Hubbry Logo
PresuppositionPresuppositionMain
Open search
Presupposition
Community hub
Presupposition
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Contribute something
Presupposition
Presupposition
Presupposition
View on Wikipedia
from Wikipedia

In linguistics and philosophy, a presupposition is an implicit assumption about the world or background belief relating to an utterance whose truth is taken for granted in discourse. Examples of presuppositions include:

  • Jane no longer writes fiction.
    • Presupposition: Jane once wrote fiction.
  • Have you stopped eating meat?
    • Presupposition: you had once eaten meat.
  • Have you talked to Hans?
    • Presupposition: Hans exists.

A presupposition is information that is linguistically presented as being mutually known or assumed by the speaker and addressee. This may be required for the utterance to be considered appropriate in context, but it is not uncommon for new information to be encoded in presuppositions without disrupting the flow of conversation (see accommodation below).[1] A presupposition remains mutually known by the speaker and addressee whether the utterance is placed in the form of an assertion, denial, or question, and can be associated with a specific lexical item or grammatical feature (presupposition trigger) in the utterance.

Crucially, negation of an expression does not change its presuppositions: I want to do it again and I don't want to do it again both presuppose that the subject has done it already one or more times; My wife is pregnant and My wife is not pregnant both presuppose that the subject has a wife. In this respect, presupposition is distinguished from entailment and implicature. For example, The president was assassinated entails that The president is dead, but if the expression is negated, the entailment is not necessarily true.

Negation of a sentence containing a presupposition

[edit]

If presuppositions of a sentence are not consistent with the actual state of affairs, then one of two approaches can be taken. Given the sentences My wife is pregnant and My wife is not pregnant when one has no wife, then either:

  1. Both the sentence and its negation are false; or
  2. Strawson's approach: Both "my wife is pregnant" and "my wife is not pregnant" use a wrong presupposition (i.e. that there exists a referent which can be described with the noun phrase my wife) and therefore can not be assigned truth values.

Bertrand Russell tries to solve this dilemma with two interpretations of the negated sentence:

  1. "There exists exactly one person, who is my wife and who is not pregnant"
  2. "There does not exist exactly one person, who is my wife and who is pregnant."

For the first phrase, Russell would claim that it is false, whereas the second would be true according to him.

Projection of presuppositions

[edit]

A presupposition of a part of an utterance is sometimes also a presupposition of the whole utterance, and sometimes not. For instance, the phrase my wife triggers the presupposition that I have a wife. The first sentence below carries that presupposition, even though the phrase occurs inside an embedded clause. In the second sentence, however, it does not. John might be mistaken about his belief that I have a wife, or he might be deliberately trying to misinform his audience, and this has an effect on the meaning of the second sentence, but, perhaps surprisingly, not on the first one.

  1. John thinks that my wife is beautiful.
  2. John said that my wife is beautiful.

Thus, this seems to be a property of the main verbs of the sentences, think and say, respectively. After work by Lauri Karttunen,[2][3] verbs that allow presuppositions to "pass up" to the whole sentence ("project") are called holes, and verbs that block such passing up, or projection of presuppositions are called plugs. Some linguistic environments are intermediate between plugs and holes: They block some presuppositions and allow others to project. These are called filters. An example of such an environment are indicative conditionals ("If-then" clauses). A conditional sentence contains an antecedent and a consequent. The antecedent is the part preceded by the word "if," and the consequent is the part that is (or could be) preceded by "then." If the consequent contains a presupposition trigger, and the triggered presupposition is explicitly stated in the antecedent of the conditional, then the presupposition is blocked. Otherwise, it is allowed to project up to the entire conditional. Here is an example:

If I have a wife, then my wife is blonde.

Here, the presupposition (that I have a wife) triggered by the expression my wife is blocked, because it is stated in the antecedent of the conditional: That sentence doesn't imply that I have a wife. In the following example, it is not stated in the antecedent, so it is allowed to project, i.e. the sentence does imply that I have a wife.

If it's already 4am, then my wife is probably angry.

Hence, conditional sentences act as filters for presuppositions that are triggered by expressions in their consequent.

A significant amount of current work in semantics and pragmatics is devoted to a proper understanding of when and how presuppositions project.

Presupposition triggers

[edit]

A presupposition trigger is a lexical item or linguistic construction which is responsible for the presupposition, and thus "triggers" it.[4] The following is a selection of presuppositional triggers following Stephen C. Levinson's classic textbook on Pragmatics, which in turn draws on a list produced by Lauri Karttunen. As is customary, the presuppositional triggers themselves are italicized, and the symbol » stands for 'presupposes'.[5]

Definite descriptions

[edit]
Main article: Definite description

Definite descriptions are phrases of the form "the X" where X represents a noun phrase. The description is said to be proper when the phrase applies to exactly one object, and conversely, it is said to be improper when either there exist more than one potential referents, as in "the senator from Ohio", or none at all, as in "the king of France". In conventional speech, definite descriptions are implicitly assumed to be proper, hence such phrases trigger the presupposition that the referent is unique and existent.

  • John saw the man with two heads.
    »there exists a man with two heads.

Factive verbs

[edit]
See also: Epistemology § Truth

In Western epistemology, there is a tradition originating with Plato of defining knowledge as justified true belief. On this definition, for someone to know X, it is required that X be true. A linguistic question thus arises regarding the usage of such phrases: does a person who states "John knows X" implicitly claim the truth of X? Steven Pinker explored this question in a popular science format in a 2007 book on language and cognition, using a widely publicized example from a speech by a U.S. president.[6] A 2003 speech by George W. Bush included the line, "British Intelligence has learned that Saddam Hussein recently sought significant quantities of uranium from Africa."[7] Over the next few years, it became apparent that this intelligence lead was incorrect. But the way the speech was phrased, using a factive verb, implicitly framed the lead as truth rather than hypothesis. There is however a strong alternative view that the factivity thesis, the proposition that relational predicates having to do with knowledge, such as knows, learn, remembers, and realized, presuppose the factual truth of their object, is incorrect.[8]

  • Martha regrets drinking John's home brew.
    • Presupposition: Martha did in fact drink John's home brew.
  • Frankenstein was aware that Dracula was there.
    • Presupposition: Dracula was in fact there.
  • John realized that he was in debt.
    • Presupposition: John was in fact in debt.
  • It was odd how proud he was.
    • Presupposition: He was in fact proud.

Some further factive predicates: know; be sorry that; be proud that; be indifferent that; be glad that; be sad that.

Implicative verbs

[edit]
  • John managed to open the door.
    »John tried to open the door.
  • John forgot to lock the door.
    »John ought to have locked, or intended to lock, the door.

Some further implicative predicates: X happened to V»X didn't plan or intend to V; X avoided Ving»X was expected to, or usually did, or ought to V, etc.

Change of state or continuation of state verbs

[edit]

With these presupposition triggers, the current unfolding situation is considered presupposed information.[9]

  • John stopped teasing his wife.
    »John had been teasing his wife.
  • Joan began teasing her husband.
    »Joan hadn't been teasing her husband.

Some further change of state verbs: start; finish; carry on; cease; take (as in X took Y from Z » Y was at/in/with Z); leave; enter; come; go; arrive; etc.

Iteratives

[edit]

These types of triggers presuppose the existence of a previous state of affairs.[9]

  • The flying saucer came again.
    »The flying saucer came before.
  • You can't get gobstoppers anymore.
    »You once could get gobstoppers.
  • Carter returned to power.
    »Carter held power before.

Further iteratives: another time; to come back; restore; repeat; for the nth time.

Temporal clauses

[edit]

The situation explained in a clause that begins with a temporal clause constructor is typically considered backgrounded information.[9]

  • Before Strawson was even born, Frege noticed presuppositions.
    »Strawson was born.
  • While Chomsky was revolutionizing linguistics, the rest of social science was asleep.
    »Chomsky was revolutionizing linguistics.
  • Since Churchill died, we've lacked a leader.
    »Churchill died.

Further temporal clause constructors: after; during; whenever; as (as in As John was getting up, he slipped).

Cleft sentences

[edit]

Cleft sentence structures highlight particular aspects of a sentence and consider the surrounding information to be backgrounded knowledge. These sentences are typically not spoken to strangers, but rather to addressees who are aware of the ongoing situation.[9]

  • Cleft construction: It was Henry that kissed Rosie.
    »Someone kissed Rosie.
  • Pseudo-cleft construction: What John lost was his wallet.
    »John lost something.

Comparisons and contrasts

[edit]

Comparisons and contrasts may be marked by stress (or by other prosodic means), by particles like "too", or by comparatives constructions.

  • Marianne called Adolph a male chauvinist, and then HE insulted HER.
    »For Marianne to call Adolph a male chauvinist would be to insult him.
  • Carol is a better linguist than Barbara.
    »Barbara is a linguist.

Counterfactual conditionals

[edit]
  • If the notice had only said 'mine-field' in Welsh as well as in English, we would never have lost poor Llewellyn.
    »The notice didn't say 'mine-field' in Welsh.

Questions

[edit]

Questions often presuppose what the assertive part of the question presupposes, but interrogative parts might introduce further presuppositions. There are three different types of questions: yes/no questions, alternative questions and WH-questions.

  • Is there a professor of linguistics at MIT?
    »Either there is a professor of linguistics at MIT or there isn't.
  • Is Newcastle in England or in Australia?
    »Newcastle is in England or Newcastle is in Australia.
  • Who is the professor of linguistics at MIT?
    »Someone is the professor of linguistics at MIT.

Possessive case

[edit]
  • John's children are very noisy.
    »John has children.

Accommodation of presuppositions

[edit]

A presupposition of a sentence must normally be part of the common ground of the utterance context (the shared knowledge of the interlocutors) in order for the sentence to be felicitous. Sometimes, however, sentences may carry presuppositions that are not part of the common ground and nevertheless be felicitous. For example, I can, upon being introduced to someone, out of the blue explain that my wife is a dentist, this without my addressee having ever heard, or having any reason to believe that I have a wife. In order to be able to interpret my utterance, the addressee must assume that I have a wife. This process of an addressee assuming that a presupposition is true, even in the absence of explicit information that it is, is usually called presupposition accommodation. We have just seen that presupposition triggers like my wife (definite descriptions) allow for such accommodation. In "Presupposition and Anaphora: Remarks on the Formulation of the Projection Problem",[10] the philosopher Saul Kripke noted that some presupposition triggers do not seem to permit such accommodation. An example of that is the presupposition trigger too. This word triggers the presupposition that, roughly, something parallel to what is stated has happened. For example, if pronounced with emphasis on John, the following sentence triggers the presupposition that somebody other than John had dinner in New York last night.

John had dinner in New York last night, too.

But that presupposition, as stated, is completely trivial, given what we know about New York. Several million people had dinner in New York last night, and that in itself doesn't satisfy the presupposition of the sentence. What is needed for the sentence to be felicitous is really that somebody relevant to the interlocutors had dinner in New York last night, and that this has been mentioned in the previous discourse, or that this information can be recovered from it. Presupposition triggers that disallow accommodation are called anaphoric presupposition triggers.

Presupposition in critical discourse analysis

[edit]

Critical discourse analysis (CDA) is a broad study belonging to not one research category. It focuses on identifying presuppositions of an abstract nature from varying perspectives. CDA is considered critical, not only in the sense of being analytical, but also in the ideological sense.[11] Through the analysis of written texts and verbal speech, Teun A. van Dijk (2003) says CDA studies power imbalances existing in both the conversational and political spectrum.[11] With the purpose of first identifying and then tackling inequality in society, van Dijk describes CDA as a nonconformist piece of work.[11] One notable feature of ideological presuppositions researched in CDA is a concept termed synthetic personalisation[12]

Logical construct

[edit]

To describe a presupposition in the context of propositional calculus and truth-bearers, Belnap defines "A sentence is a presupposition of a question if the truth of the sentence is a necessary condition of the question's having some true answer." Then referring to the semantic theory of truth, interpretations are used to formulate a presupposition: "Every interpretation which makes the question truly answerable is an interpretation which makes the presupposed sentence true as well."

A sentence that expresses a presupposition in a question may be characterized as follows: the question has some true answer if and only if the sentence is true.[13]

See also

[edit]

References

[edit]

Further reading

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Presupposition

View on Grokipedia
from Grokipedia
Presupposition is a core concept in and denoting the background assumptions or propositions that a speaker implicitly conveys as taken-for-granted common ground in an , distinct from the asserted content. These assumptions arise from specific linguistic triggers, such as definite descriptions (e.g., "the king"), factive verbs (e.g., "realize"), change-of-state predicates (e.g., "stop"), and temporal clauses (e.g., "before"), which encode information presupposed to hold independently of the utterance's . A defining feature of presuppositions is their projection behavior: unlike entailments, which are canceled under negation or questioning, presuppositions typically "project" outward, surviving embedding in such contexts—for instance, "John regrets cheating on the exam" and "John does not regret cheating on the exam" both presuppose that John cheated. This projection problem—explaining how presuppositions of complex sentences derive from those of their parts—has driven much theoretical work, with early accounts by Frege and Strawson emphasizing semantic preconditions for truth-aptness, later refined through satisfaction theories treating presuppositions as context-update requirements. Debates persist over presupposition's nature, with semantic views linking it to conventional meaning and pragmatic perspectives attributing it to speaker intentions or contextual accommodation, where hearers inferentially add unstated assumptions to maintain coherence. Empirical studies, including psycholinguistic experiments, reveal that presuppositions facilitate efficient communication by packaging old or inferable information, though failures in projection (e.g., via "soft triggers" like implicature-like items) highlight contextual filtering mechanisms. These characteristics underscore presupposition's role in causal chains of , enabling speakers to build on shared causal realities without explicit assertion.

Fundamentals

Definition and Core Properties

A presupposition is a semantic relation whereby a sentence carries an implicit that must hold true in the for the to be appropriately used, rendering the sentence infelicitous if the presupposition fails. This is backgrounded, taken for granted by the speaker, and projected as a commitment regardless of the sentence's polarity or under certain operators. Presuppositions differ from the at-issue content of an assertion, which directly contributes to the truth conditions of the sentence, as they constitute preconditions for semantic evaluation rather than entailed consequences. A defining core property of presuppositions is projection, the phenomenon where the presupposed content "projects" outward from embedded contexts, surviving negation, , or modal embedding, unlike entailments which are blocked by such operators. For example, "Mary realized that she won the race" presupposes that Mary won the race, and this presupposition persists in "Mary did not realize that she won the race" or "Did Mary realize that she won the race?". This projection arises because presuppositions impose satisfaction conditions on the local of their trigger, ensuring the overall requires the presupposed to be accommodated or already mutually accepted. Another core property is accommodation, the dynamic process by which hearers infer and add an uncontroversial presupposition to the common ground without explicit prior assertion, allowing discourse to proceed smoothly even if the presupposition was not previously shared. Accommodation occurs globally, updating the discourse context, or locally within specific embeddings, but it is constrained by contextual relevance and the hearer's willingness to accept the addition as uncontroversial. Empirical studies in experimental pragmatics confirm that projection and accommodation exhibit graded robustness, influenced by factors like trigger strength and contextual cues, rather than being binary semantic necessities. Presuppositions also exhibit local satisfaction or triviality in embedded contexts: for the complex sentence to be defined, the presupposition must be entailed by the local in which the trigger appears, preventing undefinedness from propagating unless projection filters apply. This property underpins formal semantic analyses, such as trivalent logics where sentences with unmet presuppositions yield a third value (neither true nor false), highlighting presupposition failure as a distinct category from falsity. These properties collectively distinguish presuppositions as pragmatic-semantic constraints on felicity, rooted in the management of shared knowledge in .

Distinctions from Entailment, Assertion, and

Presuppositions are distinguished from semantic entailments primarily by their behavior under operators such as , questions, and conditionals, known as the projection problem. Whereas an entailment of a sentence S is a P such that the truth of S guarantees the truth of P, but the negation of S does not, presuppositions "project" through such operators, remaining intact regardless of the polarity or modality of the context. For instance, the sentence "Tina stopped smoking" entails that Tina does not smoke now but presupposes that she used to smoke; under , "Tina did not stop smoking" retains the presupposition of prior smoking while canceling the entailment about current non-smoking. This projection holds in questions ("Did Tina stop smoking?") and conditionals ("If Tina stopped smoking, she feels better"), where the presupposition persists, unlike entailments, which are scoped within the operator. In contrast to assertion, which constitutes the at-issue or main propositional content of an that the speaker directly commits to and invites challenge, presuppositions represent background assumptions taken for granted and not part of the primary claim. Assertions form the truth-conditional core subject to direct denial (e.g., "John lying" asserts the regret, which can be negated as "John does not lying"), whereas presuppositions like the fact of lying in the example are accommodated as common ground unless explicitly contested. This distinction aligns with formal semantic treatments where assertions update the discourse context additively, while presuppositions require prior satisfaction or accommodation to render the felicitous, often leading to gaps in truth-value (e.g., Strawsonian undefinedness) if unmet. Presuppositions differ from implicatures—pragmatic inferences arising from Gricean principles or contextual expectations—in their conventional encoding and resistance to cancellation. Implicatures, such as scalar ones (e.g., "Some students passed" implying not all), are derived inferentially and can be explicitly overridden without contradiction (e.g., "Some, in fact all, passed"), remaining outside strict truth conditions. Presuppositions, however, are triggered by specific linguistic elements (e.g., factive verbs like "" or definite descriptions) and persist as non-cancellable commitments, projecting as semantic prerequisites rather than optional inferences. This semantic status makes presuppositions less context-dependent and more entrenched than conversational implicatures, though some theories treat certain presuppositions as conventional implicatures due to their non-at-issue nature.

Historical and Philosophical Foundations

Origins in Frege, Russell, and Early Logic

Gottlob Frege introduced key elements of presupposition in his 1892 essay "Über Sinn und Bedeutung" ("On Sense and Reference"), linking it to the referential requirements of singular terms like proper names and definite descriptions. He maintained that such terms presuppose the of their referents; failure of reference results in the sentence lacking a truth-value, as it fails to express a complete thought with alethic status. For example, "Kepler died in misery" presupposes Kepler's , and without it, the sentence yields a truth-value gap rather than truth or falsity. Frege distinguished these presuppositions (Voraussetzungen) as conditions external to the sentence's sense yet essential for truth-aptness, contrasting with the idealized referential purity of his logical language in (1879), where such gaps were minimized. Bertrand Russell critiqued and reformulated this framework in "On Denoting" (1905), advancing the to eliminate presuppositional issues in favor of strict bivalence. Definite descriptions, such as "the present King of ," function as incomplete symbols unpacked into existential quantifiers asserting existence, uniqueness, and predication: there is exactly one entity satisfying the description, and it bears the attributed property. Thus, "The present King of is bald" is false in 1905 due to the absence of a unique referent, rendering the entire proposition false without invoking truth-value gaps or non-referring entities. Russell's analysis scoped descriptions logically to avoid Frege's semantic imperfections, aligning with his broader logicist program in (co-authored with , 1910–1913), which prioritized propositions analyzable into truth-functional components. This Frege-Russell divergence—presupposition as referential precondition versus quantificational assertion—originated modern logical treatments of semantic phenomena, highlighting tensions between natural language's apparent gaps and formal logic's demand for exhaustive truth-values. Frege's approach tolerated defects to preserve intuitive semantics, while Russell's enforced ontological parsimony, influencing early analytic debates on and structure without reliance on Meinongian subsistent entities.

Strawson's Semantic Presupposition and Key Developments

P. F. Strawson introduced the concept of semantic presupposition in his 1950 paper "On Referring," critiquing Bertrand Russell's 1905 analysis of definite descriptions as asserting existence and uniqueness within the proposition's truth conditions. Strawson maintained that such descriptions, like "the present King of France," presuppose the existence of a unique referent as a background condition for the sentence to be truth-evaluable, rather than entail it as part of the asserted content. Under this view, if the presupposition fails—such as when no King of France exists—the utterance suffers presupposition failure, rendering the statement neither true nor false, but instead defective in a semantic sense akin to a category mistake or referential gap. Strawson distinguished presupposition from entailment by noting its projection behavior: presuppositions persist under negation, questioning, or modal embedding, unlike asserted content. For instance, "The present King of France is not bald" still presupposes a unique King, whereas denying the assertion would not. He emphasized that presuppositions arise from the referring use of expressions in context, separating the semantics of sentences (truth-conditional meaning) from performative acts of reference, which require successful presuppositional satisfaction for felicitous assertion. This semantic approach treated presuppositions as definedness conditions, necessary for assigning truth values, echoing Fregean ideas of but applied pragmatically to ordinary language. In his 1952 book Introduction to Logical Theory, Strawson extended presupposition to formal logic, arguing that traditional syllogistic inferences presuppose non-empty subject terms for validity, resolving debates over existential import without reducing them to material implications. He proposed a partial logic system where statements with failed presuppositions lack truth values, allowing preservation of ordinary deductive patterns while accommodating presuppositional defects, as in subject-predicate forms assuming class non-emptiness. This development bridged descriptive and inference theory, critiquing Russellian quantification for overlooking such semantic prerequisites in logic. Strawson's 1959 Individuals: An Essay in Descriptive Metaphysics articulated two interconnected theories of presupposition: a narrower semantic one, where presuppositions are conditions for truth-valuable statements via referential identification, and a broader performative one, linking them to the assertive enterprise in , where speakers commit to background truths enabling communication. These theories connected presupposition to , positing that identifying references (e.g., to ) presuppose a structured conceptual scheme of individuals and spatiotemporal frameworks, essential for empirical assertions. This work influenced subsequent distinctions between semantic and pragmatic dimensions of presupposition, though Strawson prioritized semantic constancy over contextual variability. Subsequent refinements in Strawson's oeuvre, such as his 1964 reply to , defended presupposition against charges of ad hoc truth-value gaps by stressing its role in preserving referential success over bivalent logic. These ideas laid groundwork for later linguistic theories, including projection and accommodation, by establishing presupposition as a distinct semantic invariant across embeddings, distinct from Gricean implicatures.

Presupposition Triggers

Definite Descriptions and Referential Expressions

Definite descriptions, such as noun phrases introduced by the definite article "the," function as key presupposition triggers by implying the and of the entity they denote. For instance, the sentence "The king of is bald" presupposes that there exists exactly one king of France at the relevant time, regardless of whether the sentence is affirmed or negated. This presupposition holds because definite descriptions encode a referential intent that assumes a unique in the context or shared knowledge, distinguishing them from indefinite descriptions like "a king of France," which lack such commitments. Bertrand Russell's 1905 analyzed definite descriptions as quantificational expressions asserting existence and uniqueness as part of the sentence's truth conditions, rendering "The king of France is bald" false in 1905 due to the absence of a unique referent. In contrast, Peter Strawson argued in his 1950 paper "On Referring" that these are presuppositions rather than assertions; failure of the presupposition results in referential failure, leaving the sentence without a rather than false. Strawson's view aligns with empirical observations of linguistic use, where speakers often treat presupposition failure as infelicitous rather than straightforwardly false, influencing subsequent semantic theories that treat uniqueness as a contextual maximality condition rather than strict logical uniqueness. Referential expressions extend beyond definite descriptions to include proper names, demonstratives (e.g., "this" or "that"), pronouns, and possessives, each triggering presuppositions of and identifiability of their referents. For example, "John regrets on the exam" presupposes John's and the prior event of cheating, with "John" as a referential anchor assuming shared familiarity. These triggers operate by invoking discourse referents that must be resolved for the to be felicitous, often relying on anaphoric links or world knowledge; pronouns like "he" presuppose a salient antecedent in . Empirical studies in confirm that such expressions project presuppositions robustly under embedding, as in questions or modals, supporting their role in maintaining referential continuity across sentences.

Factive and Implicative Verbs

Factive verbs constitute a primary lexical class of presupposition triggers, presupposing the truth of their clausal complements regardless of embedding contexts such as negation or interrogation. Examples include "know," "realize," "regret," and "discover," where utterances like "John knows that the experiment succeeded" or its negated form "John does not know that the experiment succeeded" both assume the experiment's success as background truth. This projection property distinguishes factive presuppositions from entailments, as the complement's truth survives operators that reverse polarity; empirical studies confirm that comprehenders infer these presuppositions even when the complement conflicts with prior discourse, influencing downstream interpretation. Non-factive counterparts, such as "believe" or "think," lack this presuppositional force, entailing the complement only in affirmative contexts without projection under negation.
Factive VerbExample SentencePresupposed Content
KnowShe knows the is flawed.The is flawed.
RegretHe regrets ignoring the .He ignored the .
RealizeThey realized the model failed.The model failed.
DiscoverResearchers discovered the anomaly.The anomaly exists.
Implicative verbs, such as "manage," "fail," and "forget," differ by entailing the complement's truth (or falsity under ) while presupposing ancillary conditions like effort, difficulty, or prior states. For instance, "She managed to solve the equation" entails solution achievement and presupposes non-trivial effort or obstacle, a inference that projects under as in "She did not manage to solve the equation," which entails non-achievement but retains the difficulty presupposition. Karttunen's 1971 analysis posits that implicatives presuppose necessary conditions for the complement's realization, such as attempts for "fail" or ongoing activity for "forget," yielding asymmetric inferences across polarities unlike the symmetric projection of factives. These presuppositions arise from interacting with contextual assumptions, with experimental evidence showing they activate as soft triggers, accommodable but defeasible in .
Implicative VerbAffirmative InferenceNegative InferencePresupposed Condition
Manage toEntails (VP occurred).Entails (VP did not occur).Difficulty or non-triviality.
Fail toEntails (VP did not occur).Entails (VP occurred).Attempt or expectation of VP.
Forget toEntails omission (VP did not occur).Entails occurrence (VP occurred, then remembered).Prior intention or habit.
The distinction between factive and implicative verbs highlights presupposition's sensitivity to verb semantics: factives enforce complement truth as uncontroversial background, while implicatives layer conditional presuppositions atop entailments, enabling nuanced inference patterns in . Debates persist, with some analyses questioning strict factivity for all instances due to context-dependent cancellations, though cross-linguistic data affirm their robust triggering role.

Change-of-State Verbs and Iteratives

Change-of-state s, including stop, start, continue, and cease, trigger presuppositions by entailing a prior state that holds antecedent to the event time described by the verb. For example, "The committee stopped meeting" presupposes that the committee had been meeting previously, while asserting the termination of that activity. This presupposition targets the initial state rather than the final one, as the verb's semantics lexically encodes the change from an assumed prior condition. Such triggers exhibit projection behavior, surviving , questions, and other embeddings; thus, "The committee did not stop meeting" retains the presupposition of prior meetings. These verbs are typically categorized as "soft" presupposition triggers, permitting contextual cancellation or local accommodation in scenarios where the prior state is not independently entailed or expected. Empirical studies indicate variability in projection strength, influenced by contextual probability; for instance, interrogatives like "Have you stopped beating your wife?" may weaken the presupposition if the prior state lacks antecedent support. This contrasts with "hard" triggers like definite descriptions, where presuppositions resist cancellation more robustly. Iterative expressions, such as the adverb again, presuppose a previous instantiation of the event or state denoted by the host . In "John solved the puzzle again," the presupposition is that John had solved the puzzle at least once before, with the adverb indicating repetition. Projection holds under and modals, as evidenced by "John did not solve the puzzle again," which still implies an earlier solution. Related iteratives like too presuppose the predicate's applicability to an alternative entity, e.g., "Sue left too" assumes someone else left. Unlike many change-of-state verbs, iteratives function as weak triggers, where presuppositions are processed with lower obligatory force and can be ignored in neutral contexts without high processing costs. Psycholinguistic experiments show faster comprehension and reduced confidence in accommodating again's presupposition when unsupported, compared to stronger triggers like continue, which demand accommodation (e.g., β = 1.80, p < 0.001 in targeted studies). This distinction arises from iteratives' lesser dependency on the presupposed content for semantic coherence, allowing flexible interpretation across contexts.

Structural Triggers: Questions, Clefts, and Conditionals

Wh-questions in English and other languages trigger existential presuppositions concerning the queried variable. For example, "Who won the race?" presupposes that someone won the race, as the structure demands an answer identifying an existent winner. This presupposition holds regardless of or , such as in "Who didn't win the race?" which still assumes a winner exists. Such triggers stem from the form's requirement for a complete answer entailing the background . Yes/no questions and alternative questions similarly generate presuppositions tied to their propositional content. A yes/no question like "Did John stop ?" presupposes that John previously smoked, inheriting triggers from embedded elements like change-of-state verbs, while the interrogative frame projects this assumption. Alternative questions, such as "Is John in or in ?", presuppose that John is in one of the specified locations. These structural effects ensure the question's felicity depends on the presupposed background being accommodated or satisfied. Cleft constructions, including it-clefts and pseudo-clefts, systematically presuppose the content of their . In "It was John who broke the ," the presupposition is that someone broke the , with "John" as the focused, asserted element providing exhaustive identification. Pseudo-clefts operate analogously: "The one who broke the was John" presupposes an entity broke the . This between presupposed background and asserted focus distinguishes clefts as conventional triggers, observable across languages and robust under , as in "It wasn't John who broke the ," which retains the presupposition. Empirical studies confirm speakers these as backgrounded during comprehension. Conditional structures interact with presuppositions primarily through projection from embedded clauses, rather than introducing novel triggers inherent to the connective itself. Presuppositions in the antecedent, such as definite descriptions or factives, typically project to the entire conditional: "If has a king, the king of is bald" presupposes has a king. This "hole" behavior allows presuppositions to escape the hypothetical scope, unless filtered by contextual satisfaction or local accommodation. Counterfactual conditionals, like "If John had stopped smoking, he would be healthier," may additionally imply or presuppose the antecedent's falsity via conventional , though this is debated as projection versus separate . Theoretical accounts, including satisfaction theory, explain this by requiring presuppositions to hold in the conditional's local for felicity.

The Projection Problem

Behavior Under Negation and Embedding

Presuppositions exhibit a distinctive behavior under negation, projecting outward regardless of the operator's polarity. The sentence "John stopped smoking" presupposes that John previously smoked, and this presupposition persists in its negation: "John did not stop smoking" similarly assumes prior smoking. This projection under negation, termed the negation test, serves as a primary diagnostic for identifying presuppositions, distinguishing them from entailments—which reverse or fail under negation—and conversational implicatures, which may cancel. For instance, the entailment in "John stopped smoking" that he currently does not smoke does not hold in the negated form, whereas the presupposition survives intact. Under broader embedding contexts, such as modals, questions, and attitude verbs, presuppositions often project as "holes" that permit passage, though outcomes vary by operator type. itself functions as a hole, allowing unrestricted projection, as do modals like "possibly" in "It is possible that John regrets his decision," which inherits the presupposition that John made the decision. Questions embed presuppositions similarly: "Has John stopped smoking?" presupposes prior smoking, projecting to the entire utterance. Karttunen (1973) formalized this variability, classifying embeddings into holes (e.g., , modals), which transmit all constituent presuppositions; plugs (e.g., "say" or "realize" in certain uses), which block projection by absorbing presuppositions into the speaker's commitment, as in "John said he stopped smoking" not presupposing prior smoking if the report is indirect; and filters (e.g., conjunctions or conditionals), which conditionally suppress projection if prior context entails the presupposition. Filtering mechanisms highlight conditional projection in embeddings like implications: "If John is married, his wife is happy" does not globally presuppose John's marriage, as the antecedent entails the definite description's requirement, locally satisfying it without upward projection. In conjunctions, sequential filtering applies: "John has children and his children are asleep" projects the existence presupposition only if not entailed by the first conjunct. Plugs, conversely, prevent projection by treating embedded content as non-asserted, evident in factive plugs like "know" behaving as holes ("John knows he stopped smoking" projects prior smoking) versus non-factives like "say" as plugs. These patterns underscore the projection problem's complexity, where presuppositions evade scope typical of asserted content, often requiring contextual satisfaction for felicity.

The Family-of-Sentences Test

The family-of-sentences test serves as a diagnostic tool for detecting presuppositions within the projection problem by assessing whether an inference associated with a sentence persists when the sentence is embedded under operators such as , , conditionals, modals, and conjunctions. This persistence, known as projection, distinguishes presuppositions from entailments—which fail under —and from conventional implicatures or conversational implicatures, which may cancel in such contexts. The test originates from observations in presupposition theory, including Karttunen's 1973 classification of embedding operators as "holes" (which allow projection), "plugs" (which block it), and "filters" (which conditionally permit it), providing a framework to evaluate projective behavior systematically. To apply the test, linguists generate a "family" of related sentences from an original utterance and check if the target inference remains intact and taken for granted in each variant. For example, the factive verb "regret" in "Sarah regrets that she failed the exam" triggers the presupposition that Sarah failed the exam; this holds under negation ("Sarah does not regret that she failed the exam"), questioning ("Does Sarah regret that she failed the exam?"), conditional embedding ("If Sarah regrets that she failed the exam, she will study harder"), and modal embedding ("Sarah might regret that she failed the exam"), confirming its projective status. Similarly, change-of-state verbs like "stop" presuppose a prior state, as in "The rain stopped," which projects through "The rain did not stop" and "Did the rain stop?" but may interact with filters in conditionals like "If the rain stopped, the picnic can proceed." In the context of the projection problem, the test underscores the challenge of compositionally deriving presuppositions for complex sentences, as projection is not uniform: holes like and questions typically allow full projection, while filters such as conditionals may bind or attenuate presuppositions from antecedents depending on entailment relations between clauses. Empirical applications reveal variability; for instance, certain triggers like temporal clauses ("before") project less reliably under disjunction than under conjunction, prompting refinements to projection theories. Despite its utility, the test assumes idealized contexts and may not capture pragmatic accommodations or speaker intentions that modulate projection in natural , as noted in psycholinguistic studies where projection rates vary by type and embedding depth.

Theoretical Explanations: Filter Conditions and Satisfaction Theory

Karttunen (1973) classified embedding operators as holes, plugs, or filters based on their impact on presupposition projection, with filters permitting conditional blocking of presuppositions from embedded clauses under specific entailment conditions. Holes, such as and modals, allow all presuppositions from their complements to project to the whole sentence; plugs, like factive verbs (e.g., "realize"), block projection entirely; and filters, including conjunctions and conditionals, conditionally inherit presuppositions depending on the content of adjacent clauses. For conjunctions in sentences of the form "A and B," the presuppositions of B project unless entailed by the conjunction of the global context and A, thereby filtering them locally. Similarly, in conditionals "if A then B," presuppositions of B are filtered if entailed by A alone, as A provides the local context restricting projection. These filter conditions explain cases where presuppositions fail to project, such as in "John has children and his children are bald," where the presupposition of "his children are bald" does not project because it is entailed by "John has children." For disjunctions "A or B," filtering occurs if presuppositions of B are entailed by the global context alone or by the global context conjoined with the of A. Karttunen's framework, while effective for simple connectives, requires recursive application for nested embeddings and does not fully generalize to all operators without additional stipulations. Heim (1983) developed the satisfaction theory as a dynamic semantic account that derives Karttunen's filtering effects compositionally from the requirement that presuppositions be satisfied by the input at each update step. In this theory, sentences carry both an asserted context change potential (CCP), which updates the common ground with new information, and a presupposition, which demands that the input entails the presupposed content for the update to be defined. Projection arises because unsatisfied presuppositions must hold of the global , but filters like antecedents in conditionals provide a temporary local that can satisfy embedded presuppositions, preventing their global . Under satisfaction theory, a conditional "if A then B" is felicitous in c if c entails the presuppositions of A, and for every world compatible with c updated by A, that local context satisfies B's presuppositions; only presuppositions of B not thus locally satisfied project globally. This predicts the same filtering as Karttunen for conjunctions, treating "A and B" as sequential updates where B's presuppositions must be satisfied by c updated by A. Unlike Karttunen's stipulation-based rules, Heim's approach integrates presupposition handling into a general of discourse update, extending naturally to attitudes and modals via context-shifting. Empirical support includes its ability to handle "holes" as operators demanding satisfaction in the global context without local provision, though it faces challenges with non-monotonic embeddings requiring further refinements like satisfaction in all or some branches.

Accommodation Mechanisms

Global vs. Local Accommodation

In presupposition theory, accommodation refers to the process by which a hearer updates their mental representation of the discourse context to incorporate presupposed content that is not already entailed by the prior context. Global accommodation involves adding the presupposed information directly to the global context—the overall common ground shared by interlocutors—allowing it to project outward and become available for subsequent discourse. This mechanism ensures presuppositions behave as background assumptions that hold across the entire conversation, as formalized in Irene Heim's 1983 file change semantics, where the context is treated as a file of facts updated incrementally. Local accommodation, by contrast, entails inserting the presupposed content into a subordinate or local , such as the scope of an operator like a conditional antecedent, , or attitude verb, thereby restricting its projection to that subdomain without altering the global . For instance, in the sentence "If John has children, his son is bald," the presupposition triggered by "his son" (that John has a son) can be locally accommodated within the consequent, satisfied locally by the antecedent's hypothetical assumption rather than requiring global commitment to John's parenthood. Heim (1983) proposed this as a rescue strategy when global accommodation would lead to inconsistency with the existing , such as in cases where prior denies the presupposition. Theoretical accounts emphasize a default preference for global over local accommodation to minimize contextual changes and maintain coherence, as local insertions can lead to "holes" in projection behavior that deviate from standard presupposition patterns. This preference, termed the Principle of Global Accommodation (PGA), posits that hearers opt for global updates unless they generate infelicity, such as redundancy or contradiction; empirical support comes from judgments where local accommodation feels marked or effortful in isolation. However, local accommodation proves necessary for explaining non-projecting readings, as in embedded presuppositions under factive verbs within questions ("Did John stop beating his wife?"), where global commitment might overburden the context. The distinction influences formal models of presupposition projection: , building on Heim, computes satisfaction relative to contexts derived compositionally, with accommodation as an update rule applied at the lowest feasible level to satisfy presupposition requirements. Critics note that unrestricted accommodation risks overgeneration of readings, prompting constraints like satisfaction theory, which mandates presuppositions be entailed by the union of and global contexts. Cross-linguistic data, such as variable projection in languages with softer triggers, further tests these mechanisms, revealing that global defaults align with processing efficiency in incremental interpretation.

Empirical Constraints on Accommodation

Empirical studies indicate that presupposition accommodation incurs measurable cognitive costs, manifesting as prolonged reading times in self-paced word-by-word tasks compared to cases where presuppositions are satisfied by prior context. This process unfolds incrementally and immediately upon encountering the trigger, yet accommodated content proves harder to recall than satisfied presuppositions, suggesting limits tied to demands. A primary constraint arises from contextual plausibility: accommodation falters or becomes effortful when the presupposed content conflicts with established or world knowledge. In experiments using definite descriptions and additive particles like "too," implausible presuppositions elicited significantly longer reading times (e.g., 89 ms delays) and reduced acceptance rates (40% vs. 79% in controls) in stops-making-sense judgments, whereas plausible variants imposed no such penalty. These findings, drawn from 32-item sets with native English speakers via , support the view that hearers evaluate and potentially reject accommodations requiring implausible updates, rather than applying them reflexively. Neural evidence further delineates these limits through event-related potentials (ERPs), where accommodation triggers a biphasic N400-P600 response reflecting sequential linking and model updating. The N400 component, peaking around 400 ms with centro-parietal distribution, predominates for definite descriptions due to antecedent search demands, while the later P600 (500-1000 ms, posterior) signals integration costs more acutely for change-of-state verbs. Trigger type thus modulates processing load, constraining accommodation efficacy based on semantic retrieval and contextual fit requirements. Additional factors, such as presupposition surprisingness or , prompt resistance, as hearers weigh informational updates against prior beliefs. Collectively, these psycholinguistic results impose boundaries on accommodation, emphasizing its dependence on low-cost, compatible contexts over unconstrained global adjustment.

Empirical and Experimental Evidence

Psycholinguistic Processing Studies

Psycholinguistic studies on presupposition processing primarily employ online measures such as self-paced reading, eye-tracking, and event-related potentials (ERPs) to examine how presuppositions are computed during real-time comprehension. These methods reveal that presupposition triggers, such as definite descriptions or change-of-state verbs, elicit rapid inference generation, often within 200-300 milliseconds of encounter, supporting the view that presuppositions are distinct from asserted content in their immediate accessibility. For instance, self-paced reading experiments demonstrate slower reading times at regions following satisfied presuppositions compared to unsatisfied ones, indicating active verification against prior context. Evidence for presupposition projection emerges in embedding contexts like negation or questions, where processing effects persist despite embedding, as shown in eye-tracking studies using triggers like "again" or "stop." In visual world paradigms, listeners fixate on presupposed referents faster than alternatives, even under negation (e.g., "Sam didn't realize that the duck stopped"), suggesting projection overrides embedding operators during incremental parsing. Similarly, ERP data indicate N400-like modulations for mismatched presuppositions and P600 for accommodation efforts, with factive verbs like "know" generating presuppositions that influence subsequent semantic integration regardless of polarity. Accommodation of unsupported presuppositions incurs measurable costs, with self-paced reading and eye-tracking revealing increased regression rates and fixation durations when new information must be added to the model. A 2017 study found that accommodating presuppositions from triggers like "too" in novel contexts slows processing by 50-100 ms relative to entailed content, highlighting the effortful nature of global updates versus local assertions. Recent eye-tracking work () further links this to retrieval, where presupposition resolution cues discourse-level recall, potentially explaining variability in projection strength across soft versus hard triggers. These findings align with satisfaction theory, positing filter-like mechanisms that evaluate presuppositions early, though debates persist on whether processing is fully automatic or modulated by context predictability.

Cross-Linguistic and Developmental Findings

Children demonstrate sensitivity to presuppositions as early as age 3, with preschoolers reliably interpreting triggers like "too" as presupposing a prior event or state, as shown in tasks where they infer unmentioned alternatives from utterances such as "Anna ate some cookies, and Ben ate too." Similarly, 4-year-olds understand the presupposition of verbs, rejecting continuations that deny the embedded proposition in scenarios like "Sammy buying the balloon" implying the purchase occurred. For factive verbs like "know" and "realize," third graders (around age 8) exhibit adult-like intuitions in truth-value judgments, distinguishing them from nonfactives by treating the complement as presupposed background, though younger children (under 6) may base discriminations more on subjective certainty conveyed by the verb than strict presuppositional projection. Acquisition of presupposition projection under embedding develops gradually, with children under 5 often failing to filter presuppositions in questions or negations, treating them as asserted content rather than backgrounded; for instance, in experiments with definite descriptions, 4-year-olds accommodate presuppositions locally but struggle with global projection until age 7. Bilingual children apply principles like Maximize Presupposition (preferring presuppositional forms like "the" over "a" when both are possible) comparably to monolinguals by age 4, suggesting early cross-linguistic robustness in scalar implicature-related presuppositions. Thesis-level analyses of contextual interplay indicate that presuppositional content is partially innate but refined through exposure, with delays in atypical development (e.g., autism spectrum disorder) affecting triggers like additives ("also"), where comprehension lags until school age. Cross-linguistically, the projective behavior of presupposition triggers exhibits stability, with soft triggers (e.g., implicative verbs) and hard triggers (e.g., clefts) showing consistent backgrounding and survival under negation in languages like German and Chinese, as evidenced by experimental ratings of inference strength in embedded contexts. Additive particles like "too" or equivalents mandate obligatory insertion when their existential presupposition holds, a pattern replicated in experimental comparisons across Indo-European and non-Indo-European languages, supporting universal constraints on accommodation. However, variation exists: in Salishan languages such as St'át'imcets, conventional triggers like "also," "more," and "stop" fail to generate pragmatic presuppositions in the Stalnakerian sense, instead yielding at-issue entailments or no projection, challenging claims of universality for trigger-specific effects. Overall, presupposition research reveals minimal cross-linguistic divergence in core projection mechanisms, with heterogeneity among trigger types preserved, though pragmatic realization differs in polysynthetic or focus-marking languages where existential or change-of-state implications integrate as asserted rather than presupposed. Developmental trajectories align with this, as competence in trigger interpretation precedes full mastery of projection filters, informed by empirical tasks emphasizing verification over mere lexical acquisition.

Recent Advances in Projection Prediction

Recent experimental investigations have refined predictions about presupposition projection under connectives, highlighting the role of linear order. A 2024 study tested projection from presupposition triggers embedded in conjunctions ('and') versus disjunctions ('or'), finding that presuppositions in initial positions project more consistently than in final ones for 'and', but show reversed or attenuated effects for 'or'. These results challenge symmetric predictions from satisfaction-based theories and support models incorporating processing asymmetries, such as incremental comprehension where earlier material establishes stronger contextual commitments. Pragmatic theories have advanced explanations for projection of non-anaphoric contents, like factive verbs or evidentials without explicit antecedents. In 2024 research, projection occurs when such contents meet minimal preconditions—such as independence from discourse referents—leading to global inferences unless locally satisfied by contextual alternatives. This framework predicts selective non-projection in embeddings like conditionals, corroborated by judgment data showing accommodation only when preconditions align with utterance goals, thus improving over purely semantic filters. Debates on projection versus local in reports have yielded predictive distinctions. A 2025 contrasts classic projection (where presuppositions escape embeddings) with (where they integrate into reported attitudes), predicting differential acceptance under : projecting content yields inconsistency, while admitted content allows endorsement variability. Empirical ratings from speakers confirm these patterns, with higher rejection rates for projecting cases, informing hybrid models that weigh semantic inheritance against attitudinal scoping. Computational modeling via large language models (LLMs) has emerged as a predictive tool for cross-linguistic projection. A 2025 study on conditionals in English and Mandarin revealed LLMs replicate human-like projection for definite descriptions but underproject soft triggers like 'too', enabling scalable hypothesis testing and refinement of theories like context-local satisfaction. Such approaches forecast deviations from native judgments, attributing them to training biases, and suggest augmenting formal models with probabilistic inference for better empirical fit.

Formal and Logical Treatments

Trivalent and Gap Theories

Trivalent theories treat presupposition failure as yielding an undefined truth value (#) in a three-valued semantics, alongside true (1) and false (0). This framework, building on P.F. Strawson's (1950) view that sentences like "The king of France is bald" lack truth value when the definite description fails to refer, assigns # to atomic sentences with unmet presuppositions. Complex sentences compose via trivalent connectives, such as Strong Kleene conjunction, where the result is # if either conjunct is # (regardless of the other's value) or follows bivalent rules otherwise: for inputs (1,1)→1, (1,0)→0, (0,1)→0, (0,0)→0. Negation typically preserves #: ¬# = #, while flipping 1 to 0 and vice versa, ensuring presuppositions project uniformly under embedding operators unless filtered by context. This predicts, for instance, that "John regrets failing the exam" presupposes John failed the exam even under negation ("John does not regret..."), as failure propagates # outward. Gap theories, often aligned with but distinct from strict trivalence, posit truth-value gaps—absence of truth or falsity—due to presupposition failure, without mandating a third logical value. Noel Burton-Roberts (1989) defends a semantic gap theory, arguing that presuppositions are conditions for truth-evaluability: sentences with gaps (e.g., via non-referring terms) are neither true nor false, and composition via partial functions preserves gaps without introducing undefinedness as a "value." This contrasts with trivalent implementations by avoiding symmetric treatment of # in connectives, emphasizing instead that gaps render wholes inevaluable unless presuppositions are satisfied locally. Projection follows: embedded gaps cause global inevalurability, explaining persistence under or questions, but allows pragmatic repair absent in pure semantics. Both approaches predict presupposition projection via failure propagation but differ in formalization: trivalent logics enable precise truth tables (e.g., Weak Kleene for maximal gap spread, Strong Kleene for Karttunen-like filtering), while gap theories prioritize semantic partiality, critiquing trivalence for overvaluing undefinedness. Empirical adequacy arises in handling quantifiers: trivalent repairs predict variable strength (e.g., "every" triggers universal presuppositions, "at least one" weaker ones via function deployment). Critics note trivalence's intuitive appeal for gaps but challenge its handling of non-uniform projection without ad hoc repairs, favoring dynamic alternatives for complex embeddings.

Dynamic Semantics and Context Change

In dynamic semantics, sentence meanings are defined not as truth-conditional propositions but as context change potentials (CCPs), which are functions transforming an input —typically represented as a set of possible worlds or an information state—into an output after successful utterance processing. This approach, formalized in Irene Heim's File Change Semantics (1983), models as "files" of indexed entities with associated properties, where utterances add, update, or filter file entries to reflect accumulating discourse information. Presuppositions arise as preconditions for CCP application: for a sentence φ with presupposition p to update c, c must entail p (denoted c ⊨ p); failure renders the update undefined, preserving p as a requirement on higher . Presupposition projection follows from this precondition mechanism in embedded environments. In a conditional "If A then B," where B presupposes p, the local context for B's CCP is c' = c[A] (global context c updated by A); successful projection requires c' ⊨ p, meaning either c ⊨ p (global satisfaction) or A entails p (local filtering), else p projects as a global presupposition of the whole conditional. Similarly, for conjunctions, sequential updates enforce cumulative satisfaction: the presupposition of the second must hold in the context after the first, projecting unless locally entailed. This recursive satisfaction condition unifies projection patterns across connectives, contrasting with static theories where presuppositions demand filters. Dynamic treatments extend to attitude verbs and questions via domain restrictions or two-dimensional contexts. For instance, in "John believes that his sister is happy" (presupposing John has a sister), the belief context inherits global presuppositions but may locally accommodate or filter them, with projection occurring unless the attitude operator entails satisfaction. Frank Veltman's update semantics (1996) refines CCPs using plausibility orderings on worlds, treating presuppositions as tests that eliminate incompatible worlds from the input state before assertive updates. David Beaver's framework (2001) critiques Heim's strict preconditioning for overgenerating undefinedness in "soft" triggers (e.g., factives), proposing instead a dynamic assertion-presupposition divide where presuppositions update via "informative" rather than "eliminative" changes, allowing graded satisfaction. Empirical support for these models comes from projection asymmetries, such as Karttunen's (1973) holes/plugs/filters classification, which dynamic CCPs derive without stipulation: hole operators (e.g., ) pass local contexts unchanged, propagating presuppositions; plugs (e.g., verbs of saying) reset to global, blocking projection. Challenges persist for non-monotonic embeddings or probabilistic contexts, prompting hybrid extensions like Rothschild's (2011) use of dynamic conjunction for variable-strength projections. Overall, prioritizes causal update sequences over static entailment, explaining how presuppositions dynamically constrain discourse evolution while integrating with anaphora resolution.

Applications, Implications, and Criticisms

Role in Philosophy of Language and Argumentation

In the , presuppositions are analyzed as background assumptions that must hold for an utterance to possess a determinate , distinct from asserted content or conversational implicatures. , in his 1950 analysis of definite descriptions, contended that sentences like "The present king of is bald" presuppose the and uniqueness of the referent; failure of this presupposition results in a truth-value gap, rendering the sentence neither true nor false, rather than false as Bertrand Russell's 1905 predicts, which reduces such statements to existential claims analyzable via logical quantification. This Strawsonian view underscores presuppositions as preconditions for semantic evaluation, influencing theories of reference and compositionality by highlighting how presuppositional content "projects" through embeddings like or questions—e.g., "Is the present king of bald?" still presupposes —challenging purely truth-conditional semantics. Presupposition theory thus intersects with debates on semantic innocence and context-dependence, where philosophers like Frege and later Strawson equated presupposition failure with indefinability or lack of expressed, prompting alternatives such as trivalent logics assigning a third value (neither true nor false) to gapped sentences. Critics, however, argue that speakers often accommodate failed presuppositions by updating rather than deeming utterances defective, as in dynamic semantic frameworks, though Strawson's emphasis on presuppositional constancy under operator embedding remains foundational for distinguishing presupposition from entailment. This framework reveals presuppositions' role in delimiting linguistic meaning, where ignoring them risks conflating pragmatic accommodation with semantic content. In , presuppositions operate as implicit premises that arguers rely upon without explicit defense, potentially undermining an argument's soundness if unchallenged or unreasonable. Fabrizio Macagno proposes treating presuppositions argumentatively as presumptive drawn from contextual commitments, where the felicity of a depends on the interlocutor's tacit acceptance; failure to justify a presupposition exposes fallacies, such as when it circularly assumes what it seeks to prove. For instance, an argument asserting "John regrets cheating on the exam" presupposes John's guilt, shifting the burden dialectically: challengers must refute the presupposition to invalidate the , aligning with pragma-dialectical models that view as critical discussion requiring explicitness of unshared assumptions. This argumentative lens critiques presuppositional deployment in persuasive contexts, where loaded questions like "Have you stopped beating your wife?" embed false or contestable presuppositions, committing the of presupposition by forcing acceptance of unwarranted background claims. Philosophers of argumentation, building on Austin and Searle's theory, emphasize that presuppositions' survival under or modality enables their use to frame debates subtly, as in transcendental arguments where presupposing shared conditions the possibility of coherent dispute itself. Thus, rigorous argumentation demands meta-level scrutiny of presuppositions to ensure dialectical fairness, preventing their exploitation as covert assertions.

Persuasive Uses and Potential Misapplications

Presuppositions serve persuasive functions in by embedding assumptions that recipients are more likely to accept as background rather than explicit claims requiring . Experimental demonstrates that presuppositions exert greater persuasive influence than equivalent assertions when addressees accommodate them, as the assumed content bypasses critical evaluation and integrates into the context more seamlessly. This effect arises because presupposition triggers, such as definite descriptions or factive verbs, signal shared , prompting hearers to infer rather than the embedded propositions. In ideological discourse, presuppositions facilitate persuasion by conveying novel information under the guise of commonality, thereby advancing partisan views without direct confrontation. For instance, phrases like "the ongoing crisis in border security" presuppose the existence and severity of a crisis, framing subsequent arguments to reinforce ideological alignments such as skepticism toward immigration policies. Advertising exploits this mechanism similarly, employing triggers like temporal clauses ("after using our product, you'll notice...") to presuppose efficacy, which enhances consumer persuasion by implying uncontroversial outcomes. However, these uses risk misapplication when presuppositions introduce unshared or false assumptions, constituting manipulative fallacies in argumentation. Presuppositional fallacies occur when speakers rely on unexamined to shift burden or constrain alternatives, as in loaded questions like "Have you stopped beating your ?" which presupposes prior regardless of the respondent's history. In political contexts, such tactics manifest in statements presupposing disputed facts, such as "reforming the corrupt system" assuming systemic without , thereby derailing toward concessions on the premise's validity. Empirical of persuasive texts reveals higher frequencies of these implicit strategies in ideological appeals compared to neutral communication, often inducing shallow processing that evades rational scrutiny. To counter misapplications, dialectical approaches emphasize challenging presuppositions explicitly, restoring argumentative balance by demanding justification for embedded claims.

Debates on Theoretical Status and Empirical Adequacy

Debates persist over whether presuppositions constitute a distinct semantic category or arise primarily from pragmatic mechanisms. Semantic approaches, such as those employing trivalent logics or , posit that presupposition failure renders sentences neither true nor false, or fails to update the appropriately, distinguishing them from entailments which do not project under or modals. Pragmatic theories, conversely, view presuppositions as inferences drawn from speaker commitments or common ground, without inherent truth-conditional effects, emphasizing accommodation where hearers infer unstated assumptions to maintain coherence. A key contention concerns the heterogeneity of presupposition triggers, with Lauri Karttunen characterizing them as a "" of disparate elements—including definite descriptions, factive verbs, and change-of-state verbs—rather than manifestations of a unified phenomenon. This diversity undermines attempts at a monolithic , as triggers exhibit varying projection behaviors; for instance, factives like "" robustly project, while implicatives may cancel more readily, suggesting multiple underlying relations akin to Frege's distinctions between Voraussetzung (presupposition) and Nebengedanke (side implication). Critics of semantic unification argue that pragmatic variability better explains such differences, avoiding overgeneralization in formal models. Formal semantic theories face challenges in projection adequacy, particularly Heim's satisfaction theory, which requires presuppositions to hold in local contexts derived recursively from embeddings. The "proviso problem" arises when this predicts existential projections for conditionals (e.g., "If John has children, his is bald" presupposes someone has a ), yet empirical intuitions favor stronger universal readings, necessitating adjustments like local accommodation. addresses projection via context change potentials but struggles with non-monotonic updates, where order effects in conjunctions ("and" vs. "or") yield asymmetric inferences not fully captured without pragmatic supplements. Empirical tests via psycholinguistic experiments reveal mixed adequacy for these theories. Studies on conditional embeddings show satisfaction theory's existential predictions partially align with inference rates, but deviations occur under uncertainty, supporting hybrid pragmatic influences. Recent work on linear order demonstrates stronger projection from conjuncts in "and" than disjuncts in "or," challenging symmetric semantic computations and favoring context-sensitive pragmatic resolution. Such findings indicate that while semantic cores explain robust triggers, full empirical coverage requires integrating pragmatic factors, as pure satisfaction models underpredict variability in naturalistic judgments.

References

  1. https://www.ucl.ac.uk/~ucjtudo/teaching/egg2016/1.pdf
  2. https://www.glossa-journal.org/article/id/5415/
  3. https://www.cambridge.org/core/books/linguistics-meets-philosophy/presuppositions/F132F20294A4B8FD49A8B02F7DA24F39
  4. https://dingo.sbs.arizona.edu/~langendoen/ProjectionProblemPresuppositions.pdf
  5. https://eggschool.org/wp-content/uploads/2016/06/Egg_presupposition_1-1.pdf
  6. https://florianschwarz.net/wp-content/uploads/papers/HandbookExpSemPrag_Ps.pdf
  7. https://core.ac.uk/download/210669071.pdf
  8. https://semanticsarchive.net/Archive/GFiMGNjN/Presupp%2520projection%252090.pdf
  9. https://distantreader.org/stacks/journals/llt/llt-255.pdf
  10. http://mit.edu/fintel/fintel-2000-accomm.pdf
  11. https://semprag.org/index.php/sp/article/download/sp.6.5/pdf/6019
  12. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/download/2503/2252/2691
  13. https://www.cmu.edu/dietrich/philosophy/docs/simons/Presupposition%2520and%2520cooperation.non-anonymous.pdf
  14. https://www.phil.uu.nl/~yoad/esslli2021/T8-Presuppositions.pdf
  15. https://web.stanford.edu/~cgpotts/manuscripts/potts-blackwellsemantics.pdf
  16. https://semantics.uchicago.edu/kennedy/classes/f09/semprag1/abbott00.pdf
  17. https://link.springer.com/article/10.1007/s11229-021-03346-7
  18. https://www3.nd.edu/~jspeaks/courses/2011-12/83104/handouts/russell-descriptions.pdf
  19. https://semantics.uchicago.edu/kennedy/classes/f09/semprag1/strawson50.pdf
  20. https://faculty.washington.edu/smcohen/453/StrawsonDisplay.pdf
  21. https://www.cambridge.org/core/journals/philosophy/article/introduction-to-logical-theory-by-p-f-strawson-methuen-1952-pp-x-266-15s-net/E6A1FC061A87531667ED21CF14C46F38
  22. https://www.jstor.org/stable/2252414
  23. https://www.researchgate.net/publication/322712456_Definite_Descriptions_Presuppositions_and_Reference
  24. https://academic.oup.com/book/55233/chapter/428585117
  25. https://plato.stanford.edu/entries/presupposition/
  26. https://plato.stanford.edu/entries/strawson/
  27. https://www.ling.ohio-state.edu/~croberts/pronouns.pdf
  28. https://link.springer.com/article/10.1007/s11245-014-9297-9
  29. https://pmc.ncbi.nlm.nih.gov/articles/PMC6497401/
  30. https://journals.linguisticsociety.org/proceedings/index.php/ExtendedAbs/article/download/784/560/729
  31. https://pubmed.ncbi.nlm.nih.gov/30584974/
  32. https://aristotle.rutgers.edu/joomlatools-files/docman-files/Hazlett.pdf
  33. https://pnadathur.github.io/pdfs/implicatives_final.pdf
  34. https://www.asc.ohio-state.edu/roberts.21/LSA/Roberts.LSA.lecture3.pdf
  35. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/download/26.705/3668/5312
  36. https://czasopisma.uni.lodz.pl/research/article/download/12313/11884/31051
  37. https://web.stanford.edu/group/csli_lnr/Lexical_Resources/phrasal-implicatives/simple-and-phrasal-implicatives.pdf
  38. https://www.jstor.org/stable/412084
  39. https://semanticsarchive.net/Archive/GNiYzUxN/vanLeusen2012-iv.pdf
  40. https://www.christinehowes.com/papers/AyubKhan.etal_AMLaP24_implicatives.pdf
  41. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/download/34.010/5653/12895
  42. https://www.glossa-journal.org/article/id/4984/
  43. https://www.scielo.org.mx/scielo.php?script=sci_arttext&pid=S1405-55462015000400647
  44. https://glossary.sil.org/term/presupposition-trigger
  45. https://semprag.org/index.php/sp/article/download/sp.6.6/pdf/6024
  46. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/view/2588
  47. https://semantics.uchicago.edu/kennedy/classes/f09/semprag1/karttunen73.pdf
  48. https://socialsci.libretexts.org/Bookshelves/Linguistics/Analyzing_Meaning_-_An_Introduction_to_Semantics_and_Pragmatics_%28Kroeger%29/03%253A_Truth_and_inference/3.04%253A_Presupposition
  49. https://academic.oup.com/jos/article/41/3-4/331/7821885
  50. https://judith-tonhauser.github.io/files/hofmann-etal-SuB28.pdf
  51. http://semantics.uchicago.edu/workshops/spl/anaphora/heim83.pdf
  52. https://semprag.org/article/view/sp.15.15
  53. https://mit.edu/fintel/fintel-2008-accommodation-again.pdf
  54. https://people.umass.edu/partee/MGU_2009/materials/MGU099_2up.pdf
  55. https://carleton.ca/singhr/wp-content/uploads/Global-over-local-2015.pdf
  56. https://escholarship.org/content/qt3m9028zz/qt3m9028zz_noSplash_349b3a8ef3715fa808e7332b4ba92a5b.pdf
  57. https://yimeixiang.wordpress.com/wp-content/uploads/2021/03/ho6-presuppositions.pdf
  58. https://www.researchgate.net/publication/281874946_Accommodation
  59. http://mit.edu/fintel/fintel-2008-accommodation-again.pdf
  60. https://link.springer.com/article/10.1007/s10936-017-9534-7
  61. https://tedlab.mit.edu/tedlab_website/researchpapers/Singh_et_al_2015_Cogsci.pdf
  62. https://www.sciencedirect.com/science/article/abs/pii/S0911604417300295
  63. https://ojs.ub.uni-konstanz.de/sub/index.php/sub/article/view/401
  64. https://link.springer.com/article/10.1007/s10936-019-09686-3
  65. https://www.researchgate.net/publication/267425204_Psycholinguistic_Evidence_for_Presuppositions_On---line_and_Off---line_Data
  66. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/view/24.1
  67. https://www.sciencedirect.com/science/article/abs/pii/S001002771830307X
  68. https://elitu.sites.uu.nl/2024/01/15/tijn-schmitz-memory-retrieval-in-presupposition-processing-3-eye-tracking-experiments/
  69. https://pmc.ncbi.nlm.nih.gov/articles/PMC7182549/
  70. https://www.researchgate.net/publication/360526370_The_development_of_presupposition_Pre-schoolers%27_understanding_of_regret_and_too
  71. https://www.sciencedirect.com/science/article/pii/0885201494900175
  72. https://link.springer.com/article/10.1007/s10988-022-09364-z
  73. https://www.glossa-journal.org/article/id/5424/
  74. https://dspace.mit.edu/handle/1721.1/120669
  75. https://pmc.ncbi.nlm.nih.gov/articles/PMC7656057/
  76. https://www.degruyterbrill.com/document/doi/10.1515/ip-2024-2001/html
  77. https://www.glossa-journal.org/article/id/5431/
  78. https://semanticsarchive.net/Archive/DRjNGY1Z/Cross-linguistic%2520presupposition.pdf
  79. https://link.springer.com/article/10.1007/s10988-024-09413-9
  80. https://link.springer.com/article/10.1007/s10988-025-09431-1
  81. https://openreview.net/forum?id=PbgnKfZ3N9
  82. https://www.ucl.ac.uk/~ucjtudo/teaching/SRS2018/trivalentps.pdf
  83. https://journals.linguisticsociety.org/proceedings/index.php/SALT/article/download/2472/2220/2659
  84. https://eecoppock.info/DynamicSemantics/Lectures/why-dymsem-part1.pdf
  85. https://people.umass.edu/partee/MGU_2009/materials/MGU096.pdf
  86. https://semprag.org/article/view/sp.4.3
  87. https://semanticsarchive.net/Archive/jU1MDVmZ/.book-2001.pdf
  88. https://www.tandfonline.com/doi/abs/10.4314/sajpem.v27i3.31515
  89. https://onlinelibrary.wiley.com/doi/10.1111/mila.12549
  90. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2147348
  91. https://philarchive.org/rec/MACPAA-21
  92. https://link.springer.com/article/10.1007/s10503-023-09625-6
  93. https://link.springer.com/article/10.1007/s10503-022-09566-6
  94. https://www.cambridge.org/core/journals/applied-psycholinguistics/article/presuppositions-are-more-persuasive-than-assertions-if-addressees-accommodate-them-experimental-evidence-for-philosophical-reasoning/85033A000FBF5E91C1DB36D05F4133D9
  95. https://www.sciencedirect.com/science/article/pii/S0378216623000978
  96. https://academic.oup.com/book/46548/chapter/407937135
  97. https://www.semanticscholar.org/paper/The-Persuasive-Use-of-Presupposition%253A-A-Pragmatic-Aziz-Hassan/311d440ac0b5fdd3672651260d85ebc86b1b449b
  98. https://pmc.ncbi.nlm.nih.gov/articles/PMC9389194/
  99. https://www.frontiersin.org/journals/communication/articles/10.3389/fcomm.2021.610807/full
  100. https://semprag.org/index.php/sp/article/download/sp.9.13/pdf/7057
  101. https://www.researchgate.net/publication/281266180_An_Experimental_Investigation_of_Presupposition_Projection_in_Conditional_Sentences
Add your contribution
Related Hubs
Contribute something
User Avatar
No comments yet.