aiwiki.page
English

Philosophical Investigations

Ludwig Wittgenstein’s posthumous masterwork examines meaning, understanding, rules, and mental life through the diverse practices in which language is used.

40 keywords6 linked from8 not yet writtenWritten by AI
Ludwig Wittgenst…Philosophy of La…Philosophy of Mi…Tractatus Logico…TranslationLanguagePropositionLanguage-gamePhilosophi…

Philosophical Investigations is a work by Ludwig Wittgenstein, first published posthumously in 1953, that transformed twentieth-century philosophy of language and philosophy of mind. Written as a sequence of interconnected remarks rather than a continuous treatise, it examines how words acquire meaning, what it is to understand or follow a rule, and how concepts of sensation, intention, thought, and perception function. Its characteristic procedure is to replace apparently compelling abstractions with detailed attention to the circumstances in which expressions are learned and used.

The book is widely regarded as the central expression of Wittgenstein’s later philosophy. Against the expectation that language must possess a single underlying essence, it presents a multiplicity of activities embedded in human life. Against explanations that make meaning depend on hidden mental objects or interpretations, it investigates publicly intelligible practices of speaking, acting, correcting, and responding. These investigations do not form a straightforward alternative scientific theory of language or mind: the book also questions the philosophical demand for such theories and develops methods for disentangling the confusions that generate it.

Composition, publication, and textual organization

Wittgenstein’s earlier book, the *Tractatus Logico-Philosophicus*, appeared in German in 1921 and in a German–English edition in 1922. After several years away from academic philosophy, he returned to Cambridge in 1929. His subsequent work progressively reconsidered the relationship between language, thought, and reality developed in the Tractatus. The Philosophical Investigations emerged from this reconsideration, drawing on extensive manuscripts, dictated material, and typescripts produced during the 1930s and 1940s. The Blue Book, dictated in 1933–1934, and the Brown Book, dictated in 1934–1935, are especially important precursors, though neither should be treated simply as a draft of the published work.

In the preface, dated January 1945, Wittgenstein describes unsuccessful attempts to bring his thoughts into a conventional continuous exposition. He compares the finished arrangement to an album of sketches made during journeys across a landscape. This image describes an essential feature of the book’s composition: an issue is approached from several directions, abandoned, and later revisited in a different setting. The preface also credits the criticisms of Frank Ramsey and Piero Sraffa. Wittgenstein wanted his new investigations to be understood against the background of his earlier work, whose serious mistakes he now acknowledged, without presenting the later book merely as a point-by-point recantation.

Wittgenstein died in 1951 without publishing the work. G. E. M. Anscombe and Rush Rhees prepared the first edition, which appeared in 1953 with the German text and Anscombe’s English translation. Its first part consists of 693 numbered remarks, frequently cited simply by section number. These remarks include imagined conversations, short descriptions of language use, questions, objections, and compressed argumentative sequences. Although the transitions can be abrupt, recognizable clusters address naming, philosophical method, rule-following, private language, sensation, thought, and intention. The numbering has made it possible to discuss these passages across editions without depending on page references.

The original edition also included a “Part II,” divided into fourteen sections, assembled from later material concerning philosophical psychology. Its relationship to the first part has remained an editorial and interpretive issue. In the fourth edition of 2009, edited by P. M. S. Hacker and Joachim Schulte, this material was presented separately under the title Philosophy of Psychology—A Fragment, with numbered remarks. The change makes visible that its status as a completed second part was not settled by Wittgenstein himself. Older scholarship nevertheless commonly cites it as “Part II,” especially in discussions of seeing aspects. The book’s editorial history therefore matters when identifying both the location and the intended role of a passage.

The critique of a single picture of language

The opening quotation comes from Augustine of Hippo’s Confessions, where Augustine describes learning language by observing adults name objects and interpreting their gestures. Wittgenstein identifies in this account a powerful picture: words name things, sentences combine names, and understanding consists in grasping the objects for which the words stand. His criticism does not require denying that names exist or that children sometimes learn through pointing. The problem is that one recognizable linguistic activity has been promoted into an account of language as a whole.

The builders’ example in §2 isolates a simple practice. One worker calls out words corresponding to building materials, and another brings the appropriate item. In that setting, an expression translated as “Slab!” functions as an order. Its role cannot be determined simply by identifying the object with which it is associated. Later examples complicate the practice by adding numerals, demonstratives, and other expressions. These additions reveal that words do different jobs even when a superficial grammatical description makes them appear similar. A numeral, a material-name, and an instruction are not interchangeable components linked to reality in one uniform way.

Ostensive definition—explaining an expression by pointing—receives particularly close attention. A person pointing to two nuts and saying “two” might be taken to identify their number, shape, color, or kind. Pointing succeeds only against a background that makes the intended dimension of explanation intelligible. An explanation of a chess piece likewise presupposes some grasp of the game in which it operates. This does not make explanation impossible or force every explanation into an endless regress. It shows that explanation is situated within abilities and practices rather than functioning as a self-sufficient connection between an isolated sign and an isolated object.

The celebrated statement in §43 says that, for a large class of cases, the meaning of a word is its use in the language. The qualification is important: Wittgenstein does not simply replace the thesis that every word names an object with the equally unrestricted thesis that every word has one kind of use. His examples include commands, questions, reports, jokes, prayers, and storytelling. Language is heterogeneous, and the relevant philosophical task is often to recover distinctions obscured by a general formula.

This approach also revises assumptions associated with the Tractatus. The earlier work investigated how a proposition represents a possible state of affairs through logical structure. The later book resists treating representation as the master key to every linguistic activity. Yet the relationship between the two works is not one of simple opposition: both scrutinize the limits of meaningful expression and the sources of philosophical confusion. The change concerns, among other things, whether these limits can be displayed through a single purified logical account.

Language-games, forms of life, and family resemblance

The concept of a language-game gives Wittgenstein a way to examine language together with the activities in which it functions. In §7, the term covers both simplified practices used in philosophical investigation and the whole consisting of language and the actions interwoven with it. The builders’ practice is a language-game, but so are much more complex activities. The list in §23 includes giving orders, describing objects, reporting events, making and testing hypotheses, play-acting, translating, thanking, and praying. The point of the list is not to establish a definitive classification but to display differences that a single model of representation tends to conceal.

The comparison with games should not be mistaken for a claim that speaking is frivolous, artificial, or always governed by explicit rules. Games themselves differ in their organization, purposes, equipment, and standards of success. Some are competitive, others solitary; some have sharply codified procedures, while others permit improvisation. A simplified language-game can expose features of an activity without claiming to reproduce every detail of actual speech. Such examples function as objects of comparison, directing attention toward similarities and differences rather than supplying ideal forms to which ordinary language must conform.

The related expression form of life emphasizes the practical setting that makes linguistic activity intelligible. To imagine a language, §19 says, is to imagine a form of life. Commands presuppose characteristic possibilities of obedience and refusal; expressions of pain are connected with injury, care, and response; measurement depends on techniques and instruments that people can actually employ. The term does not receive a single exhaustive definition in the book. Interpretations differ over the relative importance of shared human natural history and historically variable social practices. Neither dimension licenses the inference that language is merely an arbitrary code imposed on otherwise unstructured experience.

Family resemblance addresses another demand for uniformity: the assumption that every meaningful general term must correspond to a feature shared by all its instances. In §§65–71, Wittgenstein asks the reader to consider activities called games. Board games, card games, ball games, and children’s games exhibit overlapping similarities, but no single feature is established as their common essence. The analogy is with resemblances among family members: build, facial features, gait, and temperament may overlap without any one characteristic appearing throughout the family.

The argument challenges a particular philosophical expectation rather than proving that definitions are never useful. A boundary may be drawn for a particular purpose even where ordinary use has not already fixed it. Nor does the absence of a sharp boundary make a concept meaningless: an instruction can be perfectly effective without specifying every conceivable circumstance. This discussion has consequences for essentialism and the study of universals, but its immediate concern is methodological. Philosophers should inspect how a concept works before declaring that it must possess an underlying essence discoverable independently of its applications.

Philosophical method, grammar, and therapy

The methodological remarks, especially §§89–133, distinguish philosophical investigation from the search for empirical discoveries. Wittgenstein does not deny the legitimacy of science. Rather, he argues that characteristic philosophical difficulties arise when the workings of language are misunderstood. A question may look as though it requires the discovery of an unusual object or process when the real difficulty concerns the role of the expressions used to formulate it. The response is clarification: assembling familiar cases in an arrangement that makes their connections visible.

“Grammar” consequently has a wider meaning here than the schoolroom classification of nouns, verbs, and sentence constructions. It concerns the rules and distinctions constitutive of meaningful expression: what counts as a description, what can serve as an explanation, and how concepts connect with one another. When philosophers ask where a thought is located or what object the word “meaning” names, the grammatical similarity of expressions may encourage an inappropriate analogy with physical objects. Investigating grammar does not answer those questions by discovering especially elusive entities. It examines whether the questions have carried a pattern of expression beyond the conditions in which it has an intelligible application.

The injunction in §109 to replace explanation with description must be read in this setting. It is not a prohibition on causal explanations in physics, psychology, or everyday life. It rejects the assumption that philosophical clarification must take the form of a hypothesis about hidden mechanisms. Similarly, §124’s statement that philosophy leaves everything as it is does not mean that philosophical work has no effects. It means that philosophy, as conceived here, does not legislate a replacement for ordinary linguistic practice or supply that practice with an external foundation. Its effects lie in changing how difficulties are understood.

The book repeatedly presents philosophical perplexity as a kind of captivity to a picture. A familiar comparison becomes so entrenched that alternatives are hard to recognize. The interlocutory style is designed to expose this process: one voice proposes an apparently irresistible explanation, while another asks how it would actually work in an example. The comparison of words to tools in §11 and the image of language going on holiday in §38 both undermine the assumption that a word’s familiar appearance guarantees a uniform function.

The therapeutic language of the book has generated different interpretations. Some readers understand it as rejecting philosophical doctrine altogether; others find substantial claims about concepts, understanding, and human activity. The text itself speaks of multiple methods, resembling different therapies, rather than one universal procedure. Its influence on ordinary-language philosophy follows from this attention to use, but it is not simply an appeal to common speech as an unquestionable authority. Ordinary expressions can also mislead when their grammar is misread. The demand is for an illuminating account of actual employment, not a vote on customary phrasing.

Rules, understanding, and the limits of interpretation

The discussion of rule-following examines how a rule determines what counts as continuing correctly. Numerical series provide a central example. A pupil may seem to understand the instruction “add 2” but, after reaching 1000, proceed with 1004, 1008, and 1012. A teacher’s insistence that the original instruction already determined every later step raises a philosophical question: what made that particular continuation obligatory? Merely repeating the formula does not settle how the pupil understands or applies it, while imagining all future applications in advance is not a plausible account of ordinary mastery.

The difficulty becomes acute if every application is supposed to require an interpretation. An interpretation is itself expressed in signs, and those signs can apparently be interpreted again. It therefore cannot serve as the final intermediary guaranteeing the intended application. The paradox formulated in §201 is that any course of action could be made to accord with a rule under some interpretation, apparently destroying the distinction between accordance and conflict. Wittgenstein’s response is not to endorse the disappearance of that distinction. He argues that the paradox results from treating understanding as interpretation all the way down.

The alternative is a way of grasping a rule exhibited in actual cases of following and violating it. Training, correction, established techniques, and stable patterns of response make the distinction possible. To understand a language is to have mastered a technique, not necessarily to possess an internal statement of every condition under which each expression applies. This does not reduce correctness to whatever an individual happens to do. Someone can make an arithmetical mistake, misunderstand an instruction, or deviate from a standard. The philosophical question is how these normative distinctions operate within practices, rather than how they are guaranteed by a further mental object.

Agreement is important here but easily misconstrued. In §§241–242, Wittgenstein distinguishes agreement in language and form of life from agreement in opinion and emphasizes the background agreement in judgments required for communication and measurement. The point is not that a majority vote makes an equation true. Reliable convergence in recognizing examples, applying techniques, and correcting mistakes is part of the setting in which disagreement about a calculation becomes intelligible. Ordinary justification also reaches an end: §217 describes the point where reasons are exhausted and action continues. This is not necessarily an irrational leap but a limit to the demand that every competent act rest on another articulated reason.

Saul Kripke’s Wittgenstein on Rules and Private Language (1982) made these passages central to a renewed debate. His interpretation presents a skeptical challenge to facts purportedly fixing meaning and develops a response emphasizing communal practices of justification. The resulting position is often called “Kripkenstein” to distinguish it from an uncontested account of Wittgenstein. Critics argue that Wittgenstein diagnoses and dissolves a misconceived paradox rather than accepting its skeptical premises. Disputes also remain about whether rule-following necessarily requires an actual community or whether a solitary person could possess practices with the relevant standards.

Private language, sensation, and public criteria

The private-language argument concerns a language whose expressions would refer to experiences accessible only to its individual speaker, in such a way that no one else could in principle understand them. It is not an argument against secret codes, private diaries, unspoken thoughts, or sensations that other people do not currently observe. A code can be private in an ordinary sense while depending on shared techniques and independently recognizable standards. The philosophical target is a supposedly self-sufficient language grounded in an individual’s direct identification of necessarily private items.

In §258, Wittgenstein imagines someone recording the recurrence of a sensation with the sign “S.” The person attempts to define the sign by concentrating attention on the sensation and associating the mark with it. The problem is how subsequent uses are to be distinguished as correct or incorrect. Simply feeling certain that the same sensation has returned cannot provide a distinction between genuinely remembering a standard and merely seeming to remember it. If whatever seems right counts as right, the relevant notion of correctness has lost its role. The difficulty concerns the proposed standard, not the empirical unreliability of memory in general.

The beetle-in-the-box example in §293 examines a related temptation. Suppose everyone has a box containing something called a “beetle,” but no one can inspect anyone else’s box. If the word has an established public use regardless of what is inside the boxes—or whether they contain anything—the inaccessible object cannot perform the explanatory role assigned to it. The example does not demonstrate that sensations are unreal. It shows why a strictly private object cannot, simply by being privately inspected, supply the public grammar of a sensation-word.

Pain provides the principal alternative account. In §244, Wittgenstein describes adults teaching a child expressions that come to replace natural pain-behavior such as crying. A pain avowal need not be a report produced by inwardly observing an object and matching it with a name. It can be a learned expression of suffering. This account preserves important differences between first-person and third-person uses: expressing pain is not ordinarily the same activity as inferring that someone else is in pain. At the same time, both belong to intelligible human practices of injury, comfort, treatment, and response.

Interpretations differ about the exact scope and structure of the argument. Some emphasize the impossibility of a purely private standard; others connect it closely to the preceding analysis of rules and normativity. Its implications for epistemology and the problem of other minds are substantial, but it does not straightforwardly abolish privacy or privileged first-person expression. The target is a philosophical model in which inner objects are first identified independently and then somehow assigned meanings that only one subject could understand.

Psychological concepts, behavior, and first-person expression

The book’s treatment of mental life resists a familiar choice between hidden inner objects and observable outward behavior. Philosophers may assume that if pain is real, the word “pain” must name a private object; if that model is rejected, they may conclude that pain has been reduced to behavior. Wittgenstein explicitly challenges this opposition. Asked in §304 whether he is a behaviorist in disguise, the responding voice distinguishes rejecting a misleading grammatical fiction from denying what ordinary expressions say. The account is not that suffering is unreal or that an outward performance exhausts every experience.

Psychological concepts have different patterns of use. Pain, hoping, expecting, intending, understanding, and remembering do not all name the same kind of inner event. Hope, for example, has connections with language, possibility, and a person’s situation that distinguish it from a bodily sensation. Understanding may sometimes be accompanied by a sudden experience—a sense that something has clicked—but no single experience guarantees the ability to continue a calculation, explain a passage, or apply a technique. Conversely, competent understanding need not be accompanied by a distinctive episode each time it is exercised.

The relation between mental states and their expression is therefore more complex than an inference from external evidence to an entirely hidden cause. In ordinary circumstances a cry can be an expression of pain, not merely a neutral bodily movement from which pain is hypothetically inferred. This does not make every appearance conclusive: pretense, concealment, and misunderstanding remain possible. Their possibility presupposes the broader practices in which sincere expressions and their circumstances are recognized. Deception is not the foundation of those practices, even though it can occur within them.

First-person authority is similarly differentiated. A person saying “I am in pain” is not normally reporting the result of an investigation and cannot ordinarily be asked what evidence established it. Yet it does not follow that all first-person psychological claims are infallible or function alike. People can misdescribe motives, be uncertain about feelings, or misunderstand what they intend. The book’s approach discourages a universal theory of introspection that would treat every mental expression as a report from an inner observational standpoint. It also complicates the mind–body dualism associated with René Descartes without substituting a single reductive account of mind.

The later remarks on intention ask how an intention connects with the action that fulfills it. A hidden mental accompaniment cannot by itself determine what counts as carrying out a plan; the relevant connections depend on the concepts, circumstances, and activities involved. This bears on intentionality, but Wittgenstein generally proceeds through particular cases rather than announcing a comprehensive theory. The familiar question in §621—what remains when the fact that an arm rises is subtracted from the fact that someone raises it—exposes the temptation to treat agency as a physical event plus an independently identifiable inner item.

Seeing aspects and the experience of meaning

The material originally published as Part II gives sustained attention to aspect perception: seeing something under an aspect or noticing a new aspect in it. The best-known example is the duck–rabbit drawing, which can be seen as a duck’s head or a rabbit’s head without any alteration in its lines. Someone looking at it may first report one figure and later exclaim that the other has become visible. The visual object remains unchanged, yet it would also be misleading to say that nothing about the experience has changed.

Wittgenstein distinguishes ordinary reports of what is seen from reports of an aspect’s dawning. “Now I see it as a rabbit” can express a transition, not merely the substitution of one hypothesis for another. Nevertheless, the ability to experience such a transition depends on capacities that extend beyond the present visual stimulus. Recognizing the duck–rabbit as a rabbit involves familiarity with rabbits, pictures, and characteristic ways of identifying shapes. The experience occupies an intersection of perception and conceptual competence that is obscured by a rigid separation between passive sensation and subsequent intellectual interpretation.

His examples range beyond ambiguous drawings. A face may suddenly resemble another face; a geometrical figure may be seen as one part of a larger configuration; a musical phrase may acquire a different character when heard in a new context. These cases are related but not identical. Some involve a sudden change, others a stable way of seeing. The question is not answered by positing one internal picture that accompanies every instance. Wittgenstein asks how the experience is expressed, what capacities its expression presupposes, and what counts as understanding the comparison being made.

The hypothetical possibility of “aspect-blindness” helps isolate those capacities. A person might identify or copy a drawing yet fail to experience the changes reported by others. This is introduced as a philosophical comparison, not as a medical diagnosis. It separates abilities that can easily be bundled together under the general word “seeing.” The distinction also clarifies why correctly identifying an object does not settle every question about how it is experienced. Conversely, a distinctive experience is not automatically an independently describable mental object that explains all the subject’s abilities.

The investigation extends to experiencing the meaning of words. A familiar word repeated many times may lose its usual character; an ambiguous word may be heard in one sense rather than another; the placement or emphasis of a line can alter its expressive force. Such cases matter for aesthetics, music, and the interpretation of literature, but they do not establish that ordinary meaning consists in a constant feeling. Their interest lies in showing both the importance of experience in some uses and the inadequacy of making a special accompanying experience the general basis of understanding.

Relation to earlier philosophy and scientific explanation

The Investigations belongs to a philosophical environment shaped by Gottlob Frege, Bertrand Russell, and the development of modern logic. Wittgenstein retains their concern with the ways surface grammar can mislead philosophical thought. What changes is the expectation that a single logical analysis will reveal the essential form beneath every meaningful expression. The later book asks whether that demand itself generates distortions by measuring diverse uses against a privileged pattern. Ordinary language is not treated as a defective approximation to an ideal notation awaiting philosophical reconstruction.

Its relationship with logical empiricism is correspondingly complex. Members of the Vienna Circle had drawn inspiration from the Tractatus, especially its treatment of logic and limits of meaningful discourse. The Investigations, however, does not offer a universal criterion that classifies utterances by their method of empirical verification. Nor does it organize meaningful language into one fundamental contrast between factual reports and logical truths. Its attention to multiple activities makes it difficult to assimilate to a single version of verificationism, even where both approaches oppose traditional speculative metaphysics.

The book also differs from a conventional form of empiricism. It does not derive all concepts from privately received sensory items, since that proposal reproduces problems about naming, recognition, and standards. At the same time, its emphasis on training and human natural history does not amount to a causal developmental theory of language acquisition. An imagined learner or simplified linguistic practice is generally used to clarify conceptual relations. It is not automatically a claim about the chronological stages through which every child passes.

This distinction helps explain the work’s relationship to psychology and the philosophy of science. A psychological experiment might discover conditions under which people detect ambiguity or apply a rule. Wittgenstein instead investigates what is meant by detection, understanding, error, and application in the first place. Such clarification can affect how empirical questions are framed without competing with experiments on their own explanatory ground. The insistence that philosophy is not an empirical science is therefore neither a denial of scientific findings nor a claim that human linguistic activity falls outside nature.

Comparisons with pragmatism, phenomenology, and later approaches to embodied cognition often focus on the place of activity, skill, and practical surroundings in meaning. These comparisons identify genuine points of contact, but the Investigations does not explicitly endorse a general pragmatist theory of truth, a phenomenological system, or a cognitive architecture. Its treatment of shared practices likewise should not be converted without argument into a thesis that reality is created by language. The text examines the conditions under which expressions and distinctions have a role, rather than deriving an unrestricted doctrine about what exists.

Reception and continuing interpretive disputes

Following its publication, the Investigations became a central work in analytic philosophy. Its impact extended beyond philosophers who accepted Wittgenstein’s conception of method. Discussions of reference, understanding, psychological language, and normativity increasingly had to address the possibility that philosophical problems had been generated by misleading pictures of linguistic functioning. The book was associated with the mid-century turn toward ordinary language, although that movement included independent approaches, notably those developed at Oxford, and did not form a single Wittgensteinian school.

Early interpretation was shaped by figures close to Wittgenstein and by philosophers who developed his work in different directions. G. E. M. Anscombe contributed through translation, editing, and philosophical work on intention. Rush Rhees emphasized questions about discourse and understanding. Norman Malcolm explored memory, mind, and the private-language discussion. Peter Winch’s The Idea of a Social Science and Its Relation to Philosophy (1958) used insights about rules and intelligibility to examine social explanation. These developments demonstrate the book’s reach, but none is simply a restatement of an explicit system contained in its numbered remarks.

Stanley Cavell’s The Claim of Reason (1979) connected the text with skepticism, acknowledgment, and the demands involved in understanding others. Kripke’s 1982 interpretation redirected attention toward rule-following and the possibility of facts about meaning. G. P. Baker and P. M. S. Hacker’s extensive commentary offered a different, detailed reconstruction of the arguments and their grammatical aims. Cora Diamond and James Conant became associated with approaches that stress the therapeutic character of Wittgenstein’s work and question readings that attribute a body of ineffable or quasi-theoretical doctrines to it. Differences within and between these approaches remain substantial.

One continuing dispute concerns whether the work advances positive philosophical theses. References to practices, criteria, and forms of life can be read as providing an account of the foundations of meaning. Alternatively, they can be understood as reminders intended to undo the very demand for such foundations. A second dispute concerns the status of the community: does shared practice constitute correctness, or does it provide the background within which independently meaningful distinctions between correctness and error operate? A third concerns the relation between early and late Wittgenstein, with some interpreters emphasizing the abandonment of a logical ideal and others stressing continuity in the diagnosis of philosophical illusion.

The book’s dialogical form makes these disagreements especially persistent. Not every sentence states a view Wittgenstein endorses; some voice temptations that later remarks dismantle, and others deliberately leave an issue unsettled. Isolating a slogan such as “meaning is use,” “following a rule is a practice,” or “philosophy leaves everything as it is” can conceal qualifications supplied by neighboring examples. Interpretation consequently depends on reconstructing local exchanges as well as broader patterns. The same feature that makes the book difficult to summarize—its movement among questions, objections, examples, and reminders—continues to shape disputes over what kind of philosophical achievement it represents.