THEORY-119: An agent can be morally responsible for what she did although she could not have done otherwise, when what made the action unavoidable played no part in bringing it about, so the principle of alternate possibilities is false

The evidence, and what is unread

The argument is read first-hand: Frankfurt’s “Alternate Possibilities and Moral Responsibility” (LIT-694, NOTE-546), and the willing addict of “Freedom of the Will and the Concept of a Person” (LIT-692, NOTE-545). The objections and replies are not: Widerker, Kane, Ginet, Fischer, Pereboom, Mele and Robb, Hunt, and Frankfurt’s own later replies (1983, 2003) are held only through Robb’s SEP entry (LIT-294, NOTE-269), McKenna and Coates (LIT-291, NOTE-270) and Talbert (LIT-691, NOTE-548). The owner authorised filing it on this evidence.

Source

• Frankfurt (1969), LIT-694, read in NOTE-546: §§II–V, pp. 830–839, the case and the argument.
• Frankfurt (1971), LIT-692, read in NOTE-545: §IV, pp. 18–20, the willing addict.
• Robb, SEP Moral Responsibility and the Principle of Alternative Possibilities, LIT-294, §§3.4.3, 4 and 5, read in NOTE-269: the objections and replies.
• McKenna and Coates, SEP Compatibilism, LIT-291, §§3.2, 4.1.5 and 4.4.2, read in NOTE-270: the case’s place in compatibilism, and the new dispositionalists.
• Talbert, SEP Moral Responsibility, LIT-691, §1, read in NOTE-548: the dilemma, and Frankfurt’s later gloss.

What the sources show

The argument. LIT-694 separates what makes an action unavoidable from what brings it about. Black, who “prefers to avoid showing his hand”, will intervene only if Jones4 is about to decide otherwise, by any means a reader thinks sufficient, up to manipulating “the minute processes of Jones4’s brain”. Jones4 decides “for reasons of his own” and Black never acts. Jones4 “will bear precisely the same moral responsibility … as he would have borne if Black had not been ready”, and it would be “quite unreasonable to excuse Jones4 … or to withhold the praise to which it would normally entitle him” (pp. 835–836). The general premise is that a fact which “could have been subtracted from the situation without affecting what happened or why it happened” has no weight in assessing responsibility (p. 837). Frankfurt’s later recipe, quoted by Robb (LIT-294, §3.4.3, from 1983: 322), states the same distinction: “circumstances which make it inevitable that a person will perform some action but which do not bring it about that he performs it”. Talbert (LIT-691, §1) quotes the 2006 gloss: “making an action unavoidable is not the same thing as bringing it about that the action is performed”.

The replacement. LIT-694 proposes that “a person is not morally responsible for what he has done if he did it only because he could not have done otherwise”, and says it “does not appear to conflict” with compatibilism (pp. 838–839). It rejects the intermediate revision (“because”, without “only”) because, on the incompatibilist reading of “could have done otherwise”, it would excuse every determined act (p. 838).

A second route. LIT-692’s willing addict cannot do otherwise, since his desire “will be effective regardless”, but takes the drug “of his own free will” and “may be morally responsible” (pp. 19–20). Robb (LIT-294, §3.2) reports it as a counterexample and limits it: it is really overdetermination, with a responsibility-undermining factor present.

The objections, as Robb reports them (LIT-294, NOTE-269).

• The dilemma defence (§4.3; Kane, Widerker, Ginet, Wyma). Black needs a prior sign. Either the sign determines the decision, and the libertarian will not grant that Jones is responsible, or it does not, and Jones keeps a robust alternative. The replies are Fischer’s on the first horn, that a causally idle factor cannot be what makes the agent not responsible (1999: 113, cf. Frankfurt 2003), and, on the second, Pereboom’s buffer case, Hunt’s blockage and Mele and Robb (1998), against which the critics say blockage collapses into determination.
• Prior signs. The sign is in the original case: a twitch that precedes deciding A and never B, which Frankfurt says “does not beg the question of determinism” and does not require that the earlier phases be “deterministically related to still earlier events” (LIT-694, n. 3). That is the point the dilemma presses: the footnote leaves open whether the twitch determines the decision, and the dilemma says either answer defeats the case.
• Flickers of freedom (§4.2.2). Jones could still have shown the sign, or refrained on his own; Frankfurt concedes it is “in a way up to him whether he acts on his own or as a result of Black’s intervention” (LIT-694, p. 836). Fischer’s reply is that such alternatives are not “robust” (1994: 140–147); the counter-replies are that a flicker may mark indeterminism, or that refraining is itself a doing.
• General abilities (§4.2.1; LIT-291, §4.1.5). The new dispositionalists say a Frankfurt agent keeps the general ability to do otherwise, so PAP-general survives.
• The case is too unusual (§4.1). PAP-ordinary restricts the principle to ordinary circumstances. The reply is that some Frankfurt cases are ordinary, and that one coherent case shows the tie contingent. Frankfurt himself lets Black be “a machine” or “natural forces involving no will or design at all” (LIT-694, n. 4).

Robb’s assessment is that “not much consensus” has emerged (§5). McKenna and Coates say “No clear victor has emerged” (LIT-291, §3.2).

What this does not say

• It does not say determinism is compatible with responsibility. It removes one premise of the classical incompatibilist argument. Robb (§5.1) and Talbert (§1) report that determinism may still threaten responsibility by the way it actually produces action.
• It does not say what responsibility does require. Frankfurt’s own replacement is “only because”; Fischer and Ravizza’s is guidance control (THEORY-121); the hierarchical account is THEORY-120.
• It does not cover volitional necessity on its own. Robb (§3.3) cites Frankfurt’s 1982 essay (LIT-686) for cases like Luther’s, and reports that tracing restores PAP for them. This account rests on the counterfactual-intervener cases, not on those.
• It does not settle the dilemma. The record reads the defence and every reply only in Robb’s summary.

Relations

• rivals THEORY-118 (Wolf’s asymmetry). Wolf, as McKenna and Coates report her (LIT-291, §4.3), holds that praise needs no ability to do otherwise but blame does, the ability to have acted on the right reasons. This account says neither needs it. On praise they agree. On blame they cannot both be right: Frankfurt states his case for both excusing and withholding praise (LIT-694, p. 836), and his action is neither good nor bad, so a Frankfurt case of wrongdoing is blameworthy here and excused on Wolf’s view unless she has an answer the record has not read. Neither yet corrects the other.
• Fischer and Ravizza (THEORY-121) build on this. Their account says responsibility requires guidance control “and not regulative control over alternative possibilities”, and its Frankfurt-case basis is this argument (their sources include LIT-294 for it). That is extends from their side. It is not declared here, because their file would hold it without a sentence explaining it; the coordinator should add it there.
• THEORY-120: no relation. The 1971 willing addict serves both, but the hierarchical account is about what makes a motive one’s own, and this one is about alternatives.