Empty Names

The Problem of a Name Without a Bearer

The philosophical problem of empty names begins with the apparent coexistence of successful linguistic understanding and unsuccessful reference. A speaker can competently use “Vulcan” to discuss the planet once hypothesized to explain the anomalous motion of Mercury, while accepting that no such planet was discovered. Readers understand the name “Sherlock Holmes,” make distinctions between Holmes and Watson, and disagree about what the stories entail. Yet these practices do not establish that either expression designates an ordinary concrete individual. A theory of names must explain how that distinction is possible and what follows for truth, thought, inference, and ontology.

The difficulty becomes acute when a familiar referential theory is made explicit. If the semantic contribution of a proper name is precisely its bearer, a name without a bearer contributes no such object. A simple predication apparently containing that name will then lack the constituent required for the usual object-dependent proposition. Nevertheless, utterances involving the name may communicate something specific, support rational disagreement, figure in reports of beliefs, and produce true-looking negative existentials. The problem is therefore a test of the architecture joining reference, meaning, truth conditions, and propositional content rather than a curiosity about fictional vocabulary (Braun, 1993; Sainsbury, 2005).

The crucial inquiry is not whether every intuitive truth judgment must be vindicated by one uniform semantic mechanism. It is whether each proposed account can identify what the speaker has done, what proposition—if any—was expressed, what makes an assertion true or false, and what existence commitments its interpretation incurs. Those are distinct questions. A theory can explain why a conversation is intelligible while leaving its literal atomic sentences unevaluable; it can supply truth values while denying ordinary singular propositions; or it can recognize abstract fictional entities while distinguishing those entities from the characters as depicted.

The analysis that follows reconstructs the historical arguments, compares semantic and metaphysical solutions, and proposes a disciplined diagnostic: determine the kind of name-use and its reference conditions before assigning ontological consequences to the grammar of the sentence.

Definition, Identity, and Scope

An empty name is a name-use that purports to designate an individual but lacks a referent under the relevant interpretation and domain. The qualification “name-use” matters. The written word “Alice” can name many persons; a specific use can succeed, fail, be merely proposed, or be embedded in a fiction. It is not the string alone that is empty. A scientific hypothesis may supply a stable practice of using a name even though the hypothesis is false. A text may introduce a character whose identity is constituted by a narrative practice while never supplying a concrete object that satisfies its predicates.

Reference failure is also more precise than nonexistence. “Aristotle” refers to a deceased person; temporal absence from the present does not make the name empty. A name can refer to an abstract entity even when the bearer has no spatial position. Conversely, the stipulated description “the current king of France” may lack a satisfier, but its logical form is not thereby established to be that of a proper name. The distinction between proper names and definite descriptions is itself disputed, so theories must specify whether they are treating a name as a syntactically singular expression, a convention of designation, or a device for identifying an object through properties (Russell, 1905; Evans, 1982).

Four cases deserve separate treatment. Fictional names are introduced through narrative production; mythical names often circulate through traditions that invite belief, pretense, or both; theoretical names arise within explanatory hypotheses and may fail empirically; accidental empty names arise from perceptual error, misleading testimony, failed demonstrations, or mistaken historical records. A fifth, deliberately stipulated empty name, isolates the logical problem when everyone knows from the outset that nothing is being designated.

The scope of this dossier is the semantics and philosophical significance of these failures. The general nature of proper names belongs to Proper Names; the wider theory of referential relations belongs to Reference; the quantificational analysis of descriptions belongs to Definite Descriptions; and a complete metaphysics of fictional objects belongs to Nonexistent and Fictional Entities. Here those neighboring questions are investigated where they directly determine the explanation of empty names.

A name can also be temporarily ungrounded rather than decisively empty. An archaeologist may use a conventional label for a supposed author while the surviving documents are insufficient to decide whether the attributed author was a single person, several persons, or an invention of later tradition. A model which treats such uncertainty as settled nonexistence collapses epistemic uncertainty into semantic failure. The word “apparent” in the dossier's editorial definition therefore records an epistemic possibility; the philosophical classification of each case must eventually specify whether no bearer exists, no unique bearer has been fixed, or the evidence does not permit a decision. These alternatives have different effects on modal claims, historical identifications, and the use of identity predicates.

The problem also differs between ungrounded expressions and merely incomplete descriptions. A name can refer successfully without its speaker knowing any accurate uniquely identifying fact about the bearer. That is why the descriptive content accessible to the speaker cannot always determine whether the name is empty. A diagnosis requires attention to origins, conventions, intentions, chains of communication, and the circumstances under which a candidate counts as the bearer. No single feature is sufficient in every theory; the philosophical task is to make each proposed condition explicit.

A Preliminary Map of the Explananda

The empty-name problem is often presented through one sentence, such as “Pegasus does not exist.” A single example obscures the diversity of phenomena a theory must explain. Begin with “Vulcan is a planet.” This is an apparently atomic assertion made in the voice of scientific description. Compare “Vulcan is not a planet,” a negated predication; “Vulcan does not exist,” a negative existential; “Le Verrier believed that Vulcan existed,” an intentional report; “The name ‘Vulcan’ was introduced in astronomy,” a metalinguistic statement; and “Holmes is a fictional detective,” a claim about a literary practice or artifact. Their grammatical similarities do not ensure identical semantic interpretations.

A satisfactory account must distinguish at least six dimensions. The first is intelligibility: competent users understand the utterance. The second is semantic value: a name may provide a bearer, a sense, a rule of use, or an incomplete contribution. The third is propositional completeness: something may be communicated even if a strict singular proposition is unavailable. The fourth is truth: a sentence may be false, lack a truth value, or be evaluated relative to a story or presupposition. The fifth is ontological commitment: the best analysis may quantify over concrete persons, abstract artifacts, possible individuals, linguistic tokens, or nothing corresponding to the name. The sixth is cognitive significance: believers may reason differently with distinct names even where neither denotes.

These dimensions constrain the assessment of all theories. For example, treating the truth of a negative existential as obvious already commits one to a semantic account of what that sentence asserts; treating fictional predication as literally true risks changing the kind of entity under discussion. The main work is to show which distinctions are doing explanatory work rather than using one favored intuition to settle all six dimensions at once (Donnellan, 1974; Kripke, 2013; Hanks, 2015).

Before Analytic Semantics: Being, Thought, and Intentional Objects

The tension between thinking of something and acknowledging its absence has a philosophical prehistory in debates about being and intentionality. Ancient discussions of what can be thought or spoken about prompted the question whether meaningful thought must have an existing object. That problem should not be retroactively equated with the modern semantics of proper names: the categories of singular term, reference function, and compositional truth conditions belong to later developments. Yet the older concern supplies the general pressure that reappears in the modern case.

Alexius Meinong's 1904 “Über Gegenstandstheorie” gives the issue a distinct object-theoretical shape. In §1 he begins from the directedness of judging and representing; in §2 he challenges the tendency to restrict objects of thought to what is actual. His framework permits investigation of objects independently of their existence and offers resources for speaking about impossible or merely conceived objects (Meinong, 1904). A name apparently directed at a nonexistent object may thus retain a determinate intentional target. Such a view explains the grammar of representation without immediately identifying the target with an existing material thing.

The price is an exact account of the sense in which a nonexistent object is an object at all. If contradictory properties can be ascribed to an impossible object, how are ordinary principles of identity and predication preserved? Can “the golden mountain” or “the round square” be objects without enlarging the inventory of what exists? Later neo-Meinongian proposals distinguish modes of predication or existence from objecthood, but the historical impulse is already visible in Meinong's insistence that the domain of inquiry extends beyond the actual.

The lasting contribution is a question rather than a foregone conclusion: does the directedness of thought entail an intentional object with an identity independent of thinking? A referential semantics cannot simply infer that entailment from grammar, and a deflationary semantics must explain why apparent object-directed thinking remains so systematic.

Frege: Sense, Reference, and the Intelligibility of Failure

Gottlob Frege's “Über Sinn und Bedeutung” (1892) distinguishes Sinn, commonly translated as sense, from Bedeutung, often translated in this context as reference. The point is not exhausted by the fact that two names can designate the same individual. The apparent informational difference between a trivial identity and an informative identity shows that understanding involves a mode of presentation in addition to an object designated. This structure makes room for meaningful cognition even when the intended object is absent.

Frege explicitly discusses “Odysseus” and considers a sentence saying that Odysseus was put ashore on Ithaca while asleep. The sentence can convey a thought through its sense although the proper name is doubtful as a source of ordinary reference. Frege's account therefore provides a way to explain understanding without manufacturing a concrete bearer for the name (Frege, 1892, pp. 31–33). That result must be kept distinct from the further Fregean thesis that a complete assertoric sentence ordinarily refers to a truth value. If a component lacks customary reference, the ordinary route to a referential truth value can fail.

This gives the classic Fregean account two achievements and one difficulty. It preserves informative sense and distinguishes literature from factual inquiry, where actual reference becomes decisive. It also avoids identifying understanding with a successful act of naming. Yet the handling of negative existentials remains delicate. If “Odysseus does not exist” is a complete judgment, what truth value does it possess and how does its sense connect to what it denies? A theory cannot merely transfer the semantic incompleteness of one predication to every construction containing the same name.

Frege's treatment of reference failure has generated important interpretive debates concerning presupposition and truth-value gaps. His 1892 analysis establishes a distinction between possessing a sense and securing a referent, while subsequent applications must specify how the semantic dependence of complex expressions on their components is interpreted. The central insight remains the separation of cognitive content from achieved designation.

Frege's distinction also separates the semantic work of an expression from its success in factual inquiry. The sense of a sentence can be grasped as a thought, and the speaker can understand what would have to be the case if its constituent expressions designated appropriately. A further step is required to determine whether the sentence has the referential value appropriate to an assertion of truth. A proper name lacking a bearer can thus leave its linguistic and cognitive significance intact while frustrating the ordinary movement from the thoughts expressed to the reference of the complete sentence. This is precisely why Frege's Odysseus example matters: imaginative engagement with a narrative can persist without the question of the name's actual referent being decided.

This distinction resists two oversimplifications. One assumes that a sense is automatically equivalent to a description that every competent speaker can articulate; Frege's mode of presentation is a semantic and cognitive notion that need not be exhausted by a fixed inventory of descriptive beliefs. The other assumes that sense alone settles truth values for sentences containing empty names. The relation between the sense of a sentence, its compositional construction, and a referential truth value still needs analysis. A Fregean can explain the intelligibility of “Odysseus is asleep” and face the independent question of why “Odysseus never existed” should count as true. The negative existential must be analyzed at the appropriate level rather than treated as an ordinary predication whose purported subject has already failed to denote.

Russell: Denoting, Quantification, and Apparent Names

Bertrand Russell's “On Denoting” (1905) challenged the assumption that every grammatically singular phrase contributes an object to a proposition. The classical example “The present king of France is bald” can be analyzed through the existence and uniqueness conditions of a description: there is exactly one present king of France, and that individual is bald. In the absence of such an individual, the ordinary analysis makes the positive statement false, rather than forcing reference to a nonexistent king (Russell, 1905).

The negative counterpart displays the importance of logical scope. “The present king of France is not bald” can deny that there exists a unique present king who is bald, or it can assert that there exists a unique present king and that he is not bald. On the first reading it is true under failure of the existence condition; on the second it is false. This analysis dissolves one class of puzzles by refusing to read logical form directly from grammatical surface.

Russell's treatment of ordinary proper names can be developed descriptively: when “Vulcan” abbreviates a condition identifying a hypothesized intra-Mercurial planet, its failure is represented by the absence of a unique satisfier. A negative existential then states that no object satisfies the relevant condition. But this accommodation is not costless. Names behave differently from descriptions in modal and epistemic contexts; speakers may use the same name while associating divergent descriptions. If all content depends on an identifying condition, a mistaken or impoverished description threatens to change the subject of conversation.

Russell's achievement is a general method for avoiding ontological multiplication through analysis. His limitation, for the specific issue of empty proper names, is that a satisfactory theory of descriptions does not automatically justify a descriptive semantics of names. The distinction will become pivotal in the causal-historical critique.

Russell's distinction between logically proper names and ordinary names is indispensable here. In his mature logical program, a logically proper name is associated with a form of acquaintance that secures its bearer; ordinary names as used in everyday communication may conceal descriptions. It follows that a failed ordinary name need not be a failed logically proper name. It may instead be an apparent singular expression whose analysis never requires a corresponding object. This is a substantive philosophical thesis about logical form and knowledge, not simply a typographical change in how a sentence is written.

The existential analysis is most convincing when the identifying condition genuinely captures the communicative target. If one uses “Vulcan” for the hypothetical planet responsible for Mercury's anomalies, the failure of the planetary hypothesis can be modeled as an unsatisfied description. But the identification of a name with such a description becomes unstable when speakers disagree about the planet's supposed orbit or when a competent user knows only that a historical astronomer employed the name. Russell's strategy therefore divides into two questions: how to eliminate unwarranted ontological commitment in a particular logical reconstruction, and whether that reconstruction gives the correct general account of the meaning of names. A convincing answer to the first question does not establish the second.

Quine and the Logic of Ontological Commitment

W. V. O. Quine's “On What There Is” (1948) develops an influential response to apparent commitment to nonexistent objects. The grammatical ability to utter “Pegasus is not” does not require admitting Pegasus into a universe of strange entities. The theorist should identify the bound variables of a regimented theory and determine which objects must fall within their range for its asserted sentences to be true. Paraphrase can remove a seemingly singular reference that the logical analysis does not need (Quine, 1948).

Quine's challenge is especially clear against a strategy that permits a nonexistent object for every apparently meaningful name. If ordinary language can introduce infinitely many expressions with empty conditions, an unrestricted metaphysics of named nonentities becomes ontologically extravagant. Regimentation provides a disciplined way of distinguishing a linguistic practice from an inventory of entities.

The difficulty is that a successful paraphrase must preserve what matters to the original claim. A substitution involving “Pegasus” may sustain ordinary negative existential judgment, yet it can fail to capture the specific perspective of someone who believed Pegasus was real, the historical origin of the myth, or the difference between two stories with similar descriptions. The semantic identity of a name-using practice is not reducible in every context to a convenient bound-variable reconstruction.

The philosophical result is a distinction between ontological and semantic economy. A theory may avoid unwanted objects by translating discourse into quantificational form. It still owes an account of what makes the translation faithful in modal, intentional, and literary contexts. The existence of an effective paraphrase for one sentence gives no unrestricted license to replace every appearance of a proper name by the same description.

Quine's proposal gains force from a disciplined contrast between naming and quantifying. The fact that a language permits a term in subject position is compatible with its regimented theory making no existential commitment to a bearer. Conversely, a theory could avoid singular names while quantifying over unwanted entities. Ontological commitment depends on the structures of the theory as regimented, not on a count of the ordinary nouns and names appearing in a passage. A philosopher who describes a fictional creature without treating it as a constituent of a theory's quantified domain has not thereby asserted its existence.

The method leaves a question of expressive preservation. Imagine two historians discussing competing hypotheses called “Vulcan” and “Planet X,” both now abandoned. A regimentation that merely yields “there is no planet satisfying the proposed conditions” can preserve their common negative conclusion while losing the historical difference between the hypotheses. That loss may be harmless in one context and decisive in another. Consequently an ontologically parsimonious paraphrase must be evaluated against the claims, inferences, and identity conditions for which the sentence is being used. The relevant standard of fidelity is not identical across formal logic, scientific history, and reports of belief.

Strawson: Referring as an Act and Existential Presupposition

P. F. Strawson's “On Referring” (1950) shifts attention from a sentence considered as an abstract syntactic object to its use by a speaker on an occasion. The same sentence can be used in different settings to refer successfully or unsuccessfully. The distinction is central to empty names because reference failure may be an event in communication rather than a permanent semantic deficiency of a word type (Strawson, 1950).

Under a presuppositional treatment, ordinary utterances that depend on the existence of a specified object do not automatically become false when the object is missing. Their ordinary assertoric use presupposes a condition that has failed. Consider a speaker earnestly claiming that the nonexistent detective “is waiting outside.” The hearer may reject the background assumption that the detective exists instead of simply issuing a correction about the detective's location. That reaction supports distinguishing denial of an assertion from rejection of its presupposed referential basis.

Presupposition solves some pragmatic puzzles while posing its own. Negative existentials such as “Holmes never existed as a flesh-and-blood detective” are naturally used to challenge the very assumption that a referring use might presuppose. A general rule that all uses of empty names lack truth values would exclude assertions designed to deny existence. Consequently the account requires either special constructions, a use-sensitive diagnosis, or an explanation of how an existence denial operates at a different semantic level.

The debate initiated by Russell and Strawson shows that theories of reference must separate sentence meaning, utterance content, presupposed conditions, and the appropriate conversational response. Those distinctions remain indispensable even for theories that reject Strawson's specific analysis.

Names as Historical Practices: Evans, Donnellan, and Kripke

The movement away from descriptive equivalence developed through attention to causal history, naming practices, and communicative intentions. Gareth Evans's contribution to “The Causal Theory of Names” (1973) and posthumous The Varieties of Reference (1982) investigate the socially sustained information links that underwrite name-use. Keith Donnellan's “Speaking of Nothing” (1974) brings unsuccessful reference directly into the dispute. Saul Kripke's Naming and Necessity, first developed in lectures and published as a book in 1980, argues that identifying descriptions frequently do not determine a name's semantic contribution in the way classic descriptivism requires (Evans & Altham, 1973; Evans, 1982; Donnellan, 1974; Kripke, 1980).

Causal-historical theories explain how speakers can refer while possessing false beliefs about a bearer. The name may be passed along through a communicative chain that depends on an earlier naming event rather than on the descriptive adequacy of every later speaker. But an uninterrupted chain need not terminate in a concrete individual. A mythical founding ceremony, an invented biography, or an erroneous observation can sustain stable transmission without making the intended referent real.

The challenge to direct-reference theories therefore becomes sharper after the critique of descriptivism. Success with ordinary names appears object-dependent; failure does not prevent a public practice from developing. A critic asks whether that practice possesses genuine semantic content or only historically organized habits of discourse. The answer cannot be extracted from the causal account alone because causal origin and semantic contribution are separate theoretical jobs.

Kripke's Reference and Existence (2013) revisits these issues with fiction and negative existence judgments at the center. A particularly important lesson is that anti-descriptivist arguments do not by themselves force one theory of empty names. An account may preserve direct reference in successful cases while introducing different resources for failed and fictional uses. This plurality of possible responses is the starting point of contemporary semantic analysis.

Causal continuity has limits as a criterion of successful reference. A community may inherit a mythical name through many generations of testimony, ritual, and written record. The later users' utterances are linked by a real historical chain, but the chain begins with a story rather than with a worldly individual. If a causal account of meaning simply demands the existence of a chain, it falsely treats the myth as an ordinary name of a concrete entity. If it demands an object at the origin, it needs a theory of what happens when a founding act mistakes an appearance for a person or when different participants attach the same name to competing candidates. The transmission history explains the existence of the practice and leaves the success of reference to further conditions.

This creates a useful comparison between scientific and fictional naming. In scientific inquiry, investigators may hope to locate the object that a provisional name designates. In fiction, the author may intentionally introduce a name with no expectation of actual-world designation. A history-sensitive theory can recognize both practices as systematic while giving their success conditions different interpretations. It must nonetheless decide whether the name in an explicitly fictional narrative has no actual-world referent, refers to a created abstract artifact, or is assigned a character within a representational context. The mere fact of historical transmission cannot settle those ontological alternatives.

The Negative Existential Puzzle

The sentence “Vulcan does not exist” is the canonical stress test. It seems true because astronomers have abandoned the postulated intra-Mercurial planet. Yet if the name “Vulcan” contributes only its referent, there is no object to enter into the proposition that the object lacks existence. The intuitive truth of the sentence then appears incompatible with the theory's account of what can be asserted.

One route treats existence claims as quantificational rather than predicative. Instead of predicating nonexistence of a special nonexistent planet, the sentence says that nothing satisfies a reference-fixing condition associated with the relevant naming practice. A second route makes the negative existential metalinguistic: the name lacks a bearer. A third allows truth at the level of a partially completed or gappy proposition. A fourth analyzes the construction through failed presupposition, special negation, or a speech act whose target is the attempted reference itself. A fifth recognizes a nonconcrete object and distinguishes its existence as an artifact from the existence of an object answering to the theory's original empirical description (Braun, 1993; Salmon, 1998; Sainsbury, 2005).

These options must be compared against two conditions. First, they should explain why the proposition is about a particular failed posit rather than every nonexistent item indiscriminately. “Vulcan does not exist” differs in informative history from “Aetherius does not exist” even when neither name denotes. Second, the theory should explain how the truth of the assertion is warranted. In the scientific case the evidence concerns a hypothesis and an absence of appropriate confirming observations, not a mysterious encounter with an entity that lacks being.

My analysis favors treating the negative existential initially as a claim about an unsuccessful designation within an identifiable name-use practice. This is a diagnostic recommendation about logical interpretation, not an assumption that such practices must replace every literal semantic account. It preserves the evidentiary basis of the denial and makes the relevant target of failure explicit.

The puzzle can be formulated as an argument with three contested premises. The first says that ordinary proper names contribute only their bearers to literal propositional content. The second says that an empty name contributes no bearer. The third says that the negative existential containing that name expresses a true complete proposition of exactly the ordinary referential kind. Taken together, these premises are unstable because the first two deprive the relevant singular proposition of its supposed constituent, while the third asserts its completeness and truth. The premises identify where rival theories diverge. A Fregean rejects the exclusivity in the first premise; a gappy-proposition theorist modifies the requirement of completeness; a pragmatist revises the third premise's account of what is asserted; a description theorist rejects the assumed logical form.

The inference from a true denial to a nonexistent object is especially hazardous. In a natural-language sentence, the grammatical subject of the denial may look like a name of an entity. Logical analysis can show that its successful use involves an attempt to designate and that the claim rejects the success of the attempt. It is therefore possible to preserve the informative truth of the denial while assigning no individual to the term. Yet this does not make the problem disappear. There remains the compositional question: how does the language encode that failed attempt in the negative existential without treating every similar name as synonymous? A semantic solution must explain both the true judgment and the difference between its subject and another empty name.

Identity, Cognitive Difference, and Empty Names

Identity provides a complementary difficulty. When two empty names have different histories, competent speakers can distinguish them, even though each lacks a bearer. They can learn that a mythical figure was known under two names, doubt a proposed identification, or discover that two supposed scientific objects arose from the same mistaken inference. Such learning is intelligible even if there is no worldly individual that constitutes the shared referent.

The Fregean response assigns senses or modes of presentation to the expressions. The descriptive response supplies conditions with distinguishable informative content. A historical-practice response distinguishes public trajectories of naming, including origins, transmission, and conditions for successful reference. A gappy-proposition theory requires supplementation if it is to discriminate two propositions with otherwise identical empty positions. If every failed name contributes simply nothing, sentences differing only in empty names threaten to collapse into the same content (Braun, 1993; Spencer, 2016).

A crucial distinction arises between identity of bearer and identity of semantic role. A statement of the form “a = b” cannot report successful co-reference when neither name refers, yet it may express a conjecture that two investigations concern one and the same individual if either succeeds. It may alternatively report that two name-use practices have been unified. The kinds of evidence and the truth conditions of those interpretations differ.

What follows is that a theory of empty names needs a nontrivial principle of individuation for failed reference. A single placeholder for all failures cannot model our capacity to think separately about Vulcan, Pegasus, and an unnamed invented island. That requirement is semantic and cognitive; it does not logically entail a corresponding population of nonexistent individual objects.

Millianism and the Pressure of Missing Constituents

Millianism holds, in one influential formulation, that the contribution of an ordinary proper name to what is said is its referent itself. When a referent exists and the name functions directly, the resulting proposition may be object-dependent or singular. The theory explains why replacing a name with a co-referring descriptive phrase can alter the way a thought is presented while retaining a core object-directed content. Its challenge is what to do when no object enters the proposition.

The straightforward option is to say that an atomic sentence containing a genuinely empty name expresses no complete proposition, although its utterance can still have linguistic significance. This view distinguishes structured semantic incompleteness from unintelligibility. If “Vulcan is hot” has an open position where a designated planet should be, the hearer can identify the failed endeavor and understand the predicate without possessing a standard proposition of the form Hot(v).

The cost is explaining belief reports and negative existentials. “Le Verrier believed Vulcan existed” ordinarily attributes a historically intelligible scientific belief. A Millian theory cannot simply use the very singular proposition unavailable under its account. Nor can it automatically treat every absence of an object as false, because the distinction between negated predication and denial of existence has not yet been explained.

Solutions include gappy structured contents, pragmatic enrichment, propositional attitudes involving representational vehicles, and special interpretations of negation. David Braun's 1993 defense of empty-name treatment within a direct-reference setting is a major example. Peter Hanks's 2015 reference-act approach offers a different strategy: an empty name can inherit semantic content from a type of reference act without supplying an object. The common lesson is that direct reference in successful cases underdetermines the treatment of failures (Braun, 1993; Hanks, 2015).

The strongest Millian argument begins from successful cases. If a name actually refers to a particular individual, and if its semantic role is direct designation, then the proposition expressed by “Aristotle taught Alexander” can depend on Aristotle himself rather than on whatever descriptive beliefs a speaker happens to possess. This captures the object-dependence of a genuinely singular assertion. The pressure arises only when the apparently same construction contains an expression with no bearer. One is then forced to distinguish the rules of a semantic type from the circumstances in which a token of that type yields a complete content.

A tempting repair is to assign the name a surrogate, such as the name itself or a name-use practice. The repair can make the content complete, but it threatens to change the claim. If “Vulcan is a planet” is analyzed as predicating planethood of the word “Vulcan,” it becomes an elementary category mistake rather than a scientific hypothesis. If it predicates planethood of a linguistic practice, the assertion fails for a different reason from the hypothesis's empirical failure. The surrogate can properly enter the content of a metalinguistic statement, but it cannot simply be installed as the object of the original predication without explanation. A defensible hybrid theory must state which constructions change interpretive level and why.

Gappy Propositions and Their Limits

A gappy proposition is a structured content whose object position remains unfilled. In simplified notation the attempted thought “Vulcan is a planet” has a predicative structure with a designated argument slot but no individual occupying that slot. The structure retains more than a mere sound pattern; it identifies the relevant predication and the failed singular contribution. Braun (1993) argues that a theory of empty names can be made compatible with direct reference by attending to such partial contents.

Two issues follow. The first is whether a gappy content is truth-evaluable. A theorist may declare it truth-valueless, count a positive atomic occurrence as false, or introduce a rule specific to negation and existence. These are substantive semantic choices, not consequences of the metaphor of a gap. The second is individuation. If “Vulcan is luminous” and “Pegasus is luminous” have the same unary predicate and the same unfilled position, a naive structure predicts sameness where ordinary discourse draws differences.

Joshua Spencer's account of plenitudinous Russellianism addresses the second problem by distinguishing multiple gappy propositions rather than allowing all empty positions to collapse (Spencer, 2016). Such views must provide principled identity conditions for their gaps: mere stipulation of a separate gap for each string risks importing syntactic distinctions without explaining their semantic role. A practice-sensitive theory can give the gaps historical profiles, but then those profiles carry information the most stringent object-only semantics initially excluded.

Gappiness therefore offers a useful representation of failure without completing the whole theory. It captures the absent object while inviting a further account of truth, individuation, and cognition. It succeeds when the explanatory roles assigned to a structured gap are explicitly specified.

Free Logic: Names Without Denotation in Formal Systems

Standard classical first-order semantics typically assign every individual constant a member of the domain. Free logics relax that assumption so that singular terms may fail to denote. Their central importance is methodological: they make room for the logic of nonreferring expressions without forcing every name to stand for an existing object. Negative, positive, and neutral free logics differ chiefly in the treatment of atomic formulas containing empty terms (see the revised SEP entry “Free Logic,” 2025).

In a negative free logic, an atomic predication with an empty singular term is false. The familiar rule from F(a) to ∃x F(x) cannot be applied without a safeguard ensuring that a denotes. In a neutral free logic, the empty-termed atomic claim may lack a classical truth value; negation and composition then require an explicit nonclassical treatment. Positive free logics can allow some atomic statements with empty terms to be true, often using special valuation clauses or distinctions among domains.

Consider the seemingly simple formula a = a. In classical semantics its truth is automatic because a denotes a domain member. In a free logic that permits an empty a, the status of the identity may depend on whether identity is treated as strict and denotation-sensitive. The usual unrestricted universal instantiation and existential generalization principles must likewise be revisited if they are not to generate objects solely from the occurrence of a constant.

Free logic is not itself a single metaphysical solution. The choice of domain, the existence predicate, and the valuation of atomic sentences all express assumptions that should be declared. Its significance is to make those assumptions visible and to provide a precise setting in which competing accounts can be evaluated without relying on ambiguous natural-language intuitions.

A formal illustration makes the inferential cost visible. In classical first-order logic, an instance F(a) ordinarily supports ∃xF(x) because the interpretation function assigns a domain element to a. Allow a to be non-denoting, and the usual existential generalization can no longer be accepted without qualification. One may require E!(a) as an additional premise, where E! indicates that the term denotes in the relevant domain. This adjustment models the principle that the mere syntactic presence of a name in an assertion cannot guarantee a corresponding object. It also explains why a calculus suitable for mathematical constants may need modification when applied to fictional or failed scientific terms.

The systems differ in their treatment of expressions such as F(a) and ¬F(a). A negative free logic can assign the empty positive atom false and its negation true under two-valued semantics. A neutral free logic may assign an undefined value to the atom, raising questions about whether its negation is undefined as well. A positive free logic may license some atomic truths involving non-denoting terms under specially constructed valuation rules. There is no bare inference from the adjective “free” to one choice of valuation. A free logic supplies a formal space of possible accounts; substantive semantic and ontological commitments decide which calculus is appropriate for a given philosophical interpretation.

Negation, Scope, and the Existence Predicate

Negation interacts with empty names at several levels. There is ordinary predicate negation, as in “Vulcan is not spherical”; sentential negation of an attempted assertion, as in “It is not the case that Vulcan is spherical”; and a negative existential, as in “Vulcan does not exist.” In a classical setting these forms can appear closely related. Under reference failure they may behave differently because the existence condition or the definedness of an expression enters at a distinct logical level.

For example, if an empty-termed atom F(a) is false under negative free logic, its ordinary negation ¬F(a) is true in a classical two-valued interpretation. But one must resist the inference that this makes it literally true that a particular nonexistent individual has the property of not being F. The valuation is a rule for sentences containing a non-denoting term. It may describe the failure of successful predication rather than an attribute of a spectral individual.

An existence predicate E!(a) is often defined using identity and a quantifier, or introduced as a primitive whose behavior is tied to denotation. In negative free logic, denying E!(a) can be evaluated true without committing to a referent. In a dual-domain semantics, by contrast, one may distinguish outer objects available to discourse from inner objects counted as existent. The apparent truth of “a does not exist” then depends on which existence notion and domain are in use.

A systematic analysis must state explicitly whether “exists” is a first-order property, a second-order quantificational notion, an actuality predicate, or an expression of successful designation. Multiple coherent formalizations exist, but they answer different questions. Their plurality is evidence for the importance of semantic disambiguation rather than an invitation to treat all existential statements as interchangeable.

Fictional Discourse and the Problem of Truth in Stories

The sentence “Holmes lived on Baker Street” invites a judgment of truth within Conan Doyle's stories. That judgment can be appropriate even though Holmes does not appear in historical records as a London detective. A theory that assigns a truth value to the sentence without recording the literary context may confuse a property attributed in a fiction with a property literally instantiated by a concrete person.

A story operator provides one general strategy: “According to the Holmes stories, Holmes lived on Baker Street.” The operator changes the standard of assessment. Its interpretation can use facts about the narrative, what is explicitly said, and what follows from appropriate background assumptions. David Lewis's account of truth in fiction and Kendall Walton's theory of make-believe are prominent developments of this approach (Lewis, 1978; Walton, 1990). Story-relative evaluation is often richer than direct quotation: readers infer unmentioned details, compare incompatible passages, and discuss what follows under narrative conventions.

An operator alone does not settle every issue. The content of a story may be inconsistent, incomplete, or open-ended. A statement true in one fictional work can be false in a sequel or another adaptation. Where a narrative says that two characters share a name, the semantics must distinguish the narrative individuals without finding worldly persons that satisfy the fictional descriptions.

The right distinction is between truth according to a representation and literal truth about the represented world. Fictional discourse can support genuine reasoning under its own assumptions while leaving room for disputes over whether fictional characters also exist as abstract artifacts. One need not settle that ontological dispute to recognize the logical usefulness of evaluation relative to a specified fiction.

The semantics of story-relative truth must also address incompleteness. Doyle's narratives do not specify the exact number of molecules in Holmes's body. They may support determinate claims about the detective's address or his methods while leaving other questions unanswered. A fiction operator should therefore distinguish propositions actually entailed by the text and appropriate background assumptions from propositions that merely could have been true in some compatible imagined world. A theory that treats every undecided fact as either true or false without justification projects a completeness onto the fictional world that the literary work never established.

Inconsistent fiction generates a second pressure. A story can portray an impossible transformation or contain incompatible descriptions. If truth in a work is modeled by a set of ordinary possible worlds, no world may satisfy all the stipulated conditions. A richer account may use contextual selection, partial models, impossible worlds, or a paraconsistent treatment of story-internal consequence. Each option has formal costs, but the underlying issue is stable: the logic of statements inside a fiction need not be copied unmodified from unrestricted ordinary-world assertion. The name's apparent referential emptiness is one part of a broader problem concerning the status of represented circumstances.

Metafictional Statements and Abstract Artifacts

Ordinary criticism makes assertions apparently about fictional characters from outside the stories: “Holmes is more famous than Watson,” “Holmes was created by Conan Doyle,” or “Holmes is a fictional character.” These do not merely report what Doyle's narrator says. They concern the cultural role, origin, reception, and comparative properties of the characters as represented objects. Such uses challenge theories that interpret every occurrence of a fictional name as an empty term in a strict atomic sentence.

Amie L. Thomasson's Fiction and Metaphysics proposes that fictional characters be understood as dependent abstract artifacts, created through intentional activity and sustained by literary works and cultural practices (Thomasson, 1999). This gives a referent to many metafictional uses. A character can be created, discussed, adapted, and remembered even though no flesh-and-blood detective was thereby created. The theory connects the ontology of fiction with that of other historically constituted cultural objects.

The proposal raises an identity question. When a publisher produces an adaptation that substantially changes the character, is the character the same artifact, a derivative version, or a distinct one? What fixes the boundaries between the original textual character, its interpretive traditions, and its later cultural manifestations? Dependence on a public practice explains much but requires principles of persistence.

It also raises an ambiguity issue. If “Holmes is a detective” is true only within the story, while “Holmes is a cultural icon” is literally true of an abstract artifact, a semantic theory must explain the transition between levels. One should not infer from the real existence of the artifact that the fictional detective exists as the sort of concrete individual depicted. Thomasson's theory is powerful precisely when these two kinds of predication remain explicit.

Meinongian and Neo-Meinongian Alternatives

A distinct strategy takes apparently nonexistent intentional objects more directly. Instead of paraphrasing all discourse away or confining it to story operators, a Meinongian framework allows reference to objects that fail to exist in an ordinary sense. The motivation is descriptive adequacy: speakers appear to attribute properties to, distinguish, and compare nonexistent objects, and the theory gives those operations a corresponding semantic domain (Meinong, 1904).

The classic problem is how to avoid collapsing characterization into metaphysical commitment. If the round square is both round and square, does the ontology contain a contradiction? Responses distinguish the properties an object is characterized as possessing from properties it exemplifies in the actual world, or distinguish existence from objecthood. Such distinctions can yield coherent formal systems, but their success depends on the details of predication and identity rather than on an unrestricted slogan that every conceived object exists in some mode.

For empty names the question becomes whether the bearer of “Vulcan” is an intentional nonexistent object, a merely possible planet, an abstract construction, or no object. These should not be conflated. A named hypothesis can be historically identifiable without supplying a uniquely determined possible planet; many possible planets could satisfy a partial scientific description. A theory introducing nonexistent individuals must tell us which object was intended and what would make two such intentional targets identical.

The comparative verdict is conditional. Object theories are strongest when they give rigorous conditions of individuation and prevent arbitrary inferences from discourse to ontology. Their burden is greater when the name arose through a genuine attempt at empirical reference, where the scientific conclusion is precisely that no such entity was established.

Scientific Names: Vulcan, Lemuria, and Failed Hypotheses

Failed scientific reference differs structurally from invented fiction. The historical use of “Vulcan” aimed to identify a physical planet. The name was linked to explanatory expectations about Mercury's orbit and to an empirical search that did not secure the intended object. “Lemuria,” introduced in nineteenth-century explanations of biological distribution, similarly represents a historical theoretical posit later displaced by other scientific developments. The relevant existence claims are answerable to empirical evidence rather than to narrative authority (Fisher, 2026).

A scientific name can acquire a determinate historical role after its referential failure is accepted. Scholars can ask who introduced it, which predictions accompanied it, which evidence changed its standing, and how it differed from rival hypotheses. Such claims are often literally true because their referents are texts, acts, scientific practices, and historical episodes. It does not follow that the missing planet or continent exists as an object answering to the original hypothesis.

The crucial point is a distinction between an object of investigation and the record of an investigation. A research program can be definite and historically traceable while its supposed target is absent. The stability of the research program does not entail successful object designation. Conversely, a later theoretical reinterpretation can sometimes identify a new referent for a familiar term, but that requires a justified account of semantic continuity rather than a retroactive assertion that the original target existed all along.

Scientific failures also illuminate the relation between meaning and correction. The name's continued intelligibility is what makes the correction communicable: one must be able to say which hypothesis failed. A semantic theory that cannot represent this difference between failing to refer and failing to say anything has ignored a central function of scientific language.

Belief Reports and Intentionality Without Successful Reference

Reports of propositional attitudes introduce empty names under verbs such as believes, fears, seeks, imagines, and doubts. The historical statement “Le Verrier believed that Vulcan existed” is useful because it attributes a determinate mistaken belief. A strictly object-dependent account of propositions confronts an immediate difficulty: what proposition can be the object of belief when the putative constituent is missing?

One response distinguishes the content of the attitude from a singular proposition of the usual referential form. The scientist may have believed that the planet meeting a certain theoretical role exists. Another treats the belief as involving a public name or a mental representation, allowing the psychological state to be individuated without a corresponding planet. A third allows incomplete or nonstandard structured contents, supplemented by the history of the name. Neil Feit's treatment illustrates the importance of separating the truth of an attitude report from the availability of a proposition expressed by its embedded clause (Feit, 2009).

The same challenge occurs in everyday perception. A person who mistakes a shadow for an intruder may ask whether “that person” is dangerous. The interrogative has a cognitive role even though no person is there. Its content might be explained using perceptual demonstratives, an object-directed mode of thought, or an existentially conditional representation. Different theories will place the failure at different points in the chain from experience to reference.

The general result is that the attribution of a belief does not automatically entail that an object answering to every singular expression in its reported content exists. Yet it cannot be explained merely by reporting the speaker's uttered sounds. A satisfactory theory must retain inferential organization, historical sensitivity, and the capacity for epistemic revision.

An attitude report can remain true even when the attitude's object as conceived is absent. Suppose two researchers both deny the existence of Vulcan but differ in their understanding of the historical evidence. One correctly attributes to Le Verrier the belief that a planet would explain Mercury's observed behavior, while the other incorrectly attributes a belief that a fictional detective caused the motion. Their reports are not equivalent. Their truth conditions depend on how the scientist's representational state and its theoretical role are identified, not merely on whether an object was actually designated.

The result raises a challenge for mental-content externalism and for object-dependent theories of singular thought. If the availability of a singular content requires a successful environmental relation, what content does a misinformed thinker possess before a discovery that disconfirms a purported object? Some theorists distinguish a fully singular thought from a descriptive or otherwise general thought that the thinker mistakes for singular. Others allow the intentional structure of an unsuccessful singular attempt to count as content. The choice affects rational explanation: one must still account for why the thinker gathered observations, inferred consequences, and changed beliefs. A psychology of error cannot explain mistaken cognition by erasing its determinacy altogether.

Reference Acts, Speaker Intentions, and Metasemantics

Peter Hanks's Propositional Content (2015) develops an act-based approach to semantic content, including an account of empty names through types of reference acts. On this view, an attempt to refer can be intelligible and content-bearing despite failure to secure an object. The relevant unit is a repeatable kind of communicative act rather than exclusively the bearer that would complete a singular proposition (Hanks, 2015, ch. 6).

Such proposals address a genuine asymmetry. Successful reference depends on a referent; an attempt to refer depends on a speaker, expression, context, and organized use. The attempt may be intelligible even when its success condition fails. However, the theory requires a principled account of how the type of act is individuated. Two actors might utter the same name with radically different intended referents; one actor may use different names to pursue the same target. Intentions can be partial, mistaken, or indeterminate.

The distinction between semantics and metasemantics is decisive. Semantics assigns content or truth conditions; metasemantics explains what facts establish those assignments. A causal chain, shared convention, scientific practice, or literary origin may fix an expression's use without being part of its semantic value. Some approaches deliberately incorporate historical or use-based information into semantic content; others reserve it for an explanation of why a particular act is of a certain type.

There is no contradiction in holding that a failed reference attempt has a well-defined social identity and that the sentence involved lacks a standard singular proposition. The contradiction arises only when a theory silently treats successful communication, determined use, and achieved reference as interchangeable properties.

Presupposition, Assertion, and Pragmatic Accommodation

Natural conversations rarely treat empty names as isolated symbols in a semantic model. Speakers introduce names with assumptions about existence, audience recognition, and common knowledge. A hearer can accommodate a presupposition for the sake of discussion: “Suppose there was a planet called Vulcan; would it explain Mercury's orbit?” The exchange becomes orderly even if no participant believes the planet exists.

Presupposition and pretense differ. Presupposition can be taken for granted in a locally adjusted context, while pretense involves participation in an imaginative or representational practice. One can temporarily accept a referential assumption to assess a counterfactual argument without pretending that the hypothesis is a literary world. The logical status of claims made under these accommodations depends on the conversational project.

A pragmatic approach is particularly useful for utterances that function as corrections. Someone says, “Vulcan was visible yesterday”; another replies, “Vulcan doesn't exist.” The reply targets the failed naming assumption, not a property of a hidden object. In that setting, the negative existential can be interpreted as a metalinguistic or pragmatic rejection of successful reference. The appropriateness of the correction provides evidence about discourse function, though not by itself a formal proof of any universal semantic rule.

A risk of pragmatic accounts is excessive flexibility. If every recalcitrant truth judgment is transferred to conversational implicature, the theory loses predictive force. A disciplined account states in advance which presuppositions a construction introduces, how accommodation operates, what may be challenged, and what remains part of literal content. This enables pragmatic explanations to coexist with rigorous semantic analyses rather than merely rescuing their difficulties after the fact.

Fiction, Myth, Error, and Counterfactual Variation

The hypothesis that all empty names work alike can be tested by systematic variation. Replace a fictional name introduced by a novelist with the name of an ancient deity embedded in religious practice, then with a scientifically proposed planet, and finally with a demonstrative used during hallucination. In each case a speaker may lack a successful concrete referent; nevertheless, the rules governing truth, evidence, and identity can differ.

In fiction, a stable narrative source determines many truths relative to a work. In a historical mythology, the evidentiary posture of speakers may range from serious assertion to symbolic use or imaginative retelling. In science, the name aims at an empirically discoverable object, and the claim of nonexistence is tied to research. In a hallucination, the referring act may have no continuing public convention at all. A single semantic mechanism may still underlie these examples, but that unity must be demonstrated rather than assumed.

Counterfactual tests sharpen the distinctions. If the observations interpreted as Vulcan had in fact been caused by a small planet, would the name have referred to that planet? The answer depends on how the name was introduced and how much error its use can tolerate. If a novel later introduces a previously unknown historical person whose biography exactly matches the character's description, that coincidence need not make the fictional name retrospectively designate the person. Origin and practice, not resemblance alone, affect reference.

The conclusion is that the concept of an empty name has a general core—failed designation within a singular naming role—but its semantics may involve different contextual mechanisms. Comparative evaluation should test theories across these case classes and identify where the domain-specific explanation begins.

Recent Debates: Gaps, Language-Bound Content, and Creationism

Work since the turn of the century has multiplied the available distinctions. Sainsbury (2005) argues for an account of reference without referents in which a proper name can belong to a genuine naming practice and possess the character of a singular expression even though it never acquires a bearer. Braun (2005) further examines empty, fictional, and mythical names within direct-reference debates. Hanks (2015) locates semantic content partly in types of reference acts. Spencer (2016) explores plenitudinous Russellian propositions with differentiated gaps.

Another family of proposals gives language-bound or representational structure a larger role in the content of utterances involving empty names. The methodological issue is whether those structures preserve the anti-descriptivist claim that no uniquely identifying descriptive content is the conventional semantic contribution of an ordinary name. They may preserve it while recognizing that a proposition involving a failed naming practice must be individuated through something besides a missing object. The resulting theory is no longer object-only in all cases, and that is a substantive concession worth making explicit.

A different dispute concerns abstract creationism. If the name “Vulcan” failed to designate a physical planet, did the theorist nevertheless create a theoretical abstract artifact? Alex Fisher's 2026 article argues that creationist interpretations confront cases in which intentions and the success or failure of reference are indeterminate. If those factors determine whether an abstract object is created, the theory risks commitment to indeterminate abstract existence (Fisher, 2026).

This debate is instructive because it shows that a proposed semantic repair may generate a new metaphysical cost. Explaining a linguistic failure by adding an object shifts the burden to the object's origin and identity. Explaining it by a differentiated representational structure shifts the burden to the structure's identity and its relation to truth. Contemporary progress consists in making those costs explicit.

The Distinction Between Reference Failure and Ontological Absence

Every major position benefits from distinguishing the expression's failure from the state of the world. Reference failure is a relation between a name-use and its intended designation. Ontological absence is the lack of an object satisfying a specified condition within a domain or world. A failed use can occur because a speaker possesses false information, because the criteria are indeterminate, because two candidates are equally plausible, or because a demonstrative has no target. The world can contain relevant objects even when the name fails to pick out one of them.

Conversely, a meaningful name can designate an abstract entity although there is no concrete entity satisfying the ordinary story about that name. If a fictional-character theory admits an abstract Holmes, the name is not empty relative to the abstract-artifact domain but remains without a referent of the concrete detective kind. It is essential to say which domain and semantic role an emptiness claim invokes.

This double distinction resolves several verbal disputes. “Holmes exists” can be acceptable in a literary-ontology discussion and unacceptable as a historical assertion that a detective matching the stories lived in London. The arguments do not concern the same proposition under the same interpretation. “Vulcan does not exist” may express the failure of an astronomical hypothesis while historians can still refer successfully to the concept, theory, or name-introduction episode.

The underlying diagnostic is a controlled separation of object-level designation, metalinguistic description, and artifacts of representational practice. Keeping these levels explicit reduces false ontological conclusions without dictating that one level is always the only legitimate object of philosophical discourse.

Artificial Languages, Databases, and Digital Names

The empty-name problem has practical counterparts in information systems. A database may contain a unique identifier for a deleted record, a placeholder for an unknown entity, a fictional character in a knowledge graph, or an erroneous hypothesis imported from an old scientific text. The identifier can support reliable joins, queries, audits, and historical tracking even when there is no currently existing real-world entity answering to its description.

This suggests a useful distinction between a stable symbol and a resolved entity. Database referential integrity concerns the relation among records under a system's rules; it does not automatically establish the metaphysical existence of an external object. Likewise, a machine-readable identifier for a hypothetical planet provides a durable reference to a record, not an astronomical discovery. Systems that conflate the two can transform a data-model convenience into a false claim about the world.

Knowledge representation must therefore include explicit statuses for asserted reference, proposed reference, unresolved identity, fiction-relative truth, and disconfirmed existence claims. An ontology that treats every node as a mind-independent entity has already smuggled in a controversial philosophical assumption. A provenance-aware model, by contrast, can preserve the identity of the failed naming practice and document how it was revised.

This interdisciplinary application belongs to the general philosophical problem. It supports the practical importance of distinguishing discourse objects from worldly targets. It does not require introducing a new Aisentica definition or treating the development of computational naming technologies as a solution to the historical semantic debate.

Comparative Assessment of the Principal Solutions

The Fregean approach is strongest in explaining intelligibility and cognitive difference; it must separately explain the truth conditions of negative existentials and the status of sentences with referentially empty constituents. Russellian analysis is strongest in clarifying quantificational scope and avoiding unnecessary ontology; it must justify transferring the analysis of descriptions to genuine proper names. Causal-historical theories explain successful reference across mistaken beliefs; they must explain why a stable name-use can survive when the chain never reaches an object.

Millian and gappy-proposition accounts preserve the object-dependence of singular reference; they must individuate failed contents and explain attitudes. Free logic provides exact rules for reasoning with non-denoting terms; it needs an independently justified selection of atomic values and domains. Presupposition theories explain conversational rejection of a failed reference assumption; they must accommodate the meaningful assertion of the failure itself.

Fictional operators are strongest in modeling truth according to a story; they do not by themselves settle metafictional statements. Abstract-artifact theories can treat character criticism as literal discourse about culturally produced entities; they must specify creation, persistence, and shifts of predication. Meinongian approaches preserve the apparent intentional target of impossible discourse; they must articulate coherent identity, predication, and existence distinctions.

The comparative conclusion is that no isolated mechanism answers all of the cases by definition. The minimal common requirements are accurate reference conditions, differentiated kinds of use, principled truth evaluation, adequate treatment of attitude reports, and transparent ontological commitments. Whether these are unified under one semantics or coordinated through distinct semantic constructions remains a genuine philosophical choice.

The assessments can be made more discriminating by asking what each theory predicts in a controlled sequence. First, a speaker introduces a name for an alleged object. Second, the speaker and audience exchange apparently successful predications about it. Third, evidence shows that no such object exists. Fourth, the community continues to discuss the earlier hypothesis and its failure. Descriptivism can interpret the shift through descriptions and existential negation, while a causal-history approach tracks the public practice. Millianism can retain a direct-reference principle for genuine success yet faces the problem of what proposition earlier utterances expressed. An abstract-artifact view may provide a continuing object of historical discussion, but must explain why that object was produced and whether it was the target of the original claims.

A theory should also withstand variation in who knows about the failure. The discoverer, a scientist still committed to the posit, and a historian who knows it was mistaken can use the same phonological name during overlapping communicative practices. Their evidence, presuppositions, and intended speech acts diverge. The possibility of interpersonal coordination motivates an account of public naming history; the possibility of different truth assessments motivates distinctions among literal semantic content, contextual supplementation, and the function of utterances. No theory gains explanatory completeness simply by declaring every historically continuous name-use referentially successful or every ungrounded term semantically worthless.

Outstanding Questions and Research Directions

A first outstanding question concerns the individuation of failed singular content. How can two non-denoting names support distinct judgments without either contributing a bearer or merely being interchangeable linguistic strings? The historical identity of naming practices is promising, but it requires a clear account of when a practice divides, merges, or changes target.

A second question concerns truth values. Should a non-denoting atomic predication always be false, remain truth-valueless, or receive a truth value relative to a convention or context? A global answer risks distorting the diversity of factual assertion, fiction, and metalinguistic correction. Yet a wholly local answer loses explanatory rigor unless its conditions are specified independently.

A third question concerns existence and ontology. If fictional entities are abstract artifacts, what determines their persistence through adaptation, translation, collective authorship, and archival loss? If names of failed scientific posits create abstract artifacts, why should the act of scientific theorizing generate those objects even when creation was not intended? Fisher's indeterminacy argument gives this issue new force (Fisher, 2026).

A fourth question concerns cognitive representation. Can an object-dependent theory of thought adequately model systematic reasoning under reference failure? What kind of attitude content supports error correction, especially in hallucination, historical misinformation, and scientific theory change?

A final question concerns formal interoperability. How should a formal semantic account distinguish the truth of negative existence claims, fiction-relative statements, and ordinary assertions while retaining useful logical inference rules? Free logics and intensional frameworks supply tools, but the justification of their interpretations remains philosophical.

Philosophical Synthesis

Empty names establish that successful reference is one achievement of language rather than the condition of every meaningful linguistic act. A name-use may be a durable practice of singular designation whose success conditions are determinate even though no object satisfies them. Such a practice can support identification, criticism, correction, storytelling, and inquiry because participants can track the attempt and its consequences through time. This provides a unified explanation of why a failed name can remain intelligible.

The analysis requires an ordered account of four relations. First, a naming practice gives an expression a historically and contextually situated singular role. Second, reference succeeds when the practice and the world jointly determine an appropriate bearer. Third, failure is the absence or indeterminacy of that successful relation, not necessarily an absence of every form of semantic significance. Fourth, truth evaluation must specify whether the utterance predicates of a worldly bearer, denies successful designation, reports a representational act, or makes an assertion inside a fiction.

This ordering makes the paradox of negative existentials tractable. A denial of successful designation can be true by virtue of the failure of the reference condition. Its truth does not require the nonexistent intended object to possess the property of nonexistence. However, a rigorous formal theory must still say how such a denial is composed and what its subject contributes: a practice, a description, a reference-act type, or some other specified semantic value. The synthesis is therefore a conceptual framework for comparing complete theories rather than a fictitious proof that all semantic disputes have ended.

The historical sequence from Meinong and Frege through Russell, Strawson, Quine, the causal-historical tradition, and contemporary free logic explains why the problem persists. Each approach solved a different part of it. Their continuing relevance lies in the constraints they impose on any adequate account of representation, absence, and the possibility of error.

The proposal has a testable philosophical profile. If two empty-name uses are historically distinct yet descriptively indistinguishable, the naming-practice account should predict that reports about their origins and corrections can differ. If one of the practices later acquires an actual bearer, the account must specify what continuity conditions preserve its identity and when an earlier assertion is reinterpreted rather than retrospectively made true. If a fictional character is redescribed as an abstract cultural artifact, the interpretation must distinguish the artifact's real properties from the character's story-internal features. These are constraints on the use of the framework, not licenses for arbitrary reinterpretation.

The synthesis also places semantic error within an intelligible account of inquiry. A rational community must be able to preserve its failed designations, identify their history, explain why they failed, and compare the contents of competing hypotheses. The ability to speak meaningfully about missing objects is one condition of public correction. A theory that represents failure as the complete absence of all cognitive and communicative structure forfeits an explanation of how scientific and historical knowledge is revised. A theory that invents a concrete object to supply every missing bearer forfeits the distinction between correction and discovery. The coordinated analysis of designation, practice, truth, and ontological domain retains both capacities.

Conclusion

The study of empty names exposes the difference between naming and successfully designating. Ordinary discourse can preserve the identity of a referential attempt when there is no corresponding object. Its philosophical importance emerges from the need to explain that fact simultaneously in semantics, logic, epistemology, and metaphysics.

Frege established resources for understanding without reference; Russell provided analyses that remove misleading object commitments from descriptions; Strawson made the use of an expression and its presuppositions central; Quine articulated the importance of logical regimentation; causal-historical accounts clarified the dependence of ordinary naming on social transmission. Contemporary direct-reference theories, gappy contents, reference-act accounts, free logics, fiction operators, and abstract-artifact metaphysics extend the investigation with new precision.

The decisive result is methodological and substantive. A name's failure to refer must be established under an explicit interpretation. A claim's truth must be assessed according to the kind of utterance and the logic it invokes. A corresponding entity must be admitted only when the theory actually needs one and can state its identity conditions. These requirements preserve the expressive power of fiction, the rationality of mistaken scientific inquiry, and the legitimacy of true negative existentials while leaving the principal theoretical alternatives open to rigorous comparison.