NOTE-270: Compatibilism

Contribution

This is the standard reference map of compatibilism, organised around two incompatibilist arguments.

• The Classical Incompatibilist Argument (§1.1). Free will requires the ability to do otherwise, and determinism rules that out.
• The Source Incompatibilist Argument (§1.2). Free will requires ultimate sourcehood, and determinism rules that out.

It tells the history in three stages (§§2–4):

1. Classical compatibilism and the failure of its conditional analysis (§2).
2. Three 1960s contributions: the Consequence Argument, Frankfurt’s attack on PAP, and Strawson on the reactive attitudes (§3).
3. Five contemporary positions (§4): new dispositionalism and agential abilities; Frankfurt’s hierarchical theory; Wolf’s Reason View; Fischer’s reasons-responsiveness; Strawsonian views.

The supplement adds the manipulation-argument dialectic (Pereboom’s four-case argument, Mele’s zygote argument) and the internalist/externalist (“snapshot” versus “historical”) distinction. It then covers the later history of identification theories (Frankfurt, Watson, Velleman, Bratman), Fischer & Ravizza, Haji, Wallace, Bok and Mele.

For this batch, the contribution is §4.2 plus supplement §A. Together they give a compact, page-referenced statement of the 1971 hierarchical theory, its two structural objections, and Frankfurt’s later replies as the authors reconstruct them.

Key insight

Frankfurt’s theory places freedom in a relation between parts of the agent: the will she has and the will she wants. It does not place it in the agent’s alternatives. This buys immunity to the Consequence Argument, because no regulative control is needed. The price is two exposures.

• The hierarchical problem. Whatever higher-order attitude is said to “speak for” the agent, she can be alienated from it too (Watson 1975).
• The mesh problem. Any mesh defined only by present structure can be produced by manipulation, and then the source incompatibilist asks why determinism is different.

The entry’s organising claim is that this second exposure is shared by every mesh compatibilism: Frankfurt’s, Wolf’s, Fischer & Ravizza’s, Haji’s, Wallace’s, Bok’s and Mele’s. Each must either show what is wrong with manipulation cases or “bite the bullet”. Per supplement §A, Frankfurt bites it.

Assumptions

Premises and authorities the entry works from:

• Two conceptions of control. “Leeway” or regulative control is the ability to select among alternatives. “Sourcehood” is originating one’s action (§§1.1–1.2). Fischer’s regulative/guidance distinction (1994) is used throughout to classify views (§4.1; §4.2.1).

• The conditional analysis is dead. It was “refuted” (Chisholm 1964; van Inwagen 1983, pp. 114–9), by the Danielle/blond-Lab case (§2.2). The incompatibilists “can rightly claim to have laid [it] to rest” (§3.2).

• Frankfurt’s anti-PAP argument is a live, unsettled debate. “No clear victor has emerged” (§3.2). The entry nonetheless uses Frankfurt examples as data against agent-based reasons-responsiveness (§4.4.2).

• The manipulation-argument schema (supplement, Introduction):

  1. An agent manipulated into X-ing is not free or responsible, despite meeting the compatibilist conditions.
  2. A determined agent is no different in any relevant respect.
  3. Therefore determined agents are not free or responsible.

  Compatibilists deny (1) (internalists) or (2) (externalists).

• Readings of Frankfurt in Fischer’s vocabulary. “Guidance control” is attributed to Frankfurt “to again return to Fischer’s helpful distinction” (§4.2.1). This is the authors’ gloss, not Frankfurt’s term.

Key results

What the entry argues and reports.

Frankfurt 1971, as stated in §4.2 (current revision)

• Setting (§4.2). “On the back of his rejection of” PAP, Frankfurt (1971) developed a compatibilism that “does not appeal to regulative control in any way”. It can be read as a development of classical compatibilism, the unencumbered ability to get what one wants. It is a “hierarchical mesh theory”: free action issues from desires “nested within more encompassing elements of her self”, so that actions “emanate from her rather than from something foreign”.
• First- and second-order desires (§4.2.1). First-order desires have actions as objects (cheesecake, a movie). Second-order desires have first-order desires as objects: “If only I wanted to go to the gym today…”.
• The will. Some first-order desires do not move one to act. Frankfurt “identifies an agent’s will with her effective first-order desire”, the one moving a person “all the way to action” (1971, p. 84).
• Mere second-order desires versus second-order volitions.
  • A psychotherapist who wants to feel a craving for narcotics, to understand a patient, has a second-order desire with “no wish that this desire be effective” (1971, pp. 84–5).
  • Second-order desires for effective first-order desires, desires “that would comprise her will”, are second-order volitions. The entry’s illustration is a dieter wanting her desire for health to be effective.
• Higher orders. “There is no theoretical limit to how highly-ordered one’s desires might be.” The example is a third-order desire that one’s second-order desire not dominate one’s deliberation.
• The addicts.
  • The unwilling addict has conflicting first-order desires and a second-order volition that the desire to take the drug not be her will. Her irresistible desire is her will nonetheless.
  • The willing addict has the same first-order conflict but, “by way of a second-order volition, embraces her addictive first-order desire… She wants to be as she is and act as she does.”
• The theory. The unwilling addict does not act of her own free will, because her will “conflicts at a higher level with what she wishes it to be”. The willing addict does. Stated as a biconditional: “One acts of her own free will if and only if her action issues from the will she wants.”
• No alternatives needed. The willing addict could not do otherwise yet acts freely. The relevant freedom concerns “what an agent does do and her actual basis for doing it”, which the entry glosses as guidance control, not regulative control.

Wantons and persons (2004–2015 revisions only; removed in 2019)

In §5.3.1 of the 2004 text and §5.2.1–5.2.2 of the 2015 text:

• “According to Frankfurt, a distinctive feature of personhood is that only a person has second-order volitions. Only a person wants to be moved by different desires and motives from the ones that move her.”
• “Frankfurt calls agents who have no second-order volitions wantons. Persons care about which desires lead them to action. Wantons do not. They are passive bystanders to their wills (1971, p. 89).”
• There are three addicts. The wanton addict has the same first-order conflict “but… no higher-order volition regarding which of her first-order desires wins out. She is passive with regard to the battle of desires taking place within her.” Hence “The wanton is not a person, and so is not a candidate for freely willed action.”
• These revisions also call the view a “Source model of control”. They add that it “has been categorized as a Real Self Theory (Wolf, 1990, p. 29)”, on which acting freely “reveals her deeper self… those motives are hers… She owns them.”
• Note 22 (2015) says Watson 1975 offers “another important sort of mesh theory that is not hierarchically structured”, between “the motivational and the evaluative features of rational agency”.
• Note 23 (2015) says “All page references to Frankfurt 1971 are to the reprint in Watson (ed.) 1982.”

The two problems (§4.2.2)

• The hierarchical problem.
  • To resolve a first-order conflict, “Frankfurt tells us that a person forms a second-order desire that has as its object the first-order desire(s) she wants to move her to action”. By this she “endorses or identifies with” one desire, draws it “within the sphere of her self”, and “alienates another”.
  • But she “can also be alienated from second, or even at higher-orders (Watson 1975)”. She might both want a first-order desire to be her will and be “horrified at the thought”.
  • So the problem “can reappear at these ever ascending stages”. Frankfurt’s condition may be necessary but “appears not to be sufficient”.
  • The 2015 text framed this as conflict at higher orders (“If a person can be conflicted at the level of her first-order desires, she can also be conflicted at the second…”). The 2024 text frames it as alienation. Both cite Watson 1975.
• The mesh problem (manipulation).
  • If freedom “wholly depends on the relation between an agent’s will and her second-order volitions”, then how the mesh arose “does not matter in any way”.
  • Cases suggest it does matter (Slote 1980; Fischer & Ravizza 1998, pp. 194–206). In one, the willing addict’s second-order willingness is itself an effect of the drug. In others the agent is hypnotised, brainwashed, or “aliens zapping” her.
  • “Frankfurt seems committed to the view that such agents act of their own free will and are morally responsible.”
  • Against the Source Incompatibilist Argument, this means Frankfurt “must show that being the source of one’s motives is not needed”. Otherwise, “why wouldn’t a deterministic history do the same?”

Frankfurt’s later replies (supplement §A, unchanged since 2009)

• The earlier account’s basis. Identification “previously rested solely on an agent’s having a higher-order volitional desire that some first-order desires be effective”.

• The 1987 reply. A person “properly identifies with a lower-order desire… when she has an unopposed higher-order volition to act in accord with the lower-order desire, and she judges that further deliberation would not influence her resolve”.

• The objection to it. The unopposed desire might be one “towards which the agent is passive”, as if it had “merely ‘happened’ to the agent”. The example is a willing addict “who has simply given up the struggle and drearily resigned herself to her addiction”. She satisfies the amended theory but does not “identify in a deep sense”.

• The 1994 reply. “So Frankfurt further amended his view with the requirement that an agent’s identification must be active. Her stance towards her own will is that it determines itself (1994). She and it are fully integrated… she regards it wholeheartedly.”

• Love. “More recent work… different features of personhood (such as the emotion of love)” is said to “bolster his original appeal to identification”. No citation is given.

• Manipulation. Frankfurt is quoted (2002, p. 27, “Reply to John Martin Fischer”):

  • “What we need most essentially to look at is, rather, certain aspects of the psychic structure that is coincident with the person’s behavior…”
  • “A manipulator may succeed… in providing a person… with a new character. That person is then morally responsible for the choices and the conduct to which having this character leads… It is irrelevant whether those causes are operating by virtue of the natural forces… or… through the deliberate manipulative designs of other human agents.”

  The entry concludes: “For Frankfurt, free will and moral responsibility are snapshot properties. Frankfurt is a pure internalist… he simply bites the bullet.” Critics are Fischer & Ravizza 1998, Haji 1998 and Mele 1995. A “limited defense” is McKenna 2004.

• Watson 1975 as a rival mesh. It was “Watson who originally objected to Frankfurt’s mere appeal to higher-order desires to account for identification. This precipitated Frankfurt’s further amendments.” Watson’s own model is “Platonic”: judgment can be its own source of motivation. The mesh is between a motivational system and a valuational system, and a free agent’s motivational system “works in harmony with her valuational system”.

• Velleman 1992’s point. Watson’s valuational system “seems susceptible to the same sort of objection that he (Watson) put to Frankfurt”. For any mesh, “we can always ask why these elements are to play that role”. The entry cites Velleman 2000, Watson 2004 and Buss & Overton 2002 for the ensuing debate.

• Bratman. He “embraced the problematic framed by the controversy between Frankfurt and Watson” and answers Velleman’s challenge with an “intention-based theory” (2007, p. 239):

  • agential authority lies in “self-governing policies” of practical reasoning;
  • these policies are hierarchical “in a way that differs from Frankfurt’s”, since they govern other plans (2007, pp. 240–3);
  • they have authority because they “bind the agent’s activities across time with the same threads that bind her identity across time” (2007, pp. 243–9).

  The entry calls Bratman an externalist (historical) theorist. A Pereboom-style manipulation “from a temporal distance” could “load in” his historical requirements.

• An open verdict. “Perhaps Frankfurt’s treatment of wholeheartedness has shielded his hierarchical view from the problem of ever ascending conflicts… Or perhaps instead Watson’s or Bratman’s proposal will help.” All mesh theorists still owe an account of “real self”, “agential authority” and “speaks for”, and reasons-responsiveness theorists may explain compulsion more elegantly (Mele 1995).

Frankfurt-style cases

• §3.2. PAP is stated as “A person is morally responsible for what she does do only if she can do otherwise”. The entry gives “a close approximation to the example Frankfurt presented in his original paper”. Black secretly arranges to manipulate Jones into shooting Smith if Jones shows “any sign” of wavering; Jones shoots Smith “for his own reasons” and Black never intervenes. The entry’s conclusion: if the argument is sound, the kind of freedom determinism might threaten “is not the kind of freedom required for moral responsibility”.
• §4.1.5. New dispositionalists (Fara 2008, pp. 854–5; Nelkin 2012; Smith 2003, p. 19; Vihvelin 2000, 2004, pp. 445–8, 2013) say the Frankfurt agent could have done otherwise. The entry leaves it open whether they may restrict attention to worlds without the intervener.
• §4.4.2–4.4.3. A banjo variant (Frank Zappa, Jerry Garcia, Jimi Hendrix) shows that Frankfurt cases defeat agent-based reasons-responsiveness, which is why Fischer moves to mechanism-based guidance control.

Attributions to the paywalled originals (every one, by location)

original	what the entry attributes to it	where
Frankfurt 1969	Rejects that the freedom to do otherwise is necessary for meaningful free will (with 1971)	§1
Frankfurt 1969	Argument against PAP; the Jones/Black/Smith shooting case, given as a “close approximation”, not quoted	§3.2
Frankfurt 1969	One of three 1960s contributions that “radically altered” the debate	§3, §4
Frankfurt 1971	Developed “on the back of” rejecting PAP; no appeal to regulative control; extends classical compatibilism; “hierarchical mesh”	§4.2
Frankfurt 1971	First- and second-order desires; the will is the effective first-order desire, “all the way to action” (p. 84)	§4.2.1
Frankfurt 1971	Therapist: a mere second-order desire, not a volition (pp. 84–5); second-order volitions; no limit to orders	§4.2.1
Frankfurt 1971	Unwilling and willing addicts; free will iff the action issues from “the will she wants”; freedom concerns the actual basis of action	§4.2.1
Frankfurt 1971	Persons are defined by second-order volitions; wantons are “passive bystanders to their wills” (p. 89); the wanton addict; the wanton is not a candidate for free will	2004–2015 revisions only (§5.3.1–2 / §5.2.1–2)
Frankfurt 1971	Conflict is resolved by forming a second-order desire, which identifies with one desire and alienates another	§4.2.2
Frankfurt 1971	Committed to mesh-only freedom, so manipulated agents with the mesh are free (the entry’s inference: “seems committed”)	§4.2.2
Frankfurt 1971	Identification “rested solely” on a higher-order volition	supp. §A
Frankfurt 1987	Identification = an unopposed higher-order volition plus a judgment that further deliberation would not change one’s resolve	supp. §A
Frankfurt 1992	Nothing: not cited, not listed	—
Frankfurt 1994	Identification must be active; the will “determines itself”; full integration; wholeheartedness	supp. §A
Frankfurt (uncited, “more recent work”)	Love bolsters identification	supp. §A
Frankfurt 2002, “Reply to Fischer”	Manipulated character is no bar to responsibility; causes “irrelevant” (p. 27, quoted)	supp. §A
Watson 1975	Rejects that the freedom to do otherwise is necessary for free will	§1
Watson 1975	Alienation possible at second and higher orders, so Frankfurt’s condition is not sufficient	§4.2.2
Watson 1975	Originated the objection that “precipitated Frankfurt’s further amendments”; a Platonic motivational/valuational mesh	supp. §A
Watson 1975	A mesh theory “not hierarchically structured”	2015 n. 22
Watson 1987	Strawsonian communication-based account	§4.5
Watson 1987	The Robert Harris case: history makes us ambivalent about blame	supp. Intro
Bratman 1997, 2003, 2004, 2007	Planning theory; self-governing policies (2007, pp. 239, 240–3, 243–9); diachronic identity as the source of authority; externalist	supp. §A
Bratman 1996	Nothing: not cited, not listed	—
Velleman 1992	Watson’s valuational system is open to Watson’s own objection; the generalised “why these elements?” challenge	supp. §A

Claims

id	claim	strength	support
C1	Frankfurt identifies the will with the effective first-order desire, “all the way to action”	second-hand report with page	§4.2.1, “1971, p. 84” (Watson-1982 reprint pagination, per 2015 n. 23)
C2	Frankfurt distinguishes mere second-order desires (the therapist) from second-order volitions (desires that a first-order desire be one’s will)	second-hand report with page	§4.2.1, pp. 84–5
C3	For Frankfurt, one acts of one’s own free will iff one’s action issues from the will one wants; the willing addict acts freely, the unwilling addict does not	second-hand report; the biconditional is the authors’ formulation	§4.2.1; no page given
C4	For Frankfurt, only persons have second-order volitions; wantons are “passive bystanders to their wills” and are not candidates for free will	second-hand report with page, in the 2004–2015 revisions only	2015 §5.2.1–5.2.2, “1971, p. 89”
C5	Watson 1975 objected that the agent can be alienated from (2015: conflicted at) second and higher orders, so Frankfurt’s condition is at most necessary	second-hand report; no page	§4.2.2; supp. §A
C6	Frankfurt’s 1971 view makes freedom independent of how the mesh arose, so manipulated agents come out free	the authors’ inference (“seems committed”) backed by cited critics	§4.2.2; Slote 1980; Fischer & Ravizza 1998, pp. 194–206
C7	Frankfurt accepts this: a manipulator-given character leaves the person morally responsible	direct quotation from Frankfurt 2002, p. 27	supp. §A; the only verbatim Frankfurt passage in the entry apart from “all the way to action”
C8	Frankfurt’s 1987 reply: an unopposed higher-order volition plus a judgment that further deliberation would not change one’s resolve	second-hand report; no page, no quotation	supp. §A
C9	Frankfurt later (1994) required identification to be active, the will “determines itself”, which amounts to wholeheartedness; this was to exclude the resigned addict	second-hand reconstruction; no page, no quotation; the resigned-addict motive is the authors’	supp. §A. Contradicted in part by Frankfurt’s own 2004 text (fr1); see Connections
C10	Velleman showed that Watson’s valuational mesh faces Watson’s own objection, and that any mesh faces “why these elements?“	second-hand report	supp. §A (Velleman 1992; consistent with fr3’s reading of the original)
C11	Bratman grounds agential authority in self-governing policies whose authority comes from cross-temporal identity; his view is externalist	second-hand report with pages	supp. §A (Bratman 2007, pp. 239–49); “externalist” is the authors’ classification
C12	Every mesh compatibilism (Frankfurt, Wolf, Fischer & Ravizza, Haji, Wallace, Bok, Mele) must either explain manipulation cases away or accept manipulated agents as free, and historical conditions alone do not escape (Pereboom, Mele’s zygote argument)	informal argument, the authors’ own	§§4.2.2, 4.3, 4.4.4; supp. Intro, B.2, C–F
C13	The Jones/Black/Smith case is a counterexample to PAP if sound; the debate is unresolved	reported, with the case paraphrased (“close approximation”)	§3.2
C14	The classical conditional analysis has been refuted	assertion, citing Chisholm 1964 and van Inwagen 1983	§2.2 (the Danielle case)

Concepts

• will (Frankfurt, per §4.2.1): the effective first-order desire, the one that moves the agent “all the way to action”.
• second-order desire / second-order volition: a desire about a desire, versus a desire that a given first-order desire be one’s will.
• wanton (2004–2015 revisions only): an agent with no second-order volitions, “passive bystander” to her will; not a person.
• willing / unwilling / wanton addict: the three cases illustrating free will as having the will one wants.
• mesh theory: freedom as a fit between elements of the agent’s psychology. Frankfurt’s is hierarchical; Watson’s is between the motivational and valuational systems; Wolf’s is between action and “the True and the Good” (an external anchor); Bratman’s is between intentions and plans and self-governing policies.
• hierarchical problem / mesh problem (§4.2.2): higher-order alienation; manipulated meshes.
• regulative / guidance control (Fischer 1994): control over alternatives versus control in the actual sequence.
• internalist (snapshot) / externalist (historical) theories (supp. Intro): freedom fixed by present structure, versus partly by causal history (the analogies are weight versus sunburn or a genuine dollar bill).
• manipulation argument (supp. Intro): manipulated agents are not free; determined agents are relevantly like them; therefore determined agents are not free. Pereboom’s four-case version; Mele’s zygote argument (supp. F).
• real self theory (Wolf 1990, p. 29, as cited in the 2004–2015 revisions): Frankfurt’s view as one on which free action reveals the agent’s “deeper” or “true self”.

Connections

Within this batch. None of these readings has a code yet, so they are named in prose.

Frankfurt’s own Tanner Lectures (“Taking Ourselves Seriously”/“Getting It Right”, fr1).

• Agreement. The biconditional here, action issuing from “the will she wants”, matches Frankfurt’s 2004 statement that “we will freely when what we want is what we want to want” (p. 177, per fr1).
• What each lacks.
  • The Tanner text restates freedom of action versus freedom of the will; no revision of this entry does.
  • This entry has the second-order-volition vocabulary and the addicts; the Tanner text does not.
• A real disagreement about resignation. Supplement §A says Frankfurt amended his view so that a resigned, given-up willing addict does not identify, requiring “active” identification. Frankfurt himself in 2004 says identification is “willing acceptance… whether with a welcoming approval, or in weary resignation” (p. 173, per fr1), and that accepted states’ “force is now our force… We are active”. So on Frankfurt’s own late account, resigned acceptance is identification and is activity. The entry’s reconstruction of the 1994 amendment should not be trusted on this point without the original.

SEP “Personal Autonomy” (Buss & Westlund, fr4).

• Watson. This entry and PA differ on Watson’s role. PA files Watson 1975 only as a rival coherentist account and never presents him as objecting to Frankfurt; its regress argument is its own and uncited. This entry makes Watson 1975 the source of the higher-order objection (§4.2.2; supp. §A: “it was Watson who originally objected”), and a rival mesh.
• Frankfurt’s replies. PA’s note 1 gives them as “decisive commitments”, then “satisfaction” (an absence of “restlessness or resistance”). This entry gives 1987 unopposed volition plus settled judgment, then 1994 active self-determining wholeheartedness, and never mentions satisfaction or “The Faintest Passion”. If PA’s gloss on satisfaction is right, satisfaction is exactly what accommodates the resigned agent, which runs against the direction of the amendment this entry reports.
• Manipulation. The two agree that manipulation defeats coherence as a sufficient condition. This entry adds Frankfurt’s own acceptance of the consequence (2002, p. 27).

IEP “Autonomy” (Dryden, fr5).

• Where it agrees with this entry. IEP attributes the regress/incompleteness objection to Watson 1975, as this entry does. It also treats Frankfurt as “ahistorical”, matching this entry’s “pure internalist”.
• Disagreement 1: when the decisive reply begins. IEP says Frankfurt ends the series with a “decisive commitment” already in the 1971 paper (“Frankfurt 1988, 21”, which falls in the 1971 reprint). Velleman (fr2) likewise cites IWWCA 21 for the image of commitment resounding up the hierarchy. This entry says the 1971 identification “rested solely” on a higher-order volition, and that the first reply came in 1987.
• Disagreement 2: the date of wholeheartedness. IEP ties wholeheartedness to “Identification and Wholeheartedness” (1987; IWWCA pp. 168–175). This entry ties it to 1994.
• Terminology. IEP slips between “second-order desires” and “second-order volitions”. This entry keeps them apart, and its therapist example is the clearest statement in the batch of why the difference matters.

SEP PAP entry (Robb, fr6). Both retell Frankfurt 1969 without quoting it, and they retell it differently. Robb gives Fischer’s voting version (Fischer 1982: 26, “cp. Frankfurt 1969: 835–836”). This entry gives a shooting version as “a close approximation”. Neither is the 1969 text. Robb’s §5.1 compresses the 1971 view to identification with one’s motives. This entry supplies the apparatus behind that phrase.

Velleman, “What Happens When Someone Acts?” (fr3). The supplement’s use of Velleman 1992 is consistent with fr3’s reading: Watsonian values, like higher-order desires, are items the agent can be alienated from. The supplement presents it as a challenge to Watson more than to Frankfurt, and it omits Velleman’s own positive proposal.

Velleman, “Identification and Identity” (fr2). Only incidental overlap: the same “1987” dating slip for Frankfurt’s 1988 collection. The supplement does not reach Frankfurt’s 1990s volitional-essence view, which is Velleman’s target.

Existing record.

• LIT-202 (“Consent and the Formation of Preferences”). Per its reading, it uses a historical, mechanism-based criterion for free agency in preference formation, with a Frankfurt-style actualist point. That criterion is the kind of externalist condition this entry’s supplement argues can be defeated by manipulation that “loads in” the history (B.2, F).
• LIT-197 (AI and free will, Deferred). Per its abstract, it runs manipulation and reasons-responsiveness together. This entry’s §4.4 and supplement B separate reasons-responsiveness from the historical ownership condition. The connection is to the abstract only.

Bearing on the record

• The Frankfurt set in the reading record. This entry should be filed as the source for the 1971 apparatus (C1–C4), with an explicit note that wantons and persons come from the 2004–2015 revisions, and that the pages are Watson-1982 reprint pages.
• Frankfurt’s later development. Treat this entry’s reconstruction (C8–C9) as one reading among three. It disagrees with the SEP “Personal Autonomy” and IEP accounts on dates and content, and in part with Frankfurt’s own Tanner text.
• ML practice. It carries nothing for ML practice. The manipulation-argument dialectic is sometimes invoked in writing on AI influence over human choice (cf. LIT-197), but the entry says nothing about AI and offers no instruction that belongs in the Anthology of the SOTA.

Limitations

• Second-hand throughout.
  • The only verbatim Frankfurt text is the phrase “all the way to action” and the 2002 manipulation passage.
  • Everything else about 1969, 1971, 1987 and 1994 is paraphrase.
  • The 1969 case is an approximation.
  • The 1987 and 1994 replies have no page references.
• The current revision is thinner on 1971 than earlier ones. Wantons, persons, the wanton addict and the “real self” classification were removed in 2019. The supplement still presupposes the last of these.
• Frankfurt’s freedom of the will versus freedom of action distinction is never stated.
• The supplement is not current. Its §A has been unchanged since the 2009 revision, and its latest citation is 2008. It lacks “The Faintest Passion” (satisfaction), Necessity, Volition, and Love beyond a bibliography line, Taking Ourselves Seriously (2004/2006), Bratman 1996, and all post-2008 work on identification.
• Frankfurt is framed through Fischer’s regulative/guidance distinction and the source-incompatibilist challenge. That is the authors’ frame. Frankfurt’s own concerns (personhood, caring, love) appear only in passing.
• Bibliography and cross-reference defects are listed under corrections. The 1971 page references are uninterpretable from the current revision alone.

Open questions

• Is the resigned willing addict identified with her addiction on Frankfurt’s considered view?
  • Supplement §A says Frankfurt’s 1994 amendment excludes her.
  • Frankfurt 2004 (fr1) and “satisfaction” as SEP PA glosses it seem to include her.
  • Frankfurt 1994 (“Autonomy, Necessity, and Love”) and 1992 (“The Faintest Passion”) would settle it.
• Does the 1971 paper already contain the decisive-commitment answer to the regress (IEP; Velleman’s IWWCA 21)? This entry implies it does not. The 1971 text, around its final pages, would settle it.
• Is the “unopposed volition plus judgment that further deliberation would not change one’s resolve” formulation (C8) Frankfurt’s 1987 wording, or the authors’ paraphrase? It needs “Identification and Wholeheartedness”, pp. 159–76 of the 1988 collection.
• Does Frankfurt’s 2002 bullet-biting (C7) extend to the willing addict whose second-order willingness was produced by the drug itself (§4.2.2’s first manipulation case)? The quoted passage speaks of a new character, not of a desire installed in isolation.

Corrections to the seeded skim

• none (there was no seed or dossier)
• Authorship differs by revision. The scoping report gives “Michael McKenna; 2004; rev. 2024-04-16”, and the batch item gives “McKenna & Coates”. McKenna alone wrote the 2004 first version and the 2009 revision. Coates is co-author from the 25 Feb 2015 revision onward. Cite it as McKenna & Coates, revision of 16 Apr 2024.
• The scoping report says §4.2 gives “the 1971 apparatus”. That is only partly true of the current revision.
  • What is there: first- and second-order desires; the will; second-order volitions versus mere second-order desires; orders without limit; the unwilling and willing addicts; the biconditional.
  • What was removed: in the 26 Nov 2019 revision, the paragraph defining persons by second-order volitions and wantons by their absence, the wanton addict, and “The wanton is not a person, and so is not a candidate for freely willed action”.
  • Where to find it: the 2004–2015 revisions have it, in §5.3.1–5.3.2 (2004) and §5.2.1–5.2.2 (2015).
  • Never stated, in any revision I read: Frankfurt’s freedom of the will versus freedom of action distinction. “Freedom of action” appears only in §2.1 of the 2015 text, for Schlick.
• The 1971 page references now point nowhere.
  • The current text cites “(1971, p. 84)” and “(1971, pp.84–5)”. The bibliography gives the original as J. Phil. 68: 5–20, which has no p. 84.
  • The 2015 revision’s note 23 says “All page references to Frankfurt 1971 are to the reprint in Watson (ed.) 1982”. The current revision has no notes, so the key is lost.
  • Read these as Watson-1982 pages. Converting them to J. Phil. or Importance of What We Care About pagination is unverified.
• Stale cross-references from the 2019 renumbering.
  • §4.2 says classical compatibilism is at “Section 3.1”; it is §2.1.
  • §3.2 points to “section 3.3” for the conditional analysis; it is §2.2. Supplement E points to “3.3 above”; also §2.2.
  • §4.1.5 says “Consider a Frankfurt example (section 4.2)”; the example is in §3.2.
  • §4.1.5 points to “section 5.5 below” for Fischer & Ravizza. It is §4.4, and there is no §5.
  • §4 promises “six” positions and has five subsections.
  • Supplement §A says Frankfurt’s account “formed the basis for categorizing Frankfurt’s as a real self theory”. The main-text paragraph that did so (Wolf 1990, p. 29) was removed in 2019, so the remark now dangles.
• Slips in the text.
  • §4.1.5 says “Smith’s shooting Jones”; in the example Jones shoots Smith.
  • §4.2.1 spells “Frankurt”.
  • The supplement keeps McKenna’s first person (“as I have characterized it”, “I’ll not set out the details”).
  • In supplement §A the entry says Bratman, an externalist, “would deny the first premise” of the manipulation argument. By the supplement’s own scheme, externalists (Fischer & Ravizza, B.2) deny the second premise. The sentence is either a slip or a reading of premise (1)‘s “despite satisfying” clause; it is unclear which.
• Bibliography defects.
  • Frankfurt 1969 and 1971 are said to be “reprinted in … Frankfurt, 1987”. The collection is listed in the same bibliography as 1988 (The Importance of What We Care About). Velleman’s “Identification and Identity” (fr2) makes the same 1987 slip.
  • “Identification and Wholeheartedness” (1987) is said to be “reprinted in Frankfurt, 1999” (Necessity, Volition, and Love). The SEP “Personal Autonomy” and IEP “Autonomy” readings place it in the 1988 collection, pp. 159–76.
  • Velleman 1992 is given as Mind 101: 462–81. Crossref, per fr3, gives 461–481.
  • “McKenna 2005, ‘Reasons Reactivity & Incompatibilist Intuitions’, Philosophical Explorations 8(2): 131–43” duplicates “Fischer 2005” with the same title and venue; one of them is misattributed (unverified which).
  • Bratman 1997, “Responsibility and Planning”, is given as “Journal of Philosophy, 1 (1)”. The volume number suggests The Journal of Ethics (unverified).
  • Bratman 1996, “Identification, Decision, and Treating as a Reason”, is neither cited nor listed.
  • Frankfurt 1992 (“The Faintest Passion”) is neither cited nor listed.
• The scoping report calls this a “compact summary of the 1971 theory” and marks it optional. For this batch it is more than that. It is the only reading so far that states the will as the effective desire, the therapist case, the second-order desire/volition distinction, both addicts and the free-will biconditional, with page references, and (in 2004–2015) the wanton. It is also the only one that quotes Frankfurt directly on manipulation (2002, p. 27).