Metaphysical Realism and Anti-Realism

The Contested Thesis of an Independent World

The central dispute concerns the relation between reality and the conditions under which reality is represented. A metaphysical realist maintains, in the relevant domain, that what exists and how it is are not constituted by our judgments, descriptions, beliefs, or discoveries. An anti-realist challenges some component of that independence, often by arguing that truth, objecthood, or the bounds of meaningful assertion cannot be articulated apart from human or other cognitive practices. This formulation distinguishes two questions that philosophical discussions repeatedly join: what constitutes the world and what warrants our assertions about it.

Consider a mountain that existed before the emergence of human observers. The realist's immediate claim is that its existence and geological processes did not await its description. Yet the word "mountain," the criteria that distinguish a mountain from a hill, and the practice of treating a geological formation as one countable object are products of conceptual organization. The first fact supports a straightforward account of temporal and causal independence. The latter facts complicate the thesis that reality is composed of one uniquely privileged inventory of all objects. Recognizing the complication does not require attributing the mountain's geological history to linguistic activity.

The example discloses a crucial division. An independent causal order may constrain true descriptions, while multiple admissible schemes count and classify that order differently. A realist who argues for the former cannot automatically infer the unique correctness of every ontological partition. An anti-realist who establishes the dependence of classification on conceptual practice cannot automatically infer that the classified processes depend on that practice for their existence. The movement from one claim to another requires an additional premise, which becomes a principal object of examination throughout this study.

The contemporary expression "metaphysical realism" has several meanings. Hilary Putnam's influential formulation ties it to a fixed totality of mind-independent objects, a uniquely complete true description, and correspondence understood from an external vantage point (Putnam, 1981, chap. 3). Other realists defend mind-independence while rejecting the ideal of a single maximally complete description. Contemporary accounts therefore distinguish general realism from the particularly demanding package targeted in Putnam's criticism. The Stanford Encyclopedia of Philosophy's discussions of realism and challenges to metaphysical realism document this plurality; the philosophical work here is to identify the premises on which each version depends.

The problem is also historically expansive. Plato's Forms, Aristotle's substances, medieval universals, Berkeley's ideas, Kant's appearances, the logical empiricists' frameworks, and contemporary theories of grounding do not all address the same proposition. They nevertheless illuminate recurring relationships among being, thought, language, and objectivity. An adequate history traces both continuity of problems and the transformation of the terms in which those problems become intelligible.

Realism's Conceptual Grammar

Realism is ordinarily indexed to a subject matter. Realism about ordinary physical objects asserts that such objects exist independently of beliefs about them. Realism about universals commits to repeatable features or kinds in a sense denied by nominalism. Moral realism often combines a claim about moral facts with theses concerning their truth or independence from endorsement. Scientific realism takes the theoretical descriptions of successful science to disclose at least some unobservable structure. These claims require different arguments; success in one domain offers no immediate proof in another.

Four dimensions should be distinguished. Existential independence concerns whether an entity exists without being thought about. Constitutive independence concerns whether its identity and nature are fixed by factors other than representational practices. Alethic independence concerns whether the truth of propositions depends on our ability to recognize or verify them. Epistemic independence, a potentially misleading label, concerns whether the criteria used to justify claims are themselves independent of the methods and capacities of knowers. Independence in the first sense need not entail independence in the fourth. Indeed, any justification available to us must be pursued through some cognitive and evidential means.

A fifth dimension is explanatory priority. One may argue that facts explain the success and failure of our classifications while acknowledging that classifications furnish the conceptual vocabulary in which those facts are reported. An opposing constitutive account may hold that some objects, notably legal institutions or currencies, owe their existence to historically specific rules and practices. Here dependence is an ordinary feature of what the objects are; the existence of practice-dependent objects is compatible with an independently existing physical and social history in which the practices arise.

Truth requires separate treatment. The correspondence family of accounts links truth to how things stand independently of an assertion. Minimalist and deflationary accounts challenge the need for a substantial correspondence property without thereby denying mountains, stars, or other independently existing entities. Semantic realism in Michael Dummett's sense concerns whether the meaning of sentences can be specified by potentially recognition-transcendent truth conditions. Its opponent can accept a robust physical world while disputing the relevant account of linguistic understanding (Dummett, 1963/1978). Thus, semantic anti-realism and metaphysical idealism are distinct positions.

Objectivity itself also admits degrees and kinds. An objective judgment may be one supported by evidence available to different investigators, one whose truth is independent of assent, one invariant under transformation of standpoint, or one that correctly tracks an independently structured world. These notions overlap without becoming equivalent. Theoretical discussion becomes clearer when a claim to "objectivity" says which form is at issue and explains why the associated norms are justified.

The central analytic discipline is consequently to resist inflation: from the independence of some phenomena to the independence of all categories; from the pluralism of descriptions to the unreality of their objects; from the limitations of verification to the indeterminacy of truth; or from methodological dependence to universal ontological construction. These are substantive inferences, each requiring an argument.

Ancient Foundations: Being, Forms, and Substance

Ancient philosophy posed the question of what is real through disputes concerning change, appearance, knowledge, and the principles of being. Parmenides' argument that thought and being cannot straightforwardly be separated challenges accounts that treat what-is as subject to generation and disappearance. The surviving poem, conventionally cited as fragments in Diels–Kranz numbering, must be interpreted against the distinction between the way of truth and the opinions of mortals. The text is fragmentary, and the interpretation of its ontological commitments remains contested (Palmer, 2025; historical textual overview). It is an anachronism to attribute to Parmenides a modern correspondence theory; his importance lies in making the intelligibility of being a problem for reason itself.

Plato differentiates stable intelligible objects of knowledge from changing perceptible particulars. In Republic VI–VII, particularly the divided line at 509d–511e and the cave at 514a–517a, the distinction between intelligible and visible orders connects levels of cognition with different modes of intelligibility. In the Phaedo and other dialogues, Forms function as standards or explanatory grounds for properties and judgments. A Platonic commitment to realities not reducible to any individual act of perception is a substantial historical source for realism, especially in the dispute over universals. Yet his project is not simply the claim that everyday physical things persist when unobserved; the hierarchical relation between forms and sensibles complicates such an assimilation.

Aristotle reworks the problem through inquiry into substance, form, matter, potentiality, and actuality. Metaphysics Γ investigates being as being, including the principle of noncontradiction at 1005b19–20; Metaphysics Ζ examines what makes something the entity it is; the Categories differentiates primary substances from what is predicated of them. His account treats determinate beings as objects of inquiry, but the explanation of their identities is internal to a theory of substance and form rather than a contemporary theory of mind-independent description. The challenge is to explain what in things makes predication, individuation, and change possible.

The ancient contribution is twofold. First, the question of reality already includes a problem of explanatory structure: appearances must be understood through what makes them intelligible. Second, the philosophical demand for knowledge arises from a world whose determinations are not settled by anyone's preferences. This combination makes Plato and Aristotle relevant to later realism even where their metaphysical inventories sharply diverge. The continuity lies in the search for determinants of truth and being, not in an unchanged doctrine extending from antiquity to the present.

A further caution is important for comparative scholarship. Distinctions between being and appearance have analogues in several philosophical traditions, but their expressions in Greek, Sanskrit, and classical Chinese texts belong to different argumentative contexts. The general inquiry into the relation between appearance and reality is cross-cultural; it does not authorize translating every doctrine of ultimate reality directly into the analytic realism/anti-realism opposition. A wider comparison requires close attention to each tradition's own forms of argument.

Medieval Realisms and the Scope of Independence

In medieval philosophy, "realism" characteristically concerns universals: are common natures or repeatable forms grounded in reality, and in what way do they exist? This debate inherits problems from Aristotle's Categories and Porphyry's Isagoge, then develops through Boethius, Abelard, Aquinas, Scotus, and Ockham. The historical label therefore denotes a more focused issue than the contemporary dispute over the mind-independence of the world as a whole.

A strong realist takes common natures to possess some status in reality not exhausted by acts of conception or by linguistic predicates. A nominalist resists multiplying entities beyond particulars and the means by which they are signified. Intermediate views distinguish a nature as existing in individual things from its universality as conceived by intellect. These alternatives make a general lesson visible: the existence of independent particulars does not by itself decide whether universals are independent entities. A philosopher may be a realist about stones while remaining a nominalist about stoniness.

Aquinas' account of intellectual abstraction situates the intelligibility of natures in material things and the activity of the intellect (*Summa theologiae* I, q. 85, a. 1); his treatment of divine ideas belongs to a distinct theological explanation of archetypes (I, q. 15). Scotus's account of common nature and formal distinction raises different questions about unity and individuation. Ockham's semantic and ontological analyses articulate nominalist economies without eliminating the reality of individual entities. These differences are carefully reconstructed in Klima's historical survey (2022) and should be investigated in their particular texts rather than collapsed into three fixed schools; the terms "realism," "nominalism," and "conceptualism" designate families of interpretations, and medieval positions are sensitive to the form of the question asked.

The medieval history also clarifies an ambiguity of dependence. A universal conceived as universal may depend on intellectual comparison; its instances can still exist independently of that comparison. An account that distinguishes the mode of conception from the mode of existence is conceptually different from a theory that reduces objects themselves to ideas. Later controversies over whether reality possesses an independently determinate classification reproduce part of this distinction at a new logical and semantic level.

For the present dossier, the medieval material establishes the genealogy of disputed ontological articulation. The full textual history of the universals controversy belongs to its own research subject; here its role is to demonstrate that ontological realism must specify which entities and modes of existence it defends.

Early Modern Experience and the External World

The early modern debate transforms the problem by placing representation, certainty, and the causal relation between mind and world at its center. Descartes' Meditations (1641), especially the First, Second, Third, and Sixth Meditations, stages systematic doubt concerning sensory belief, establishes the certainty of the thinking subject in the very act of doubt, and attempts to restore warranted belief in an external material world. The skeptical possibility of indistinguishable appearances reveals that certainty about present experience and certainty about its external causes have different logical structures.

Descartes does not conclude that external bodies are mere thoughts. His epistemic route through ideas and divine nondeception is intended to justify claims concerning material substance. The problem thereby becomes acute: how can states immediately available to consciousness secure reference to what is independent of that consciousness? Later philosophy inherits this epistemological challenge even when it rejects the theological component of Descartes' solution.

Locke's Essay Concerning Human Understanding (1690), particularly Book II, chapter viii, distinguishes primary from secondary qualities. Solidity, extension, and figure are given a role different from color, taste, and sound as experienced. His distinction is often used to support a qualified representational realism: our ideas may vary according to perceivers while features of bodies have a reality that explains sensory production. The challenge is to explain how the alleged resemblance or correspondence of representation to independent qualities can be known.

The argument from perceptual variation exposes the difficulty. What looks red in one illumination may look brown in another; a distant object appears small, then larger upon approach. Variation does not entail that there is no object. It invites a theory separating the character of an appearance from the underlying causal structure. A realist interprets the variation as evidence for a structured environment and perceptual processes; an idealist can interpret it as evidence that sensory qualities exist only as experienced. The shared data underdetermine the metaphysical conclusion until an account of perception and explanation is supplied.

Modern perceptual science strengthens the need to distinguish world-dependence from representational mediation. Signals are filtered, transformed, and integrated by biological sensory systems. The dependence of perceived color or spatial appearance on these processes can coexist with independent electromagnetic, chemical, and spatial conditions. An anti-realist who emphasizes perceptual construction therefore raises a real epistemic challenge; a realist who appeals to causal constraints must explain how reliable reference and correction emerge from fallible representations.

Berkeley: Immaterialism and the Reality of Experience

George Berkeley's A Treatise Concerning the Principles of Human Knowledge (1710), especially Part I, §§1–7 and §§18–24, attacks the philosophical posit of material substance existing independently of every mind. For Berkeley, sensible objects are complexes of ideas, and an idea's existence belongs to its being perceived. He also recognizes active perceiving spirits, which are not themselves merely passive ideas. His immaterialism consequently requires an ontology of minds and their ideas, with divine perception underwriting the order and continuity of experience.

Berkeley's argument from resemblance denies that an idea could resemble an unperceived material archetype in a way accessible to the perceiver. His attack on abstraction questions whether we can form a coherent idea of an object stripped of all sensible qualities. If the alleged material substrate is presented only as the unknowable support of sensible qualities, Berkeley argues that the explanatory purpose of the posit becomes obscure (Berkeley, 1710, §§3–4, 8–24). These are philosophical arguments about what can be conceived and explained, not a report that the experienced world is chaotic or imaginary.

Berkeley's strongest point concerns the explanatory cost of postulating an inaccessible material substratum while claiming that all immediate content of perception is mental. His weakest transition is from the premise that we directly apprehend ideas to the conclusion that a mind-independent material world is incoherent. A realist may argue that mediated knowledge remains knowledge of a cause, that resemblance is not the only possible relation between perception and object, and that the independence of a thing does not require a perceiver to form an image of its independence.

The familiar objection asks whether the world vanishes when human observers leave a room. Berkeley's answer appeals to divine perception; it would misrepresent him to attribute that consequence as a thesis he accepted. The deeper point is that the relevant independence of objects is reinterpreted through a metaphysical account in which finite human experience is embedded within an ordered divine activity. The disagreement concerns the ultimate bearers of reality, not the practical possibility of ordinary distinctions between illusion and veridical experience.

Berkeley also illuminates a recurring philosophical hazard. A thesis about the conditions under which objects are known may be combined with a thesis about the conditions under which objects exist. The combination is substantive. The realist's reply must give an account of access; the idealist's reply must justify the move from perceptual immediacy to ontological dependence. Subsequent theories refine these competing burdens rather than eliminating them.

Kant: Empirical Reality and Transcendental Conditions

Kant's Critique of Pure Reason appeared in 1781 and was substantially revised in 1787. It transforms the realism problem by asking how objects of possible experience can be known under universal and necessary conditions. The Transcendental Aesthetic argues that space and time are forms of human sensible intuition. The Transcendental Analytic relates objective experience to the categories of understanding; the Transcendental Deduction treats the conditions under which representations can belong to the cognition of objects. These arguments distinguish objects as appearances from things considered independently of our mode of cognition.

Kant's formula joins empirical realism to transcendental idealism. Empirical objects are objects in space and time, subject to public experience and empirical investigation. Their reality does not fluctuate with individual wishes. Yet the spatial and temporal form of their appearance belongs to the conditions of finite sensibility rather than being established as a determination of things in themselves. In the second edition's Refutation of Idealism (B274–279), Kant argues that determinate temporal consciousness requires experience of something persistent in space. This is a response to Cartesian doubts about external objects, not a defense of Berkeley's immaterialism (Kant, 1781/1787, B274–279; see also Kant's Transcendental Arguments).

The resulting position is structurally different from a global claim that reality is produced by individual cognition. Kant places the possibility of objectively valid experience within conditions supplied by the knowing subject, while treating the distinction between appearance and thing in itself as a limit on claims to know reality apart from those conditions. Exactly how to interpret things in themselves remains disputed. A two-aspect reading construes the distinction as contrasting ways of considering the same things; a two-world reading assigns greater independence to the noumenal and phenomenal domains. The choice changes the strength of Kant's departure from metaphysical realism.

Kant's reasoning can be reconstructed as follows. Objective experience requires a unified spatiotemporal order and rules for synthesizing representations. These rules are necessary for the kind of cognition we possess. Therefore claims concerning objects of possible experience are valid under those conditions, while the conditions cannot simply be projected as cognized properties of things considered apart from possible experience. The premises concern the structure of cognition; the conclusion concerns its legitimate scope. The controversial step is the assertion that necessary features of our experience must originate in our faculties rather than reflect features of things independently. Later Kantians and realists debate precisely that inference.

The lesson is a distinction between empirical independence and transcendental conditions. A scientific object can be independent of the preferences of any particular scientist yet be knowable only under conditions of spatiotemporal representation and conceptual judgment. To conflate these two levels is to obscure both Kant's realism about empirical inquiry and the force of his critical restriction on metaphysical knowledge. Kant is therefore a major source of the modern debate about conceptual contribution without being reducible to either side of the simplest realism/anti-realism division.

After Kant: Idealism, Objectivity, and Common Sense

Post-Kantian idealists challenge the stable separation of determinate being from self-conscious activity. Fichte treats the activity of the I as central to the constitution of experience; Schelling develops accounts of nature and spirit whose relationship cannot be captured by a merely passive knower encountering a complete inventory; Hegel places the intelligibility of determinate objects within a dialectical account of conceptual development. Their differences are decisive: the expression "German idealism" identifies a historical field of debate, not one uniform thesis that everything is a private mental image.

Hegel's Phenomenology of Spirit (1807) traces transformations in forms of consciousness, while his Science of Logic (1812–1816; later revised) examines categories of being, essence, and concept. A Hegelian response to simple metaphysical realism asks whether the determination of an object can be intelligible independently of the relations through which the object is identified and explained. The argument concerns the conceptual articulation of determinacy. A realist can accept that concepts are relationally organized while questioning the inference that being itself must be conceptually constituted in the strongest sense.

British idealism, represented by figures including F. H. Bradley and Bernard Bosanquet, develops a further critique of the adequacy of isolated relations and finite appearances to ultimate reality. Bradley's Appearance and Reality (1893) disputes whether ordinary relational thought delivers the whole structure of what is real. This discussion belongs to a distinctive metaphysical account of the Absolute. It should not be read as a twentieth-century linguistic relativism transferred backward in time.

At the turn of the twentieth century, G. E. Moore's "The Refutation of Idealism" (1903) attacks a central inference from the presence of consciousness in experience to the identity of its object with consciousness. Moore's defense of common sense and later "Proof of an External World" (1939) treat ordinary knowledge of material objects as a legitimate starting point for philosophical analysis. The strategy reverses the skeptical burden: philosophical premises producing denial of everyday knowledge may be less secure than the ordinary judgments they are supposed to defeat.

Bertrand Russell's The Problems of Philosophy (1912) distinguishes sense-data from physical objects and asks what reason we have to infer an external world. His appeal to the simplicity and explanatory strength of positing enduring physical objects retains a representational difficulty without endorsing idealism. Moore and Russell demonstrate that realism can operate both as a methodological confidence in ordinary existence claims and as a metaphysical theory requiring further articulation. The first can resist skeptical conclusions before the second has supplied a full account of reference.

These developments mark an enduring change. Philosophers increasingly ask whether arguments from the dependence of knowing on concepts or consciousness illicitly transform epistemic claims into claims of ontological constitution. Yet the counterclaim requires care: a realist cannot merely assert that independence is obvious when the relevant issue concerns the nature, individuation, or truth conditions of what is claimed to exist.

Truth and Verification: Dummett's Reframing

Michael Dummett's essay "Realism" (1963; collected in Truth and Other Enigmas, 1978) proposes that disputes traditionally described as realism versus anti-realism can be investigated through the theory of meaning. What changes is the unit of analysis. Rather than first asking whether a certain class of objects exists, Dummett asks what competence in understanding statements about that class involves. A realist understanding takes their truth conditions to outrun our capacities for recognizing whether those conditions obtain. An anti-realist account connects meaning more intimately with demonstrability, verification, or warranted assertibility.

The famous mathematical example of a presently undecided proposition reveals the stakes. If a sentence is meaningful and has a determinate truth value despite the absence of any method guaranteed to find it, then semantic realism permits recognition-transcendent truth. An anti-realist who insists that meaning must be manifested in recognizable use challenges the intelligibility of this gap. The debate bears on bivalence and logical laws but is not identical with the claim that a physical object disappears when unobserved.

The anti-realist argument may be reconstructed in four moves. Understanding an expression is a practical competence whose possession must be manifest in use. A truth condition that could in principle never be recognized seems to make an essential aspect of understanding inaccessible to any manifestation of that competence. Meaning should instead be explained in terms of publicly accessible abilities and evidential conditions. Consequently, unrestricted realism about truth conditions becomes suspect in domains where decisive verification is unavailable. Each transition requires defense; the key dispute is whether manifestation requires finite verification of truth or only the ability to reason with and apply the expression under a wider range of circumstances.

A realist reply separates knowing a truth condition from knowing whether it is satisfied. We can understand a proposition while remaining unable to settle it, just as we can understand claims about remote historical events whose decisive traces have disappeared. Public practice may fix the role of a sentence sufficiently for it to have a truth value, even if no one can determine that value. The challenge then returns: by what account of language use is this distinction intelligible rather than stipulative?

Dummett's framework also shows why anti-realism varies by domain. One might accept classical truth conditions for everyday objects but advocate intuitionistic restrictions in a fragment of mathematics, or propose verification-oriented semantics for statements about the remote past. Any attempt to infer universal metaphysical idealism from domain-specific semantic arguments exceeds their immediate force. Conversely, a realist who treats bivalence as self-evidently appropriate for every discourse must engage arguments concerning vagueness, future contingents, and mathematical proof.

The central result is that realism can be tested as a thesis about the relationship between truth and evidence without first resolving every question concerning the ultimate ontology of the world. This semantic turn broadens the debate and also creates the risk that the word "realism" shifts among multiple distinct projects. The updated SEP Realism entry is especially useful for keeping these forms apart.

Language Frameworks and Ontological Commitment

Rudolf Carnap's "Empiricism, Semantics, and Ontology" (1950) distinguishes internal questions asked within a linguistic framework from external questions concerning the adoption of the framework itself. Inside an ordinary thing-framework, whether there is a desk in a room is a factual question answerable through inquiry. Inside mathematics, whether there are prime numbers with specified properties is answered by appropriate methods. Carnap is skeptical of purportedly factual external questions that demand the reality of a whole framework's entities from a standpoint beyond all rules of description (Carnap, 1950).

Carnap's proposal is frequently called ontologically tolerant or deflationary. It does not claim that desks are created by saying that they exist. It distinguishes the practical choice of a representational scheme from the factual inquiries pursued within it. The anti-metaphysical conclusion depends on a semantic diagnosis: that an external question, when detached from framework-bound criteria, lacks the cognitive content its traditional formulation appears to promise. The realist may challenge that diagnosis while accepting the importance of explicit criteria.

W. V. Quine's "On What There Is" (1948) offers a contrasting approach. A theory's ontological commitments are clarified by analyzing what its bound variables must range over when the theory is regimented in a canonical notation. Quine's slogan concerning the values of variables is best understood as a criterion of commitment rather than a proof that the entities quantified over exist. Whether a theory is true, whether its quantifications are indispensable, and what objects satisfy them remain further substantive questions (Quine, 1948).

The Carnap–Quine controversy places a methodological issue at the center of the realist debate. If changing a framework changes what sentences count as meaningful, then some apparent existential disputes may concern language choice. If the best overall theory is assessed holistically and involves indispensable quantification over certain entities, ontological inquiry can reemerge as continuous with theory choice. Quine's challenge to a sharp analytic–synthetic distinction complicates Carnap's effort to distinguish merely framework-internal truths from practical external decisions.

Contemporary metaontology extends these questions. Eli Hirsch's quantifier variance suggests that some disputes about what exists may reflect equally legitimate uses of existential expressions. Theodore Sider, by contrast, emphasizes that reality may possess an objective structure and that some forms of quantification may carve closer to its joints than others. The dispute is therefore more demanding than asking whether participants have different dictionaries. It concerns whether there are objective constraints on the most fundamental ways of describing and quantifying over the world.

The analytic payoff is a distinction between verbal disputes, substantive ontological disputes, and mixed cases. Two theorists may agree on every empirical fact and differ only about whether a complex arrangement counts as one additional entity. They may also differ about genuine causal or modal features of that arrangement. Resolving which kind of disagreement is present calls for a comparison of inferential commitments, explanatory roles, and the proposed dependence relations, not a slogan concerning the emptiness or importance of metaphysics.

Conceptual Schemes and Their Critics

The concept of a conceptual scheme expresses the possibility that cognition organizes experience according to different categories, predicates, or rules of individuation. It becomes an argument against metaphysical realism when scheme plurality is taken to undermine the idea of a uniquely structured, scheme-independent world. Nelson Goodman's Ways of Worldmaking (1978) develops a striking form of pluralism, treating versions and their criteria of rightness as central to the construction of worlds. The philosophical challenge concerns how conflicting but adequate world-versions can be related.

A modest scheme-pluralist notes that different descriptions partition the same field of investigation differently. A geological region can be treated as one mountain, several peaks, a material aggregate, or a stage in a process. These are context-dependent individuations. The stronger metaphysical thesis says that there is no further fact concerning which entities or structures exist independently of admissible conceptual arrangements. To move from the modest to the stronger thesis one must show that no scheme-independent constraints can arbitrate the relevant differences.

Donald Davidson's "On the Very Idea of a Conceptual Scheme" (1974) attacks the dualism of an organizing scheme and an uninterpreted content waiting to be organized. His argument draws on the problem of making sense of a radically untranslatable language as a language. If meaningful attribution of belief already requires standards of interpretation, an imagined conceptual scheme wholly disconnected from ours may be an incoherent explanatory posit. Davidson does not thereby establish that every existing vocabulary is equally successful; he challenges a picture in which scheme and content are independently identifiable entities (Davidson, 1974, pp. 5–20).

A central objection to Davidson holds that partial or significant conceptual difference may be real even if total unintelligibility is incoherent. Different scientific ontologies and cultural practices may diverge over what entities to count or how to classify their relations, while retaining sufficient points of contact to support translation and disagreement. The rejection of a totally uninterpretable scheme therefore leaves substantial questions about comparative ontology, conceptual change, and the limits of commensurability.

The realist's strongest answer is to treat conceptual variation as a feature of representations operating under shared constraints. The anti-realist's strongest answer is to insist that the phrase "the same underlying world" requires an account of cross-scheme identity, and that this account must itself employ some conceptual resources. Neither reply decides in advance whether all admissible schemes are equally adequate. Constraints may be causal, inferential, explanatory, predictive, practical, or combinations of these.

The deeper issue is whether comparison presupposes a conceptually neutral standpoint. A requirement of complete neutrality would be self-defeating if any account of comparison must be linguistically expressed. But the failure of absolute neutrality does not eliminate objectivity. Cross-perspectival correction can work through shared observations, causal interactions, translation practices, and the possibility of revising both schemes. Objectivity may be an achievement within reciprocal inquiry while its object remains causally and constitutively resistant to that inquiry.

Putnam: Model Theory and Internal Realism

Hilary Putnam's Reason, Truth and History (1981) supplies a decisive modern articulation of metaphysical realism and one of its most influential challenges. Putnam characterizes the externalist image as involving mind-independent objects, one complete true description, and a correspondence relation conceived from outside our representational practices. His internalist response treats questions about what objects there are as meaningful within theory or description. The expression "internal realism" identifies a period of Putnam's changing philosophical trajectory; it should not be presented as an unchanged doctrine governing all of his later work (Putnam, 1981, chap. 3).

Putnam's model-theoretic challenge exploits the gap between formal satisfaction and intended reference. Roughly, a sufficiently rich, consistent theory may possess models with different interpretations, so purely formal constraints may fail to select the intended relation between language and objects. If the metaphysical realist treats reference as a determinate relation fixed independently of all epistemic and practical constraints, the skeptic can ask what makes one interpretation uniquely intended. Model-theoretic facts alone do not refute a causal or naturalistic theory of reference; their philosophical force depends on the claim that all the realist's admissible constraints have already been exhausted.

The argument should be separated from a claim that logic proves there is no external world. It instead attacks an explanatory ideal: that a fully described, language-independent correspondence relation can be invoked to fix intended reference without explaining how that fixing occurs. A realist can propose causal histories, use, natural properties, explanatory success, and worldly interactions as reference-determining constraints. The challenge is then whether such constraints have been specified noncircularly and whether they genuinely rule out unintended models.

Putnam's conceptual-relativity examples raise a second issue. Depending on admissible mereological descriptions, a domain may be counted as containing a different number of objects. One account treats a collection of simples as just those simples; another admits their fusions as further objects. Both may agree about the ordinary observations. This suggests that numerical answers to unrestricted existence questions can depend on conventions of individuation. The realist counterargument distinguishes conventionally available ways of counting from the question whether there are objective facts about composition.

Putnam's brain-in-a-vat argument, meanwhile, concerns semantic externalism. If reference requires appropriate causal or historical relations to the referents, a permanently envatted subject's utterance "I am a brain in a vat" would not automatically refer to ordinary brains and vats as it does in our language. The argument's target is a certain globally skeptical scenario under particular semantic assumptions. It should not be conflated with a proof of internal realism or an argument that all entities depend on conceptual schemes.

The decisive philosophical contribution of Putnam's challenge is to insist that the realist account of objectivity must explain reference, determinacy, and the relationship of theoretical success to truth. A bare assertion that the world fixes meanings is incomplete if no intelligible account is given of how it does so. The converse warning is equally important: inability to give a wholly external description of reference does not establish that there are no external referents. The question is whether explanatory dependence on representational practices entails metaphysical dependence of what is represented.

Contemporary Realist Reconstructions

The realist answer to modern anti-realism can be developed without assuming an exhaustive view of the universe. A realist may defend that entities, events, and constraints exist independently of representation while remaining fallibilist about how they are described. This position rejects the inference from fallibility to dependence: that our knowledge can be corrected is intelligible precisely because some of what corrects it is not settled by our current beliefs.

Thomas Nagel's The View from Nowhere (1986) investigates how thought becomes more objective by compensating for the limits of particular standpoints. The movement toward objectivity is an achievement of situated agents; it does not guarantee a complete standpoint free of all limitations. Nagel's position explicitly allows reality to exceed what any possible expansion of our objectifying capacities could grasp. This is a realist recognition of cognitive finitude, not an abandonment of objectivity (Nagel, 1986).

Michael Devitt's Realism and Truth (1984; later editions) defends a naturalistic metaphysical realism in which the world exists largely independently of mental operations. Devitt separates metaphysical realism from a particular theory of truth. This separation addresses an important weakness in the externalist package criticized by Putnam: commitment to an external world does not force commitment to a single controversial correspondence semantics. A metaphysical realist can examine theories of truth and reference without making the independent existence of tables or electrons depend on the eventual choice among them.

Kit Fine's "The Question of Realism" (2001) presses the dispute beyond simple existential claims. A philosopher who says that numbers exist within ordinary arithmetic may still ask whether numbers are part of reality's fundamental constitution or whether talk of numbers is reducible or derivative. Analogously, saying that moral facts exist in a permissive sense may leave open whether those facts possess the realist's required standing. Fine's work demonstrates that the reality of a domain can concern its metaphysical status, not merely the truth of an existential sentence (Fine, 2001).

This brings grounding, fundamentality, and the distinction between fundamental and derivative existence into contact with the realism debate. A social institution can exist and play an objective causal role while depending for its constitution on rules and collective practices. Some properties may be objective yet derivative upon more basic relations. Realism about derivative entities is compatible with denying that they belong to the most fundamental inventory. What matters is whether the dependence relations themselves can be justified and whether they preserve explanatory distinctions rather than multiplying metaphysical vocabulary without work.

A demanding form of realism maintains a uniquely privileged joint-carving description. Theodore Sider's Writing the Book of the World (2011) develops an account of metaphysical structure intended to distinguish more natural or fundamental ways of representing reality from merely convenient classifications. The thesis is stronger than the claim that reality constrains inquiry. It requires arguments showing that reality itself privileges certain quantifiers, predicates, or structural relations. A naturalist may favor scientific explanatory roles; a metaontological critic may object that the choice of fundamental vocabulary remains underdetermined by ordinary truth conditions.

A responsible realist therefore distinguishes three projects. An independence thesis claims that there is a reality whose existence and many features do not depend on our conceptions. A determination thesis claims that reality settles which descriptions are true. A privileged-structure thesis claims that some conceptual articulations reflect objective joints in reality itself. Arguments for one may assist the others, but none becomes automatic. A realist synthesis grows more persuasive when these commitments are defended separately.

Metametaphysical realism in the 2025–2026 debate

A further distinction has become central in recent work: realism in metaphysics concerns which entities and facts exist, whereas realism about metaphysics concerns the cognitive status of metaphysical inquiry itself. Mark Balaguer's 2025 survey "Realism and Anti-Realism about Metaphysics" examines proposals that some apparently profound ontological disputes are trivial, merely verbal, or lack a factual answer. The distinction is logically indispensable. Someone may affirm that there are tables and nevertheless regard the theoretical controversy over their composition as merely verbal. Conversely, someone may deny composite objects while insisting that the denial states a substantive truth about the world's structure. The first-order answer and second-order assessment of the question vary independently (Balaguer, 2025).

One family of deflationary strategies appeals to ordinary linguistic rules and the circumstances in which existential expressions are correctly applied. Amie L. Thomasson's Ontology Made Easy (2015) develops an "easy approach" to existence questions that can be answered through competent conceptual use together with relevant empirical facts. Her Rethinking Metaphysics (2025) presses a wider challenge to traditional ambitions of discovering deep metaphysical structure and proposes a constructive role for conceptual engineering. The argument has an important methodological consequence: philosophers should examine whether they have located a substantive disagreement before pursuing an elaborate metaphysical resolution. But a conclusion that some questions are answerable through ordinary rules leaves open whether those rules adequately express the phenomena, and whether the criteria themselves deserve revision.

Timothy Williamson's January 2026 review of Rethinking Metaphysics argues that Thomasson's criticism may underestimate explanatory and abductive methods in metaphysics. In particular, forms of unification, systematicity, and noncausal explanation can be epistemically significant even when they are not experimental causal hypotheses. Williamson also questions whether nominalization rules can straightforwardly secure an ontology of properties without confronting logical difficulties associated with unrestricted abstraction (Williamson, 2026). This reply strengthens the methodological case for substantive metaphysics but does not guarantee that every alleged metaphysical dispute is well posed.

The disagreement can be tested through a concrete example. If two speakers differ only over whether to apply the word "object" to a particular fusion, and neither predicts or explains anything differently, a merely verbal or framework-relative diagnosis gains force. If their theories differ over persistence conditions, causal interaction, the existence of essential relations, or the grounds of truth, then the dispute may have substantive commitments even if the surface vocabulary looks interchangeable. The appropriate test is therefore to reconstruct competing claims, their explanatory consequences, and the principles fixing the relevant quantifiers and predicates.

The most significant outcome of the recent debate is a more differentiated view of philosophical legitimacy. A realist about metaphysics need not maintain that every ontological question is deep. A defender of conceptual engineering need not claim that all reality is linguistically constituted. Both programs become more precise when they specify which issues are about the entities and relations of the world, which concern the constitutive conventions of an institution or discourse, and which ask how inquiry should be organized. This is the natural continuation of Carnap's framework distinction, Quine's criticism of that distinction, and later disputes about structure and fundamentality.

Realism, Science, and Structure

Scientific inquiry offers the realism debate a particularly disciplined setting because experimental intervention and predictive success yield resistance to arbitrary interpretation. The no-miracles argument holds that the sustained success of mature scientific theories is best explained by their being at least approximately true and by the successful reference of central theoretical terms. Its strength lies in the explanatory relationship between success and an independent target. The argument is abductive: it does not deductively exclude every non-realist explanation of success.

Bas van Fraassen's The Scientific Image (1980) develops constructive empiricism as a rival. A theory may be accepted as empirically adequate without belief in its full account of unobservable entities. Van Fraassen does not deny the literal meaningfulness of theoretical statements or declare every unobservable nonexistent. His distinction concerns the epistemic commitments appropriate to scientific acceptance. This establishes why scientific anti-realism can be compatible with metaphysical realism about the external world (van Fraassen, 1980).

The underdetermination argument begins from the possibility that incompatible theories accommodate the same evidence. If evidence is insufficient to distinguish them, acceptance of their full ontologies demands additional justification. The realist can appeal to explanatory virtues, future intervention, simplicity, unification, and improved testing. The anti-realist can reply that these virtues do not guarantee truth and may reflect pragmatic priorities. The dispute is about the warrant for unobservable commitments, not merely a dispute over what observations have occurred.

History complicates the realist inference from success to truth. Larry Laudan's critique of convergent realism (1981) emphasizes past empirically successful theories whose central ontological commitments were later abandoned. Such examples challenge the unrestricted claim that empirical success reliably guarantees approximate truth. The realist has responses: refine the class of theories counted as mature, distinguish retained from discarded components, and explain why referential or structural continuity can survive radical conceptual change. These strategies require detailed historical case studies rather than retrospective claims that every successful past theory was approximately correct (Realism and Theory Change in Science).

John Worrall's "Structural Realism: The Best of Both Worlds?" (1989) develops an intermediate position based on continuity in mathematical structure across theory change. Epistemic structural realism restricts commitment to relations or structure while suspending judgment about the intrinsic nature of the unobservables. Ontic structural realism makes stronger claims about the primacy or sufficiency of structure itself. These positions address different questions; an epistemic restriction on what may be known should not be silently presented as a metaphysical claim that objects are nothing but relations (Structural Realism).

The Newman problem is a critical difficulty for structural accounts. A purely abstract characterization of relational structure may be satisfied by too many domains of suitable cardinality, rendering the claim of knowledge of structure insufficiently informative. A structural realist must therefore specify substantive physical, causal, modal, or otherwise privileged relations and explain how scientific theories identify them. The same problem echoes Putnam's model-theoretic challenge: formal models alone may be too permissive to secure intended worldly content.

Scientific realism, then, strengthens the case for objective constraints while preserving a serious debate about inference beyond observable evidence. Its central philosophical relevance is methodological. Experiment, prediction, and intervention exemplify how representational systems confront outcomes that can defeat expectations. They do not settle by themselves which ultimate metaphysical account of structure, causation, or theoretical entities must be correct.

Domains of Dependence: Mathematics, Morality, and Social Reality

The realism question changes character across domains, and the differences are philosophically informative. Mathematical realism raises the issue of abstract objects and mathematical truths, neither of which is ordinarily thought to be causally located like a stone. Platonist accounts posit mathematical entities independent of human construction; nominalist, formalist, structuralist, and constructivist positions challenge different components of that picture. Benacerraf's "Mathematical Truth" (1973) sharpens the epistemic problem: if mathematical objects are causally inert, how are our mathematical beliefs reliably connected to them? The argument challenges one theory of knowledge of abstracta without directly refuting the independence of physical nature.

In moral philosophy, a realist may hold that at least some moral judgments are truth-apt and answerable to reasons or facts that transcend current endorsement. An expressivist can explain moral language through attitudes and practical commitments while allowing rich forms of argumentative discipline. Constructivist accounts relate principles to procedures of practical reasoning. None of these positions follows automatically from an account of whether mountains exist. The evaluative authority of a moral reason must be argued in normative terms, even if the debate uses the same language of independence and objectivity.

Social ontology supplies especially clear cases of constitution by practice. Money, citizenship, universities, and legal offices exist through complex systems of conventions, institutions, powers, and historically situated coordination. Their dependence on collective practices does not render them fictional; it makes their causal and normative roles intelligible. A bank account can affect an individual's possibilities, even though the account is an institutionally sustained entity rather than a fundamental physical particle. The philosophical task is to describe dependence relations accurately.

John Searle's The Construction of Social Reality (1995) analyzes institutional facts and the assignment of status functions. Later theories dispute whether collective intentionality, joint commitment, rule-governed practice, or broader social structures best explain institutional existence. An institutional realist can affirm that institutions are real and objectively consequential while recognizing that their identity depends on specific practices. An anti-realist about a purportedly intrinsic social essence may share these claims without denying institutional facts.

Aesthetic judgments present another configuration. The properties attributed to artworks involve material arrangements, historically situated conventions, perceptual responses, and evaluative norms. Whether aesthetic properties are response-dependent does not determine whether there are correct and incorrect aesthetic judgments. Responsiveness to practices can coexist with objective standards internal to, or justified across, those practices. Here the words "objective" and "independent" need especially careful specification.

These domains establish a general principle: constitutive dependence is sometimes part of the best account of an entity's reality. A global doctrine that treats every dependence on cognitive or social activity as an objection to existence obscures this fact. A global doctrine that infers the dependence of everything from some practice-dependent objects commits an opposite overgeneralization. Domain-sensitive analysis preserves the metaphysical difference between geological processes, numbers, institutions, and normative claims.

Argumentative Balance and the Limits of Global Positions

The strongest argument for realism is the possibility of error and correction by a reality that exceeds present judgment. A proposition can be asserted confidently and turn out false; an intervention can frustrate a theory; previously unknown evidence can revise an accepted history. The realist explains these events by appeal to the difference between how things are and how they are thought to be. This explanation becomes especially persuasive where independent causal pathways, converging evidence, and reproducible interventions are available.

The anti-realist can accept every instance of correction while questioning whether its possibility requires a scheme-independent total inventory of objects. Correction occurs through standards of evidence, meaning, inference, and practice, all of which have histories and may themselves be revised. The realist's appeal to constraint establishes that inquiry is answerable to something, but the exact metaphysical account of this answerability requires further argument. A conception of mind-independence that requires a complete "view from nowhere" risks building an impossible ideal of representation into its definition.

The strongest anti-realist argument is accordingly not the claim that anyone may make the world whatever they like. It is that identity, truth conditions, and descriptions often depend on rules of representation in ways that cannot be eliminated by merely invoking an unarticulated world. A determinate answer to "how many objects are there?" presupposes a criterion of what counts as an object. A complete true description depends upon a language in which descriptions are formulated. Such observations are decisive against an unqualified inference from independent existence to uniquely privileged description.

The realist can accept these points and still defend metaphysical independence. A rule of counting does not cause the geological events to which it is applied. The need for a language to state that stars existed before language does not imply that those stars depended on language for their existence. The anti-realist's passage from conditions of articulation to conditions of existence is defensible in some social and conceptual domains but demands special argument in the case of physical history. The distinction between metaphysical constitution and epistemic access is therefore central.

One recurrent difficulty is the self-application of radical anti-realism. If every purported statement of reality is true only relative to a scheme, then the thesis expressing universal relativity must explain its own status. Is it true across schemes, or merely from within a particular one? The dilemma does not refute all forms of constructivism; a constructivist can offer a locally scoped thesis or a reflective account of its own conditions. It does show that global universalization adds a burden absent from more carefully delimited positions.

A parallel difficulty confronts maximal realism. If one uniquely complete description determines the full inventory and structure of reality, the realist must say what individuates that description and why competing descriptions are merely partial, derivative, or mistaken. Appealing to an intended interpretation without an account of reference may assume what Putnam contests. Appealing to science without acknowledging theory change may exaggerate what explanatory success establishes. A realist who relaxes maximal demands loses some rhetorical simplicity and gains conceptual precision.

The debate can be organized through four diagnostic questions. Which entities or facts are at issue? What precise form of dependence or independence is being claimed? What account of truth, meaning, or evidence is required? What explanatory consequence would differ if the disputed thesis were false? When answers to these questions are clear, some disagreements dissolve into differences of vocabulary while others reveal genuinely incompatible theses concerning constitution, structure, or truth.

Open Problems and Metaphysical Synthesis

One open problem concerns structural determinacy. Can reality constrain the truth of descriptions without privileging a unique fundamental division into objects, properties, and relations? Scientific practice suggests that multiple models may capture different aspects of one target, but the metaphysical distinction between useful modeling and objective structure remains difficult. Accounts of naturalness, grounding, causal powers, and structural realism seek to specify the constraint. Their success depends on explaining why certain structural distinctions deserve priority.

A second problem concerns reference. Causal interaction can help anchor words to objects, yet scientific concepts and abstract terms often have histories more complex than simple causal naming. Formal semantic models offer rigor while leaving room for many interpretations. A complete explanation requires coordination among practices of use, worldly interaction, inferential roles, and conditions of successful correction. The challenge is to formulate this coordination without presupposing an already uniquely interpreted language.

A third concerns the limits of truth. Truth may transcend evidence even when justification never does. Some domains contain statements whose truth values remain inaccessible; others involve indeterminacy, vagueness, or dependence on decision procedures. Global bivalence is stronger than the independence of ordinary objects. A unified philosophical account must explain when recognition-transcendent truth is intelligible and where the introduction of determinate truth values would outpace the content of a discourse.

A fourth concerns conceptual plurality. Different scientific and ordinary schemes may be jointly informative while apparently committing to different object inventories. The most promising comparison asks which differences alter empirical prediction, intervention, explanation, or modal commitment, and which are matters of equally legitimate articulation. Neither mere disagreement over terms nor agreement on observations provides a universal decision procedure for ontology.

A fifth concerns perspective. Thomas Nagel's account of the drive toward objectivity suggests that overcoming local distortions is a real achievement even though no investigator attains an unrestricted standpoint. Artificial systems, scientific instruments, collective institutions, and biological investigators may process evidence through different representational architectures. This broadens the practical range of perspectives without establishing that all perspectives disclose different worlds. The philosophical challenge is to characterize invariance, correction, and shared reference across unlike cognitive realizations.

Within the wider Artificial Era, the distinction has methodological relevance. The emergence of nonbiological systems capable of producing descriptions and arguments makes it increasingly urgent to distinguish the identity of the knower, the form of representation, and the object of inquiry. A geological fact is not made true by the biological or artificial origin of its description; the legitimacy of a particular attribution still requires an account of evidence, reference, and interpretation. This is an application of the dossier's general analysis, not a revision of Aisentica's canonical terms or a claim that a new ontology has been experimentally established.

The philosophical synthesis follows from the foregoing distinctions. Mind-independent reality is a defensible thesis wherever the existence and explanatory role of entities survive changes in our beliefs and representational practices. The thesis carries no automatic entitlement to a uniquely exhaustive ontology, an unrestricted semantics of bivalence, or a privileged formulation of every scientific theory. Conceptual mediation is an indispensable fact about knowledge and articulation. Its contribution to an entity's existence is domain-specific and must be shown by analysis of constitution rather than inferred from mediation alone.

This synthesis can be formulated as a rule of inquiry: evaluate independence, individuation, truth, and epistemic warrant separately, then establish their actual relations. Realism earns its force by accounting for worldly constraint, independence, and the possibility of error. Anti-realism earns its force by exposing the conditions under which objects and claims become determinate for thought and practice. Their enduring controversy advances metaphysics when these different achievements are confronted within a common argumentative framework.

Conclusion

Metaphysical realism and anti-realism form a family of disputes about the independence of being, the articulation of reality, the conditions of truth, and the authority of human and other cognitive representations. The historical record shows successive transformations of this problem: ancient investigations of stable being and substance, medieval accounts of universals, early modern theories of perception, Kant's critical account of possible experience, idealist challenges to isolated objecthood, semantic approaches to truth, and contemporary arguments about reference, science, and metaphysical structure.

The resulting analysis establishes five conclusions. The existence of mind-independent entities is logically distinct from the claim that reality has one uniquely complete description. The dependence of knowledge on conceptual resources does not by itself constitute the things known. Truth conditions can exceed available verification without requiring that every discourse obey the same semantic constraints. Some real entities—especially social institutions—are constituted through practices, making dependence an objective feature of their ontology. Scientific success and correction provide powerful evidence of constraint while leaving substantive questions about theoretical reference and fundamental structure open.

The central philosophical result is a disciplined account of independence and dependence. Claims of realism become determinate when they identify an entity or domain, specify what is independent, and explain what this independence contributes to truth or explanation. Anti-realist challenges become determinate when they identify the practices or conditions on which something depends and show whether the dependence is semantic, epistemic, constitutive, or explanatory. A systematic metaphysics advances by establishing these distinctions and testing their implications against history, argument, and the actual practices of inquiry.