criminal behavior is analogous to disease. Dangerous individuals may be detained for public safety—but not punished. Detention must be minimally restrictive, non-punitive, rehabilitation-focused, and subject to periodic risk assessment.[7] Paraphrase

The Norway Evidence

Norway reorganized around rehabilitation and normalization. Result: reconviction dropped from 60–70% to 18% within two years. The US rate is approximately 75%.[22] Secondary The empirical case for de-emphasizing retribution is strong regardless of who is right about free will.

The Open Problems

The rehabilitated monster: If detention is justified only by ongoing danger, and a serial killer is genuinely rehabilitated, the model's own logic demands release. Caruso's pragmatic response—that danger presumption is "functionally permanent" for the worst offenders—may be retributive intuition smuggled in through the back door.[7] Inference

Fischer's performative contradiction: If no one deserves anything, on what basis do victims deserve protection? The right to public safety that justifies quarantine-model detention is itself a moral claim that seems to require some framework of desert.[23] Secondary

The parole judge finding—real but contested: Sapolsky presents the finding that judges grant ~65% parole after eating and ~0% after hours without food as a slam dunk against rational justice. But Glockner (2016) showed that simulations of rational time-management produce similar patterns, and case ordering may not have been random.[24] Secondary The study is real but the magnitude is disputed.

Sapolsky's Epilepsy Argument

We used to burn epileptics for "choosing Satan." Now we medicate them. Sapolsky: the only difference between an epileptic driver who kills a pedestrian and a road-rage driver who does the same is that we can currently see the epileptic's causal chain more clearly.[21] Paraphrase This remains his most powerful single argument for eliminating moral blame from criminal justice.

X. The Deep History: Nothing Is New

The modern free will debate is 2,500 years old. Very little in it is genuinely new. Inference

Modern ClaimHistorical Precedent
Harris: "Man can do what he wills, but he cannot will what he wills"Harris is quoting Schopenhauer (1839), who derived it from Kant[25]
Dennett: free will is acting from your own character without coercionChrysippus and the Stoics (~3rd c. BC): actions "up to you" if from your own nature[26]
Kane: genuine openness requires physical indeterminismEpicurus (~300 BC): the atomic "swerve" breaks the causal chain[26]
Caruso: nobody "deserves" punishment since they didn't choose their characterSpinoza (1677): "Men are conscious of their actions and ignorant of the causes"[26]
Sapolsky: your brain is determined by biology you didn't chooseAugustine's anti-Pelagian turn (~412 AD): the will is corrupted and cannot choose good[27]

The Riskin Thesis

Stanford historian Jessica Riskin argued in the New York Review of Books (2025) that Sapolsky's scientific determinism "remains what it has been from the beginning, a [Christian] theology."[28] Secondary Her claim: hard determinists define free will using Augustine's impossible standard (causa sui—uncaused self-causation), then declare it impossible. The standard was invented to make humans dependent on God's grace. Secular determinists imported the framing without the God.

Compatibilists never accepted that standard. They trace through Hobbes → Hume → Mill → Dennett—a 400-year tradition that never required escaping causation.[26] Paraphrase

What IS genuinely new

Two things: (1) The neuroscience evidence, even though it turned out not to settle the question. Before Libet, the debate was purely philosophical. After Libet–Schurger–Maoz, we know that conscious intention is not the simple first cause of action—but this doesn't eliminate free will, it constrains how it operates. (2) List's emergentist framework, which provides a philosophical vocabulary for saying "determined at one level, free at another" without it being a contradiction. Whether this is a genuine advance or a repackaging of Dennett is debated. Inference

XI. Beyond the West

The Western debate assumes individual selves, physical causation, and a binary question (free or not free). Other traditions dissolve these assumptions entirely. Inference

Buddhism: Agentless Agency

The Buddha explicitly rejected determinism, indeterminism, and theistic determinism 2,500 years ago.[29] Secondary His alternative—dependent origination—holds that all phenomena arise from conditions, but those conditions are themselves mutable through intentional action. Karma requires effort. But there is no enduring self (anatta) who makes the effort. This produces what the 2017 Repetti anthology calls "agentless agency": genuine causal efficacy of mental events without a substantial agent.[30] Secondary

Harris claims meditation confirms determinism. Buddhist scholars contest this. The tradition holds that meditative seeing transforms the causal chain—the witnessing itself is the intervention, not passive observation of a fixed process.[29] Secondary

Hindu Philosophy: Six Schools, Six Answers

There is no single "Hindu position" on free will. Advaita Vedanta dissolves agency as illusion (neither Harris's determinism nor freedom, since the causal chain is itself illusory). Dvaita affirms full individual agency. Samkhya separates consciousness (witness) from nature (doer).[31] Secondary

Islam: The Oldest Free Will Debate

The Ash'ari/Mu'tazili dispute (9th–10th century) may be the oldest sustained philosophical debate about free will, predating the Western Reformation debate by centuries. Al-Ghazali's critique of causation anticipated Hume by 650 years.[32] Secondary

Confucianism: Self-Cultivation as Highest Agency

Confucian philosophy reframes the question. Wu wei (effortless action) is not the absence of agency but its highest expression. The sage acts without deliberation because their character has been so thoroughly cultivated that the right action flows naturally. This resembles Dennett's "well-tuned deliberation" but adds a developmental trajectory: freedom is an achievement, not a given.[32] Secondary

Ubuntu: Agency Is Communal

African Ubuntu philosophy holds that "a person is a person through other persons." Individual free will is a category error. Agency is constitutively communal. This dissolves rather than answers the Western framing.[32] Secondary

The Parochialism Verdict

The Western debate is conducted almost entirely by American and European men arguing about individual selves in a physicalist framework. Buddhism offers "conditioned but not predetermined" agency with no self. Hinduism spans the full spectrum within a single tradition. Islam debated the question centuries before the Reformation. Confucianism asks not "are you free?" but "have you cultivated yourself?" Ubuntu asks not "are you free?" but "through whom are you free?" Any synthesis that ignores these perspectives is parochial. Inference

XII. Where the Evidence Points

Having surveyed 200+ sources across six major thinkers, four philosophical positions, 2,500 years of history, and five non-Western traditions, here is where the weight of evidence lands: Inference

What Can Be Said With High Confidence

1. Libertarian free will (soul-based, contra-causal) is almost certainly false. No credible evidence supports the existence of an uncaused causer outside the physical world. Kane's event-causal libertarianism is the most sophisticated attempt to rescue genuine openness, but the luck objection remains unresolved after 50 years. The Stanford Encyclopedia verdict: "the luck objection has not been met."[10] Secondary

2. The neuroscience evidence does not disprove free will. The readiness potential probably reflects stochastic noise, not unconscious decisions.[14] The fMRI "predictions" were barely above chance with heavily selected data.[16] The RP doesn't appear for meaningful decisions.[17] Neuroscience killed the Cartesian picture but left the philosophical question untouched.

3. Sapolsky's book has a fatal logical flaw. He defines free will as requiring causal indeterminism at the neuron level (page 15), then spends 500 pages proving determinism. Every philosophical reviewer identified this circularity.[20] His biological evidence is magnificent and his temporal causation framework is genuinely illuminating, but the philosophical argument reduces to a definitional move that begs the question against compatibilism.

4. Harris is a compatibilist in everything but name. On every practical question—consequentialist responsibility, coercion/non-coercion distinctions, the causal efficacy of deliberation—Harris holds compatibilist positions. His insistence that free will is "an illusion" has no practical implications beyond the ones Dennett already endorses.[5]

5. The debate is partly but not entirely semantic. The Harris–Dennett dispute is largely semantic (they agree on practices, disagree on labels). The Dennett–Caruso dispute is NOT semantic—they disagree on whether causal history defeats desert, which is a genuine philosophical question with genuine practical consequences.

What Remains Genuinely Open

6. Whether "basic desert" is coherent. This is the load-bearing question. If you didn't choose your genes, your parents, your childhood, or the neural machinery that makes you "reason-responsive"—can you deserve anything in the backward-looking sense? Caruso says no. Dennett says the question is confused. Neither has won. Inference

7. Whether determined capacities can ground genuine agency. Dennett's Game of Life demonstrations show that deterministic systems can exhibit real avoidance behavior. But Kane's "marionette who loves the strings" captures something that Dennett's framework never quite addresses: the intuition that endorsing your own determined outputs isn't the same as authoring them. Inference

8. What we should do with this. Norway's 18% recidivism vs America's 75% makes the empirical case for reform overwhelming, independent of the free will question. But whether we frame that reform as "they don't deserve punishment" (Caruso) vs "they deserve better designed punishment" (Dennett) has real institutional implications. Inference

The "What Would Change Your Mind" Test

A final diagnostic: Inference

PositionWhat Would Falsify It
Hard incompatibilism (Harris)Harris has essentially said he cannot conceive of evidence for free will. This is a sign of an unfalsifiable position.[17]
Libertarianism (Kane)Specific physics findings (e.g., ruling out quantum effects at neural scale). At least in principle falsifiable.
Compatibilism (Dennett)The "bypass threat"—proof that conscious processes are causally inert. This is empirically testable, and hasn't been demonstrated.[17]
Hard incompatibilism (Caruso)A convincing account of how determined capacities can ground basic desert without circular reasoning.

Only compatibilism has a genuinely falsifiable empirical commitment—and it hasn't been falsified.

Personal synthesis: The evidence favors a position in the neighborhood of Dennett/List—compatibilism enriched by emergence theory—with Caruso's criminal justice reforms adopted on consequentialist grounds regardless. The "free will is an illusion" narrative is popular but philosophically weak: it depends on defining free will using a theological standard (Augustine's causa sui) that no serious philosopher has defended in centuries, then triumphantly proving that standard can't be met. The compatibilist tradition—from the Stoics through Hume through Dennett—never accepted that standard. And the neuroscience that was supposed to settle the matter has, upon careful examination, settled nothing.

But Caruso is right that the question of basic desert remains genuinely open and genuinely important. Even if compatibilism is the best framework for understanding agency, the criminal justice implications of taking causal history seriously are enormous. Norway already shows what's possible. The question isn't "do we have free will?"—it's "what kind of freedom actually matters, and what follows from it?" Inference

Bibliography

[1] Cross-analysis of Harris–Dennett exchange (2014–2016), Kane deep dive, and Caruso deep research files. Multiple agents independently identified the "different questions" dynamic.

[2] Harris, Sam. Free Will. Free Press, 2012. 96 pages. Also: "The Marionette's Lament" (blog response to Dennett, 2014). Also: 557 Substack posts and 72 transcripts analyzed from Making Sense archive.

[3] Dennett, Daniel C. Freedom Evolves. Viking, 2003. Also: Elbow Room: The Varieties of Free Will Worth Wanting. MIT Press, 1984 (revised 2015). Deep research compilation at Dennett/dennett-deep-research.md.

[4] Kane, Robert. The Significance of Free Will. Oxford University Press, 1996. Also: The Complex Tapestry of Free Will. OUP, 2024 (posthumous). Deep dive at Kane/Kane-Libertarian-Free-Will.md, Kane/Kane-SFA-Depth.md.

[5] Dennett, Daniel C. "Reflections on Free Will" (review of Harris). Published at samharris.org, January 2014. ~14,000 words. Also: Harris–Dennett podcast, Making Sense Episode #39, "Free Will Revisited," July 2016. TED Summit, Banff, Canada.

[6] Kane died April 20, 2024 at age 85. Dennett died April 19, 2024 at age 82. Sources: multiple news outlets and academic tributes, cited in Kane deep dive and Dennett deep research.

[7] Caruso, Gregg D. Rejecting Retributivism: Free Will, Punishment, and Criminal Justice. Cambridge University Press, 2020. APA Gittler Award, 2022. Deep research at Caruso/caruso-deep-research.md. 30+ sources.

[8] Dennett, Daniel C. & Caruso, Gregg D. Just Deserts: Debating Free Will. Polity Press, 2021. 206 pages, 107 exchanges across 3 conversations. #1 New Philosophy Release on Amazon.

[9] Bourget, David & Chalmers, David J. "Philosophers on Philosophy: The 2020 PhilPapers Survey." Philosophers' Imprint, 2023. Results: 62% compatibilist (up from 59% in 2009), 13% libertarian, 10% no free will (down from 12%). philpapers.org/surveys/

[10] O'Connor, Timothy & Franklin, Christopher. "Free Will." Stanford Encyclopedia of Philosophy, rev. 2022. Quote: "the luck objection has not been met." Also: Kane luck objection deep dive at Kane/Luck-Objection-Deep-Dive.md. plato.stanford.edu/entries/freewill/

[11] List, Christian. Why Free Will Is Real. Harvard University Press, 2019. Analysis at List/List-Landscape-Position.md. 10+ sources.

[12] Mitchell, Kevin J. Free Agents: How Evolution Gave Us Free Will. Princeton University Press, 2023. Comparison with Sapolsky identified in Sapolsky deep dive and independently by Kane deep dive agents.

[13] Libet, Benjamin. "Unconscious cerebral initiative and the role of conscious will in voluntary action." Behavioral and Brain Sciences 8(4): 529–566, 1985. Also: Libet, Mind Time: The Temporal Factor in Consciousness. Harvard University Press, 2004. Comprehensive analysis at Neuroscience/neuroscience-evidence-deep-dive.md.

[14] Schurger, Aaron, Sitt, Jacobo D., & Dehaene, Stanislas. "An accumulator model for spontaneous neural activity prior to self-initiated movement." Proceedings of the National Academy of Sciences 109(42), 2012. Also: Schurger & Hu, "What Is the Readiness Potential?" Trends in Cognitive Sciences, 2021. Open access: PMC8192467.

[15] Schurger, Aaron. "Probing for Intentions" (2025). Citation from neuroscience deep dive research agents.

[16] Soon, Chun Siong et al. "Unconscious determinants of free decisions in the human brain." Nature Neuroscience 11(5): 543–545, 2008. Critique: Lages, Martin & Jaworska, Katarzyna. "How Predictable Are 'Spontaneous Decisions' and 'Hidden Intentions'?" PLOS ONE, 2012.

[17] Maoz, Uri et al. "Neural precursors of decisions that matter: an ERP study of deliberate and arbitrary choice." eLife, 2019. Also: Trevena & Miller, 2010; Matsuhashi & Hallett, 2008. Comprehensive analysis at Neuroscience/neuroscience-evidence-deep-dive.md and Neuroscience/Can-Neuroscience-Settle-The-Debate.md.

[18] Lavazza, Andrea. "Free Will and Neuroscience: From Explaining Freedom Away to New Ways of Operationalizing and Measuring It." Frontiers in Human Neuroscience, 2016. Also: Mele, Alfred. "Free Will and Neuroscience." In Routledge Companion to Free Will, 2017.

[19] Nichols, Shaun & Knobe, Joshua. "Moral Responsibility and Determinism: The Cognitive Science of Folk Intuitions." Noûs 41(4): 663–685, 2007.

[20] Fischer, John Martin. Review of Determined by Robert Sapolsky. Notre Dame Philosophical Reviews, 2024. Setiya, Kieran. Review of Determined. London Review of Books, 2024.

[21] Sapolsky, Robert. Determined: A Science of Life Without Free Will. Penguin Press, 2023. 528 pages. NYT bestseller, Bill Gates pick. Deep dive at Sapolsky/DEEP-DIVE.md. 30+ sources, 28 citations.

[22] Norwegian Correctional Service statistics, cited in Sapolsky Determined and Caruso Rejecting Retributivism. Recidivism comparison: Norway ~18% within 2 years vs. US ~75% within 5 years (Bureau of Justice Statistics, 2018).

[23] Fischer's performative contradiction argument, from NDPR review of Determined and engagement in Just Deserts secondary analysis.

[24] Danziger, Shai, Levav, Jonathan, & Avnaim-Pesso, Liora. "Extraneous factors in judicial decisions." PNAS, 2011. Critique: Glöckner, Andreas. "The irrational hungry judge effect revisited." Judgment and Decision Making 11(6), 2016. Also: Sapolsky biological mechanisms agent analysis.

[25] Schopenhauer, Arthur. On the Freedom of the Will (Uber die Freiheit des menschlichen Willens). Royal Norwegian Society of Sciences essay prize, 1839. Harris quotes the "will what he wills" line as epigraph-level framing in Free Will. plato.stanford.edu/entries/schopenhauer/

[26] Timeline and lineage analysis compiled from: O'Connor & Franklin, "Free Will," SEP (2022); McKenna & Coates, "Compatibilism," SEP (2024); "Epicurus," SEP; IEP "Free Will." Full timeline at TIMELINE.md. plato.stanford.edu/entries/compatibilism/

[27] "Augustine." Stanford Encyclopedia of Philosophy. Also: "Jonathan Edwards," SEP. Theological determinism lineage analysis at TIMELINE.md.

[28] Riskin, Jessica. Review of Determined. New York Review of Books, 2025. Paywalled at nybooks.com. Content reconstructed from secondary citations and Sapolsky's public response.

[29] Eastern-Contemplative-Perspectives.md research compilation. Primary sources: Repetti (ed.), Buddhism and Free Will, 2017; Goodman, "Redeçtion and Living Well" (SEP Buddhist Ethics); Harvey, An Introduction to Buddhist Ethics, 2000. 10 sources cited.

[30] Repetti, Rick (ed.). Buddhist Perspectives on Free Will: Agentless Agency? Routledge, 2017. Landmark anthology.

[31] Hindu philosophy analysis from Eastern-Contemplative-Perspectives.md. Primary sources: Rambachan, "The Advaita Worldview" (SUNY Press); Sharma, Classical Indian Philosophy; Potter (ed.), Encyclopaedia of Indian Philosophies.

[32] Non-Western perspectives compiled from: Islamic-African-Confucian-Free-Will.md. Sources: Ormsby, Ghazali and Free Will; Metz, "Ubuntu as a Moral Theory"; Slingerland, Effortless Action: Wu-wei as Conceptual Metaphor. 11 sources cited.

This synthesis was compiled from 75 key findings across 25+ research files, 200+ academic sources, deep dives on 6 major thinkers, and coverage of Western, Buddhist, Hindu, Islamic, African, Confucian, and Daoist perspectives. Research conducted March 2026. Full research corpus at Personal/Research/Free-Will-Investigation/.