第3章

23

Itisageneralruleindealingwithargumentsthatdependonlanguagethatthesolutionalwaysfollowstheoppositeofthepointonwhichtheargumentturns：e。g。iftheargumentdependsuponcombination，thenthesolutionconsistsindivision；ifupondivision，thenincombination。Again，ifitdependsonanacuteaccent，thesolutionisagraveaccent；ifonagraveaccent，itisanacute。Ifitdependsonambiguity，onecansolveitbyusingtheoppositeterm；e。g。ifyoufindyourselfcallingsomethinginanimate，despiteyourpreviousdenialthatitwasso，showinwhatsenseitisalive：

if，ontheotherhand，onehasdeclaredittobeinanimateandthesophisthasprovedittobeanimate，sayhowitisinanimate。Likewisealsoinacaseofamphiboly。Iftheargumentdependsonlikenessofexpression，theoppositewillbethesolution。’Couldamangivewhathehasnotgot？’No，notwhathehasnotgot；buthecouldgiveitinawayinwhichhehasnotgotit，e。g。onediebyitself。’

Doesamanknoweitherbylearningorbydiscoveryeachthingthatheknows，singly？butnotthethingsthatheknows，collectively。’

Alsoamantreads，perhaps，onanythinghewalksthrough，butnotonthetimehewalksthrough。Likewisealsointhecaseoftheotherexamples。

24

IndealingwithargumentsthatdependonAccident，oneandthesamesolutionmeetsallcases。Forsinceitisindeterminatewhenanattributeshouldbeascribedtoathing，incaseswhereitbelongstotheaccidentofthething，andsinceinsomecasesitisgenerallyagreedandpeopleadmitthatitbelongs，whileinotherstheydenythatitneedbelong，weshouldtherefore，assoonastheconclusionhasbeendrawn，sayinanswertothemallalike，thatthereisnoneedforsuchanattributetobelong。Onemust，however，bepreparedtoadduceanexampleofthekindofattributemeant。AllargumentssuchasthefollowingdependuponAccident。’DoyouknowwhatIamgoingtoaskyou？youknowthemanwhoisapproaching’，or’themaninthemask’？’Isthestatueyourworkofart？’or’Isthedogyourfather？’’Istheproductofasmallnumberwithasmallnumberasmallnumber？’Foritisevidentinallthesecasesthatthereisnonecessityfortheattributewhichistrueofthething’saccidenttobetrueofthethingaswell。Foronlytothingsthatareindistinguishableandoneinessenceisitgenerallyagreedthatallthesameattributesbelong；whereasinthecaseofagoodthing，tobegoodisnotthesameastobegoingtobethesubjectofaquestion；norinthecaseofamanapproaching，orwearingamask，is’tobeapproaching’thesamethingas’tobeCoriscus’，sothatsupposeIknowCoriscus，butdonotknowthemanwhoisapproaching，itstillisn’tthecasethatIbothknowanddonotknowthesameman；

nor，again，ifthisismineandisalsoaworkofart，isitthereforemyworkofart，butmypropertyorthingorsomethingelse。（Thesolutionisafterthesamemannerintheothercasesaswell。）

Somesolvetheserefutationsbydemolishingtheoriginalpropositionasked：fortheysaythatitispossibletoknowandnottoknowthesamething，onlynotinthesamerespect：accordingly，whentheydon’tknowthemanwhoiscomingtowardsthem，butdoknowCorsicus，theyassertthattheydoknowanddon’tknowthesameobject，butnotinthesamerespect。Yet，aswehavealreadyremarked，thecorrectionofargumentsthatdependuponthesamepointoughttobethesame，whereasthisonewillnotstandifoneadoptsthesameprincipleinregardnottoknowingsomething，buttobeing，ortobeingisainacertainstate，e。g。supposethatXisfather，andisalsoyours：forifinsomecasesthisistrueanditispossibletoknowandnottoknowthesamething，yetwiththatcasethesolutionstatedhasnothingtodo。Certainlythereisnothingtopreventthesameargumentfromhavinganumberofflaws；butitisnottheexpositionofanyandeveryfaultthatconstitutesasolution：foritispossibleforamantoshowthatafalseconclusionhasbeenproved，butnottoshowonwhatitdepends，e。g。inthecaseofZeno’sargumenttoprovethatmotionisimpossible。Sothatevenifanyoneweretotrytoestablishthatthisdoctrineisanimpossibleone，hestillismistaken，andevenifheprovedhiscasetenthousandtimesover，stillthisisnosolutionofZeno’sargument：forthesolutionwasallalonganexpositionoffalsereasoning，showingonwhatitsfalsitydepends。Ifthenhehasnotprovedhiscase，oristryingtoestablishevenatrueproposition，orafalseone，inafalsemanner，topointthisoutisatruesolution。Possibly，indeed，thepresentsuggestionmayverywellapplyinsomecases：butinthesecases，atanyrate，noteventhiswouldbegenerallyagreed：forheknowsboththatCoriscusisCoriscusandthattheapproachingfigureisapproaching。Toknowandnottoknowthesamethingisgenerallythoughttobepossible，whene。g。oneknowsthatXiswhite，butdoesnotrealizethatheismusical：forinthatwayhedoesknowandnotknowthesamething，thoughnotinthesamerespect。ButastotheapproachingfigureandCoriscusheknowsboththatitisapproachingandthatheisCoriscus。

Alikemistaketothatofthosewhomwehavementionedisthatofthosewhosolvetheproofthateverynumberisasmallnumber：forif，whentheconclusionisnotproved，theypassthisoverandsaythataconclusionhasbeenprovedandistrue，onthegroundthateverynumberisbothgreatandsmall，theymakeamistake。

Somepeoplealsousetheprincipleofambiguitytosolvetheaforesaidreasonings，e。g。theproofthat’Xisyourfather’，or’son’，or’slave’。Yetitisevidentthatiftheappearanceaproofdependsuponapluralityofmeanings，theterm，ortheexpressioninquestion，oughttobearanumberofliteralsenses，whereasnoonespeaksofAasbeing’B’schild’intheliteralsense，ifBisthechild’smaster，butthecombinationdependsuponAccident。’IsA

yours？’’Yes。’’AndisAachild？’’Yes。’’ThenthechildAisyours，’

becausehehappenstobebothyoursandachild；butheisnot’yourchild’。

Thereisalsotheproofthat’something\"ofevils\"isgood’；forwisdomisa’knowledge\"ofevils\"’。Buttheexpressionthatthisis’ofsoand—so’（=’so—and—so’s’）hasnotanumberofmeanings：itmeansthatitis’so—and—so’sproperty’。Wemaysupposeofcourse，ontheotherhand，thatithasanumberofmeanings—forwealsosaythatmanis’oftheanimals’，thoughnottheirproperty；andalsothatanytermrelatedto’evils’inawayexpressedbyagenitivecaseisonthataccountaso—and—so’ofevils’，thoughitisnotoneoftheevils—butinthatcasetheapparentlydifferentmeaningsseemtodependonwhetherthetermisusedrelativelyorabsolutely。’Yetitisconceivablypossibletofindarealambiguityinthephrase\"Somethingofevilsisgood\"。’Perhaps，butnotwithregardtothephraseinquestion。Itwouldoccurmorenearly，supposethat’A

servantisgoodofthewicked’；thoughperhapsitisnotquitefoundeventhere：forathingmaybe’good’andbe’X’s’withoutbeingatthesametime’X’sgood’。Noristhesayingthat’Manisoftheanimals’aphrasewithanumberofmeanings：foraphrasedoesnotbecomepossessedofanumberofmeaningsmerelysupposeweexpressitelliptically：forweexpress’GivemetheIliad’byquotinghalfalineofit，e。g。’Giveme\"Sing，goddess，ofthewrath……\"’

25

Thoseargumentswhichdependuponanexpressionthatisvalidofaparticularthing，orinaparticularrespect，orplace，ormanner，orrelation，andnotvalidabsolutely，shouldbesolvedbyconsideringtheconclusioninrelationtoitscontradictory，toseeifanyofthesethingscanpossiblyhavehappenedtoit。Foritisimpossibleforcontrariesandoppositesandanaffirmativeandanegativetobelongtothesamethingabsolutely；thereis，however，nothingtopreventeachfrombelonginginaparticularrespectorrelationormanner，ortopreventoneofthemfrombelonginginaparticularrespectandtheotherabsolutely。Sothatifthisonebelongsabsolutelyandthatoneinaparticularrespect，thereisasyetnorefutation。Thisisafeatureonehastofindintheconclusionbyexaminingitincomparisonwithitscontradictory。

Allargumentsofthefollowingkindhavethisfeature：’Isitpossibleforwhatis—nottobe？\"No。\"But，yousee，itissomething，despiteitsnotbeing。’Likewisealso，Beingwillnotbe；foritwillnothesomeparticularformofbeing。Isitpossibleforthesamemanatthesametimetobeakeeperandabreakerofhisoath？’’Canthesamemanatthesametimebothobeyanddisobeythesameman？’

Orisn’titthecasethatbeingsomethinginparticularandBeingarenotthesame？Ontheotherhand，Not—being，evenifitbesomething，neednotalsohaveabsolute’being’aswell。Norifamankeepshisoathinthisparticularinstanceorinthisparticularrespect，isheboundalsotobeakeeperofoathsabsolutely，buthewhoswearsthathewillbreakhisoath，andthenbreaksit，keepsthisparticularoathonly；heisnotakeeperofhisoath：noristhedisobedientman’obedient’，thoughheobeysoneparticularcommand。

Theargumentissimilar，also，asregardstheproblemwhetherthesamemancanatthesametimesaywhatisbothfalseandtrue：butitappearstobeatroublesomequestionbecauseitisnoteasytoseeinwhichofthetwoconnexionstheword’absolutely’istoberendered—with’true’orwith’false’。Thereis，however，nothingtopreventitfrombeingfalseabsolutely，thoughtrueinsomeparticularrespectorrelation，i。e。beingtrueinsomethings，thoughnot’true’

absolutely。Likewisealsoincasesofsomeparticularrelationandplaceandtime。Forallargumentsofthefollowingkinddependuponthis。’Ishealth，orwealth，agoodthing？’’Yes。’’Buttothefoolwhodoesnotuseitarightitisnotagoodthing：thereforeitisbothgoodandnotgood。’’Ishealth，orpoliticalpower，agoodthing？’’Yes。\"Butsometimesitisnotparticularlygood：thereforethesamethingisbothgoodandnotgoodtothesameman。’Orratherthereisnothingtopreventathing，thoughgoodabsolutely，beingnotgoodtoaparticularman，orbeinggoodtoaparticularman，andyetnotgoodorhere。’Isthatwhichtheprudentmanwouldnotwish，anevil？’’Yes。’’Buttogetridof，hewouldnotwishthegood：

thereforethegoodisanevil。’Butthatisamistake；foritisnotthesamethingtosay’Thegoodisanevil’and’togetridofthegoodisanevil’。Likewisealsotheargumentofthethiefismistaken。

Foritisnotthecasethatifthethiefisanevilthing，acquiringthingsisalsoevil：whathewishes，therefore，isnotwhatisevilbutwhatisgood；fortoacquiresomethinggoodisgood。Also，diseaseisanevilthing，butnottogetridofdisease。’Isthejustpreferabletotheunjust，andwhattakesplacejustlytowhattakesplaceunjustly？’Yes。’’Buttotobeputtodeathunjustlyispreferable。’’Isitjustthateachshouldhavehisown？’’Yes。’’Butwhateverdecisionsamancomestoonthestrengthofhispersonalopinion，evenifitbeafalseopinion，arevalidinlaw：thereforethesameresultisbothjustandunjust。’Also，shouldonedecideinfavourofhimwhosayswhatisunjust？’’Theformer。’’Butyousee，itisjustfortheinjuredpartytosayfullythethingshehassuffered；

andthesearefallacies。Forbecausetosufferathingunjustlyispreferable，unjustwaysarenotthereforepreferable，thoughinthisparticularcasetheunjustmayverywellbebetterthanthejust。

Also，tohaveone’sownisjust，whiletohavewhatisanother’sisnotjust：allthesame，thedecisioninquestionmayverywellbeajustdecision，whateveritbethattheopinionofthemanwhogavethedecisionsupports：forbecauseitisjustinthisparticularcaseorinthisparticularmanner，itisnotalsojustabsolutely。Likewisealso，thoughthingsareunjust，thereisnothingtopreventthespeakingofthembeingjust：forbecausetospeakofthingsisjust，thereisnonecessitythatthethingsshouldbejust，anymorethanbecausetospeakofthingsbeofuse，thethingsneedbeofuse。

Likewisealsointhecaseofwhatisjust。Sothatitisnotthecasethatbecausethethingsspokenofareunjust，thevictorygoestohimwhospeaksunjustthings：forhespeaksofthingsthatarejusttospeakof，thoughabsolutely，i。e。tosuffer，theyareunjust。

26

Refutationsthatdependonthedefinitionofarefutationmust，accordingtotheplansketchedabove，bemetbycomparingtogethertheconclusionwithitscontradictory，andseeingthatitshallinvolvethesameattributeinthesamerespectandrelationandmannerandtime。Ifthisadditionalquestionbeputatthestart，youshouldnotadmitthatitisimpossibleforthesamethingtobebothdoubleandnotdouble，butgrantthatitispossible，onlynotinsuchawayaswasagreedtoconstitutearefutationofyourcase。Allthefollowingargumentsdependuponapointofthatkind。’DoesamanwhoknowsAtobeA，knowthethingcalledA？’andinthesameway，’isonewhoisignorantthatAisAignorantofthethingcalledA？’

’Yes。’’ButonewhoknowsthatCoriscusisCoriscusmightbeignorantofthefactthatheismusical，sothathebothknowsandisignorantofthesamething。’Isathingfourcubitslonggreaterthanathingthreecubitslong？’’Yes。’’Butathingmightgrowfromthreetofourcubitsinlength；’nowwhatis’greater’isgreaterthana’less’：accordinglythethinginquestionwillbebothgreaterandlessthanitselfinthesamerespect。

27

Astorefutationsthatdependonbeggingandassumingtheoriginalpointtobeproved，supposethenatureofthequestiontobeobvious，oneshouldnotgrantit，eventhoughitbeaviewgenerallyheld，butshouldtellhimthetruth。Suppose，however，thatitescapesone，then，thankstothebadnessofargumentsofthatkind，oneshouldmakeone’serrorrecoiluponthequestioner，andsaythathehasbroughtnoargument：forarefutationmustbeprovedindependentlyoftheoriginalpoint。Secondly，oneshouldsaythatthepointwasgrantedundertheimpressionthatheintendednottouseitasapremiss，buttoreasonagainstit，intheoppositewayfromthatadoptedinrefutationsonsideissues。

28

Also，thoserefutationsthatbringonetotheirconclusionthroughtheconsequentyoushouldshowupinthecourseoftheargumentitself。Themodeinwhichconsequencesfollowistwofold。Fortheargumenteitheristhatastheuniversalfollowsonitsparticular—as（e。g。）’animal’followsfrom’man’—sodoestheparticularonitsuniversal：fortheclaimismadethatifAisalwaysfoundwithB，thenBalsoisalwaysfoundwithA。Orelseitproceedsbywayoftheoppositesofthetermsinvolved：forifA

followsB，itisclaimedthatA’soppositewillfollowB’sopposite。

OnthislatterclaimtheargumentofMelissusalsodepends：forheclaimsthatbecausethatwhichhascometobehasabeginning，thatwhichhasnotcometobehasnone，sothatiftheheavenhasnotcometobe，itisalsoeternal。Butthatisnotso；forthesequenceisviceversa。

29

Inthecaseofanyrefutationswhosereasoningdependsonsomeaddition，lookandseeifuponitssubtractiontheabsurdityfollowsnonetheless：andthenifso，theanswerershouldpointthisout，andsaythathegrantedtheadditionnotbecausehereallythoughtit，butforthesakeoftheargument，whereasthequestionerhasnotuseditforthepurposeofhisargumentatall。

30

Tomeetthoserefutationswhichmakeseveralquestionsintoone，oneshoulddrawadistinctionbetweenthemstraightawayatthestart。Foraquestionmustbesingletowhichthereisasingleanswer，sothatonemustnotaffirmordenyseveralthingsofonething，noronethingofmany，butoneofone。Butjustasinthecaseofambiguousterms，anattributebelongstoatermsometimesinbothitssenses，andsometimesinneither，sothatasimpleanswerdoesone，asithappens，noharmdespitethefactthatthequestionisnotsimple，soitisinthesecasesofdoublequestionstoo。Whenever，then，theseveralattributesbelongtotheonesubject，ortheonetothemany，themanwhogivesasimpleanswerencountersnoobstacleeventhoughhehascommittedthismistake：butwheneveranattributebelongstoonesubjectbutnottotheother，orthereisaquestionofanumberofattributesbelongingtoanumberofsubjectsandinonesensebothbelongtoboth，whileinanothersense，again，theydonot，thenthereistrouble，sothatonemustbewareofthis。Thus（e。g。）inthefollowingarguments：SupposingtobegoodandBevil，youwill，ifyougiveasingleansweraboutboth，becompelledtosaythatitistruetocallthesegood，andthatitistruetocallthemevilandlikewisetocallthemneithergoodnorevil（foreachofthemhasnoteachcharacter），sothatthesamethingwillbebothgoodandevilandneithergoodnorevil。Also，sinceeverythingisthesameasitselfanddifferentfromanythingelse，inasmuchasthemanwhoanswersdoublequestionssimplycanbemadetosaythatseveralthingsare’thesame’notasotherthingsbut’asthemselves’，andalsothattheyaredifferentfromthemselves，itfollowsthatthesamethingsmustbeboththesameasanddifferentfromthemselves。Moreover，ifwhatisgoodbecomesevilwhilewhatisevilisgood，thentheymustbothbecometwo。Sooftwounequalthingseachbeingequaltoitself，itwillfollowthattheyarebothequalandunequaltothemselves。

Nowtheserefutationsfallintotheprovinceofothersolutionsaswell：for’both’and’all’havemorethanonemeaning，sothattheresultingaffirmationanddenialofthesamethingdoesnotoccur，exceptverbally：andthisisnotwhatwemeantbyarefutation。Butitisclearthatiftherebenotputasinglequestiononanumberofpoints，buttheanswererhasaffirmedordeniedoneattributeonlyofonesubjectonly，theabsurditywillnotcometopass。

31

Withregardtothosewhodrawoneintorepeatingthesamethinganumberoftimes，itisclearthatonemustnotgrantthatpredicationsofrelativetermshaveanymeaninginabstractionbythemselves，e。g。that’double’isasignificanttermapartfromthewholephrase’doubleofhalf’merelyonthegroundthatitfiguresinit。Fortenfiguresin’tenminusone’andin’notdo’，andgenerallytheaffirmationinthenegation；butforallthat，supposeanyoneweretosay，’Thisisnotwhite’，hedoesnotsaythatitiswhite。Thebareword’double’，onemayperhapssay，hasnotevenanymeaningatall，anymorethanhas’the’in’thehalf’：andevenifithasameaning，yetithasnotthesamemeaningasinthecombination。Noris’knowledge’thesamethinginaspecificbranchofit（supposeit，e。g。tobe’medicalknowledge’）asitisingeneral：foringeneralitwasthe’knowledgeoftheknowable’。Inthecaseoftermsthatarepredicatedofthetermsthroughwhichtheyaredefined，youshouldsaythesamething，thatthetermdefinedisnotthesameinabstractionasitisinthewholephrase。For’concave’hasageneralmeaningwhichisthesameinthecaseofasnubnose，andofabandyleg，butwhenaddedtoeithersubstantivenothingpreventsitfromdifferentiatingitsmeaning；infactitbearsonesenseasappliedtothenose，andanotherasappliedtotheleg：forintheformerconnexionitmeans’snub’andinthelatter’bandyshaped’；i。e。itmakesnodifferencewhetheryousay’asnubnose’or’aconcavenose’。

Moreover，theexpressionmustnotbegrantedinthenominativecase：

foritisafalsehood。Forsnubnessisnotaconcavenosebutsomething（e。g。anaffection）belongingtoanose：hence，thereisnoabsurdityinsupposingthatthesnubnoseisanosepossessingtheconcavitythatbelongstoanose。

32

Withregardtosolecisms，wehavepreviouslysaidwhatitisthatappearstobringthemabout；themethodoftheirsolutionwillbeclearinthecourseoftheargumentsthemselves。Solecismistheresultaimedatinallargumentsofthefollowingkind：’Isathingtrulythatwhichyoutrulycallit？’’Yes’。’But，speakingofastone，youcallhimreal：thereforeofastoneitfollowsthat\"himisreal\"。’No：rather，talkingofastonemeansnotsayingwhich’but’whom’，andnot’that’but’him’。If，then，anyoneweretoask，’Isastonehimwhomyoutrulycallhim？’hewouldbegenerallythoughtnottobespeakinggoodGreek，anymorethanifheweretoask，’Ishewhatyoucallher？’Speakinthiswayofa’stick’oranyneuterword，andthedifferencedoesnotbreakout。Forthisreason，also，nosolecismisincurred，supposeanyoneasks，’Isathingwhatyousayittobe？’’Yes’。’But，speakingofastick，youcallitreal：

therefore，ofastickitfollowsthatitisreal。’’Stone’，however，and’he’havemasculinedesignations。Nowsupposesomeoneweretoask，’Can\"he\"beashe\"（afemale）？’，andthenagain，’Well，butisnotheCoriscus？’andthenweretosay，’Thenheisa\"she\"，’hehasnotprovedthesolecism，evenifthename’Coriscus’doessignifya’she’，if，ontheotherhand，theanswererdoesnotgrantthis：thispointmustbeputasanadditionalquestion：whileifneitherisitthefactnordoeshegrantit，thenthesophisthasnotprovedhiscaseeitherinfactorasagainstthepersonhehasbeenquestioning。Inlikemanner，then，intheaboveinstanceaswellitmustbedefinitelyputthat’he’meansthestone。If，however，thisneitherissonorisgranted，theconclusionmustnotbestated：

thoughitfollowsapparently，becausethecase（theaccusative），thatisreallyunlike，appearstobelikethenominative。’Isittruetosaythatthisobjectiswhatyoucallitbyname？’’Yes’。’Butyoucallitbythenameofashield：thisobjectthereforeis\"ofashield\"。’No：notnecessarily，becausethemeaningof’thisobject’isnot’ofashield’but’ashield’：’ofashield’wouldbethemeaningof’thisobject’s’。Noragainif’Heiswhatyoucallhimbyname’，while’thenameyoucallhimbyisCleon’s’，ishetherefore’Cleon’s’：forheisnot’Cleon’s’，forwhatwassaidwasthat’He，nothis，iswhatIcallhimbyname’。Forthequestion，ifputinthelatterway，wouldnotevenbeGreek。’Doyouknowthis？’’Yes。’

’Butthisishe：thereforeyouknowhe’。No：rather’this’hasnotthesamemeaningin’Doyouknowthis？’asin’Thisisastone’；inthefirstitstandsforanaccusative，inthesecondforanominativecase。’Whenyouhaveunderstandingofanything，doyouunderstandit？’

’Yes。’’Butyouhaveunderstandingofastone：thereforeyouunderstandofastone。’No：theonephraseisinthegenitive，’ofastone’，whiletheotherisintheaccusative，’astone’：andwhatwasgrantedwasthat’youunderstandthat，notofthat，ofwhichyouhaveunderstanding’，sothatyouunderstandnot’ofastone’，but’thestone’。

Thusthatargumentsofthiskinddonotprovesolecismbutmerelyappeartodoso，andbothwhytheysoappearandhowyoushouldmeetthem，isclearfromwhathasbeensaid。

33

Wemustalsoobservethatofalltheargumentsaforesaiditiseasierwithsometoseewhyandwherethereasoningleadsthehearerastray，whilewithothersitismoredifficult，thoughoftentheyarethesameargumentsastheformer。Forwemustcallanargumentthesameifitdependsuponthesamepoint；butthesameargumentisapttobethoughtbysometodependondiction，byothersonaccident，andbyothersonsomethingelse，becauseeachofthem，whenworkedwithdifferentterms，isnotsoclearasitwas。Accordingly，justasinfallaciesthatdependonambiguity，whicharegenerallythoughttobethesilliestformoffallacy，somearecleareventothemaninthestreet（forhumorousphrasesnearlyalldependondiction；e。g。’Themangotthecartdownfromthestand’；and’Whereareyoubound？’

’Totheyardarm’；and’Whichcowwillcalveafore？’’Neither，butbothbehind；’and’IstheNorthwindclear？’’No，indeed；forithasmurderedthebeggarandthemerchant。\"IsheaGoodenough—King？’’No，indeed；aRob—son’：andsowiththegreatmajorityoftherestaswell），whileothersappeartoeludethemostexpert（anditisasymptomofthisthattheyoftenfightabouttheirterms，e。g。

whetherthemeaningof’Being’and’One’isthesameinalltheirapplicationsordifferent；forsomethinkthat’Being’and’One’

meanthesame；whileotherssolvetheargumentofZenoandParmenidesbyassertingthat’One’and’Being’areusedinanumberofsenses），likewisealsoasregardsfallaciesofAccidentandeachoftheothertypes，someoftheargumentswillbeeasiertoseewhileothersaremoredifficult；alsotograsptowhichclassafallacybelongs，andwhetheritisarefutationornotarefutation，isnotequallyeasyinallcases。

Anincisiveargumentisonewhichproducesthegreatestperplexity：forthisistheonewiththesharpestfang。Nowperplexityistwofold，onewhichoccursinreasonedarguments，respectingwhichofthepropositionsaskedoneistodemolish，andtheotherincontentiousarguments，respectingthemannerinwhichoneistoassenttowhatispropounded。Thereforeitisinsyllogisticargumentsthatthemoreincisiveonesproducethekeenestheart—searching。Nowasyllogisticargumentismostincisiveiffrompremissesthatareasgenerallyacceptedaspossibleitdemolishesaconclusionthatisacceptedasgenerallyaspossible。Fortheoneargument，ifthecontradictoryischangedabout，makesalltheresultingsyllogismsalikeincharacter：foralwaysfrompremissesthataregenerallyaccepteditwillproveaconclusion，negativeorpositiveasthecasemaybe，thatisjustasgenerallyaccepted；andthereforeoneisboundtofeelperplexed。Anargument，then，ofthiskindisthemostincisive，viz。theonethatputsitsconclusiononallfourswiththepropositionsasked；andsecondcomestheonethatarguesfrompremisses，allofwhichareequallyconvincing：forthiswillproduceanequalperplexityastowhatkindofpremiss，ofthoseasked，oneshoulddemolish。Hereinisadifficulty：foronemustdemolishsomething，butwhatonemustdemolishisuncertain。Ofcontentiousarguments，ontheotherhand，themostincisiveistheonewhich，inthefirstplace，ischaracterizedbyaninitialuncertaintywhetherithasbeenproperlyreasonedornot；andalsowhetherthesolutiondependsonafalsepremissoronthedrawingofadistinction；while，oftherest，thesecondplaceisheldbythatwhosesolutionclearlydependsuponadistinctionorademolition，andyetitdoesnotrevealclearlywhichitisofthepremissesasked，whosedemolition，orthedrawingofadistinctionwithinit，willbringthesolutionabout，butevenleavesitvaguewhetheritisontheconclusionorononeofthepremissesthatthedeceptiondepends。

Nowsometimesanargumentwhichhasnotbeenproperlyreasonedissilly，supposingtheassumptionsrequiredtobeextremelycontrarytothegeneralvieworfalse；butsometimesitoughtnottobeheldincontempt。Forwheneversomequestionisleftout，ofthekindthatconcernsboththesubjectandthenerveoftheargument，thereasoningthathasbothfailedtosecurethisaswell，andalsofailedtoreasonproperly，issilly；butwhenwhatisomittedissomeextraneousquestion，thenitisbynomeanstobelightlydespised，buttheargumentisquiterespectable，thoughthequestionerhasnotputhisquestionswell。

Justasitispossibletobringasolutionsometimesagainsttheargument，atothersagainstthequestionerandhismodeofquestioning，andatothersagainstneitherofthese，likewisealsoitispossibletomarshalone’squestionsandreasoningbothagainstthethesis，andagainsttheanswererandagainstthetime，wheneverthesolutionrequiresalongertimetoexaminethantheperiodavailable。

34

Astothenumber，then，andkindofsourceswhencefallaciesariseindiscussion，andhowwearetoshowthatouropponentiscommittingafallacyandmakehimutterparadoxes；moreover，bytheuseofwhatmaterialssolescismisbroughtabout，andhowtoquestionandwhatisthewaytoarrangethequestions；moreover，astothequestionwhatuseisservedbyallargumentsofthiskind，andconcerningtheanswerer’spart，bothasawholeingeneral，andinparticularhowtosolveargumentsandsolecisms—onallthesethingslettheforegoingdiscussionsuffice。Itremainstorecallouroriginalproposalandtobringourdiscussiontoaclosewithafewwordsuponit。

Ourprogrammewas，then，todiscoversomefacultyofreasoningaboutanythemeputbeforeusfromthemostgenerallyacceptedpremissesthatthereare。Forthatistheessentialtaskoftheartofdiscussion（dialectic）andofexamination（peirastic）。Inasmuch，however，asitisannexedtoit，onaccountofthenearpresenceoftheartofsophistry（sophistic），notonlytobeabletoconductanexaminationdialecticallybutalsowithashowofknowledge，wethereforeproposedforourtreatisenotonlytheaforesaidaimofbeingabletoexactanaccountofanyview，butalsotheaimofensuringthatinstandinguptoanargumentweshalldefendourthesisinthesamemannerbymeansofviewsasgenerallyheldaspossible。

Thereasonofthiswehaveexplained；forthis，too，waswhySocratesusedtoaskquestionsandnottoanswerthem；forheusedtoconfessthathedidnotknow。Wehavemadeclear，inthecourseofwhatprecedes，thenumberbothofthepointswithreferencetowhich，andofthematerialsfromwhich，thiswillbeaccomplished，andalsofromwhatsourceswecanbecomewellsuppliedwiththese：wehaveshown，moreover，howtoquestionorarrangethequestioningasawhole，andtheproblemsconcerningtheanswersandsolutionstobeusedagainstthereasoningsofthequestioner。Wehavealsocleareduptheproblemsconcerningallothermattersthatbelongtothesameinquiryintoarguments。InadditiontothiswehavebeenthroughthesubjectofFallacies，aswehavealreadystatedabove。

Thatourprogramme，then，hasbeenadequatelycompletedisclear。

Butwemustnotomittonoticewhathashappenedinregardtothisinquiry。Forinthecaseofalldiscoveriestheresultsofpreviouslaboursthathavebeenhandeddownfromothershavebeenadvancedbitbybitbythosewhohavetakenthemon，whereastheoriginaldiscoveriesgenerallymakeadvancethatissmallatfirstthoughmuchmoreusefulthanthedevelopmentwhichlaterspringsoutofthem。

Foritmaybethatineverything，asthesayingis，’thefirststartisthemainpart’：andforthisreasonalsoitisthemostdifficult；forinproportionasitismostpotentinitsinfluence，soitissmallestinitscompassandthereforemostdifficulttosee：

whereaswhenthisisoncediscovered，itiseasiertoaddanddeveloptheremainderinconnexionwithit。Thisisinfactwhathashappenedinregardtorhetoricalspeechesandtopracticallyalltheotherarts：forthosewhodiscoveredthebeginningsofthemadvancedtheminallonlyalittleway，whereasthecelebritiesofto—dayaretheheirs（sotospeak）ofalongsuccessionofmenwhohaveadvancedthembitbybit，andsohavedevelopedthemtotheirpresentform，Tisiascomingnextafterthefirstfounders，thenThrasymachusafterTisias，andTheodorusnexttohim，whileseveralpeoplehavemadetheirseveralcontributionstoit：andthereforeitisnottobewonderedatthatthearthasattainedconsiderabledimensions。Ofthisinquiry，ontheotherhand，itwasnotthecasethatpartoftheworkhadbeenthoroughlydonebefore，whileparthadnot。Nothingexistedatall。ForthetraininggivenbythepaidprofessorsofcontentiousargumentswaslikethetreatmentofthematterbyGorgias。Fortheyusedtohandoutspeechestobelearnedbyheart，somerhetorical，othersintheformofquestionandanswer，eachsidesupposingthattheirargumentsoneithersidegenerallyfallamongthem。Andthereforetheteachingtheygavetheirpupilswasreadybutrough。Fortheyusedtosupposethattheytrainedpeoplebyimpartingtothemnottheartbutitsproducts，asthoughanyoneprofessingthathewouldimpartaformofknowledgetoobviateanypaininthefeet，werethennottoteachamantheartofshoe—makingorthesourceswhencehecanacquireanythingofthekind，butweretopresenthimwithseveralkindsofshoesofallsorts：

forhehashelpedhimtomeethisneed，buthasnotimpartedanarttohim。Moreover，onthesubjectofRhetoricthereexistsmuchthathasbeensaidlongago，whereasonthesubjectofreasoningwehadnothingelseofanearlierdatetospeakofatall，butwerekeptatworkforalongtimeinexperimentalresearches。If，then，itseemstoyouafterinspectionthat，suchbeingthesituationasitexistedatthestart，ourinvestigationisinasatisfactoryconditioncomparedwiththeotherinquiriesthathavebeendevelopedbytradition，theremustremainforallofyou，orforourstudents，thetaskofextendingusyourpardonfortheshortcomingsoftheinquiry，andforthediscoveriesthereofyourwarmthanks。