Promises and Performative Utterances

The Problem of Words That Bind

A promise is a particularly revealing form of human action because its central effect is constituted through communication. When someone says to another person, under ordinary favorable circumstances, that she promises to return a book tomorrow, a future event has not thereby occurred; a present normative relationship has. The addressee can subsequently demand an explanation for nonperformance, can organize expectations around the undertaking, and may have the standing to release the speaker from it. The statement of what will happen and the undertaking of what one will do are distinguishable achievements of the same utterance. Their separation is the philosophical starting point of the problem.

Three questions structure the inquiry. First, what makes an utterance a promise rather than a prediction, reassurance, declaration of intention, or threat? Second, how can the act generate a duty that was absent before it? Third, what gives that duty its distinctive direction toward a promisee and explains its limits, exceptions, and modes of discharge? An adequate theory must answer these questions without collapsing linguistic recognition into moral authority or reducing the moral claim to a hearer's momentary psychological expectation. The institutional and ethical dimensions interact, but successful analysis requires that each explanatory burden be placed where it belongs.

The familiar contrast between words and deeds is itself misleading. Words are deeds when they are used by agents to address others, exercise authorized powers, make commitments, or transform relations. Yet no sequence of sounds has these effects irrespective of setting: a theatrical rehearsal, an overheard soliloquy, an authorized declaration, and a message accepted as an undertaking have different practical statuses. Austin's treatment of the total speech situation remains the decisive corrective to the idea that a verb contains the whole act (Austin, 1962/1975, Lectures I–III and XI–XII). The lasting puzzle concerns how participation in a practice supplies words with a normative consequence and how morality should evaluate the consequence. Austin, How to Do Things with Words.

Definition, Scope, and the Central Distinctions

A promise is a communicatively manifested undertaking by an agent, addressed to an actual or contextually identifiable party, concerning an action or course of conduct for which the agent presents herself as answerable. A paradigmatic promise places a directed claim in the hands of a promisee: the speaker has undertaken to perform, and the promisee has a special standing to seek performance or explanation. This formulation describes the phenomenon to be explained; whether every successful promise creates a moral duty, a practice-relative duty, or both remains disputed.

The content of a promise is usually future-directed and concerns conduct sufficiently related to the promisor's agency. The words can be explicit, such as "I promise to send the draft," or contextually decisive, such as "You have my word; it will be on your desk tomorrow." A bare prediction that the draft will arrive and a promise to deliver it could have identical propositional content while differing in illocutionary force. Content says what is represented; force says what act is being performed with that content. The grammatical future tense, therefore, supplies neither a sufficient mark nor a necessary mark of promising.

Several analytical levels must be kept distinct. Performance is whether the promising act occurred. Felicity is whether it was performed in the relevant circumstances with sufficient procedural propriety. Sincerity concerns the agent's intentions or attitudes. Bindingness concerns the normative consequence of the undertaking. Fulfillment concerns the later conduct satisfying its terms. Excuse, justification, and release concern different ways in which accountability may change after the original act. An insincere promise may be genuinely made and morally binding; a sincerely intended declaration may fail to amount to a promise; a valid promise may later become impossible to perform; performance may be prohibited by an overriding moral duty. Treating any two of these as synonymous creates false solutions.

The distinctions also establish the monograph's boundary. Speech acts in general constitute the wider field; promising is a specific commissive case with unusual relevance to future agency and directed duties. Contract law, testimony, oath-taking, and institutional declarations are investigated here insofar as they reveal the nature of promising. They remain independent domains whose full theories require separate studies. The contemporary taxonomy of promises and promissory obligation is mapped in the Stanford Encyclopedia of Philosophy's Promises and Speech Acts.

Ancient Fidelity and the Conditionality of Keeping One's Word

In classical moral thought, the obligation to keep one's word is situated within an account of justice, practical wisdom, and social fidelity rather than an autonomous theory of linguistic acts. Cicero's De officiis furnishes an especially important case. In Book I, sections 31–32, he recognizes that circumstances can make the literal fulfillment of an undertaking harmful to the person to whom it was made. The point is not a convenient escape from fidelity: practical judgment must establish what justice requires when the background conditions that made keeping the promise appropriate have changed. Cicero thereby distinguishes fidelity from an unconditional mechanical rule that every uttered undertaking be performed whatever its consequences.

This early discussion anticipates two modern questions. Is the obligation generated simply by a prior speech act, or does its continuing force depend on the reasons that made the undertaking legitimate? And who determines whether the circumstances justify nonperformance? If only the promisor may judge, promising offers weak security to recipients; if the original formula binds regardless of consequences, promising may perpetuate harm. Cicero's answer locates the duty within the larger moral order of avoiding injury and preserving shared interests (Cicero, De officiis I.10.31–32, Walter Miller translation). Latin text and English rendering.

Ancient discussions help explain why fidelity became central to ethical life before performativity became a specialized category in twentieth-century philosophy of language. They also caution against projecting Austin's later classification backward as the vocabulary of Cicero. The ancient problem concerns the moral priority and content of keeping faith; the modern speech-act problem asks in addition what act a speaker performs and how the undertaking becomes publicly intelligible. The second can be understood as an extension of the first, not its historical replacement.

Medieval Vows and the Structure of Voluntary Binding

Thomas Aquinas distinguishes an intention to act from the additional commitment created by a vow. In Summa theologiae II–II, question 88, article 1, he analyzes a vow as involving deliberation, purpose, and promise; a purpose alone does not exhaust its binding form. The vow's addressee is God, and the theological setting changes how communication is understood: Aquinas holds that inward thought can suffice before God, whereas a promise between persons needs words or another outward sign. The contrast makes manifest the epistemic role of publicity in interpersonal undertakings. Aquinas, Summa theologiae, II–II, q. 88.

Aquinas also treats the content and bindingness of vows. Deliberation concerns whether an undertaking is knowingly made; the promised good concerns whether its content can be an appropriate object of commitment; and impossibility raises a question of what remains owed when the literal action can no longer be done. These are distinct difficulties. The very possibility of promising depends on agents who can deliberate, but the moral value of a promise is not exhausted by the subjective firmness of the promisor's will. In article 3, Aquinas examines fulfillment when circumstances change; this anticipates the modern discussion of discharge, substitute performance, and culpable as opposed to blameless inability.

The medieval case is important conceptually because it separates an inner practical resolution from an outward or publicly accountable undertaking while recognizing a special theological exception. It establishes that a promise is a way of directing future conduct, and that its force is not explained solely by the intensity of an intention. Treating the vow as simply identical with ordinary interpersonal promising would erase Aquinas's specifically theological conception of the addressee, obligation, and the good.

Hobbes: Covenant, Right, and Signs of Will

Thomas Hobbes treats promising within a theory of right, contract, and mutual dependence. In Leviathan (1651), chapter 14, he distinguishes a present transfer of right from a mere expression that one will give something in the future. In a free gift, future-tense words by themselves may fail to transfer right. In a covenant supported by reciprocal conduct, future performance can be binding: where one party has performed and trusts the other to perform subsequently, the undertaking acquires determinate practical standing. This is a precise demonstration that a sentence's tense does not determine its normative effect in isolation. Hobbes, Leviathan, ch. 14.

Hobbes locates the sense and force of an undertaking within an institutional and motivational context. His distinction between a promise, covenant, and transfer of right shows why a speaker cannot change the normative positions of others merely by naming a future event. The wider structure of covenant depends on trust, exchange, and socially intelligible signs of will. This is also why Hobbes's familiar account of the state of nature matters: the status of covenants and the reasons for performance depend on security, reciprocity, and political conditions. Those arguments cannot be replaced by an isolated maxim that a promise always creates enforceable duties.

For contemporary analysis, Hobbes contributes a division between saying something about a future action and performing an act that presently transfers or undertakes a right. That division appears in more recent theories of normative powers. The historical kinship is conceptual; it need not be asserted as a direct line of influence upon every later theorist.

Hume: Convention Before Obligation

David Hume gives the conventionalist problem one of its most forceful statements in A Treatise of Human Nature, Book III, Part II, section 5, "Of the Obligation of Promises." He argues first that a promise is unintelligible prior to a human convention and second that, even if it were intelligible, a moral obligation would not arise from a bare private mental act. He rejects the reduction of promising to an intention, desire, or resolution, since any of these can occur without undertaking a promissory duty (Hume, 1739–1740/2007, T 3.2.5.1–4). The Hume Texts Online critical presentation permits checking these passages against their stable Treatise numbering.

The argument's structure is more subtle than a simple appeal to the social usefulness of promises. Hume begins by testing candidate mental acts that might supposedly make a promise binding. A resolution cannot by itself generate the interpersonal duty. The suggestion that one wills the obligation appears circular when the obligation is precisely what must be explained. Human beings nevertheless have a powerful common interest in coordinating actions across time; practices of verbal undertaking arise within this need. The conventional structure gives the sign its sense, and moral attitudes toward fidelity develop around the practice.

Hume thereby offers an explanatory genealogy: the public significance of a promissory formula emerges from arrangements that enable cooperation, rather than from the natural production of obligation by words considered independently. A contemporary objection asks whether social usefulness explains why a particular person is owed performance, as opposed to explaining why the institution is valuable overall. The distinction is decisive. A beneficial practice can support general reasons to comply, but the promissory relation ordinarily involves a directed claim belonging to a particular recipient. Hume's account is strongest as an account of the institution's intelligibility and social origin; its relation to directed moral rights requires further analysis.

Kant: The False Promise as a Test of Practical Reason

Kant transforms the moral problem into a test of the rational permissibility of maxims. In the Groundwork of the Metaphysics of Morals (1785), he considers making a false promise to escape difficulty. At Akademie 4:402–403 the question is whether deceiving by promising is consistent with duty; at 4:422 he develops the case of borrowing money on a promise one intends not to honor as an illustration of the formula of universal law. When the maxim of false promising is universalized, the reliance on promising that the deceit presupposes is undermined. The agent would attempt to profit from a practice whose practical significance the maxim itself would destroy. Groundwork, Akademie text references.

This is not merely the prediction that widespread dishonesty would gradually reduce trust. The stronger argument concerns practical contradiction: a maxim of obtaining cooperation through a promise that is systematically not to be kept requires the recognized force of promising while authorizing its universal frustration. Whether this amounts to a contradiction in conception or requires supplementary assumptions about recognition and common knowledge is a substantive issue in Kant scholarship. Analyses of the false-promise example examine precisely this structure (Contradiction and Kant's Formula of Universal Law).

Kant supplies a moral test that is partly independent of the social value of promissory conventions. Even in cases where deception would be advantageous to the deceiver, practical reason assesses the maxim's universal form. This explains the special wrong in manipulating the addressee's rational reliance. Still, the argument alone does not give a complete semantic analysis of the act of promising, nor does it settle whether all promises made under coercion, deception, or immoral objectives bind. A theory of the practice and a theory of moral justification must cooperate.

Nietzsche: Memory, Responsibility, and the Capacity to Promise

Nietzsche's On the Genealogy of Morality (1887), Second Essay, sections 1–3, approaches promising as an achievement of historical formation. The capacity to make commitments across time requires the formation of memory, calculability, self-command, and an identity that others can count upon. Promising becomes exemplary of the power to stand behind one's future action. Its conditions are not purely logical: they are linked in Nietzsche's genealogy to discipline, punishment, and the historical production of responsibility (Second Essay, Ian Johnston translation).

Nietzsche's achievement is to locate promising within temporal agency. A promise binds a present agent to a future performance, although motives, capacities, and circumstances can change. Reliability requires more than an isolated moment of sincerity: it concerns the durability of a practical identity through time. His genealogy also raises a critical political issue. Social institutions that make agents predictable can cultivate responsibility but can also produce domination. The capacity to promise is therefore tied to historically organized relations of power and discipline.

The Nietzschean contribution differs from the linguistic problem of illocutionary classification. He asks what sort of agent can credibly bind a future self and how such agents are socially produced. Austin and Searle ask how an undertaking is performed. Both questions are necessary to a general philosophy of promises, because a successful utterance can create an undertaking whose fulfillment depends on an agent's capacity to carry commitments forward through time.

The Linguistic Turn and Austin's Initial Discovery

J. L. Austin's How to Do Things with Words, based on the William James Lectures of 1955 and first published posthumously in 1962, brings the phenomenon of promising to the center of ordinary-language philosophy. The second edition was edited by J. O. Urmson and Marina Sbisà (1975). The editors' prefaces matter: the text derives from lecture notes and editorial reconstruction rather than a finished monograph prepared by Austin for publication. Second-edition bibliographic record; accessible second-edition excerpts and editorial preface.

Austin begins by challenging the assumption that the primary philosophical function of sentences is to describe facts and be assessed as true or false. Uttering certain sentences under appropriate circumstances is itself to perform an act: betting, naming, appointing, marrying, or promising. In an explicit performative, the present-tense first-person formula announces the act being undertaken. The speaker who appropriately says "I promise" does not ordinarily report an already completed inner event; the utterance contributes to performing the promise.

The qualification "under appropriate circumstances" carries much of the theory. The same words could be quoted in a language lesson or uttered in a stage performance, where their force differs. Moreover, a promise may be made without an explicit performative verb. Austin distinguishes primary performatives from explicit ones, recognizing that "I will be there" can become a promise in context while remaining merely a prediction in another. His discovery concerns action through language, not the magical power of a fixed verbal formula.

Austin's Theory of Infelicity

Austin calls the conditions governing successful performatives felicity conditions and analyzes ways in which they can fail. His early scheme distinguishes misfires, where a purported act fails to come off, from abuses, where an act is performed but is defective because an appropriate intention, feeling, or subsequent conduct is absent (Austin, 1962/1975, Lectures II–IV). Misfires can involve the absence of a recognized procedure, an unsuitable person or circumstance, or incorrect or incomplete execution. Abuses encompass insincerity and failures to honor commitments undertaken.

The difference between an insincere promise and an unsuccessful attempt to promise is fundamental. A speaker who promises while secretly intending to break her word may nonetheless have made an actual promise; her deception constitutes an abuse, not necessarily a nullity. This allows moral blame for breaking a promise that was not sincerely intended from its inception. If insincerity rendered every promise void, deceitful promisors could escape the very obligations they misrepresented as undertaking. Austin explicitly notes the difference between a promise given in bad faith and an utterance that fails to count as a promise.

Felicity conditions are neither exhaustive moral commandments nor a complete theory of the legal validity of contracts. They are analytical criteria for understanding the circumstances and defects of linguistic actions. Some acts demand specific institutional authorization; everyday promising often requires less formal procedure. The classification must be responsive to variations among acts rather than treated as a single rigid universal checklist.

Locution, Illocution, and Perlocution

Austin's later framework analyzes a speech event into locutionary, illocutionary, and perlocutionary aspects (Lectures VIII–X). Locution concerns saying something meaningful. Illocution concerns what one does in saying it, such as promising, warning, or ordering. Perlocution concerns what one brings about by saying it, such as persuading, reassuring, frightening, or inducing someone to plan differently. In promising, the undertaking belongs centrally to illocution, while increased confidence or reliance belongs to typical but contingent perlocutionary consequences.

This distinction explains why trust or belief is not identical with a promise. A sincere undertaking may be understood but doubted, in which case the promisee may decline to rely on it. Conversely, a confident prediction can generate strong reliance without imposing an undertaking on the speaker. To identify promising with the production of expectation is to mistake a possible effect for the act's normative force.

At the same time, illocution and perlocution interact. Promises conventionally invite reliance, and speakers may exploit that invitation. The philosophy of promises must therefore account for the relation between a socially intelligible act of undertaking and its variable practical effects. Austin's framework supports that investigation by refusing to infer the type of act simply from its consequences.

The Fate of the Performative-Constative Distinction

Austin ultimately finds his initial contrast between performatives and constatives inadequate as an exhaustive classification. Statements themselves are actions, and assertions have conditions of appropriate performance, truthfulness, authority, and accountability. A promise can imply beliefs or predictions, while an assertion can commit a speaker to standards of evidence. Consequently, the mature theory investigates the illocutionary force of utterances generally rather than dividing all sentences into two incompatible kinds (Austin, 1962/1975, Lectures XI–XII).

The result matters for the philosophical interpretation of truth and normativity. A performative can have propositional content that is true or false in an embedded clause, while the act of undertaking it is assessed according to additional standards. For instance, the promise "I promise that I will return" involves a future-directed content without being reducible to a present report of an intention. Conversely, the phrase "I hereby state" can make explicit the force of assertion. A semantics that deals with propositional truth alone cannot specify what commitments speakers undertake by using propositions.

The mature Austinian position permits more fruitful comparisons between assertion and promising: both involve answerability, but the kinds of answerability differ. Assertion primarily subjects a speaker to norms concerning whether what is said is true and warranted; promising subjects the speaker to standards of future performance. The vocabulary of performativity remains valuable when understood as a dimension of use and force rather than an isolated list of special words.

Rebecca Kukla and Mark Lance's 'Yo!' and 'Lo!': The Pragmatic Topography of the Space of Reasons (2009) broadens this point by treating pragmatic speech acts as constitutive participants in the space of reasons. Their account challenges the assumption that declarative assertion is the privileged norm-bearing form of linguistic activity. The relevance to promising is direct: the speaker who undertakes an obligation performs a recognizable normative move whose significance depends on the roles of speaker, addressee, and shared conversational position. Their pragmatics thus helps explain why promising cannot be reduced to the semantics of a future-tense proposition (Kukla & Lance, 2009).

Searle's Conditions for Promising

John R. Searle's Speech Acts: An Essay in the Philosophy of Language (1969), chapter 3, pages 54–71, offers a systematic analysis of promising as a paradigm illocutionary act. The act has propositional-content, preparatory, sincerity, and essential conditions. In the standard case the promised action belongs to the speaker's future conduct. The hearer has an interest in its performance, and it is not already an obvious matter that the speaker would perform it regardless of the undertaking. The sincerity condition concerns the speaker's intention to do the action. The essential condition is that the utterance count as the undertaking of an obligation. Searle, chapter 3, Cambridge University Press.

Searle's nine-condition analysis is an idealization of successful, nondefective promising. The conditions should not be read as claiming that every departure from the ideal destroys the illocutionary act. A person can promise insincerely, and speech can be effective without satisfying the full set of conditions for a paradigmatic case. This distinction preserves the explanatory significance of defective promises and the difference between illocutionary success and moral propriety.

The theoretical advantage of Searle's reconstruction lies in making explicit the different jobs performed by the conditions. Propositional-content conditions delimit what is undertaken; preparatory conditions explain its practical point; sincerity identifies the characteristic psychological state; and the essential condition identifies the kind of act. The principal difficulty is that the essential condition already describes promising as undertaking an obligation. It clarifies the act's structure while leaving open the further question why the recognized act should generate a moral obligation. That second question is not resolved merely by repeating the definition.

Constitutive Rules, Institutional Facts, and Deontic Force

Searle distinguishes constitutive rules, which make certain activities possible, from regulative rules, which guide activities already intelligible independently of those rules. A rule of the form "X counts as Y in context C" describes how a recognizable event acquires a status in an institutional practice. When a properly situated utterance counts as promising, its force depends on a structure of mutual recognition and rule-governed significance. The convention does more than attach a label: it identifies what participants are doing and what can reasonably be demanded afterward (Searle, 1969, chapters 2–3).

The analogy with games is illuminating but requires restraint. Rules of chess constitute what moving a bishop is; rules of a legal institution establish the status of certain documents and declarations. Everyday promises are less centralized than either practice, and a theory must explain why participants can recognize undertakings without a formal authority establishing each case. Some promising occurs in informal relationships whose expectations are negotiable. Constitutive structure can be distributed across shared linguistic dispositions and moral understandings rather than embodied in a written procedural code.

There is a deeper philosophical issue: a constitutive rule explaining why an act counts as a promise need not independently justify the moral obligation to keep it. Institutional status and moral validity are distinguishable. An illegitimate institution may assign a status while failing to generate a justified moral demand. Conversely, a binding assurance may emerge where no explicit institutional formula is available. The explanatory task is to connect status assignment, reciprocal agency, and reasons for action without assuming the desired bridge as a premise.

The Derivation of Obligation from the Fact of a Promise

In "How to Derive 'Ought' from 'Is'" (1964), Searle uses promising to challenge a simplistic prohibition on deriving any normative conclusion from apparently descriptive premises. In schematic form, an agent utters an appropriate promissory formula under appropriate conditions; therefore the agent promised; promising involves undertaking an obligation; therefore the agent is obligated, at least prima facie, to perform. The controversial inference lies in the account of what is already included in the institutional fact of promising (Searle, The Philosophical Review 73(1), 43–58).

The objection is that the supposedly descriptive premise already presupposes a normative practice. To say that a promise was made is to classify an event under rules involving obligations; it is not to describe bare acoustic events and psychological states. The argument demonstrates that socially constituted facts can have normative entailments relative to the constitutive rules. It does not establish that a wholly norm-free description entails an independent moral command. R. I. Sikora made this distinction in his discussion of the derivation, accepting the relation between promise and obligation while questioning the fact-only premise (Sikora, 1975).

This dispute shows why two senses of "ought" must be considered. One concerns the internal standards of undertaking a promissory role. Another concerns the moral reasons an agent ultimately has in view of all relevant facts, including harm, prior duties, invalidating coercion, and release. The second may qualify or defeat a prima facie obligation even when the first is established. Philosophical progress comes from explaining the relationship between these normative levels rather than declaring the gap between facts and values closed by definition.

Speaker Meaning, Recognition, and Communicative Intentions

Paul Grice's theory of speaker meaning makes intentions and their recognition central to communicative acts. In "Meaning" (1957), he examines an agent's intention to produce an effect through an audience's recognition of that very intention. Although promising is not simply the production of a belief, Grice's model provides a powerful tool for explaining how hearers identify what speakers are doing beyond what sentence form alone specifies. Later work develops distinctions among speaker meaning, sentence meaning, and implication (SEP, Paul Grice).

Applied to promises, intention-based accounts explain the importance of communicative transparency. When someone says "I'll handle it" in response to a request for help, the utterance may be recognized as an undertaking because the social context makes an offer of responsibility the salient intention. In other circumstances identical words communicate a plan or prediction. A speaker's intention to be recognized as binding herself is therefore relevant, but cannot by itself establish a moral right in the addressee. A private intention unmanifested in any communicatively accessible way is insufficient for an ordinary interpersonal promise.

Bach and Harnish's Linguistic Communication and Speech Acts (1979) develops an intention-centered pragmatic account within a systematic analysis of communicative acts. Their approach illuminates how a hearer reasons from an utterance, contextual assumptions, and conversational expectations to the speaker's intended illocutionary force. The limitation shared by purely intention-based analyses is normative: successful recognition explains how the act is understood, while the authority of the undertaking needs a distinct account of interpersonal reasons and normative practices. MIT Press bibliographic record.

Explicit, Implicit, Conditional, and Indirect Promises

A promise need not contain the word "promise." An explicit performative announces the act by means of a verb. An implicit promise can be made through a contextually recognized assurance. A conditional promise binds performance to a specified condition, as in an undertaking to provide assistance if an event occurs. An indirect promise can be conveyed through a formula whose literal sentence type conventionally performs another function. These variations reveal that the action is constituted through the interaction of expression, communicative purpose, and normative context.

Conditionality requires careful analysis. "If the committee approves the grant, I will provide the equipment" can operate as a prediction, statement of policy, or conditional undertaking. If it is a promise, the condition specifies when performance is due; it does not necessarily suspend the creation of the undertaking until the condition occurs. The addressee may acquire a present conditional claim. Whether an agent is allowed to prevent the condition from occurring in order to avoid performance is a further question that depends on what the promise reasonably commits her not to do.

Indirect promises create difficult boundary cases. "You can count on me" may place strong assurance behind an undertaking while leaving the specific content implicit. A valid commitment needs an interpretation sufficiently determinate for accountability; otherwise the hearer could assign arbitrary duties to a vaguely reassuring expression. Contextual interpretation is therefore part of the act's normative limits. The task of moral analysis is to establish which reasonable expectations the speaker authorized, not to maximize obligations from every ambiguous phrase.

Uptake, Address, Acceptance, and the Promisee

An illocutionary act is performed in a communicative situation. Austin emphasizes understanding and uptake in the ordinary success of many acts; a hearer who cannot identify the force may never become the appropriate participant in the undertaking. A promise whispered unheard to an absent person raises a different question from a promise addressed by writing and read later. In the latter case a publicly manifested undertaking can be recognized asynchronously, while the exact moment at which the directed relation takes effect may depend on the relevant practice.

The distinction between uptake and acceptance is essential. A hearer can understand that a promise has been made without welcoming its terms. Some promises may impose a morally important intrusion: an unwanted promise of help can compromise the recipient's autonomy if acceptance is assumed. A purely unilateral theory risks treating the addressee as a passive beneficiary. An entirely acceptance-dependent theory, however, risks eliminating commonplace promises whose recipients have clearly understood the undertaking but have not yet explicitly responded. The correct model must distinguish recognition of force, willingness to rely, and consent to a proposed alteration of rights.

A directed duty differs from a free-floating duty because someone stands in a characteristic relation to its performance. That party may have the authority to waive performance, complain of breach, or insist that the obligation be discharged in a particular way. The presence of a promisee also raises questions about third-party beneficiaries and promises made publicly to a group. The speaker's communication must establish whom she is accountable to, even if the beneficiaries of fulfillment extend beyond that party.

Conventional Practices and the Rawlsian Account

John Rawls's "Two Concepts of Rules" (1955) distinguishes the justification of a social practice from the justification of a particular act within it. His discussion of promises argues that the usefulness of having a reliable practice of undertaking can justify its rules, while the justification of a particular promise-keeping action should be understood through the rule-governed practice itself. To break an individual promise whenever doing so yields locally better consequences would undermine the point of having the practice whose benefits motivated the justification (Rawls, 1955, The Philosophical Review 64(1), 3–32).

The insight is especially powerful against direct act-utilitarian reasoning. The value of promising is not secured if each promisor may treat her commitment as merely one adjustable factor in a private calculation. Recipients need assurance that particular undertakings have standing across ordinary variations in advantage. Yet a practice-based explanation must specify which practice counts as the relevant one, how its rules are justified, and why specific recipients possess claims rather than simply being beneficiaries of a useful convention.

Rawls's distinction avoids a simplistic inference from social benefit to individual obligation. General justification can authorize rules that subsequently structure the reasons for particular action. It nevertheless leaves open whether a promise can morally bind outside a familiar social practice and whether the moral force of promising is fully dependent on a convention. Those questions drive the comparison with Scanlon's account of expectations and with theories of normative power.

Expectations, Assurance, and Scanlon's Moral Principle

T. M. Scanlon's What We Owe to Each Other (1998), chapter 7, "Promises," treats the duty to keep a promise as a special case of broader principles governing expectations that agents intentionally induce. The moral wrong of breaking an undertaking can lie in deliberately leading another person to form an expectation of performance under circumstances in which that expectation matters to the person. Scanlon's contractualism evaluates principles by whether persons could reasonably reject them. The resulting analysis places the promisee's standing within a broader morality of assurance, reliance, and the wrong of misleading others. Scanlon, chapter 7 bibliographic record.

The expectation account possesses a major advantage: it can explain why certain assurances bind even in the absence of explicit ritual or established promise language. If one deliberately invites another to organize her plans on the assurance of assistance, the moral reason for keeping it may already be powerful. It also explains why deception within promises is especially serious: the promisee is encouraged to depend on the speaker's undertaken course.

A standard challenge concerns promises that create rights without inducing actual reliance. A recipient may prefer the promisor to fulfill a promise while remaining emotionally and practically prepared for breach. The fact that the recipient planned nothing on the promise does not obviously destroy her complaint against its violation. The normative-power approach takes this challenge as evidence that promising is more than the management of expectations. Scanlon's theory can answer that the wrong concerns reasonable expectations knowingly induced, rather than demonstrable economic loss. The central dispute is whether that answer fully captures the promisee's authority over performance and release.

Special Rights and Normative Powers

H. L. A. Hart's "Are There Any Natural Rights?" (1955) characterizes promises as transactions that create special rights and corresponding obligations between particular parties. Unlike general duties owed to everyone, promissory obligations are directed: one person's voluntary undertaking gives another a standing that did not previously exist. Joseph Raz subsequently develops the idea of a normative power, a recognized capacity to change obligations and rights through an intentional act. Such powers make cooperative projects and self-directed life planning possible. The SEP analysis of normative-power theories situates Hart and Raz in this tradition.

A promisor's act can be represented through a transition: prior to undertaking, there may be no directed duty concerning action A; after a valid undertaking, the promisor owes A to the promisee under the terms of the promise. This is an alteration in normative relations, not necessarily in physical circumstances. The essential explanatory question is why agents possess the authority to make such alterations. Appeal to autonomy is one answer: people should have practical power to bind themselves for the sake of projects they share with others. But unrestricted self-binding would also permit undertakings that harm third parties or contradict overriding duties.

A mature theory therefore treats normative powers as conditioned. Their scope depends on the moral standing of parties, the subject matter, and the kind of authority being transferred. The ability to grant another a claim is not equivalent to the unlimited ability to create any obligation whatsoever. The power perspective is particularly effective in explaining waiver and directed duties, while relying on an additional account of what legitimizes that power.

Authority, Waiver, and Ownership of the Promise

David Owens argues that the interest served by promising includes an authority interest: a person may value the standing to decide whether another is obliged to do something, independently of predicting or controlling the other's actual behavior. In Shaping the Normative Landscape (2012), chapter 6, "The Promissory Interest," he develops this account against reductions of promissory authority to mere coordination. Oxford University Press, chapter 6.

Consider a promise to attend a friend's event. The friend might already expect the attendance and have incurred no costs. A breach can still be a wrong to that friend because a specific directed claim was created. The friend might later release the promisor, thereby changing what is owed. Neither actual reliance nor the magnitude of disappointment fully explains this authority to release. The control belongs to the normative relation, not simply to causal consequences.

Owens's approach illuminates the difference between apology, compensation, performance, and release. These responses do different things: compensation addresses loss; apology acknowledges wrongdoing; performance satisfies the undertaking; release changes the standing requirement to perform. The challenge is to establish the boundaries of the promisee's authority. If a promisee demands a harmful action, the demand does not automatically override independent moral constraints. The promissory authority is substantial, directed, and defeasible within a larger moral order.

Joint Commitment and Collective Promising

Promises often occur inside collaborative agency. Margaret Gilbert's account of joint commitment investigates how parties form shared obligations through public manifestations of willingness to be jointly committed. Her discussion "Three Dogmas About Promising" challenges assumptions that every promissory obligation is merely an application of a general moral law or can be understood without reference to directedness (Gilbert, 2013, pp. 296–323).

The difference between two parallel individual intentions and a joint undertaking is structurally important. Two people may independently plan to meet without promising one another anything. They may also form a shared commitment in which each has standing to demand coordination and justification. In the second case the relation is not reducible to the prediction that two independent plans will converge. Participants recognize a shared practical object and assume reciprocal accountability for its execution.

Collective promises also expose questions of representation. When an organization pledges to perform, who has the authority to speak for it, what role does institutional acceptance play, and how is the duty transmitted through changes in personnel? A spokesperson may utter the relevant words, but the organization can be the bearer of the undertaking if the institution's rules confer such authority. Conversely, an unauthorized spokesperson's assurance might produce individual duties without binding the group. The question is one of attributed agency and authorized force, not merely how many people hear the sentence.

Sincerity, Deception, and Promissory Trust

The morality of promising must preserve the distinction between a commitment's existence and the speaker's sincere intention to fulfill it. Suppose a person promises to deliver a needed object while secretly deciding to keep it. Under ordinary circumstances the utterance can still perform the promise. This makes the person answerable both for the original deceit and for the subsequent breach. If sincere intention were constitutive of the duty's very existence, the deceiver could argue that the words never bound her. Austin's analysis of abuses resists that consequence (Austin, 1962/1975, Lecture I).

Trust is also more complex than optimistic expectation. One can trust a promisor while acknowledging uncertainty; one can rely on a mechanical system without regarding it as trustworthy in the interpersonal sense. Promissory trust ordinarily involves treating another as an agent who recognizes a claim and who can be asked to explain failure. The recipient's confidence may be modest, yet the relationship can still make a moral demand.

There are different wrongs when promises are broken. A breach may cause avoidable harm; it may violate a particular person's legitimate authority; it may exploit trust; it may express disregard; or it may destabilize a practice of assurance. These wrongs can coincide without being conceptually identical. Their separation matters in cases of harmless breaches, failed reliance, and justified departures from an undertaking.

Conflict, Impossibility, Coercion, and Release

Promises confront changing circumstances. A promised action may become physically impossible, prohibited, gravely harmful, or inconsistent with a more demanding duty. In such cases the speaker's obligation to perform must be examined separately from duties to notify, explain, mitigate harm, apologize, or offer appropriate substitute performance. The failure of literal performance does not automatically establish moral wrongdoing, and an excused breach need not eliminate every residual duty created through the earlier undertaking.

Coercion raises a distinct problem about authorization. A promise extracted through credible threats may be intelligibly uttered and understood while failing to generate the ordinary moral right that a free undertaking would confer. The analysis requires specifying which threats undermine valid exercise of a normative power and which pressures belong to unavoidable conditions of social choice. A generic demand for perfectly unpressured consent would invalidate most ordinary human arrangements; a complete disregard for coercion would allow domination to masquerade as a voluntary obligation.

Release provides the positive counterpart to invalidity. If a promisee has standing to waive performance, the promissory relation has an interpersonal structure different from a general moral command. Yet a promisee cannot always release duties owed also to third parties. A surgeon's undertaking to a patient, for instance, may coexist with professional or public duties not exhausted by the patient's private waiver. The philosophy of promising must differentiate directed promissory duties from independently operative moral and institutional requirements.

Promises, Contracts, Oaths, Threats, and Declarations

A promise is a commissive act: it presents the speaker as undertaking a course of future conduct. A threat also concerns future behavior, but typically presents that behavior as adverse to the addressee and as conditional leverage. The difference depends on communicative purpose, recipient interest, and context; a seemingly benevolent promise can be coercive, and a threatened action may be desirable under peculiar circumstances. The labels should follow the actual action performed, not an isolated verb.

Contracts involve one or more undertakings situated within a legal order. Some moral promises never become legally enforceable; some enforceable obligations arise through procedures and doctrines that exceed ordinary promising. Legal consideration, formalities, capacity, remedies, and jurisdiction may regulate enforceability without determining whether the speaker made an intelligible moral promise. An oath invokes additional institutions of solemn assurance or attestation; a vow can address a deity or a comprehensive life commitment; a declaration, such as an authorized appointment, alters status by a different kind of institutional power.

These neighboring acts demonstrate why performativity is a family resemblance across heterogeneous practices. Every performative utterance must be classified according to its force and authority conditions. The same act may occupy several descriptions: an oath can also be a promise, and a contract can include declarations and commitments. The appropriate distinction concerns what normative change each act effects, who is entitled to perform it, and who acquires standing afterward.

Iterability, Political Performativity, and Social Power

Jacques Derrida's "Signature Event Context" (1972; English translation in Margins of Philosophy, 1982) challenges any account that makes successful force fully dependent on a closed, determinate original context or a speaker's self-present intention. Signs can be repeated in new circumstances; a performative formula's recognizability depends partly on its citability. This iterability makes institutional repetition possible while also opening the possibility of altered, contested, or appropriated force. Stanford-hosted text of "Signature Event Context".

Judith Butler's Excitable Speech: A Politics of the Performative (1997) explores speech, injury, institutional authority, and political contestation. The language of performativity extends here beyond ordinary individual promising to consider how words participate in reproducing social relations and how agency can emerge through the contested repetition of established terms. The relevance to promising is methodological: no theory of felicitous speech can ignore unequal authority, public recognition, and the histories through which some speakers gain or lose standing (Butler, 1997, Routledge bibliographic record).

Derrida and Butler should not be assimilated to conventional accounts of promissory obligation. Their central concerns include writing, iterability, discourse, vulnerability, and power. Their contribution is to identify limits of an idealized two-person exchange in which the parties possess equal authority and unproblematic communicative access. Public promises, especially political promises and declarations of institutional intent, can be repeated, quoted, contested, and attributed through systems in which performative force is unevenly distributed.

Temporal Agency and the Politics of Promising

Hannah Arendt identifies promising and forgiveness as important capacities of action in The Human Condition (1958), chapter V, sections 33–34. Promising addresses the unpredictability inherent in plural human action by establishing limited islands of reliability within a future that no agent can fully control. Forgiveness responds to the irreversibility of actions already performed. This pairing places promising within a philosophy of political plurality: to act among others is to enter consequences that cannot be wholly mastered by an individual will. University of Chicago Press book and contents.

Arendt's account helps explain why promising has value beyond private exchange. Political association requires participants to undertake and sustain joint projects amid uncertainty and changing interests. The ability to commit is a source of shared action precisely because the future cannot be secured by complete prediction. But promises that purport to determine every aspect of the future can become instruments of domination or rigidity. The practice's virtue includes the capacity to limit uncertainty without claiming comprehensive control.

This perspective intersects with Nietzsche's genealogy without duplicating it. Nietzsche asks how a durable promising agent becomes possible; Arendt asks how promise-making supports a plural political world. Both accounts locate promises in temporal continuity, while giving different emphases to self-mastery, plurality, and political freedom. The illocutionary act joins these dimensions by making a future-oriented relation publicly accountable at a definite moment.

Digital Communication and Artificial Agents

Digital media separate utterance, transmission, receipt, and later recognition. A promise can be written, scheduled, archived, digitally signed, or delivered through a messaging service. The philosophical questions remain familiar but acquire new operational details: what identifies the promisor, what constitutes authorized transmission, when is the act received, which edits alter the undertaking, and what evidence establishes its terms? Durable digital records may improve proof of an utterance while leaving its normative validity dependent on context, authority, and interpretation.

Artificially generated promissory language introduces a further distinction between the apparent first-person form of an utterance and the agent or institution actually authorized to undertake its consequences. A software interface can display "I promise to issue a refund." The relevant questions include whether the system is authorized to bind the operator, whether users reasonably understand the undertaking as institutional, and whether a responsible party can be identified for fulfillment, waiver, and breach. Linguistic fluency alone cannot settle these matters. Conversely, an institution can take responsibility for an automated utterance when its rules and conduct confer appropriate authority.

The Artificial Era makes the general philosophical problem sharper: the study of promising cannot assume that all relevant utterances originate in the same kind of biological speaker or that the surface grammar of personhood resolves accountability. A substantive theory of nonbiological agency would require its own independently articulated ontology and account of normative standing. This monograph establishes the conditional question rather than attributing unverified capacities or liabilities to specific contemporary systems. Its argument is continuous with the classical premise that meaningful commitments require determinable parties, authority, and procedures for accountability.

Comparative Argument Reconstruction

Four arguments organize contemporary disagreement. The conventionalist argument begins from the unintelligibility of promissory force outside social practices, identifies conventions that assign the status of undertaking to recognizable acts, and concludes that promising is constituted by such practices. Its strength is explanatory economy: it can explain shared recognition. Its weakness is the further step from a convention's existence and usefulness to a particular person's morally justified claim. Hume and the practice-centered Rawlsian approach mark different stages of this argument.

The expectation argument begins from the moral importance of intentionally induced reliance, identifies a promissory act as an especially clear form of assurance, and concludes that breach violates duties not to manipulate another's reasonable expectations. Its strength is the explanation of why reliance and deception matter. Its weakness appears in cases where a promisee has made no consequential plans, yet retains a distinctive right to complain. Scanlon offers a sophisticated version built around principles no one could reasonably reject, not a simple calculation of actual economic loss.

The normative-power argument begins from the importance of voluntary normative control, treats promising as a permitted way of changing directed rights and duties, and concludes that a valid undertaking creates a claim controlled in relevant respects by the promisee. Its strength is its account of directed obligation, waiver, and harmless breach. Its explanatory challenge is why this power exists and what constrains it. Hart, Raz, and Owens provide related but differing answers grounded in special rights, autonomy, and authority interests.

The illocutionary argument begins from the observation that uttering a sentence may be an action, identifies the constitutive and communicative conditions under which the action is promising, and concludes that a promising act is a publicly intelligible undertaking. Austin and Searle disclose essential structures of this account, but the derivation of a fully justified moral duty remains a distinct issue. The strongest philosophical synthesis therefore refuses to make any one explanatory level do the work of all the others.

A representative test case clarifies the differences. Suppose a researcher promises a colleague to send a manuscript, the colleague understands but does not rely, and the researcher later fails to send it despite having no serious excuse. An illocutionary theory can identify the original undertaking. A conventionalist can explain the shared practice. An expectations theorist must specify the morally relevant assurance even without material reliance. A normative-power theorist can explain the colleague's distinctive standing to complain or release. The case does not compel one foundational theory, but it specifies what each must explain.

Open Philosophical Problems

The central unresolved question concerns moral generation: what justifies the move from a publicly recognized undertaking to a directed duty that survives ordinary changes in advantage? A satisfactory account must explain why promises bind without treating every social convention as morally authoritative. It must also explain how promisees can possess legitimate power to release others from duties, while recognizing duties that cannot be privately waived.

A second question concerns the boundary between promising and assurance. Some statements of intention invite reliance without granting the recipient control over performance. Some formal promises are made in circumstances in which reliance is impossible. Identifying whether these cases share one core moral relation requires further work on practical authority, communicative recognition, and the normative role of explicit uptake.

A third question concerns interpretation under uncertainty. A promise often leaves implicit what counts as adequate performance, what resources the promisor is expected to expend, and what circumstances justify alteration. A complete theory must explain how common ground, reasonable expectations, the promisee's understanding, and background moral norms jointly determine content without turning the promisor's undertaking into an unlimited blank check.

Finally, collective and artificial communication raises the problem of attributed agency. Which bodies can undertake directed obligations? What qualifies as authorization when utterances are generated or transmitted automatically? How are promises maintained across institutional change? These are questions about continuity of accountability and the ontology of communicative agents, rather than simply about the words displayed by an interface.

Philosophical Synthesis

The research supports a layered conception of promising. At the semantic level, an utterance carries content specifying an action or condition. At the pragmatic level, it communicates an undertaking rather than a prediction. At the institutional level, recognizable practices or authorized procedures make that undertaking publicly intelligible. At the normative level, the undertaking places an agent in a directed relation to another party. At the moral level, the obligation must be assessed under principles concerning authority, respect, fairness, deception, and the justified limits of self-binding. These layers are interconnected; their distinction makes their interaction intelligible.

The resulting general thesis is that promising is an exercise of publicly intelligible practical self-commitment that proposes and, where appropriately authorized, establishes a directed normative relation concerning future conduct. Publicity supplies recognizability; uptake situates the act in communication; practical agency supplies the capacity to undertake a course of action; normative authority supplies the conditions under which the undertaking can bind. A promise's moral importance emerges from its role in enabling one agent to take another's word as a reason for planning, demanding accountability, and participating in shared projects.

The synthesis yields three consequential distinctions. First, a performative can be successfully performed yet insincere: deception and genuine liability are compatible. Second, an undertaking can be institutionally valid while its demanded performance is morally defeated: conditional or defeasible obligation requires its own analysis. Third, the promisee's directed standing explains why breach may wrong someone even absent measurable harm. Each distinction is grounded in phenomena the historical theories illuminate but no single theory exhausts.

The history of this problem also exhibits a substantive development in philosophical attention. Cicero and Aquinas investigated fidelity and binding commitments; Hobbes and Hume analyzed covenant and convention; Kant tested the morality of false promising; Nietzsche asked how agents capable of fidelity are formed; Austin and Searle located the undertaking within communicative action; and contemporary theories explain its moral and institutional force. The development does not represent simple replacement. These traditions examine different explanatory dimensions of the same remarkable social capacity: to make one's present word a source of future accountability.

Conclusion

A promise is a present act through which an agent assumes future-directed answerability to another. The historical, linguistic, and moral questions surrounding promising are jointly necessary to explain that act. The history establishes that fidelity was recognized as a moral and political problem well before twentieth-century performative theory. Austin's analysis makes clear how words function as actions and why defective performances must be classified according to the circumstances of their utterance. Searle clarifies the structure of commissive force and constitutive rules. Hume, Rawls, and Scanlon disclose the importance of convention, social coordination, and assurance. Hart, Raz, Owens, and Gilbert reveal the directed rights and practical authority that promising can create. Nietzsche and Arendt explain its temporal and political significance.

The decisive philosophical conclusion is that the fact of a promise, its intelligibility, its normative force, and its moral justification require related but distinct arguments. A valid promise can be insincere, its performance can become impossible, and its ordinary demand can be overridden by serious competing obligations. These possibilities preserve the core meaning of promising: a freely undertaken relation in which future conduct becomes accountable to another under intelligible conditions. The ongoing work of philosophy is to specify and justify the powers that make such relations possible, especially when communications and institutions change the identity and continuity of those who speak.