Categories: Historical and Systematic Essays (Studies in Philosophy and the History of Philosophy, Volume 41)

9. Language-Games as Categories: An Aristotelian Theme in Wittgenstein’s Later Thought

9 Language-Games as Categories: An Aristotelian Theme in Wittgenstein’s Later Thought NEWTON GARVER 1. predicates and predications Aristotle฀and฀Kant฀agree฀that฀the฀species฀of฀predicates฀(or฀concepts) go฀hand฀in฀hand฀with฀the฀species฀of฀predications฀(or฀judgments).฀Why these฀two฀classifications฀go฀hand฀in฀hand฀is฀not,฀in฀either฀case,฀a฀matte of฀empirical฀discovery฀but฀has฀to฀do฀instead฀with฀the฀ver y฀nature฀of฀the project.฀ The฀project,฀in฀Aristotle’ s฀ terms,฀in฀one฀of฀making฀sense฀of “things฀that฀are฀said.”1 In฀Kant’s฀terms฀it฀is฀how฀judgments฀are฀possible, a฀judgment฀being฀in฀the฀first฀instance฀the฀determination฀that฀an฀objec falls฀under฀a฀concept,฀that฀is,฀how฀a฀concept฀can฀be฀true฀of฀an฀object.2 In these฀general฀terms฀the฀two฀projects฀are฀ver y฀much฀alike,฀and฀Kant฀says that฀he฀is฀clarifying฀what฀Aristotle฀did.฀Both฀Aristotle฀and฀Kant฀proceed in฀terms฀of฀subject-predicate฀logic,฀and฀the฀project฀is฀one฀of฀specifying at฀least฀some฀of฀the฀ways฀in฀which฀predicates฀can฀be฀true฀of฀subjects. The฀two฀classifications฀have฀to฀go฀hand฀in฀hand฀because฀they฀are฀bot aimed฀at฀classifying฀things฀that฀are฀said.฀There฀are฀not฀in฀practice฀separate ฀criteria฀for฀classifying฀predicates฀and฀for฀classifying฀predications. They฀also฀largely฀agree฀that฀parts,฀logically฀and฀metaphysically,฀come before฀wholes:฀that฀one฀focuses฀on฀the฀classification฀of฀predicates฀o concepts฀ in฀order฀to฀achieve฀a฀classification฀of฀predications฀or฀judg ments.฀It฀is฀true฀that฀Kant฀f rst฀presents฀a฀table฀of฀judgments฀that฀is฀derived ,฀he฀claims,฀from฀the฀logically฀possible฀forms฀of฀judgments,฀and then฀derives฀his฀table฀of฀categories฀from฀the฀table฀of฀judgments.฀But the฀table฀of฀judgments฀is฀hardly฀convincing,฀and฀the฀general฀line฀of฀his 136 1.฀For฀other฀reflections฀on฀Aristotle s฀categories,฀see฀the฀essays฀by฀Sanford,฀Lang,฀and Sim,฀Chapters฀1,฀2,฀and฀5 of฀this฀volume. 2.฀For฀further฀reflection฀on฀Kant s฀categories,฀and฀on฀his฀practical฀categories฀in฀par ticular ,฀see฀Quinn’s฀essay,฀Chapter฀6 of฀this฀volume. argument฀is฀that฀different฀sorts฀of฀concepts฀(different฀categories)฀synthesize ฀our฀intuitions฀and฀perceptions฀in฀different฀ways.฀He฀does฀not begin฀by฀differentiating฀the฀different฀sorts฀of฀synthesis. The฀theoretical฀reasons฀for฀beginning฀with฀parts฀rather฀than฀wholes are฀complex฀and฀I฀will฀return฀to฀them฀toward฀the฀end฀of฀this฀essay .฀The issue฀is฀partly฀dependent฀on฀one’s฀conception฀of฀the฀nature฀of฀logic,฀and especially฀on฀what฀one฀takes฀to฀be฀the฀most฀elementar y฀logical฀forms. Aristotle’s฀logic฀is฀a฀logic฀of฀terms,฀especially฀of฀subjects฀and฀predicates. These฀are฀parts฀of฀statements,฀the฀statements฀being฀constructed฀out฀of them,฀ and฀so฀Aristotle฀naturally฀focuses฀on฀terms.฀Kant฀accepts฀Aristotelian ฀logic฀and฀therefore฀follows฀him฀in฀this฀regard,฀although฀Kant’ s presentation฀ of฀the฀possible฀forms฀of฀subject-predicate฀judgment฀is somewhat฀idiosyncratic.฀The฀main฀point฀is฀that฀a฀term฀logic,฀focusing฀on subjects฀and฀predicates,฀differs฀significantly฀from฀truth-functional฀logic like฀that฀of฀Frege฀and฀Russell.฀T ruth฀and฀falsity฀apply฀to฀judgments฀or predications฀rather฀than฀to฀their฀parts.฀A฀main฀shift฀in฀the฀philosophy฀of logic฀ and฀the฀philosophy฀of฀language฀occurred฀with฀Frege’ s฀ focus฀on propositions฀ (Gedanken) as฀ the฀primar y฀ or฀primitive฀elements฀of฀language ฀ and฀logic.฀In฀spite฀of฀this฀shift฀in฀the฀philosophy฀of฀language, there฀remains฀an฀attractiveness฀to฀starting฀with฀words,฀since฀we฀are฀conscious ฀ of฀building฀our฀sentences฀out฀of฀words,฀and฀it฀seems฀natural (metaphysically฀speaking)฀to฀derive฀sentences฀from฀words฀rather฀than the฀other฀way฀around. The฀consequence฀of฀starting฀with฀kinds฀of฀words฀(predicates฀or฀concepts )฀ rather฀than฀with฀kinds฀of฀statements฀(predications฀or฀judgments )฀is฀that฀the฀categories฀one฀derives฀are฀bound฀to฀be฀semantic฀categories .฀They฀are฀semantic฀in฀the฀sense฀that฀they฀derive฀from฀and฀apply to฀the฀meanings฀of฀words.฀Jerrold฀Katz,฀who฀accepts฀Frege’ s฀logic,฀has presented฀a฀sophisticated฀exposition฀of฀Aristotle’ s฀categories฀as฀semantic ฀ categories,฀making฀use฀of฀linguistic฀methods—he฀is฀indebted฀to Chomsky—that฀are฀thoroughly฀alive฀to฀modern฀logic.฀Katz฀begins฀with words,฀understood฀as฀signs฀together฀with฀their฀meaning,฀and฀explicates meaning฀in฀terms฀of฀abstract฀entities฀called฀“semantic฀markers.”฀Logical ฀ relations฀hold฀among฀semantic฀markers,฀which฀are฀therefore฀the most฀basic฀logical฀forms฀as฀well฀as฀elements฀of฀meaning.฀Categories฀are the฀ultimate฀semantic฀markers,฀that฀is,฀those฀that฀do฀not฀entail฀any฀fur ther ฀semantic฀markers.3 They...