,,! ,,problems ,,( ,,philosophy ,by ,b]tr& ,russell ,,preface ,9 ! foll[+ pages ,i h conf9$ myself 9 ! ma9 to ^? problems ( philosophy 9 reg>d to :i* ,i ?\ght x possible to say some?+ positive & con/ructive1 s9ce m]ely negative criticism seem$ out ( place4 ,= ?is reason1 !ory ( k occupies a l>g] space ?an metaphysics 9 ! pres5t volume1 & "s topics mu* discuss$ by philosoph]s are treat$ v briefly1 if at all4 ,i h d]iv$ valuable assi/ance f unpubli%$ writ+s ( ,g4 ,e4 ,moore & ,j4 ,m4 ,keynes3 f ! =m]1 z reg>ds ! relations ( s5se-data to physical objects1 & f ! latt] z reg>ds probability & induction4 ,i h also pr(it$ greatly by ! criticisms & sugge/ions ( ,pr(essor ,gilb]t ,murray4 #aiab ,,*apt] ,i4 ,,appe>ance ,,& ,,reality ,is !re any k 9 ! _w :i* is s c]ta9 t no reasonable man c\ld d\bt x8 ,?is "q1 :i* at fir/ sight might n seem difficult1 is really "o ( ! mo/ difficult t c be ask$4 ,:5 we h realiz$ ! ob/acles 9 ! way ( a /raight=w>d & confid5t answ]1 we %all be well laun*$ on ! /udy ( philosophy--= philosophy is m]ely ! attempt to answ] su* ultimate que/ions1 n c>elessly & dogmatically1 z we d 9 ord9>y life & ev5 9 ! sci5ces1 b critically1 aft] explor+ all t makes su* que/ions puzzl+1 & aft] realiz+ all ! vagu5ess & confusion t und]lie our ord9>y ideas4 ,9 daily life1 we assume z c]ta9 _m ?+s :i*1 on a clos] scrut9y1 are f\nd to be s full ( app>5t contradictions t only a great am\nt ( ?\ght enables u to "k :at x is t we really may believe4 ,9 ! se>* = c]ta9ty1 x is natural to beg9 ) our pres5t exp]i5ces1 & 9 "s s5se1 no d\bt1 k is to be d]iv$ f !m4 ,b any /atem5t z to :at x is t our imm$iate exp]i5ces make u "k is v likely to be wrong4 ,x seems to me t ,i am n[ sitt+ 9 a *air1 at a table ( a c]ta9 %ape1 on :i* ,i see %eets ( pap] ) writ+ or pr9t4 ,by turn+ my head ,i see out ( ! w9d[ build+s & cl\ds & ! sun4 ,i believe t ! sun is ab\t n9ety-?ree million miles f ! e>?2 t x is a hot globe _m times bigg] ?an ! e>?2 t1 ow+ to ! e>?'s rotation1 x rises e morn+1 & w cont9ue to d s = an indef9ite "t 9 ! future4 ,i believe t1 if any o!r normal p]son comes into my room1 he w see ! same *airs & tables & books & pap]s z ,i see1 & t ! table :i* ,i see is ! same z ! table :i* ,i feel press+ aga9/ my arm4 ,all ?is seems to be s evid5t z to be h>dly wor? /at+1 except 9 answ] to a man :o d\bts :e!r ,i "k any?+4 ,yet all ?is may be reasonably d\bt$1 & all ( x requires mu* c>eful discussion be=e we c be sure t we h /at$ x 9 a =m t is :olly true4 ,to make our difficulties pla91 let u conc5trate att5tion on ! table4 ,to ! eye x is oblong1 br[n & %9y1 to ! t\* x is smoo? & cool & h>d2 :5 ,i tap x1 x gives out a wood5 s\nd4 ,any "o else :o sees & feels & he>s ! table w agree ) ?is description1 s t x might seem z if no difficulty w\ld arise2 b z soon z we try to be m precise our tr\bles beg94 ,al?\gh ,i believe t ! table is 'really' ( ! same col\r all ov]1 ! p>ts t reflect ! light look mu* bright] ?an ! o!r p>ts1 & "s p>ts look :ite because ( reflect$ light4 ,i "k t1 if ,i move1 ! p>ts t reflect ! light w be diff]5t1 s t ! app>5t di/ribution ( col\rs on ! table w *ange4 ,x foll[s t if sev]al p are look+ at ! table at ! same mom5t1 no two ( !m w see exactly ! same di/ribution ( col\rs1 because no two c see x f exactly ! same po9t ( view1 & any *ange 9 ! po9t ( view makes "s *ange 9 ! way ! light is reflect$4 ,= mo/ practical purposes ^! diff]5ces are unimportant1 b to ! pa9t] !y are all-important3 ! pa9t] has to unle>n ! habit ( ?9k+ t ?+s seem to h ! col\r :i* common s5se says !y 'really' h1 & to le>n ! habit ( see+ ?+s z !y appe>4 ,"h we h already ! beg9n+ ( "o ( ! di/9ctions t cause mo/ tr\ble 9 philosophy--! di/9ction betwe5 'appe>ance' & 'reality'1 betwe5 :at ?+s seem to be & :at !y are4 ,! pa9t] wants to "k :at ?+s seem to be1 ! practical man & ! philosoph] want to "k :at !y are2 b ! philosoph]'s wi% to "k ?is is /rong] ?an ! practical man's1 & is m tr\bl$ by k z to ! difficulties ( answ]+ ! "q4 ,to return to ! table4 ,x is evid5t f :at we h f\nd1 t !re is no col\r :i* pre-em95tly appe>s to be ! col\r ( ! table1 or ev5 ( any "o p>ticul> "p ( ! table--x appe>s to be ( diff]5t col\rs f diff]5t po9ts ( view1 & !re is no reason = reg>d+ "s ( ^! z m really its col\r ?an o!rs4 ,& we "k t ev5 f a giv5 po9t ( view ! col\r w seem diff]5t by artificial light1 or to a col\r-bl9d man1 or to a man we>+ blue spectacles1 :ile 9 ! d>k !re w be no col\r at all1 ?\gh to t\* & he>+ ! table w be un*ang$4 ,?is col\r is n some?+ :i* is inh]5t 9 ! table1 b some?+ dep5d+ ^u ! table & ! spectator & ! way ! light falls on ! table4 ,:51 9 ord9>y life1 we speak ( ! col\r ( ! table1 we only mean ! sort ( col\r :i* x w seem to h to a normal spectator f an ord9>y po9t ( view "u usual conditions ( light4 ,b ! o!r col\rs :i* appe> "u o!r conditions h j z good a "r to be consid]$ real2 & !re=e1 to avoid fav\ritism1 we are compell$ to d5y t1 9 itself1 ! table has any "o p>ticul> col\r4 ,! same ?+ applies to ! texture4 ,) ! nak$ eye "o c see ! gra91 b o!rwise ! table looks smoo? & ev54 ,if we look$ at x ?r\gh a microscope1 we %\ld see r\ghnesses & hills & valleys1 & all sorts ( diff]5ces t are imp]ceptible to ! nak$ eye4 ,:i* ( ^! is ! 'real' table8 ,we are naturally tempt$ to say t :at we see ?r\gh ! microscope is m real1 b t 9 turn w\ld be *ang$ by a /ill m p[]ful microscope4 ,if1 !n1 we _c tru/ :at we see ) ! nak$ eye1 :y %\ld we tru/ :at we see ?r\gh a microscope8 ,?us1 aga91 ! confid5ce 9 our s5ses ) :i* we began des]ts u4 ,! %ape ( ! table is no bett]4 ,we are all 9 ! habit ( judg+ z to ! 'real' %apes ( ?+s1 & we d ?is s unreflect+ly t we come to ?9k we actually see ! real %apes4 ,b1 9 fact1 z we all h to le>n if we try to draw1 a giv5 ?+ looks diff]5t 9 %ape f e diff]5t po9t ( view4 ,if our table is 'really' rectangul>1 x w look1 f almo/ all po9ts ( view1 z if x _h two acute angles & two obtuse angles4 ,if opposite sides are p>allel1 !y w look z if !y conv]g$ to a po9t away f ! spectator2 if !y are ( equal l5g?1 !y w look z if ! ne>] side 7 long]4 ,all ^! ?+s are n commonly notic$ 9 look+ at a table1 because exp]i5ce has taught u to con/ruct ! 'real' %ape f ! app>5t %ape1 & ! 'real' %ape is :at int]e/s u z practical m54 ,b ! 'real' %ape is n :at we see2 x is some?+ inf]r$ f :at we see4 ,& :at we see is con/antly *ang+ 9 %ape z we move ab\t ! room2 s t "h aga9 ! s5ses seem n to give u ! tru? ab\t ! table itself1 b only ab\t ! appe>ance ( ! table4 ,simil> difficulties arise :5 we consid] ! s5se ( t\*4 ,x is true t ! table always gives u a s5sation ( h>dness1 & we feel t x resi/s pressure4 ,b ! s5sation we obta9 dep5ds ^u h[ h>d we press ! table & also ^u :at "p ( ! body we press )2 ?us ! v>i\s s5sations due to v>i\s pressures or v>i\s p>ts ( ! body _c be suppos$ to reveal directly any def9ite prop]ty ( ! table1 b at mo/ to be signs ( "s prop]ty :i* p]haps causes all ! s5sations1 b is n actually app>5t 9 any ( !m4 ,& ! same applies /ill m obvi\sly to ! s\nds :i* c be elicit$ by rapp+ ! table4 ,?us x becomes evid5t t ! real table1 if !re is "o1 is n ! same z :at we imm$iately exp]i5ce by sight or t\* or he>+4 ,! real table1 if !re is "o1 is n imm$iately kn[n to u at all1 b mu/ be an inf]5ce f :at is imm$iately kn[n4 ,h5ce1 two v difficult que/ions at once arise2 namely1 7#a7 ,is !re a real table at all8 7#b7 ,if s1 :at sort ( object c x be8 ,x w help u 9 consid]+ ^! que/ions to h a few simple t]ms ( :i* ! mean+ is def9ite & cle>4 ,let u give ! "n ( 's5se-data' to ! ?+s t are imm$iately kn[n 9 s5sation3 su* ?+s z col\rs1 s\nds1 smells1 h>dnesses1 r\ghnesses1 & s on4 ,we %all give ! "n 's5sation' to ! exp]i5ce ( be+ imm$iately aw>e ( ^! ?+s4 ,?us1 :5ev] we see a col\r1 we h a s5sation ( ! col\r1 b ! col\r itself is a s5se-datum1 n a s5sation4 ,! col\r is t ( :i* we are imm$iately aw>e1 & ! aw>5ess itself is ! s5sation4 ,x is pla9 t if we are to "k any?+ ab\t ! table1 x mu/ be by means ( ! s5se-data--br[n col\r1 oblong %ape1 smoo?ness1 etc4--:i* we associate ) ! table2 b1 = ! reasons :i* h be5 giv51 we _c say t ! table is ! s5se-data1 or ev5 t ! s5se-data are directly prop]ties ( ! table4 ,?us a problem arises z to ! relation ( ! s5se-data to ! real table1 suppos+ !re is su* a ?+4 ,! real table1 if x exi/s1 we w call a 'physical object'4 ,?us we h to consid] ! relation ( s5se-data to physical objects4 ,! collection ( all physical objects is call$ 'matt]'4 ,?us our two que/ions may be re-/at$ z foll[s3 7#a7 ,is !re any su* ?+ z matt]8 7#b7 ,if s1 :at is its nature8 ,! philosoph] :o fir/ br\ght prom95tly =w>d ! reasons = reg>d+ ! imm$iate objects ( our s5ses z n exi/+ indep5d5tly ( u 0 ,bi%op ,b]keley 7#afhe-agec74 ,8 ,?ree ,dialogues betwe5 ,hylas & ,philon\s1 9 ,opposition to ,sceptics & ,a!i/s1 und]take to prove t !re is no su* ?+ z matt] at all1 & t ! _w consi/s ( no?+ b m9ds & _! ideas4 ,hylas has hi!rto believ$ 9 matt]1 b he is no mat* = ,philon\s1 :o m]cilessly drives him into contradictions & p>adoxes1 & makes 8 own d5ial ( matt] seem1 9 ! end1 z if x 7 almo/ common s5se4 ,! argum5ts employ$ are ( v diff]5t value3 "s are important & s\nd1 o!rs are confus$ or quibbl+4 ,b ,b]keley reta9s ! m]it ( hav+ %[n t ! exi/5ce ( matt] is capable ( be+ d5i$ )\t absurdity1 & t if !re are any ?+s t exi/ indep5d5tly ( u !y _c be ! imm$iate objects ( our s5sations4 ,!re are two diff]5t que/ions involv$ :5 we ask :e!r matt] exi/s1 & x is important to keep !m cle>4 ,we commonly mean by 'matt]' some?+ :i* is oppos$ to 'm9d'1 some?+ :i* we ?9k ( z occupy+ space & z radically incapable ( any sort ( ?\ght or consci\sness4 ,x is *iefly 9 ?is s5se t ,b]keley d5ies matt]2 t is to say1 he does n d5y t ! s5se-data :i* we commonly take z signs ( ! exi/5ce ( ! table are really signs ( ! exi/5ce ( some?+ indep5d5t ( u1 b he does d5y t ?is some?+ is non-m5tal1 t x is nei!r m9d nor ideas ent]ta9$ by "s m9d4 ,he admits t !re mu/ be some?+ :i* cont9ues to exi/ :5 we g out ( ! room or %ut our eyes1 & t :at we call see+ ! table does really give u reason = believ+ 9 some?+ :i* p]si/s ev5 :5 we are n see+ x4 ,b he ?9ks t ?is some?+ _c be radically diff]5t 9 nature f :at we see1 & _c be indep5d5t ( see+ altoge!r1 ?\gh x mu/ be indep5d5t ( our see+4 ,he is ?us l$ to reg>d ! 'real' table z an idea 9 ! m9d ( ,god4 ,su* an idea has ! requir$ p]man5ce & indep5d5ce ( ourselves1 )\t be+--z matt] w\ld o!rwise be--some?+ q unkn[able1 9 ! s5se t we c only inf] x1 & c nev] be directly & imm$iately aw>e ( x4 ,o!r philosoph]s s9ce ,b]keley h also held t1 al?\gh ! table does n dep5d = its exi/5ce ^u be+ se5 by me1 x does dep5d ^u be+ se5 7or o!rwise appreh5d$ 9 s5sation7 by "s m9d--n necess>ily ! m9d ( ,god1 b m (t5 ! :ole collective m9d ( ! univ]se4 ,?is !y hold1 z ,b]keley does1 *iefly because !y ?9k !re c be no?+ real--or at any rate no?+ kn[n to be real except m9ds & _! ?\ghts & feel+s4 ,we might /ate ! argum5t by :i* !y support _! view 9 "s su* way z ?is3 ',:atev] c be ?\ght ( is an idea 9 ! m9d ( ! p]son ?9k+ ( x2 !re=e no?+ c be ?\ght ( except ideas 9 m9ds2 !re=e any?+ else is inconceivable1 & :at is inconceivable _c exi/4' ,su* an argum5t1 9 my op9ion1 is fallaci\s2 & ( c\rse ^? :o advance x d n put x s %ortly or s crudely4 ,b :e!r valid or n1 ! argum5t has be5 v widely advanc$ 9 "o =m or ano!r2 & v _m philosoph]s1 p]haps a majority1 h held t !re is no?+ real except m9ds & _! ideas4 ,su* philosoph]s are call$ 'ideali/s'4 ,:5 !y come to expla9+ matt]1 !y ei!r say1 l ,b]keley1 t matt] is really no?+ b a collection ( ideas1 or !y say1 l ,leibniz 7#afdf-agaf71 t :at appe>s z matt] is really a collection ( m or less rudim5t>y m9ds4 ,b ^! philosoph]s1 ?\gh !y d5y matt] z oppos$ to m9d1 nev]!less1 9 ano!r s5se1 admit matt]4 ,x w be rememb]$ t we ask$ two que/ions2 namely1 7#a7 ,is !re a real table at all8 7#b7 ,if s1 :at sort ( object c x be8 ,n[ bo? ,b]keley & ,leibniz admit t !re is a real table1 b ,b]keley says x is c]ta9 ideas 9 ! m9d ( ,god1 & ,leibniz says x is a colony ( s\ls4 ,?us bo? ( !m answ] our fir/ "q 9 ! affirmative1 & only div]ge f ! views ( ord9>y mortals 9 _! answ] to our second "q4 ,9 fact1 almo/ all philosoph]s seem to be agre$ t !re is a real table3 !y almo/ all agree t1 h[ev] mu* our s5se-data--col\r1 %ape1 smoo?ness1 etc4--may dep5d ^u u1 yet _! occurr5ce is a sign ( some?+ exi/+ indep5d5tly ( u1 some?+ diff]+1 p]haps1 completely f our s5se-data1 & yet to be reg>d$ z caus+ ^? s5se-data :5ev] we are 9 a suitable relation to ! real table4 ,n[ obvi\sly ?is po9t 9 :i* ! philosoph]s are agre$--! view t !re is a real table1 :atev] its nature may be--is vitally important1 & x w be wor? :ile to consid] :at reasons !re are = accept+ ?is view be=e we g on to ! fur!r "q z to ! nature ( ! real table4 ,our next *apt]1 !re=e1 w be conc]n$ ) ! reasons = suppos+ t !re is a real table at all4 ,be=e we g f>!r x w be well to consid] = a mom5t :at x is t we h discov]$ s f>4 ,x has appe>$ t1 if we take any common object ( ! sort t is suppos$ to be kn[n by ! s5ses1 :at ! s5ses imm$iately tell u is n ! tru? ab\t ! object z x is ap>t f u1 b only ! tru? ab\t c]ta9 s5se-data :i*1 s f> z we c see1 dep5d ^u ! relations betwe5 u & ! object4 ,?us :at we directly see & feel is m]ely 'appe>ance'1 :i* we believe to be a sign ( "s 'reality' beh9d4 ,b if ! reality is n :at appe>s1 h we any means ( kn[+ :e!r !re is any reality at all8 ,& if s1 h we any means ( f9d+ out :at x is l8 ,su* que/ions are bewild]+1 & x is difficult to "k t ev5 ! /range/ hypo!ses may n be true4 ,?us our famili> table1 :i* has r\s$ b ! slighte/ ?\ghts 9 u hi!rto1 has become a problem full ( surpris+ possibilities4 ,! "o ?+ we "k ab\t x is t x is n :at x seems4 ,beyond ?is mode/ result1 s f>1 we h ! mo/ complete lib]ty ( conjecture4 ,leibniz tells u x is a community ( s\ls3 ,b]keley tells u x is an idea 9 ! m9d ( ,god2 sob] sci5ce1 sc>cely less wond]ful1 tells u x is a va/ collection ( electric *>ges 9 viol5t motion4 ,among ^! surpris+ possibilities1 d\bt sugge/s t p]haps !re is no table at all4 ,philosophy1 if x _c answ] s _m que/ions z we c\ld wi%1 has at lea/ ! p[] ( ask+ que/ions :i* increase ! int]e/ ( ! _w1 & %[ ! /rang5ess & wond] ly+ j bel[ ! surface ev5 9 ! commone/ ?+s ( daily life4 ,,*apt] ,,ii4 ,,! ,,exi/5ce ,,( ,,matt] ,9 ?is *apt] we h to ask ourselves :e!r1 9 any s5se at all1 !re is su* a ?+ z matt]4 ,is !re a table :i* has a c]ta9 intr9sic nature1 & cont9ues to exi/ :5 ,i am n look+1 or is ! table m]ely a product ( my imag9ation1 a dream-table 9 a v prolong$ dream8 ,?is "q is ( ! greate/ importance4 ,= if we _c be sure ( ! indep5d5t exi/5ce ( objects1 we _c be sure ( ! indep5d5t exi/5ce ( o!r p's bodies1 & !re=e /ill less ( o!r p's m9ds1 s9ce we h no gr\nds = believ+ 9 _! m9ds except su* z are d]iv$ f obs]v+ _! bodies4 ,?us if we _c be sure ( ! indep5d5t exi/5ce ( objects1 we %all be left alone 9 a des]t--x may be t ! :ole out] _w is no?+ b a dream1 & t we alone exi/4 ,?is is an uncom=table possibility2 b al?\gh x _c be /rictly prov$ to be false1 !re is n ! slighte/ reason to suppose t x is true4 ,9 ?is *apt] we h to see :y ?is is ! case4 ,be=e we emb>k ^u d\btful matt]s1 let u try to f9d "s m or less fix$ po9t f :i* to />t4 ,al?\gh we are d\bt+ ! physical exi/5ce ( ! table1 we are n d\bt+ ! exi/5ce ( ! s5se-data :i* made u ?9k !re 0 a table2 we are n d\bt+ t1 :ile we look1 a c]ta9 col\r & %ape appe> to u1 & :ile we press1 a c]ta9 s5sation ( h>dness is exp]i5c$ by u4 ,all ?is1 :i* is psy*ological1 we are n call+ 9 "q4 ,9 fact1 :atev] else may be d\btful1 "s at lea/ ( our imm$iate exp]i5ces seem absolutely c]ta94 ,desc>tes 7#aeif-afej71 ! f\nd] ( mod]n philosophy1 inv5t$ a me?od :i* may /ill be us$ ) pr(it--! me?od ( sy/ematic d\bt4 ,he det]m9$ t he w\ld believe no?+ :i* he did n see q cle>ly & di/9ctly to be true4 ,:atev] he c\ld br+ himself to d\bt1 he w\ld d\bt1 until he saw reason = n d\bt+ x4 ,by apply+ ?is me?od he gradually became conv9c$ t ! only exi/5ce ( :i* he c\ld be q c]ta9 0 8 own4 ,he imag9$ a deceitful demon1 :o pres5t$ unreal ?+s to 8 s5ses 9 a p]petual phantasmagoria2 x might be v improbable t su* a demon exi/$1 b /ill x 0 possible1 & !re=e d\bt conc]n+ ?+s p]ceiv$ by ! s5ses 0 possible4 ,b d\bt conc]n+ 8 own exi/5ce 0 n possible1 = if he did n exi/1 no demon c\ld deceive him4 ,if he d\bt$1 he mu/ exi/2 if he _h any exp]i5ces :atev]1 he mu/ exi/4 ,?us 8 own exi/5ce 0 an absolute c]ta9ty to him4 ',i ?9k1 !re=e ,i am1' he said 7,cogito1 ergo sum72 & on ! basis ( ?is c]ta9ty he set to "w to build up aga9 ! _w ( k :i* 8 d\bt _h laid 9 ru9s4 ,by inv5t+ ! me?od ( d\bt1 & by %[+ t subjective ?+s are ! mo/ c]ta91 ,desc>tes p]=m$ a great s]vice to philosophy1 & "o :i* makes him /ill useful to all /ud5ts ( ! subject4 ,b "s c>e is ne$$ 9 us+ ,desc>tes' argum5t4 ',i ?9k1 !re=e ,i am' says r m ?an is /rictly c]ta94 ,x might seem z ?\gh we 7 q sure ( be+ ! same p]son to-day z we 7 ye/]day1 & ?is is no d\bt true 9 "s s5se4 ,b ! real ,self is z h>d to arrive at z ! real table1 & does n seem to h t absolute1 conv9c+ c]ta9ty t belongs to p>ticul> exp]i5ces4 ,:5 ,i look at my table & see a c]ta9 br[n col\r1 :at is q c]ta9 at once is n ',i am see+ a br[n col\r'1 b r1 'a br[n col\r is be+ se5'4 ,?is ( c\rse involves some?+ 7or somebody7 :i* 7or :o7 sees ! br[n col\r2 b x does n ( itself involve t m or less p]man5t p]son :om we call ',i'4 ,s f> z imm$iate c]ta9ty goes1 x might be t ! some?+ :i* sees ! br[n col\r is q mom5t>y1 & n ! same z ! some?+ :i* has "s diff]5t exp]i5ce ! next mom5t4 ,?us x is our p>ticul> ?\ghts & feel+s t h primitive c]ta9ty4 ,& ?is applies to dreams & halluc9ations z well z to normal p]ceptions3 :5 we dream or see a gho/1 we c]ta9ly d h ! s5sations we ?9k we h1 b = v>i\s reasons x is held t no physical object corresponds to ^! s5sations4 ,?us ! c]ta9ty ( our k ( our own exp]i5ces does n h to be limit$ 9 any way to all[ = exceptional cases4 ,"h1 !re=e1 we h1 = :at x is wor?1 a solid basis f :i* to beg9 our pursuit ( k4 ,! problem we h to consid] is ?is3 ,grant$ t we are c]ta9 ( our own s5se-data1 h we any reason = reg>d+ !m z signs ( ! exi/5ce ( some?+ else1 :i* we c call ! physical object8 ,:5 we h enum]at$ all ! s5se-data :i* we %\ld naturally reg>d z connect$ ) ! table1 h we said all !re is to say ab\t ! table1 or is !re /ill some?+ else--some?+ n a s5se-datum1 some?+ :i* p]si/s :5 we g out ( ! room8 ,common s5se unhesitat+ly answ]s t !re is4 ,:at c be b\ght & sold & pu%$ ab\t & h a clo? laid on x1 & s on1 _c be a m]e collection ( s5se-data4 ,if ! clo? completely hides ! table1 we %all d]ive no s5se-data f ! table1 & !re=e1 if ! table 7 m]ely s5se-data1 x w\ld h ceas$ to exi/1 & ! clo? w\ld be susp5d$ 9 empty air1 re/+1 by a miracle1 9 ! place :]e ! table =m]ly 04 ,?is seems pla9ly absurd2 b :oev] wi%es to become a philosoph] mu/ le>n n to be fright5$ by absurdities4 ,"o great reason :y x is felt t we mu/ secure a physical object 9 addition to ! s5se-data1 is t we want ! same object = diff]5t p4 ,:5 t5 p are sitt+ r\nd a d9n]-table1 x seems prepo/]\s to ma9ta9 t !y are n see+ ! same tableclo?1 ! same knives & =ks & spoons & glasses4 ,b ! s5se-data are private to ea* sep>ate p]son2 :at is imm$iately pres5t to ! sight ( "o is n imm$iately pres5t to ! sight ( ano!r3 !y all see ?+s f slightly diff]5t po9ts ( view1 & !re=e see !m slightly diff]5tly4 ,?us1 if !re are to be public neutral objects1 :i* c be 9 "s s5se kn[n to _m diff]5t p1 !re mu/ be some?+ ov] & above ! private & p>ticul> s5se-data :i* appe> to v>i\s p4 ,:at reason1 !n1 h we = believ+ t !re are su* public neutral objects8 ,! fir/ answ] t naturally occurs to "o is t1 al?\gh diff]5t p may see ! table slightly diff]5tly1 /ill !y all see m or less simil> ?+s :5 !y look at ! table1 & ! v>iations 9 :at !y see foll[ ! laws ( p]spective & reflection ( light1 s t x is easy to arrive at a p]man5t object und]ly+ all ! diff]5t p's s5se-data4 ,i b\ght my table f ! =m] occupant ( my room2 ,i c\ld n buy 8 s5se-data1 :i* di$ :5 he w5t away1 b ,i c\ld & did buy ! confid5t expectation ( m or less simil> s5se-data4 ,?us x is ! fact t diff]5t p h simil> s5se-data1 & t "o p]son 9 a giv5 place at diff]5t times has simil> s5se-data1 :i* makes u suppose t ov] & above ! s5se-data !re is a p]man5t public object :i* und]lies or causes ! s5se-data ( v>i\s p at v>i\s times4 ,n[ 9 s f> z ! above consid]ations dep5d ^u suppos+ t !re are o!r p besides ourselves1 !y beg ! v "q at issue4 ,o!r p are repres5t$ to me by c]ta9 s5se-data1 su* z ! sight ( !m or ! s\nd ( _! voices1 & if ,i _h no reason to believe t !re 7 physical objects indep5d5t ( my s5se-data1 ,i %\ld h no reason to believe t o!r p exi/ except z "p ( my dream4 ,?us1 :5 we are try+ to %[ t !re mu/ be objects indep5d5t ( our own s5se-data1 we _c appeal to ! te/imony ( o!r p1 s9ce ?is te/imony itself consi/s ( s5se-data1 & does n reveal o!r p's exp]i5ces unless our own s5se-data are signs ( ?+s exi/+ indep5d5tly ( u4 ,we mu/ !re=e1 if possible1 f9d1 9 our own purely private exp]i5ces1 *>act]i/ics :i* %[1 or t5d to %[1 t !re are 9 ! _w ?+s o!r ?an ourselves & our private exp]i5ces4 ,9 "o s5se x mu/ be admitt$ t we c nev] prove ! exi/5ce ( ?+s o!r ?an ourselves & our exp]i5ces4 ,no logical absurdity results f ! hypo!sis t ! _w consi/s ( myself & my ?\ghts & feel+s & s5sations1 & t ev]y?+ else is m]e fancy4 ,9 dreams a v complicat$ _w may seem to be pres5t1 & yet on wak+ we f9d x 0 a delusion2 t is to say1 we f9d t ! s5se-data 9 ! dream d n appe> to h correspond$ ) su* physical objects z we %\ld naturally inf] f our s5se-data4 7,x is true t1 :5 ! physical _w is assum$1 x is possible to f9d physical causes = ! s5se-data 9 dreams3 a door bang+1 = in/ance1 may cause u to dream ( a naval engagem5t4 ,b al?\gh1 9 ?is case1 !re is a physical cause = ! s5se-data1 !re is n a physical object correspond+ to ! s5se-data 9 ! way 9 :i* an actual naval battle w\ld correspond47 ,!re is no logical impossibility 9 ! supposition t ! :ole ( life is a dream1 9 :i* we ourselves create all ! objects t come be=e u4 ,b al?\gh ?is is n logically impossible1 !re is no reason :atev] to suppose t x is true2 & x is1 9 fact1 a less simple hypo!sis1 view$ z a means ( acc\nt+ = ! facts ( our own life1 ?an ! common-s5se hypo!sis t !re really are objects indep5d5t ( u1 ^: action on u causes our s5sations4 ,! way 9 :i* simplicity comes 9 f suppos+ t !re really are physical objects is easily se54 ,if ! cat appe>s at "o mom5t 9 "o "p ( ! room1 & at ano!r 9 ano!r "p1 x is natural to suppose t x has mov$ f ! "o to ! o!r1 pass+ ov] a s]ies ( int]m$iate positions4 ,b if x is m]ely a set ( s5se-data1 x _c h ev] be5 9 any place :]e ,i did n see x2 ?us we %all h to suppose t x did n exi/ at all :ile ,i 0 n look+1 b sudd5ly sprang into be+ 9 a new place4 ,if ! cat exi/s :e!r ,i see x or n1 we c und]/& f our own exp]i5ce h[ x gets hungry betwe5 "o meal & ! next2 b if x does n exi/ :5 ,i am n see+ x1 x seems odd t appetite %\ld gr[ dur+ non-exi/5ce z fa/ z dur+ exi/5ce4 ,& if ! cat consi/s only ( s5se-data1 x _c be hungry1 s9ce no hung] b my own c be a s5se-datum to me4 ,?us ! behavi\r ( ! s5se-data :i* repres5t ! cat to me1 ?\gh x seems q natural :5 reg>d$ z an expression ( hung]1 becomes utt]ly inexplicable :5 reg>d$ z m]e movem5ts & *anges ( pat*es ( col\r1 :i* are z incapable ( hung] z a triangle is ( play+ football4 ,b ! difficulty 9 ! case ( ! cat is no?+ comp>$ to ! difficulty 9 ! case ( human be+s4 ,:5 human be+s speak--t is1 :5 we he> c]ta9 noises :i* we associate ) ideas1 & simultane\sly see c]ta9 motions ( lips & expressions ( face--x is v difficult to suppose t :at we he> is n ! expression ( a ?\ght1 z we "k x w\ld be if we emitt$ ! same s\nds4 ,( c\rse simil> ?+s happ5 9 dreams1 :]e we are mi/ak5 z to ! exi/5ce ( o!r p4 ,b dreams are m or less sugge/$ by :at we call wak+ life1 & are capable ( be+ m or less acc\nt$ = on sci5tific pr9ciples if we assume t !re really is a physical _w4 ,?us e pr9ciple ( simplicity urges u to adopt ! natural view1 t !re really are objects o!r ?an ourselves & our s5se-data :i* h an exi/5ce n dep5d5t ^u our p]ceiv+ !m4 ,( c\rse x is n by argum5t t we orig9ally come by our belief 9 an indep5d5t ext]nal _w4 ,we f9d ?is belief ready 9 ourselves z soon z we beg9 to reflect3 x is :at may be call$ an in/9ctive belief4 ,we %\ld nev] h be5 l$ to "q ?is belief b = ! fact t1 at any rate 9 ! case ( sight1 x seems z if ! s5se-datum itself 7 in/9ctively believ$ to be ! indep5d5t object1 :]eas argum5t %[s t ! object _c be id5tical ) ! s5se-datum4 ,?is discov]y1 h[ev]--:i* is n at all p>adoxical 9 ! case ( ta/e & smell & s\nd1 & only slightly s 9 ! case ( t\*--leaves undim9i%$ our in/9ctive belief t !re are objects correspond+ to our s5se-data4 ,s9ce ?is belief does n lead to any difficulties1 b on ! contr>y t5ds to simplify & sy/ematize our acc\nt ( our exp]i5ces1 !re seems no good reason = reject+ x4 ,we may !re=e admit--?\gh ) a slight d\bt d]iv$ f dreams--t ! ext]nal _w does really exi/1 & is n :olly dep5d5t = its exi/5ce ^u our cont9u+ to p]ceive x4 ,! argum5t :i* has l$ u to ?is conclusion is d\btless less /rong ?an we c\ld wi%1 b x is typical ( _m philosophical argum5ts1 & x is !re=e wor? :ile to consid] briefly its g5]al "* & validity4 ,all k1 we f9d1 mu/ be built up ^u our in/9ctive beliefs1 & if ^! are reject$1 no?+ is left4 ,b among our in/9ctive beliefs "s are mu* /rong] ?an o!rs1 :ile _m h1 by habit & association1 become entangl$ ) o!r beliefs1 n really in/9ctive1 b falsely suppos$ to be "p ( :at is believ$ in/9ctively4 ,philosophy %\ld %[ u ! hi]>*y ( our in/9ctive beliefs1 beg9n+ ) ^? we hold mo/ /rongly1 & pres5t+ ea* z mu* isolat$ & z free f irrelevant additions z possible4 ,x %\ld take c>e to %[ t1 9 ! =m 9 :i* !y are f9ally set =?1 our in/9ctive beliefs d n cla%1 b =m a h>moni\s sy/em4 ,!re c nev] be any reason = reject+ "o in/9ctive belief except t x cla%es ) o!rs2 ?us1 if !y are f\nd to h>monize1 ! :ole sy/em becomes wor?y ( acceptance4 ,x is ( c\rse possible t all or any ( our beliefs may be mi/ak51 & !re=e all ought to be held ) at lea/ "s slight elem5t ( d\bt4 ,b we _c h reason to reject a belief except on ! gr\nd ( "s o!r belief4 ,h5ce1 by organiz+ our in/9ctive beliefs & _! consequ5ces1 by consid]+ :i* among !m is mo/ possible1 if necess>y1 to modify or ab&on1 we c arrive1 on ! basis ( accept+ z our sole data :at we in/9ctively believe1 at an ord]ly sy/ematic organization ( our k1 9 :i*1 ?\gh ! possibility ( error rema9s1 its likelihood is dim9i%$ by ! int]relation ( ! p>ts & by ! critical scrut9y :i* has prec$$ acquiesc5ce4 ,?is function1 at lea/1 philosophy c p]=m4 ,mo/ philosoph]s1 rightly or wrongly1 believe t philosophy c d mu* m ?an ?is--t x c give u k1 n o!rwise atta9able1 conc]n+ ! univ]se z a :ole1 & conc]n+ ! nature ( ultimate reality4 ,:e!r ?is be ! case or n1 ! m mode/ function we h spok5 ( c c]ta9ly be p]=m$ by philosophy1 & c]ta9ly suffices1 = ^? :o h once begun to d\bt ! adequacy ( common s5se1 to ju/ify ! ardu\s & difficult lab\rs t philosophical problems involve4 ,,*apt] ,,iii4 ,,! ,,nature ,,( ,,matt] ,9 ! prec$+ *apt] we agre$1 ?\gh )\t be+ able to f9d demon/rative reasons1 t x is rational to believe t our s5se-data--= example1 ^? :i* we reg>d z associat$ ) my table--are really signs ( ! exi/5ce ( some?+ indep5d5t ( u & our p]ceptions4 ,t is to say1 ov] & above ! s5sations ( col\r1 h>dness1 noise1 & s on1 :i* make up ! appe>ance ( ! table to me1 ,i assume t !re is some?+ else1 ( :i* ^! ?+s are appe>ances4 ,! col\r ceases to exi/ if ,i %ut my eyes1 ! s5sation ( h>dness ceases to exi/ if ,i remove my arm f contact ) ! table1 ! s\nd ceases to exi/ if ,i cease to rap ! table ) my knuckles4 ,b ,i d n believe t :5 all ^! ?+s cease ! table ceases4 ,on ! contr>y1 ,i believe t x is because ! table exi/s cont9u\sly t all ^! s5se-data w reappe> :5 ,i op5 my eyes1 replace my arm1 & beg9 aga9 to rap ) my knuckles4 ,! "q we h to consid] 9 ?is *apt] is3 ,:at is ! nature ( ?is real table1 :i* p]si/s indep5d5tly ( my p]ception ( x8 ,to ?is "q physical sci5ce gives an answ]1 some:at incomplete x is true1 & 9 "p /ill v hypo!tical1 b yet des]v+ ( respect s f> z x goes4 ,physical sci5ce1 m or less unconsci\sly1 has drift$ into ! view t all natural ph5om5a ought to be r$uc$ to motions4 ,light & heat & s\nd are all due to wave-motions1 :i* travel f ! body emitt+ !m to ! p]son :o sees light or feels heat or he>s s\nd4 ,t :i* has ! wave-motion is ei!r ae!r or 'gross matt]'1 b 9 ei!r case is :at ! philosoph] w\ld call matt]4 ,! only prop]ties :i* sci5ce assigns to x are position 9 space1 & ! p[] ( motion accord+ to ! laws ( motion4 ,sci5ce does n d5y t x may h o!r prop]ties2 b if s1 su* o!r prop]ties are n useful to ! man ( sci5ce1 & 9 no way assi/ him 9 expla9+ ! ph5om5a4 ,x is sometimes said t 'light is a =m ( wave-motion'1 b ?is is mislead+1 = ! light :i* we imm$iately see1 :i* we "k directly by means ( our s5ses1 is n a =m ( wave-motion1 b some?+ q diff]5t--some?+ :i* we all "k if we are n bl9d1 ?\gh we _c describe x s z to convey our k to a man :o is bl9d4 ,a wave-motion1 on ! contr>y1 c\ld q well be describ$ to a bl9d man1 s9ce he c acquire a k ( space by ! s5se ( t\*2 & he c exp]i5ce a wave-motion by a sea voyage almo/ z well z we c4 ,b ?is1 :i* a bl9d man c und]/&1 is n :at we mean by light3 we mean by light j t :i* a bl9d man c nev] und]/&1 & :i* we c nev] describe to him4 ,n[ ?is some?+1 :i* all ( u :o are n bl9d "k1 is n1 accord+ to sci5ce1 really to be f\nd 9 ! out] _w3 x is some?+ caus$ by ! action ( c]ta9 waves ^u ! eyes & n]ves & bra9 ( ! p]son :o sees ! light4 ,:5 x is said t light is waves1 :at is really meant is t waves are ! physical cause ( our s5sations ( light4 ,b light itself1 ! ?+ :i* see+ p exp]i5ce & bl9d p d n1 is n suppos$ by sci5ce to =m any "p ( ! _w t is indep5d5t ( u & our s5ses4 ,& v simil> rem>ks w\ld apply to o!r k9ds ( s5sations4 ,x is n only col\rs & s\nds & s on t are abs5t f ! sci5tific _w ( matt]1 b also space z we get x ?r\gh sight or t\*4 ,x is ess5tial to sci5ce t its matt] %\ld be 9 a space1 b ! space 9 :i* x is _c be exactly ! space we see or feel4 ,to beg9 )1 space z we see x is n ! same z space z we get x by ! s5se ( t\*2 x is only by exp]i5ce 9 infancy t we le>n h[ to t\* ?+s we see1 or h[ to get a sight ( ?+s :i* we feel t\*+ u4 ,b ! space ( sci5ce is neutral z betwe5 t\* & sight2 ?us x _c be ei!r ! space ( t\* or ! space ( sight4 ,aga91 diff]5t p see ! same object z ( diff]5t %apes1 accord+ to _! po9t ( view4 ,a circul> co91 = example1 ?\gh we %\ld always judge x to be circul>1 w look oval unless we are /raight 9 front ( x4 ,:5 we judge t x is circul>1 we are judg+ t x has a real %ape :i* is n its app>5t %ape1 b belongs to x intr9sically ap>t f its appe>ance4 ,b ?is real %ape1 :i* is :at conc]ns sci5ce1 mu/ be 9 a real space1 n ! same z anybody's app>5t space4 ,! real space is public1 ! app>5t space is private to ! p]cipi5t4 ,9 diff]5t p's private spaces ! same object seems to h diff]5t %apes2 ?us ! real space1 9 :i* x has its real %ape1 mu/ be diff]5t f ! private spaces4 ,! space ( sci5ce1 !re=e1 ?\gh connect$ ) ! spaces we see & feel1 is n id5tical ) !m1 & ! mann] ( its connexion requires inve/igation4 ,we agre$ provisionally t physical objects _c be q l our s5se-data1 b may be reg>d$ z caus+ our s5sations4 ,^! physical objects are 9 ! space ( sci5ce1 :i* we may call 'physical' space4 ,x is important to notice t1 if our s5sations are to be caus$ by physical objects1 !re mu/ be a physical space conta9+ ^! objects & our s5se-organs & n]ves & bra94 ,we get a s5sation ( t\* f an object :5 we are 9 contact ) x2 t is to say1 :5 "s "p ( our body occupies a place 9 physical space q close to ! space occupi$ by ! object4 ,we see an object 7r\ghly speak+7 :5 no opaque body is betwe5 ! object & our eyes 9 physical space4 ,simil>ly1 we only he> or smell or ta/e an object :5 we are suffici5tly ne> to x1 or :5 x t\*es ! tongue1 or has "s suitable position 9 physical space relatively to our body4 ,we _c beg9 to /ate :at diff]5t s5sations we %all d]ive f a giv5 object "u diff]5t circum/ances unless we reg>d ! object & our body z bo? 9 "o physical space1 = x is ma9ly ! relative positions ( ! object & our body t det]m9e :at s5sations we %all d]ive f ! object4 ,n[ our s5se-data are situat$ 9 our private spaces1 ei!r ! space ( sight or ! space ( t\* or su* vagu] spaces z o!r s5ses may give u4 ,if1 z sci5ce & common s5se assume1 !re is "o public all-embrac+ physical space 9 :i* physical objects are1 ! relative positions ( physical objects 9 physical space mu/ m or less correspond to ! relative positions ( s5se-data 9 our private spaces4 ,!re is no difficulty 9 suppos+ ?is to be ! case4 ,if we see on a road "o h\se ne>] to u ?an ano!r1 our o!r s5ses w be> out ! view t x is ne>]2 = example1 x w be rea*$ soon] if we walk along ! road4 ,o!r p w agree t ! h\se :i* looks ne>] to u is ne>]2 ! ordnance map w take ! same view2 & ?us ev]y?+ po9ts to a spatial relation betwe5 ! h\ses correspond+ to ! relation betwe5 ! s5se-data :i* we see :5 we look at ! h\ses4 ,?us we may assume t !re is a physical space 9 :i* physical objects h spatial relations correspond+ to ^? :i* ! correspond+ s5se-data h 9 our private spaces4 ,x is ?is physical space :i* is dealt ) 9 geometry & assum$ 9 physics & a/ronomy4 ,assum+ t !re is physical space1 & t x does ?us correspond to private spaces1 :at c we "k ab\t x8 ,we c "k only :at is requir$ 9 ord] to secure ! correspond5ce4 ,t is to say1 we c "k no?+ ( :at x is l 9 itself1 b we c "k ! sort ( arrangem5t ( physical objects :i* results f _! spatial relations4 ,we c "k1 = example1 t ! e>? & moon & sun are 9 "o /raight l9e dur+ an eclipse1 ?\gh we _c "k :at a physical /raight l9e is 9 itself1 z we "k ! look ( a /raight l9e 9 our visual space4 ,?us we come to "k mu* m ab\t ! relations ( di/ances 9 physical space ?an ab\t ! di/ances !mselves2 we may "k t "o di/ance is great] ?an ano!r1 or t x is along ! same /raight l9e z ! o!r1 b we _c h t imm$iate acqua9tance ) physical di/ances t we h ) di/ances 9 our private spaces1 or ) col\rs or s\nds or o!r s5se-data4 ,we c "k all ^? ?+s ab\t physical space :i* a man born bl9d might "k ?r\gh o!r p ab\t ! space ( sight2 b ! k9d ( ?+s :i* a man born bl9d c\ld nev] "k ab\t ! space ( sight we also _c "k ab\t physical space4 ,we c "k ! prop]ties ( ! relations requir$ to pres]ve ! correspond5ce ) s5se-data1 b we _c "k ! nature ( ! t]ms betwe5 :i* ! relations hold4 ,) reg>d to "t1 our feel+ ( duration or ( ! lapse ( "t is notori\sly an unsafe guide z to ! "t t has elaps$ by ! clock4 ,times :5 we are bor$ or suff]+ pa9 pass sl[ly1 times :5 we are agreeably occupi$ pass quickly1 & times :5 we are sleep+ pass almo/ z if !y did n exi/4 ,?us1 9 s f> z "t is con/itut$ by duration1 !re is ! same necessity = di/+ui%+ a public & a private "t z !re 0 9 ! case ( space4 ,b 9 s f> z "t consi/s 9 an ord] ( be=e & aft]1 !re is no ne$ to make su* a di/9ction2 ! "t-ord] :i* ev5ts seem to h is1 s f> z we c see1 ! same z ! "t-ord] :i* !y d h4 ,at any rate no reason c be giv5 = suppos+ t ! two ord]s are n ! same4 ,! same is usually true ( space3 if a regim5t ( m5 are m>*+ along a road1 ! %ape ( ! regim5t w look diff]5t f diff]5t po9ts ( view1 b ! m5 w appe> arrang$ 9 ! same ord] f all po9ts ( view4 ,h5ce we reg>d ! ord] z true also 9 physical space1 :]eas ! %ape is only suppos$ to correspond to ! physical space s f> z is requir$ = ! pres]vation ( ! ord]4 ,9 say+ t ! "t-ord] :i* ev5ts seem to h is ! same z ! "t-ord] :i* !y really h1 x is necess>y to gu>d aga9/ a possible misund]/&+4 ,x mu/ n be suppos$ t ! v>i\s /ates ( diff]5t physical objects h ! same "t-ord] z ! s5se-data :i* con/itute ! p]ceptions ( ^? objects4 ,consid]$ z physical objects1 ! ?und] & lightn+ are simultane\s2 t is to say1 ! lightn+ is simultane\s ) ! di/urbance ( ! air 9 ! place :]e ! di/urbance beg9s1 namely1 :]e ! lightn+ is4 ,b ! s5se-datum :i* we call he>+ ! ?und] does n take place until ! di/urbance ( ! air has travell$ z f> z to :]e we are4 ,simil>ly1 x takes ab\t eight m9utes = ! sun's light to rea* u2 ?us1 :5 we see ! sun we are see+ ! sun ( eight m9utes ago4 ,s f> z our s5se-data af=d evid5ce z to ! physical sun !y af=d evid5ce z to ! physical sun ( eight m9utes ago2 if ! physical sun _h ceas$ to exi/ )9 ! la/ eight m9utes1 t w\ld make no diff]5ce to ! s5se-data :i* we call 'see+ ! sun'4 ,?is af=ds a fre% illu/ration ( ! necessity ( di/+ui%+ betwe5 s5se-data & physical objects4 ,:at we h f\nd z reg>ds space is mu* ! same z :at we f9d 9 relation to ! correspond5ce ( ! s5se-data ) _! physical c\nt]p>ts4 ,if "o object looks blue & ano!r r$1 we may reasonably presume t !re is "s correspond+ diff]5ce betwe5 ! physical objects2 if two objects bo? look blue1 we may presume a correspond+ simil>ity4 ,b we _c hope to be acqua9t$ directly ) ! quality 9 ! physical object :i* makes x look blue or r$4 ,sci5ce tells u t ?is quality is a c]ta9 sort ( wave-motion1 & ?is s\nds famili>1 because we ?9k ( wave-motions 9 ! space we see4 ,b ! wave-motions mu/ really be 9 physical space1 ) :i* we h no direct acqua9tance2 ?us ! real wave-motions h n t famili>ity :i* we might h suppos$ !m to h4 ,& :at holds = col\rs is closely simil> to :at holds = o!r s5se-data4 ,?us we f9d t1 al?\gh ! relations ( physical objects h all sorts ( kn[able prop]ties1 d]iv$ f _! correspond5ce ) ! relations ( s5se-data1 ! physical objects !mselves rema9 unkn[n 9 _! intr9sic nature1 s f> at lea/ z c be discov]$ by means ( ! s5ses4 ,! "q rema9s :e!r !re is any o!r me?od ( discov]+ ! intr9sic nature ( physical objects4 ,! mo/ natural1 ?\gh n ultimately ! mo/ def5sible1 hypo!sis to adopt 9 ! fir/ in/ance1 at any rate z reg>ds visual s5se-data1 w\ld be t1 ?\gh physical objects _c1 = ! reasons we h be5 consid]+1 be exactly l s5se-data1 yet !y may be m or less l4 ,accord+ to ?is view1 physical objects w1 = example1 really h col\rs1 & we might1 by good luck1 see an object z ( ! col\r x really is4 ,! col\r :i* an object seems to h at any giv5 mom5t w 9 g5]al be v simil>1 ?\gh n q ! same1 f _m diff]5t po9ts ( view2 we might ?us suppose ! 'real' col\r to be a sort ( m$ium col\r1 int]m$iate betwe5 ! v>i\s %ades :i* appe> f ! diff]5t po9ts ( view4 ,su* a !ory is p]haps n capable ( be+ def9itely refut$1 b x c be %[n to be gr\ndless4 ,to beg9 )1 x is pla9 t ! col\r we see dep5ds only ^u ! nature ( ! light-waves t /rike ! eye1 & is !re=e modifi$ by ! m$ium int]v5+ betwe5 u & ! object1 z well z by ! mann] 9 :i* light is reflect$ f ! object 9 ! direction ( ! eye4 ,! int]v5+ air alt]s col\rs unless x is p]fectly cle>1 & any /rong reflection w alt] !m completely4 ,?us ! col\r we see is a result ( ! ray z x rea*es ! eye1 & n simply a prop]ty ( ! object f :i* ! ray comes4 ,h5ce1 also1 provid$ c]ta9 waves rea* ! eye1 we %all see a c]ta9 col\r1 :e!r ! object f :i* ! waves />t has any col\r or n4 ,?us x is q gratuit\s to suppose t physical objects h col\rs1 & !re=e !re is no ju/ification = mak+ su* a supposition4 ,exactly simil> argum5ts w apply to o!r s5se-data4 ,x rema9s to ask :e!r !re are any g5]al philosophical argum5ts enabl+ u to say t1 if matt] is real1 x mu/ be ( su* & su* a nature4 ,z expla9$ above1 v _m philosoph]s1 p]haps mo/1 h held t :atev] is real mu/ be 9 "s s5se m5tal1 or at any rate t :atev] we c "k any?+ ab\t mu/ be 9 "s s5se m5tal4 ,su* philosoph]s are call$ 'ideali/s'4 ,ideali/s tell u t :at appe>s z matt] is really some?+ m5tal2 namely1 ei!r 7z ,leibniz held7 m or less rudim5t>y m9ds1 or 7z ,b]keley cont5d$7 ideas 9 ! m9ds :i*1 z we %\ld commonly say1 'p]ceive' ! matt]4 ,?us ideali/s d5y ! exi/5ce ( matt] z some?+ intr9sically diff]5t f m9d1 ?\gh !y d n d5y t our s5se-data are signs ( some?+ :i* exi/s indep5d5tly ( our private s5sations4 ,9 ! foll[+ *apt] we %all consid] briefly ! reasons--9 my op9ion fallaci\s--:i* ideali/s advance 9 fav\r ( _! !ory4 ,,*apt] ,,iv4 ,,idealism ,! ^w 'idealism' is us$ by diff]5t philosoph]s 9 some:at diff]5t s5ses4 ,we %all und]/& by x ! doctr9e t :atev] exi/s1 or at any rate :atev] c be kn[n to exi/1 mu/ be 9 "s s5se m5tal4 ,?is doctr9e1 :i* is v widely held among philosoph]s1 has sev]al =ms1 & is advocat$ on sev]al diff]5t gr\nds4 ,! doctr9e is s widely held1 & s int]e/+ 9 itself1 t ev5 ! briefe/ survey ( philosophy mu/ give "s acc\nt ( x4 ,^? :o are unaccu/om$ to philosophical speculation may be incl9$ to dismiss su* a doctr9e z obvi\sly absurd4 ,!re is no d\bt t common s5se reg>ds tables & *airs & ! sun & moon & mat]ial objects g5]ally z some?+ radically diff]5t f m9ds & ! cont5ts ( m9ds1 & z hav+ an exi/5ce :i* might cont9ue if m9ds ceas$4 ,we ?9k ( matt] z hav+ exi/$ long be=e !re 7 any m9ds1 & x is h>d to ?9k ( x z a m]e product ( m5tal activity4 ,b :e!r true or false1 idealism is n to be dismiss$ z obvi\sly absurd4 ,we h se5 t1 ev5 if physical objects d h an indep5d5t exi/5ce1 !y mu/ diff] v widely f s5se-data1 & c only h a correspond5ce ) s5se-data1 9 ! same sort ( way 9 :i* a catalogue has a correspond5ce ) ! ?+s catalogu$4 ,h5ce common s5se leaves u completely 9 ! d>k z to ! true intr9sic nature ( physical objects1 & if !re 7 good reason to reg>d !m z m5tal1 we c\ld n legitimately reject ?is op9ion m]ely because x /rikes u z /range4 ,! tru? ab\t physical objects mu/ be /range4 ,x may be unatta9able1 b if any philosoph] believes t he has atta9$ x1 ! fact t :at he (f]s z ! tru? is /range ought n to be made a gr\nd ( objection to 8 op9ion4 ,! gr\nds on :i* idealism is advocat$ are g5]ally gr\nds d]iv$ f ! !ory ( k1 t is to say1 f a discussion ( ! conditions :i* ?+s mu/ satisfy 9 ord] t we may be able to "k !m4 ,! fir/ s]i\s attempt to e/abli% idealism on su* gr\nds 0 t ( ,bi%op ,b]keley4 ,he prov$ fir/1 by argum5ts :i* 7 l>gely valid1 t our s5se-data _c be suppos$ to h an exi/5ce indep5d5t ( u1 b mu/ be1 9 "p at lea/1 '9' ! m9d1 9 ! s5se t _! exi/5ce w\ld n cont9ue if !re 7 no see+ or he>+ or t\*+ or smell+ or ta/+4 ,s f>1 8 cont5tion 0 almo/ c]ta9ly valid1 ev5 if "s ( 8 argum5ts 7 n s4 ,b he w5t on to argue t s5se-data 7 ! only ?+s ( ^: exi/5ce our p]ceptions c\ld assure u2 & t to be kn[n is to be '9' a m9d1 & !re=e to be m5tal4 ,h5ce he conclud$ t no?+ c ev] be kn[n except :at is 9 "s m9d1 & t :atev] is kn[n )\t be+ 9 my m9d mu/ be 9 "s o!r m9d4 ,9 ord] to und]/& 8 argum5t1 x is necess>y to und]/& 8 use ( ! ^w 'idea'4 ,he gives ! "n 'idea' to any?+ :i* is imm$iately kn[n1 z1 = example1 s5se-data are kn[n4 ,?us a p>ticul> col\r :i* we see is an idea2 s is a voice :i* we he>1 & s on4 ,b ! t]m is n :olly conf9$ to s5se-data4 ,!re w also be ?+s rememb]$ or imag9$1 = ) su* ?+s also we h imm$iate acqua9tance at ! mom5t ( rememb]+ or imag9+4 ,all su* imm$iate data he calls 'ideas'4 ,he !n proce$s to consid] common objects1 su* z a tree1 = in/ance4 ,he %[s t all we "k imm$iately :5 we 'p]ceive' ! tree consi/s ( ideas 9 8 s5se ( ! ^w1 & he argues t !re is n ! slighte/ gr\nd = suppos+ t !re is any?+ real ab\t ! tree except :at is p]ceiv$4 ,its be+1 he says1 consi/s 9 be+ p]ceiv$3 9 ! ,lat9 ( ! s*oolm5 its 'esse' is 'p]cipi'4 ,he fully admits t ! tree mu/ cont9ue to exi/ ev5 :5 we %ut our eyes or :5 no human be+ is ne> x4 ,b ?is cont9u$ exi/5ce1 he says1 is due to ! fact t ,god cont9ues to p]ceive x2 ! 'real' tree1 :i* corresponds to :at we call$ ! physical object1 consi/s ( ideas 9 ! m9d ( ,god1 ideas m or less l ^? we h :5 we see ! tree1 b diff]+ 9 ! fact t !y are p]man5t 9 ,god's m9d s long z ! tree cont9ues to exi/4 ,all our p]ceptions1 accord+ to him1 consi/ 9 a p>tial p>ticipation 9 ,god's p]ceptions1 & x is because ( ?is p>ticipation t diff]5t p see m or less ! same tree4 ,?us ap>t f m9ds & _! ideas !re is no?+ 9 ! _w1 nor is x possible t any?+ else %\ld ev] be kn[n1 s9ce :atev] is kn[n is necess>ily an idea4 ,!re are 9 ?is argum5t a good _m fallacies :i* h be5 important 9 ! hi/ory ( philosophy1 & :i* x w be z well to br+ to light4 ,9 ! fir/ place1 !re is a confusion eng5d]$ by ! use ( ! ^w 'idea'4 ,we ?9k ( an idea z ess5tially some?+ 9 somebody's m9d1 & ?us :5 we are told t a tree consi/s entirely ( ideas1 x is natural to suppose t1 if s1 ! tree mu/ be entirely 9 m9ds4 ,b ! notion ( be+ '9' ! m9d is ambigu\s4 ,we speak ( be>+ a p]son 9 m9d1 n mean+ t ! p]son is 9 our m9ds1 b t a ?\ght ( him is 9 our m9ds4 ,:5 a man says t "s bus9ess he _h to arrange w5t clean out ( 8 m9d1 he does n mean to imply t ! bus9ess itself 0 ev] 9 8 m9d1 b only t a ?\ght ( ! bus9ess 0 =m]ly 9 8 m9d1 b aft]w>ds ceas$ to be 9 8 m9d4 ,& s :5 ,b]keley says t ! tree mu/ be 9 our m9ds if we c "k x1 all t he really has a "r to say is t a ?\ght ( ! tree mu/ be 9 our m9ds4 ,to argue t ! tree itself mu/ be 9 our m9ds is l argu+ t a p]son :om we be> 9 m9d is himself 9 our m9ds4 ,?is confusion may seem too gross to h be5 really committ$ by any compet5t philosoph]1 b v>i\s att5dant circum/ances r5d]$ x possible4 ,9 ord] to see h[ x 0 possible1 we mu/ g m deeply into ! "q z to ! nature ( ideas4 ,be=e tak+ up ! g5]al "q ( ! nature ( ideas1 we mu/ dis5tangle two entirely sep>ate que/ions :i* arise conc]n+ s5se-data & physical objects4 ,we saw t1 = v>i\s reasons ( detail1 ,b]keley 0 "r 9 treat+ ! s5se-data :i* con/itute our p]ception ( ! tree z m or less subjective1 9 ! s5se t !y dep5d ^u u z mu* z ^u ! tree1 & w\ld n exi/ if ! tree 7 n be+ p]ceiv$4 ,b ?is is an entirely diff]5t po9t f ! "o by :i* ,b]keley seeks to prove t :atev] c be imm$iately kn[n mu/ be 9 a m9d4 ,= ?is purpose argum5ts ( detail z to ! dep5d5ce ( s5se-data ^u u are useless4 ,x is necess>y to prove1 g5]ally1 t by be+ kn[n1 ?+s are %[n to be m5tal4 ,?is is :at ,b]keley believes himself to h done4 ,x is ?is "q1 & n our previ\s "q z to ! diff]5ce betwe5 s5se-data & ! physical object1 t mu/ n[ conc]n u4 ,tak+ ! ^w 'idea' 9 ,b]keley's s5se1 !re are two q di/9ct ?+s to be consid]$ :5ev] an idea is be=e ! m9d4 ,!re is on ! "o h& ! ?+ ( :i* we are aw>e--say ! col\r ( my table--& on ! o!r h& ! actual aw>5ess itself1 ! m5tal act ( appreh5d+ ! ?+4 ,! m5tal act is und\bt$ly m5tal1 b is !re any reason to suppose t ! ?+ appreh5d$ is 9 any s5se m5tal8 ,our previ\s argum5ts conc]n+ ! col\r did n prove x to be m5tal2 !y only prov$ t its exi/5ce dep5ds ^u ! relation ( our s5se organs to ! physical object--9 our case1 ! table4 ,t is to say1 !y prov$ t a c]ta9 col\r w exi/1 9 a c]ta9 light1 if a normal eye is plac$ at a c]ta9 po9t relatively to ! table4 ,!y did n prove t ! col\r is 9 ! m9d ( ! p]cipi5t4 ,b]keley's view1 t obvi\sly ! col\r mu/ be 9 ! m9d1 seems to dep5d = its plausibility ^u confus+ ! ?+ appreh5d$ ) ! act ( appreh5sion4 ,ei!r ( ^! might be call$ an 'idea'2 probably ei!r w\ld h be5 call$ an idea by ,b]keley4 ,! act is und\bt$ly 9 ! m9d2 h5ce1 :5 we are ?9k+ ( ! act1 we readily ass5t to ! view t ideas mu/ be 9 ! m9d4 ,!n1 =gett+ t ?is 0 only true :5 ideas 7 tak5 z acts ( appreh5sion1 we transf] ! proposition t 'ideas are 9 ! m9d' to ideas 9 ! o!r s5se1 i4e4 to ! ?+s appreh5d$ by our acts ( appreh5sion4 ,?us1 by an unconsci\s equivocation1 we arrive at ! conclusion t :atev] we c appreh5d mu/ be 9 our m9ds4 ,?is seems to be ! true analysis ( ,b]keley's argum5t1 & ! ultimate fallacy ^u :i* x re/s4 ,?is "q ( ! di/9ction betwe5 act & object 9 our appreh5d+ ( ?+s is vitally important1 s9ce our :ole p[] ( acquir+ k is b\nd up ) x4 ,! faculty ( be+ acqua9t$ ) ?+s o!r ?an itself is ! ma9 *>act]i/ic ( a m9d4 ,acqua9tance ) objects ess5tially consi/s 9 a relation betwe5 ! m9d & some?+ o!r ?an ! m9d2 x is ?is t con/itutes ! m9d's p[] ( kn[+ ?+s4 ,if we say t ! ?+s kn[n mu/ be 9 ! m9d1 we are ei!r unduly limit+ ! m9d's p[] ( kn[+1 or we are utt]+ a m]e tautology4 ,we are utt]+ a m]e tautology if we mean by '9 ! m9d' ! same z by 'be=e ! m9d'1 i4e4 if we mean m]ely be+ appreh5d$ by ! m9d4 ,b if we mean ?is1 we %all h to admit t :at1 9 ?is s5se1 is 9 ! m9d1 may nev]!less be n m5tal4 ,?us :5 we realize ! nature ( k1 ,b]keley's argum5t is se5 to be wrong 9 sub/ance z well z 9 =m1 & 8 gr\nds = suppos+ t 'ideas'--i4e4 ! objects appreh5d$--mu/ be m5tal1 are f\nd to h no validity :atev]4 ,h5ce 8 gr\nds 9 fav\r ( idealism may be dismiss$4 ,x rema9s to see :e!r !re are any o!r gr\nds4 ,x is (t5 said1 z ?\gh x 7 a self-evid5t truism1 t we _c "k t any?+ exi/s :i* we d n "k4 ,x is inf]r$ t :atev] c 9 any way be relevant to our exp]i5ce mu/ be at lea/ capable ( be+ kn[n by u2 :5ce x foll[s t if matt] 7 ess5tially some?+ ) :i* we c\ld n become acqua9t$1 matt] w\ld be some?+ :i* we c\ld n "k to exi/1 & :i* c\ld h = u no importance :atev]4 ,x is g5]ally also impli$1 = reasons :i* rema9 obscure1 t :at c h no importance = u _c be real1 & t !re=e matt]1 if x is n compos$ ( m9ds or ( m5tal ideas1 is impossible & a m]e *ima]a4 ,to g into ?is argum5t fully at our pres5t /age w\ld be impossible1 s9ce x raises po9ts requir+ a consid]able prelim9>y discussion2 b c]ta9 reasons = reject+ ! argum5t may be notic$ at once4 ,to beg9 at ! end3 !re is no reason :y :at _c h any practical importance = u %\ld n be real4 ,x is true t1 if !oretical importance is includ$1 ev]y?+ real is ( "s importance to u1 s9ce1 z p]sons desir\s ( kn[+ ! tru? ab\t ! univ]se1 we h "s int]e/ 9 ev]y?+ t ! univ]se conta9s4 ,b if ?is sort ( int]e/ is includ$1 x is n ! case t matt] has no importance = u1 provid$ x exi/s ev5 if we _c "k t x exi/s4 ,we c1 obvi\sly1 suspect t x may exi/1 & wond] :e!r x does2 h5ce x is connect$ ) our desire = k1 & has ! importance ( ei!r satisfy+ or ?w>t+ ?is desire4 ,aga91 x is by no means a truism1 & is 9 fact false1 t we _c "k t any?+ exi/s :i* we d n "k4 ,! ^w '"k' is "h us$ 9 two diff]5t s5ses4 7#a7 ,9 its fir/ use x is applicable to ! sort ( k :i* is oppos$ to error1 ! s5se 9 :i* :at we "k is true1 ! s5se :i* applies to our beliefs & convictions1 i4e4 to :at are call$ judgem5ts4 ,9 ?is s5se ( ! ^w we "k t some?+ is ! case4 ,?is sort ( k may be describ$ z k ( tru?s4 7#b7 ,9 ! second use ( ! ^w '"k' above1 ! ^w applies to our k ( ?+s1 :i* we may call acqua9tance4 ,?is is ! s5se 9 :i* we "k s5se-data4 7,! di/9ction involv$ is r\ghly t betwe5 savoir & connatre 9 ,fr5*1 or betwe5 wiss5 & k5n5 9 ,g]man47 ,?us ! /atem5t :i* seem$ l a truism becomes1 :5 re-/at$1 ! foll[+3 ',we c nev] truly judge t some?+ ) :i* we are n acqua9t$ exi/s4' ,?is is by no means a truism1 b on ! contr>y a palpable falsehood4 ,i h n ! hon\r to be acqua9t$ ) ! ,emp]or ( ,*9a1 b ,i truly judge t he exi/s4 ,x may be said1 ( c\rse1 t ,i judge ?is because ( o!r p's acqua9tance ) him4 ,?is1 h[ev]1 w\ld be an irrelevant retort1 s9ce1 if ! pr9ciple 7 true1 ,i c\ld n "k t any "o else is acqua9t$ ) him4 ,b fur!r3 !re is no reason :y ,i %\ld n "k ( ! exi/5ce ( some?+ ) :i* nobody is acqua9t$4 ,?is po9t is important1 & dem&s elucidation4 ,if ,i am acqua9t$ ) a ?+ :i* exi/s1 my acqua9tance gives me ! k t x exi/s4 ,b x is n true t1 conv]sely1 :5ev] ,i c "k t a ?+ ( a c]ta9 sort exi/s1 ,i or "s "o else mu/ be acqua9t$ ) ! ?+4 ,:at happ5s1 9 cases :]e ,i h true judgem5t )\t acqua9tance1 is t ! ?+ is kn[n to me by description1 & t1 9 virtue ( "s g5]al pr9ciple1 ! exi/5ce ( a ?+ answ]+ to ?is description c be inf]r$ f ! exi/5ce ( some?+ ) :i* ,i am acqua9t$4 ,9 ord] to und]/& ?is po9t fully1 x w be well fir/ to deal ) ! diff]5ce betwe5 k by acqua9tance & k by description1 & !n to consid] :at k ( g5]al pr9ciples1 if any1 has ! same k9d ( c]ta9ty z our k ( ! exi/5ce ( our own exp]i5ces4 ,^! subjects w be dealt ) 9 ! foll[+ *apt]s4 ,,*apt] ,v4 ,,k ,,by ,,acqua9tance ,,& ,,k ,,by ,,description ,9 ! prec$+ *apt] we saw t !re are two sorts ( k3 k ( ?+s1 & k ( tru?s4 ,9 ?is *apt] we %all be conc]n$ exclusively ) k ( ?+s1 ( :i* 9 turn we %all h to di/+ui% two k9ds4 ,k ( ?+s1 :5 x is ( ! k9d we call k by acqua9tance1 is ess5tially simpl] ?an any k ( tru?s1 & logically indep5d5t ( k ( tru?s1 ?\gh x w\ld be ra% to assume t human be+s ev]1 9 fact1 h acqua9tance ) ?+s )\t at ! same "t kn[+ "s tru? ab\t !m4 ,k ( ?+s by description1 on ! contr>y1 always involves1 z we %all f9d 9 ! c\rse ( ! pres5t *apt]1 "s k ( tru?s z its s\rce & gr\nd4 ,b fir/ ( all we mu/ make cle> :at we mean by 'acqua9tance' & :at we mean by 'description'4 ,we %all say t we h acqua9tance ) any?+ ( :i* we are directly aw>e1 )\t ! int]m$i>y ( any process ( inf]5ce or any k ( tru?s4 ,?us 9 ! pres5ce ( my table ,i am acqua9t$ ) ! s5se-data t make up ! appe>ance ( my table--its col\r1 %ape1 h>dness1 smoo?ness1 etc42 all ^! are ?+s ( :i* ,i am imm$iately consci\s :5 ,i am see+ & t\*+ my table4 ,! p>ticul> %ade ( col\r t ,i am see+ may h _m ?+s said ab\t x--,i may say t x is br[n1 t x is r d>k1 & s on4 ,b su* /atem5ts1 ?\gh !y make me "k tru?s ab\t ! col\r1 d n make me "k ! col\r itself any bett] ?an ,i did be=e s f> z conc]ns k ( ! col\r itself1 z oppos$ to k ( tru?s ab\t x1 ,i "k ! col\r p]fectly & completely :5 ,i see x1 & no fur!r k ( x itself is ev5 !oretically possible4 ,?us ! s5se-data :i* make up ! appe>ance ( my table are ?+s ) :i* ,i h acqua9tance1 ?+s imm$iately kn[n to me j z !y are4 ,my k ( ! table z a physical object1 on ! contr>y1 is n direct k4 ,su* z x is1 x is obta9$ ?r\gh acqua9tance ) ! s5se-data t make up ! appe>ance ( ! table4 ,we h se5 t x is possible1 )\t absurdity1 to d\bt :e!r !re is a table at all1 :]eas x is n possible to d\bt ! s5se-data4 ,my k ( ! table is ( ! k9d :i* we %all call 'k by description'4 ,! table is '! physical object :i* causes su*-&-su* s5se-data'4 ,?is describes ! table by means ( ! s5se-data4 ,9 ord] to "k any?+ at all ab\t ! table1 we mu/ "k tru?s connect+ x ) ?+s ) :i* we h acqua9tance3 we mu/ "k t 'su*-&-su* s5se-data are caus$ by a physical object'4 ,!re is no /ate ( m9d 9 :i* we are directly aw>e ( ! table2 all our k ( ! table is really k ( tru?s1 & ! actual ?+ :i* is ! table is n1 /rictly speak+1 kn[n to u at all4 ,we "k a description1 & we "k t !re is j "o object to :i* ?is description applies1 ?\gh ! object itself is n directly kn[n to u4 ,9 su* a case1 we say t our k ( ! object is k by description4 ,all our k1 bo? k ( ?+s & k ( tru?s1 re/s ^u acqua9tance z its f\ndation4 ,x is !re=e important to consid] :at k9ds ( ?+s !re are ) :i* we h acqua9tance4 ,s5se-data1 z we h already se51 are among ! ?+s ) :i* we are acqua9t$2 9 fact1 !y supply ! mo/ obvi\s & /rik+ example ( k by acqua9tance4 ,b if !y 7 ! sole example1 our k w\ld be v mu* m re/rict$ ?an x is4 ,we %\ld only "k :at is n[ pres5t to our s5ses3 we c\ld n "k any?+ ab\t ! pa/--n ev5 t !re 0 a pa/--nor c\ld we "k any tru?s ab\t our s5se-data1 = all k ( tru?s1 z we %all %[1 dem&s acqua9tance ) ?+s :i* are ( an ess5tially diff]5t "* f s5se-data1 ! ?+s :i* are sometimes call$ 'ab/ract ideas'1 b :i* we %all call 'univ]sals'4 ,we h !re=e to consid] acqua9tance ) o!r ?+s besides s5se-data if we are to obta9 any tol]ably adequate analysis ( our k4 ,! fir/ ext5sion beyond s5se-data to be consid]$ is acqua9tance by memory4 ,x is obvi\s t we (t5 rememb] :at we h se5 or he>d or _h o!rwise pres5t to our s5ses1 & t 9 su* cases we are /ill imm$iately aw>e ( :at we rememb]1 9 spite ( ! fact t x appe>s z pa/ & n z pres5t4 ,?is imm$iate k by memory is ! s\rce ( all our k conc]n+ ! pa/3 )\t x1 !re c\ld be no k ( ! pa/ by inf]5ce1 s9ce we %\ld nev] "k t !re 0 any?+ pa/ to be inf]r$4 ,! next ext5sion to be consid]$ is acqua9tance by introspection4 ,we are n only aw>e ( ?+s1 b we are (t5 aw>e ( be+ aw>e ( !m4 ,:5 ,i see ! sun1 ,i am (t5 aw>e ( my see+ ! sun2 ?us 'my see+ ! sun' is an object ) :i* ,i h acqua9tance4 ,:5 ,i desire food1 ,i may be aw>e ( my desire = food2 ?us 'my desir+ food' is an object ) :i* ,i am acqua9t$4 ,simil>ly we may be aw>e ( our feel+ pleasure or pa91 & g5]ally ( ! ev5ts :i* happ5 9 our m9ds4 ,?is k9d ( acqua9tance1 :i* may be call$ self-consci\sness1 is ! s\rce ( all our k ( m5tal ?+s4 ,x is obvi\s t x is only :at goes on 9 our own m9ds t c be ?us kn[n imm$iately4 ,:at goes on 9 ! m9ds ( o!rs is kn[n to u ?r\gh our p]ception ( _! bodies1 t is1 ?r\gh ! s5se-data 9 u :i* are associat$ ) _! bodies4 ,b = our acqua9tance ) ! cont5ts ( our own m9ds1 we %\ld be unable to imag9e ! m9ds ( o!rs1 & !re=e we c\ld nev] arrive at ! k t !y h m9ds4 ,x seems natural to suppose t self-consci\sness is "o ( ! ?+s t di/+ui% m5 f animals3 animals1 we may suppose1 ?\gh !y h acqua9tance ) s5se-data1 nev] become aw>e ( ?is acqua9tance4 ,i d n mean t !y d\bt :e!r !y exi/1 b t !y h nev] become consci\s ( ! fact t !y h s5sations & feel+s1 nor !re=e ( ! fact t !y1 ! subjects ( _! s5sations & feel+s1 exi/4 ,we h spok5 ( acqua9tance ) ! cont5ts ( our m9ds z self-consci\sness1 b x is n1 ( c\rse1 consci\sness ( our self3 x is consci\sness ( p>ticul> ?\ghts & feel+s4 ,! "q :e!r we are also acqua9t$ ) our b>e selves1 z oppos$ to p>ticul> ?\ghts & feel+s1 is a v difficult "o1 ^u :i* x w\ld be ra% to speak positively4 ,:5 we try to look into ourselves we always seem to come ^u "s p>ticul> ?\ght or feel+1 & n ^u ! ',i' :i* has ! ?\ght or feel+4 ,nev]!less !re are "s reasons = ?9k+ t we are acqua9t$ ) ! ',i'1 ?\gh ! acqua9tance is h>d to dis5tangle f o!r ?+s4 ,to make cle> :at sort ( reason !re is1 let u consid] = a mom5t :at our acqua9tance ) p>ticul> ?\ghts really involves4 ,:5 ,i am acqua9t$ ) 'my see+ ! sun'1 x seems pla9 t ,i am acqua9t$ ) two diff]5t ?+s 9 relation to ea* o!r4 ,on ! "o h& !re is ! s5se-datum :i* repres5ts ! sun to me1 on ! o!r h& !re is t :i* sees ?is s5se-datum4 ,all acqua9tance1 su* z my acqua9tance ) ! s5se-datum :i* repres5ts ! sun1 seems obvi\sly a relation betwe5 ! p]son acqua9t$ & ! object ) :i* ! p]son is acqua9t$4 ,:5 a case ( acqua9tance is "o ) :i* ,i c be acqua9t$ 7z ,i am acqua9t$ ) my acqua9tance ) ! s5se-datum repres5t+ ! sun71 x is pla9 t ! p]son acqua9t$ is myself4 ,?us1 :5 ,i am acqua9t$ ) my see+ ! sun1 ! :ole fact ) :i* ,i am acqua9t$ is ',self-acqua9t$-)-s5se-datum'4 ,fur!r1 we "k ! tru? ',i am acqua9t$ ) ?is s5se-datum'4 ,x is h>d to see h[ we c\ld "k ?is tru?1 or ev5 und]/& :at is meant by x1 unless we 7 acqua9t$ ) some?+ :i* we call ',i'4 ,x does n seem necess>y to suppose t we are acqua9t$ ) a m or less p]man5t p]son1 ! same to-day z ye/]day1 b x does seem z ?\gh we mu/ be acqua9t$ ) t ?+1 :atev] its nature1 :i* sees ! sun & has acqua9tance ) s5se-data4 ,?us1 9 "s s5se x w\ld seem we mu/ be acqua9t$ ) our ,selves z oppos$ to our p>ticul> exp]i5ces4 ,b ! "q is difficult1 & complicat$ argum5ts c be adduc$ on ei!r side4 ,h5ce1 al?\gh acqua9tance ) ourselves seems probably to occur1 x is n wise to ass]t t x und\bt$ly does occur4 ,we may !re=e sum up z foll[s :at has be5 said conc]n+ acqua9tance ) ?+s t exi/4 ,we h acqua9tance 9 s5sation ) ! data ( ! out] s5ses1 & 9 introspection ) ! data ( :at may be call$ ! inn] s5se--?\ghts1 feel+s1 desires1 etc42 we h acqua9tance 9 memory ) ?+s :i* h be5 data ei!r ( ! out] s5ses or ( ! inn] s5se4 ,fur!r1 x is probable1 ?\gh n c]ta91 t we h acqua9tance ) ,self1 z t :i* is aw>e ( ?+s or has desires t[>ds ?+s4 ,9 addition to our acqua9tance ) p>ticul> exi/+ ?+s1 we also h acqua9tance ) :at we %all call univ]sals1 t is to say1 g5]al ideas1 su* z :it5ess1 div]sity1 bro!rhood1 & s on4 ,e complete s5t5ce mu/ conta9 at lea/ "o ^w :i* /&s = a univ]sal1 s9ce all v]bs h a mean+ :i* is univ]sal4 ,we %all return to univ]sals lat] on1 9 ,*apt] ,,ix2 = ! pres5t1 x is only necess>y to gu>d aga9/ ! supposition t :atev] we c be acqua9t$ ) mu/ be some?+ p>ticul> & exi/5t4 ,aw>5ess ( univ]sals is call$ conceiv+1 & a univ]sal ( :i* we are aw>e is call$ a concept4 ,x w be se5 t among ! objects ) :i* we are acqua9t$ are n includ$ physical objects 7z oppos$ to s5se-data71 nor o!r p's m9ds4 ,^! ?+s are kn[n to u by :at ,i call 'k by description'1 :i* we mu/ n[ consid]4 ,by a 'description' ,i mean any phrase ( ! =m 'a s-&-s' or '! s-&-s'4 ,a phrase ( ! =m 'a s-&-s' ,i %all call an 'ambigu\s' description2 a phrase ( ! =m '! s-&-s' 79 ! s+ul>7 ,i %all call a 'def9ite' description4 ,?us 'a man' is an ambigu\s description1 & '! man ) ! iron mask' is a def9ite description4 ,!re are v>i\s problems connect$ ) ambigu\s descriptions1 b ,i pass !m by1 s9ce !y d n directly conc]n ! matt] we are discuss+1 :i* is ! nature ( our k conc]n+ objects 9 cases :]e we "k t !re is an object answ]+ to a def9ite description1 ?\gh we are n acqua9t$ ) any su* object4 ,?is is a matt] :i* is conc]n$ exclusively ) def9ite descriptions4 ,i %all !re=e1 9 ! sequel1 speak simply ( 'descriptions' :5 ,i mean 'def9ite descriptions'4 ,?us a description w mean any phrase ( ! =m '! s-&-s' 9 ! s+ul>4 ,we %all say t an object is 'kn[n by description' :5 we "k t x is '! s-&-s'1 i4e4 :5 we "k t !re is "o object1 & no m1 hav+ a c]ta9 prop]ty2 & x w g5]ally be impli$ t we d n h k ( ! same object by acqua9tance4 ,we "k t ! man ) ! iron mask exi/$1 & _m propositions are kn[n ab\t him2 b we d n "k :o he 04 ,we "k t ! c&idate :o gets ! mo/ votes w be elect$1 & 9 ?is case we are v likely also acqua9t$ 79 ! only s5se 9 :i* "o c be acqua9t$ ) "s "o else7 ) ! man :o is1 9 fact1 ! c&idate :o w get mo/ votes2 b we d n "k :i* ( ! c&idates he is1 i4e4 we d n "k any proposition ( ! =m ',a is ! c&idate :o w get mo/ votes' :]e ,a is "o ( ! c&idates by "n4 ,we %all say t we h 'm]ely descriptive k' ( ! s-&-s :51 al?\gh we "k t ! s-&-s exi/s1 & al?\gh we may possibly be acqua9t$ ) ! object :i* is1 9 fact1 ! s-&-s1 yet we d n "k any proposition 'a is ! s-&-s'1 :]e a is some?+ ) :i* we are acqua9t$4 ,:5 we say '! s-&-s exi/s'1 we mean t !re is j "o object :i* is ! s-&-s4 ,! proposition 'a is ! s-&-s' means t a has ! prop]ty s-&-s1 & no?+ else has4 ',mr4 ,a4 is ! ,unioni/ c&idate = ?is con/itu5cy' means ',mr4 ,a4 is a ,unioni/ c&idate = ?is con/itu5cy1 & no "o else is'4 ',! ,unioni/ c&idate = ?is con/itu5cy exi/s' means '"s "o is a ,unioni/ c&idate = ?is con/itu5cy1 & no "o else is'4 ,?us1 :5 we are acqua9t$ ) an object :i* is ! s-&-s1 we "k t ! s-&-s exi/s2 b we may "k t ! s-&-s exi/s :5 we are n acqua9t$ ) any object :i* we "k to be ! s-&-s1 & ev5 :5 we are n acqua9t$ ) any object :i*1 9 fact1 is ! s-&-s4 ,common words1 ev5 prop] names1 are usually really descriptions4 ,t is to say1 ! ?\ght 9 ! m9d ( a p]son us+ a prop] "n correctly c g5]ally only be express$ explicitly if we replace ! prop] "n by a description4 ,moreov]1 ! description requir$ to express ! ?\ght w v>y = diff]5t p1 or = ! same p]son at diff]5t times4 ,! only ?+ con/ant 7s long z ! "n is rightly us$7 is ! object to :i* ! "n applies4 ,b s long z ?is rema9s con/ant1 ! p>ticul> description involv$ usually makes no diff]5ce to ! tru? or falsehood ( ! proposition 9 :i* ! "n appe>s4 ,let u take "s illu/rations4 ,suppose "s /atem5t made ab\t ,bism>ck4 ,assum+ t !re is su* a ?+ z direct acqua9tance ) oneself1 ,bism>ck himself might h us$ 8 "n directly to designate ! p>ticul> p]son ) :om he 0 acqua9t$4 ,9 ?is case1 if he made a judgem5t ab\t himself1 he himself might be a con/itu5t ( ! judgem5t4 ,"h ! prop] "n has ! direct use :i* x always wi%es to h1 z simply /&+ = a c]ta9 object1 & n = a description ( ! object4 ,b if a p]son :o knew ,bism>ck made a judgem5t ab\t him1 ! case is diff]5t4 ,:at ?is p]son 0 acqua9t$ ) 7 c]ta9 s5se-data :i* he connect$ 7rightly1 we w suppose7 ) ,bism>ck's body4 ,8 body1 z a physical object1 & /ill m 8 m9d1 7 only kn[n z ! body & ! m9d connect$ ) ^! s5se-data4 ,t is1 !y 7 kn[n by description4 ,x is1 ( c\rse1 v mu* a matt] ( *ance :i* *>act]i/ics ( a man's appe>ance w come into a fri5d's m9d :5 he ?9ks ( him2 ?us ! description actually 9 ! fri5d's m9d is accid5tal4 ,! ess5tial po9t is t he kn[s t ! v>i\s descriptions all apply to ! same entity1 9 spite ( n be+ acqua9t$ ) ! entity 9 "q4 ,:5 we1 :o did n "k ,bism>ck1 make a judgem5t ab\t him1 ! description 9 our m9ds w probably be "s m or less vague mass ( hi/orical k--f> m1 9 mo/ cases1 ?an is requir$ to id5tify him4 ,b1 = ! sake ( illu/ration1 let u assume t we ?9k ( him z '! fir/ ,*ancellor ( ! ,g]man ,empire'4 ,"h all ! words are ab/ract except ',g]man'4 ,! ^w ',g]man' w1 aga91 h diff]5t mean+s = diff]5t p4 ,to "s x w recall travels 9 ,g]many1 to "s ! look ( ,g]many on ! map1 & s on4 ,b if we are to obta9 a description :i* we "k to be applicable1 we %all be compell$1 at "s po9t1 to br+ 9 a ref]5ce to a p>ticul> ) :i* we are acqua9t$4 ,su* ref]5ce is involv$ 9 any m5tion ( pa/1 pres5t1 & future 7z oppos$ to def9ite dates71 or ( "h & !re1 or ( :at o!rs h told u4 ,?us x w\ld seem t1 9 "s way or o!r1 a description kn[n to be applicable to a p>ticul> mu/ involve "s ref]5ce to a p>ticul> ) :i* we are acqua9t$1 if our k ab\t ! ?+ describ$ is n to be m]ely :at foll[s logically f ! description4 ,= example1 '! mo/ long-liv$ ( m5' is a description involv+ only univ]sals1 :i* mu/ apply to "s man1 b we c make no judgem5ts conc]n+ ?is man :i* involve k ab\t him beyond :at ! description gives4 ,if1 h[ev]1 we say1 ',! fir/ ,*ancellor ( ! ,g]man ,empire 0 an a/ute diplomati/'1 we c only be assur$ ( ! tru? ( our judgem5t 9 virtue ( some?+ ) :i* we are acqua9t$--usually a te/imony he>d or read4 ,ap>t f ! in=mation we convey to o!rs1 ap>t f ! fact ab\t ! actual ,bism>ck1 :i* gives importance to our judgem5t1 ! ?\ght we really h conta9s ! "o or m p>ticul>s involv$1 & o!rwise consi/s :olly ( concepts4 ,all names ( places--,london1 ,engl&1 ,europe1 ! ,e>?1 ! ,sol> ,sy/em--simil>ly involve1 :5 us$1 descriptions :i* />t f "s "o or m p>ticul>s ) :i* we are acqua9t$4 ,i suspect t ev5 ! ,univ]se1 z consid]$ by metaphysics1 involves su* a connexion ) p>ticul>s4 ,9 logic1 on ! contr>y1 :]e we are conc]n$ n m]ely ) :at does exi/1 b ) :atev] might or c\ld exi/ or be1 no ref]5ce to actual p>ticul>s is involv$4 ,x w\ld seem t1 :5 we make a /atem5t ab\t some?+ only kn[n by description1 we (t5 int5d to make our /atem5t1 n 9 ! =m involv+ ! description1 b ab\t ! actual ?+ describ$4 ,t is to say1 :5 we say any?+ ab\t ,bism>ck1 we %\ld l1 if we c\ld1 to make ! judgem5t :i* ,bism>ck alone c make1 namely1 ! judgem5t ( :i* he himself is a con/itu5t4 ,9 ?is we are necess>ily defeat$1 s9ce ! actual ,bism>ck is unkn[n to u4 ,b we "k t !re is an object ,b1 call$ ,bism>ck1 & t ,b 0 an a/ute diplomati/4 ,we c ?us describe ! proposition we %\ld l to affirm1 namely1 ',b 0 an a/ute diplomati/'1 :]e ,b is ! object :i* 0 ,bism>ck4 ,if we are describ+ ,bism>ck z '! fir/ ,*ancellor ( ! ,g]man ,empire'1 ! proposition we %\ld l to affirm may be describ$ z '! proposition ass]t+1 conc]n+ ! actual object :i* 0 ! fir/ ,*ancellor ( ! ,g]man ,empire1 t ?is object 0 an a/ute diplomati/'4 ,:at enables u to communicate 9 spite ( ! v>y+ descriptions we employ is t we "k !re is a true proposition conc]n+ ! actual ,bism>ck1 & t h[ev] we may v>y ! description 7s long z ! description is correct7 ! proposition describ$ is /ill ! same4 ,?is proposition1 :i* is describ$ & is kn[n to be true1 is :at int]e/s u2 b we are n acqua9t$ ) ! proposition itself1 & d n "k x1 ?\gh we "k x is true4 ,x w be se5 t !re are v>i\s /ages 9 ! removal f acqua9tance ) p>ticul>s3 !re is ,bism>ck to p :o knew him2 ,bism>ck to ^? :o only "k ( him ?r\gh hi/ory2 ! man ) ! iron mask2 ! longe/-liv$ ( m54 ,^! are progressively fur!r remov$ f acqua9tance ) p>ticul>s2 ! fir/ comes z ne> to acqua9tance z is possible 9 reg>d to ano!r p]son2 9 ! second1 we %all /ill be said to "k ':o ,bism>ck 0'2 9 ! ?ird1 we d n "k :o 0 ! man ) ! iron mask1 ?\gh we c "k _m propositions ab\t him :i* are n logically d$ucible f ! fact t he wore an iron mask2 9 ! f\r?1 f9ally1 we "k no?+ beyond :at is logically d$ucible f ! def9ition ( ! man4 ,!re is a simil> hi]>*y 9 ! region ( univ]sals4 ,_m univ]sals1 l _m p>ticul>s1 are only kn[n to u by description4 ,b "h1 z 9 ! case ( p>ticul>s1 k conc]n+ :at is kn[n by description is ultimately r$ucible to k conc]n+ :at is kn[n by acqua9tance4 ,! fundam5tal pr9ciple 9 ! analysis ( propositions conta9+ descriptions is ?is3 ,e proposition :i* we c und]/& mu/ be compos$ :olly ( con/itu5ts ) :i* we are acqua9t$4 ,we %all n at ?is /age attempt to answ] all ! objections :i* may be urg$ aga9/ ?is fundam5tal pr9ciple4 ,= ! pres5t1 we %all m]ely po9t out t1 9 "s way or o!r1 x mu/ be possible to meet ^! objections1 = x is sc>cely conceivable t we c make a judgem5t or ent]ta9 a supposition )\t kn[+ :at x is t we are judg+ or suppos+ ab\t4 ,we mu/ atta* "s mean+ to ! words we use1 if we are to speak significantly & n utt] m]e noise2 & ! mean+ we atta* to our words mu/ be some?+ ) :i* we are acqua9t$4 ,?us :51 = example1 we make a /atem5t ab\t ,julius ,caes>1 x is pla9 t ,julius ,caes> himself is n be=e our m9ds1 s9ce we are n acqua9t$ ) him4 ,we h 9 m9d "s description ( ,julius ,caes>3 '! man :o 0 assass9at$ on ! ,ides ( ,m>*'1 '! f\nd] ( ! ,roman ,empire'1 or1 p]haps1 m]ely '! man ^: "n 0 ,julius ,caes>'4 7,9 ?is la/ description1 ,julius ,caes> is a noise or %ape ) :i* we are acqua9t$47 ,?us our /atem5t does n mean q :at x seems to mean1 b means some?+ involv+1 in/ead ( ,julius ,caes>1 "s description ( him :i* is compos$ :olly ( p>ticul>s & univ]sals ) :i* we are acqua9t$4 ,! *ief importance ( k by description is t x enables u to pass beyond ! limits ( our private exp]i5ce4 ,9 spite ( ! fact t we c only "k tru?s :i* are :olly compos$ ( t]ms :i* we h exp]i5c$ 9 acqua9tance1 we c yet h k by description ( ?+s :i* we h nev] exp]i5c$4 ,9 view ( ! v n>r[ range ( our imm$iate exp]i5ce1 ?is result is vital1 & until x is und]/ood1 mu* ( our k mu/ rema9 my/]i\s & !re=e d\btful4 ,,*apt] ,,vi4 ,,on ,,induction ,9 almo/ all our previ\s discussions we h be5 conc]n$ 9 ! attempt to get cle> z to our data 9 ! way ( k ( exi/5ce4 ,:at ?+s are !re 9 ! univ]se ^: exi/5ce is kn[n to u ow+ to our be+ acqua9t$ ) !m8 ,s f>1 our answ] has be5 t we are acqua9t$ ) our s5se-data1 &1 probably1 ) ourselves4 ,^! we "k to exi/4 ,& pa/ s5se-data :i* are rememb]$ are kn[n to h exi/$ 9 ! pa/4 ,?is k supplies our data4 ,b if we are to be able to draw inf]5ces f ^! data--if we are to "k ( ! exi/5ce ( matt]1 ( o!r p1 ( ! pa/ be=e our individual memory beg9s1 or ( ! future1 we mu/ "k g5]al pr9ciples ( "s k9d by means ( :i* su* inf]5ces c be drawn4 ,x mu/ be kn[n to u t ! exi/5ce ( "s "o sort ( ?+1 ,a1 is a sign ( ! exi/5ce ( "s o!r sort ( ?+1 ,b1 ei!r at ! same "t z ,a or at "s e>li] or lat] "t1 z1 = example1 ?und] is a sign ( ! e>li] exi/5ce ( lightn+4 ,if ?is 7 n kn[n to u1 we c\ld nev] ext5d our k beyond ! sph]e ( our private exp]i5ce2 & ?is sph]e1 z we h se51 is exce$+ly limit$4 ,! "q we h n[ to consid] is :e!r su* an ext5sion is possible1 & if s1 h[ x is effect$4 ,let u take z an illu/ration a matt] ab\t :i* none ( u1 9 fact1 feel ! slighte/ d\bt4 ,we are all conv9c$ t ! sun w rise to-morr[4 ,:y8 ,is ?is belief a m]e bl9d outcome ( pa/ exp]i5ce1 or c x be ju/ifi$ z a reasonable belief8 ,x is n easy to f9d a te/ by :i* to judge :e!r a belief ( ?is k9d is reasonable or n1 b we c at lea/ asc]ta9 :at sort ( g5]al beliefs w\ld suffice1 if true1 to ju/ify ! judgem5t t ! sun w rise to-morr[1 & ! _m o!r simil> judgem5ts ^u :i* our actions are bas$4 ,x is obvi\s t if we are ask$ :y we believe t ! sun w rise to-morr[1 we %all naturally answ] ',because x always has ris5 e day'4 ,we h a firm belief t x w rise 9 ! future1 because x has ris5 9 ! pa/4 ,if we are *all5g$ z to :y we believe t x w cont9ue to rise z h]et(ore1 we may appeal to ! laws ( motion3 ! e>?1 we %all say1 is a freely rotat+ body1 & su* bodies d n cease to rotate unless some?+ int]f]es f outside1 & !re is no?+ outside to int]f]e ) ! e>? betwe5 n[ & to-morr[4 ,( c\rse x might be d\bt$ :e!r we are q c]ta9 t !re is no?+ outside to int]f]e1 b ?is is n ! int]e/+ d\bt4 ,! int]e/+ d\bt is z to :e!r ! laws ( motion w rema9 9 op]ation until to-morr[4 ,if ?is d\bt is rais$1 we f9d ourselves 9 ! same position z :5 ! d\bt ab\t ! sunrise 0 fir/ rais$4 ,! only reason = believ+ t ! laws ( motion w rema9 9 op]ation is t !y h op]at$ hi!rto1 s f> z our k ( ! pa/ enables u to judge4 ,x is true t we h a great] body ( evid5ce f ! pa/ 9 fav\r ( ! laws ( motion ?an we h 9 fav\r ( ! sunrise1 because ! sunrise is m]ely a p>ticul> case ( fulfilm5t ( ! laws ( motion1 & !re are c\ntless o!r p>ticul> cases4 ,b ! real "q is3 ,d any numb] ( cases ( a law be+ fulfill$ 9 ! pa/ af=d evid5ce t x w be fulfill$ 9 ! future8 ,if n1 x becomes pla9 t we h no gr\nd :atev] = expect+ ! sun to rise to-morr[1 or = expect+ ! bread we %all eat at our next meal n to poison u1 or = any ( ! o!r sc>cely consci\s expectations t control our daily lives4 ,x is to be obs]v$ t all su* expectations are only probable2 ?us we h n to seek = a pro( t !y mu/ be fulfill$1 b only = "s reason 9 fav\r ( ! view t !y are likely to be fulfill$4 ,n[ 9 deal+ ) ?is "q we mu/1 to beg9 )1 make an important di/9ction1 )\t :i* we %\ld soon become involv$ 9 hopeless confusions4 ,exp]i5ce has %[n u t1 hi!rto1 ! frequ5t repetition ( "s uni=m succession or coexi/5ce has be5 a cause ( our expect+ ! same succession or coexi/5ce on ! next occasion4 ,food t has a c]ta9 appe>ance g5]ally has a c]ta9 ta/e1 & x is a sev]e %ock to our expectations :5 ! famili> appe>ance is f\nd to be associat$ ) an unusual ta/e4 ,?+s :i* we see become associat$1 by habit1 ) c]ta9 tactile s5sations :i* we expect if we t\* !m2 "o ( ! horrors ( a gho/ 79 _m gho/-/ories7 is t x fails to give u any s5sations ( t\*4 ,un$ucat$ p :o g abroad = ! fir/ "t are s surpris$ z to be incr$ul\s :5 !y f9d _! native language n und]/ood4 ,& ?is k9d ( association is n conf9$ to m52 9 animals also x is v /rong4 ,a horse :i* has be5 (t5 driv5 along a c]ta9 road resi/s ! attempt to drive him 9 a diff]5t direction4 ,dome/ic animals expect food :5 !y see ! p]son :o usually fe$s !m4 ,we "k t all ^! r crude expectations ( uni=mity are liable to be mislead+4 ,! man :o has f$ ! *ick5 e day ?r\gh\t its life at la/ wr+s its neck in/ead1 %[+ t m ref9$ views z to ! uni=mity ( nature w\ld h be5 useful to ! *ick54 ,b 9 spite ( ! mislead+ness ( su* expectations1 !y nev]!less exi/4 ,! m]e fact t some?+ has happ5$ a c]ta9 numb] ( times causes animals & m5 to expect t x w happ5 aga94 ,?us our in/9cts c]ta9ly cause u to believe t ! sun w rise to-morr[1 b we may be 9 no bett] a position ?an ! *ick5 :i* unexpect$ly has its neck wrung4 ,we h !re=e to di/+ui% ! fact t pa/ uni=mities cause expectations z to ! future1 f ! "q :e!r !re is any reasonable gr\nd = giv+ weight to su* expectations aft] ! "q ( _! validity has be5 rais$4 ,! problem we h to discuss is :e!r !re is any reason = believ+ 9 :at is call$ '! uni=mity ( nature'4 ,! belief 9 ! uni=mity ( nature is ! belief t ev]y?+ t has happ5$ or w happ5 is an in/ance ( "s g5]al law to :i* !re are no exceptions4 ,! crude expectations :i* we h be5 consid]+ are all subject to exceptions1 & !re=e liable to disappo9t ^? :o ent]ta9 !m4 ,b sci5ce habitually assumes1 at lea/ z a work+ hypo!sis1 t g5]al rules :i* h exceptions c be replac$ by g5]al rules :i* h no exceptions4 ',unsupport$ bodies 9 air fall' is a g5]al rule to :i* balloons & a]oplanes are exceptions4 ,b ! laws ( motion & ! law ( gravitation1 :i* acc\nt = ! fact t mo/ bodies fall1 also acc\nt = ! fact t balloons & a]oplanes c rise2 ?us ! laws ( motion & ! law ( gravitation are n subject to ^! exceptions4 ,! belief t ! sun w rise to-morr[ might be falsifi$ if ! e>? came sudd5ly into contact ) a l>ge body :i* de/roy$ its rotation2 b ! laws ( motion & ! law ( gravitation w\ld n be infr+$ by su* an ev5t4 ,! bus9ess ( sci5ce is to f9d uni=mities1 su* z ! laws ( motion & ! law ( gravitation1 to :i*1 s f> z our exp]i5ce ext5ds1 !re are no exceptions4 ,9 ?is se>* sci5ce has be5 rem>kably successful1 & x may be conc$$ t su* uni=mities h held hi!rto4 ,?is br+s u back to ! "q3 ,h we any reason1 assum+ t !y h always held 9 ! pa/1 to suppose t !y w hold 9 ! future8 ,x has be5 argu$ t we h reason to "k t ! future w resemble ! pa/1 because :at 0 ! future has con/antly become ! pa/1 & has always be5 f\nd to resemble ! pa/1 s t we really h exp]i5ce ( ! future1 namely ( times :i* 7 =m]ly future1 :i* we may call pa/ futures4 ,b su* an argum5t really begs ! v "q at issue4 ,we h exp]i5ce ( pa/ futures1 b n ( future futures1 & ! "q is3 ,w future futures resemble pa/ futures8 ,?is "q is n to be answ]$ by an argum5t :i* />ts f pa/ futures alone4 ,we h !re=e /ill to seek = "s pr9ciple :i* %all enable u to "k t ! future w foll[ ! same laws z ! pa/4 ,! ref]5ce to ! future 9 ?is "q is n ess5tial4 ,! same "q arises :5 we apply ! laws t "w 9 our exp]i5ce to pa/ ?+s ( :i* we h no exp]i5ce--z1 = example1 9 geology1 or 9 !ories z to ! orig9 ( ! ,sol> ,sy/em4 ,! "q we really h to ask is3 ',:5 two ?+s h be5 f\nd to be (t5 associat$1 & no in/ance is kn[n ( ! "o occurr+ )\t ! o!r1 does ! occurr5ce ( "o ( ! two1 9 a fre% in/ance1 give any good gr\nd = expect+ ! o!r8' ,on our answ] to ?is "q mu/ dep5d ! validity ( ! :ole ( our expectations z to ! future1 ! :ole ( ! results obta9$ by induction1 & 9 fact practically all ! beliefs ^u :i* our daily life is bas$4 ,x mu/ be conc$$1 to beg9 )1 t ! fact t two ?+s h be5 f\nd (t5 toge!r & nev] ap>t does n1 by itself1 suffice to prove demon/ratively t !y w be f\nd toge!r 9 ! next case we exam9e4 ,! mo/ we c hope is t ! (t5] ?+s are f\nd toge!r1 ! m probable x becomes t !y w be f\nd toge!r ano!r "t1 & t1 if !y h be5 f\nd toge!r (t5 en\gh1 ! probability w am\nt almo/ to c]ta9ty4 ,x c nev] q rea* c]ta9ty1 because we "k t 9 spite ( frequ5t repetitions !re sometimes is a failure at ! la/1 z 9 ! case ( ! *ick5 ^: neck is wrung4 ,?us probability is all we ought to seek4 ,x might be urg$1 z aga9/ ! view we are advocat+1 t we "k all natural ph5om5a to be subject to ! reign ( law1 & t sometimes1 on ! basis ( obs]vation1 we c see t only "o law c possibly fit ! facts ( ! case4 ,n[ to ?is view !re are two answ]s4 ,! fir/ is t1 ev5 if "s law :i* has no exceptions applies to our case1 we c nev]1 9 practice1 be sure t we h discov]$ t law & n "o to :i* !re are exceptions4 ,! second is t ! reign ( law w\ld seem to be itself only probable1 & t our belief t x w hold 9 ! future1 or 9 unexam9$ cases 9 ! pa/1 is itself bas$ ^u ! v pr9ciple we are exam9+4 ,! pr9ciple we are exam9+ may be call$ ! pr9ciple ( induction1 & its two p>ts may be /at$ z foll[s3 7a7 ,:5 a ?+ ( a c]ta9 sort ,a has be5 f\nd to be associat$ ) a ?+ ( a c]ta9 o!r sort ,b1 & has nev] be5 f\nd dissociat$ f a ?+ ( ! sort ,b1 ! great] ! numb] ( cases 9 :i* ,a & ,b h be5 associat$1 ! great] is ! probability t !y w be associat$ 9 a fre% case 9 :i* "o ( !m is kn[n to be pres5t2 7b7 ,"u ! same circum/ances1 a suffici5t numb] ( cases ( association w make ! probability ( a fre% association ne>ly a c]ta9ty1 & w make x approa* c]ta9ty )\t limit4 ,z j /at$1 ! pr9ciple applies only to ! v]ification ( our expectation 9 a s+le fre% in/ance4 ,b we want also to "k t !re is a probability 9 fav\r ( ! g5]al law t ?+s ( ! sort ,a are always associat$ ) ?+s ( ! sort ,b1 provid$ a suffici5t numb] ( cases ( association are kn[n1 & no cases ( failure ( association are kn[n4 ,! probability ( ! g5]al law is obvi\sly less ?an ! probability ( ! p>ticul> case1 s9ce if ! g5]al law is true1 ! p>ticul> case mu/ also be true1 :]eas ! p>ticul> case may be true )\t ! g5]al law be+ true4 ,nev]!less ! probability ( ! g5]al law is increas$ by repetitions1 j z ! probability ( ! p>ticul> case is4 ,we may !re=e repeat ! two p>ts ( our pr9ciple z reg>ds ! g5]al law1 ?us3 7a7 ,! great] ! numb] ( cases 9 :i* a ?+ ( ! sort ,a has be5 f\nd associat$ ) a ?+ ( ! sort ,b1 ! m probable x is 7if no cases ( failure ( association are kn[n7 t ,a is always associat$ ) ,b2 b7 ,"u ! same circum/ances1 a suffici5t numb] ( cases ( ! association ( ,a ) ,b w make x ne>ly c]ta9 t ,a is always associat$ ) ,b1 & w make ?is g5]al law approa* c]ta9ty )\t limit4 ,x %\ld be not$ t probability is always relative to c]ta9 data4 ,9 our case1 ! data are m]ely ! kn[n cases ( coexi/5ce ( ,a & ,b4 ,!re may be o!r data1 :i* might be tak5 into acc\nt1 :i* w\ld gravely alt] ! probability4 ,= example1 a man :o _h se5 a great _m :ite swans might argue1 by our pr9ciple1 t on ! data x 0 probable t all swans 7 :ite1 & ?is might be a p]fectly s\nd argum5t4 ,! argum5t is n disprov$ by ! fact t "s swans are black1 because a ?+ may v well happ5 9 spite ( ! fact t "s data r5d] x improbable4 ,9 ! case ( ! swans1 a man might "k t col\r is a v v>iable *>act]i/ic 9 _m species ( animals1 & t1 !re=e1 an induction z to col\r is peculi>ly liable to error4 ,b ?is k w\ld be a fre% datum1 by no means prov+ t ! probability relatively to our previ\s data _h be5 wrongly e/imat$4 ,! fact1 !re=e1 t ?+s (t5 fail to fulfil our expectations is no evid5ce t our expectations w n probably be fulfill$ 9 a giv5 case or a giv5 class ( cases4 ,?us our inductive pr9ciple is at any rate n capable ( be+ disprov$ by an appeal to exp]i5ce4 ,! inductive pr9ciple1 h[ev]1 is equally incapable ( be+ prov$ by an appeal to exp]i5ce4 ,exp]i5ce might conceivably confirm ! inductive pr9ciple z reg>ds ! cases t h be5 already exam9$2 b z reg>ds unexam9$ cases1 x is ! inductive pr9ciple alone t c ju/ify any inf]5ce f :at has be5 exam9$ to :at has n be5 exam9$4 ,all argum5ts :i*1 on ! basis ( exp]i5ce1 argue z to ! future or ! unexp]i5c$ p>ts ( ! pa/ or pres5t1 assume ! inductive pr9ciple2 h5ce we c nev] use exp]i5ce to prove ! inductive pr9ciple )\t begg+ ! "q4 ,?us we mu/ ei!r accept ! inductive pr9ciple on ! gr\nd ( its intr9sic evid5ce1 or =go all ju/ification ( our expectations ab\t ! future4 ,if ! pr9ciple is uns\nd1 we h no reason to expect ! sun to rise to-morr[1 to expect bread to be m n\ri%+ ?an a /one1 or to expect t if we ?r[ ourselves (f ! ro( we %all fall4 ,:5 we see :at looks l our be/ fri5d approa*+ u1 we %all h no reason to suppose t 8 body is n inhabit$ by ! m9d ( our wor/ enemy or ( "s total /rang]4 ,all our conduct is bas$ ^u associations :i* h work$ 9 ! pa/1 & :i* we !re=e reg>d z likely to "w 9 ! future2 & ?is likelihood is dep5d5t = its validity ^u ! inductive pr9ciple4 ,! g5]al pr9ciples ( sci5ce1 su* z ! belief 9 ! reign ( law1 & ! belief t e ev5t mu/ h a cause1 are z completely dep5d5t ^u ! inductive pr9ciple z are ! beliefs ( daily life ,all su* g5]al pr9ciples are believ$ because mank9d h f\nd innum]able in/ances ( _! tru? & no in/ances ( _! falsehood4 ,b ?is af=ds no evid5ce = _! tru? 9 ! future1 unless ! inductive pr9ciple is assum$4 ,?us all k :i*1 on a basis ( exp]i5ce tells u some?+ ab\t :at is n exp]i5c$1 is bas$ ^u a belief :i* exp]i5ce c nei!r confirm nor confute1 yet :i*1 at lea/ 9 its m concrete applications1 appe>s to be z firmly root$ 9 u z _m ( ! facts ( exp]i5ce4 ,! exi/5ce & ju/ification ( su* beliefs--= ! inductive pr9ciple1 z we %all see1 is n ! only example--raises "s ( ! mo/ difficult & mo/ debat$ problems ( philosophy4 ,we w1 9 ! next *apt]1 consid] briefly :at may be said to acc\nt = su* k1 & :at is its scope & its degree ( c]ta9ty4 ,,*apt] ,,vii4 ,,on ,,our ,,k ,,( ,,g5]al ,,pr9ciples ,we saw 9 ! prec$+ *apt] t ! pr9ciple ( induction1 :ile necess>y to ! validity ( all argum5ts bas$ on exp]i5ce1 is itself n capable ( be+ prov$ by exp]i5ce1 & yet is unhesitat+ly believ$ by e "o1 at lea/ 9 all its concrete applications4 ,9 ^! *>act]i/ics ! pr9ciple ( induction does n /& alone4 ,!re are a numb] ( o!r pr9ciples :i* _c be prov$ or disprov$ by exp]i5ce1 b are us$ 9 argum5ts :i* />t f :at is exp]i5c$4 ,"s ( ^! pr9ciples h ev5 great] evid5ce ?an ! pr9ciple ( induction1 & ! k ( !m has ! same degree ( c]ta9ty z ! k ( ! exi/5ce ( s5se-data4 ,!y con/itute ! means ( draw+ inf]5ces f :at is giv5 9 s5sation2 & if :at we inf] is to be true1 x is j z necess>y t our pr9ciples ( inf]5ce %\ld be true z x is t our data %\ld be true4 ,! pr9ciples ( inf]5ce are apt to be ov]look$ because ( _! v obvi\sness--! assumption involv$ is ass5t$ to )\t our realiz+ t x is an assumption4 ,b x is v important to realize ! use ( pr9ciples ( inf]5ce1 if a correct !ory ( k is to be obta9$2 = our k ( !m raises int]e/+ & difficult que/ions4 ,9 all our k ( g5]al pr9ciples1 :at actually happ5s is t fir/ ( all we realize "s p>ticul> application ( ! pr9ciple1 & !n we realize t ! p>ticul>ity is irrelevant1 & t !re is a g5]ality :i* may equally truly be affirm$4 ,?is is ( c\rse famili> 9 su* matt]s z tea*+ ari?metic3 'two & two are f\r' is fir/ le>nt 9 ! case ( "s p>ticul> pair ( c\ples1 & !n 9 "s o!r p>ticul> case1 & s on1 until at la/ x becomes possible to see t x is true ( any pair ( c\ples4 ,! same ?+ happ5s ) logical pr9ciples4 ,suppose two m5 are discuss+ :at day ( ! mon? x is4 ,"o ( !m says1 ',at lea/ y w admit t if ye/]day 0 ! #ae? to-day mu/ be ! #af?4' ',yes'1 says ! o!r1 ',i admit t4' ',& y "k'1 ! fir/ cont9ues1 't ye/]day 0 ! #ae?1 because y d9$ ) ,jones1 & y\r di>y w tell y t 0 on ! #ae?4' ',yes'1 says ! second2 '!re=e to-day is ! #af?4' ,n[ su* an argum5t is n h>d to foll[2 & if x is grant$ t its premisses are true 9 fact1 no "o w d5y t ! conclusion mu/ also be true4 ,b x dep5ds = its tru? ^u an in/ance ( a g5]al logical pr9ciple4 ,! logical pr9ciple is z foll[s3 ',suppose x kn[n t if ?is is true1 !n t is true4 ,suppose x also kn[n t ?is is true1 !n x foll[s t t is true4' ,:5 x is ! case t if ?is is true1 t is true1 we %all say t ?is 'implies' t1 & t t 'foll[s f' ?is4 ,?us our pr9ciple /ates t if ?is implies t1 & ?is is true1 !n t is true4 ,9 o!r words1 'any?+ impli$ by a true proposition is true'1 or ':atev] foll[s f a true proposition is true'4 ,?is pr9ciple is really involv$--at lea/1 concrete in/ances ( x are involv$--9 all demon/rations4 ,:5ev] "o ?+ :i* we believe is us$ to prove some?+ else1 :i* we consequ5tly believe1 ?is pr9ciple is relevant4 ,if any "o asks3 ',:y %\ld ,i accept ! results ( valid argum5ts bas$ on true premisses8' we c only answ] by appeal+ to our pr9ciple4 ,9 fact1 ! tru? ( ! pr9ciple is impossible to d\bt1 & its obvi\sness is s great t at fir/ sight x seems almo/ trivial4 ,su* pr9ciples1 h[ev]1 are n trivial to ! philosoph]1 = !y %[ t we may h indubitable k :i* is 9 no way d]iv$ f objects ( s5se4 ,! above pr9ciple is m]ely "o ( a c]ta9 numb] ( self-evid5t logical pr9ciples4 ,"s at lea/ ( ^! pr9ciples mu/ be grant$ be=e any argum5t or pro( becomes possible4 ,:5 "s ( !m h be5 grant$1 o!rs c be prov$1 ?\gh ^! o!rs1 s long z !y are simple1 are j z obvi\s z ! pr9ciples tak5 = grant$4 ,= no v good reason1 ?ree ( ^! pr9ciples h be5 s+l$ out by tradition "u ! "n ( ',laws ( ,?\ght'4 ,!y are z foll[s3 7#a7 ,! law ( id5tity3 ',:atev] is1 is4' 7#b7 ,! law ( contradiction3 ',no?+ c bo? be & n be4' 7#c7 ,! law ( exclud$ middle3 ',ev]y?+ mu/ ei!r be or n be4' ,^! ?ree laws are samples ( self-evid5t logical pr9ciples1 b are n really m fundam5tal or m self-evid5t ?an v>i\s o!r simil> pr9ciples3 = in/ance1 ! "o we consid]$ j n[1 :i* /ates t :at foll[s f a true premiss is true4 ,! "n 'laws ( ?\ght' is also mislead+1 = :at is important is n ! fact t we ?9k 9 accordance ) ^! laws1 b ! fact t ?+s behave 9 accordance ) !m2 9 o!r words1 ! fact t :5 we ?9k 9 accordance ) !m we ?9k truly4 ,b ?is is a l>ge "q1 to :i* we mu/ return at a lat] /age4 ,9 addition to ! logical pr9ciples :i* enable u to prove f a giv5 premiss t some?+ is c]ta9ly true1 !re are o!r logical pr9ciples :i* enable u to prove1 f a giv5 premiss1 t !re is a great] or less probability t some?+ is true4 ,an example ( su* pr9ciples--p]haps ! mo/ important example is ! inductive pr9ciple1 :i* we consid]$ 9 ! prec$+ *apt]4 ,"o ( ! great hi/oric controv]sies 9 philosophy is ! controv]sy betwe5 ! two s*ools call$ respectively 'empirici/s' & 'rationali/s'4 ,! empirici/s--:o are be/ repres5t$ by ! ,briti% philosoph]s1 ,locke1 ,b]keley1 & ,hume--ma9ta9$ t all our k is d]iv$ f exp]i5ce2 ! rationali/s--:o are repres5t$ by ! ,cont95tal philosoph]s ( ! sev5te5? c5tury1 especially ,desc>tes & ,leibniz--ma9ta9$ t1 9 addition to :at we "k by exp]i5ce1 !re are c]ta9 'innate ideas' & 'innate pr9ciples'1 :i* we "k indep5d5tly ( exp]i5ce4 ,x has n[ become possible to decide ) "s confid5ce z to ! tru? or falsehood ( ^! oppos+ s*ools4 ,x mu/ be admitt$1 = ! reasons already /at$1 t logical pr9ciples are kn[n to u1 & _c be !mselves prov$ by exp]i5ce1 s9ce all pro( presupposes !m4 ,9 ?is1 !re=e1 :i* 0 ! mo/ important po9t ( ! controv]sy1 ! rationali/s 7 9 ! "r4 ,on ! o!r h&1 ev5 t "p ( our k :i* is logically indep5d5t ( exp]i5ce 79 ! s5se t exp]i5ce _c prove x7 is yet elicit$ & caus$ by exp]i5ce4 ,x is on occasion ( p>ticul> exp]i5ces t we become aw>e ( ! g5]al laws :i* _! connexions exemplify4 ,x w\ld c]ta9ly be absurd to suppose t !re are innate pr9ciples 9 ! s5se t babies are born ) a k ( ev]y?+ :i* m5 "k & :i* _c be d$uc$ f :at is exp]i5c$4 ,= ?is reason1 ! ^w 'innate' w\ld n n[ be employ$ to describe our k ( logical pr9ciples4 ,! phrase 'a priori' is less objectionable1 & is m usual 9 mod]n writ]s4 ,?us1 :ile admitt+ t all k is elicit$ & caus$ by exp]i5ce1 we %all nev]!less hold t "s k is a priori1 9 ! s5se t ! exp]i5ce :i* makes u ?9k ( x does n suffice to prove x1 b m]ely s directs our att5tion t we see its tru? )\t requir+ any pro( f exp]i5ce4 ,!re is ano!r po9t ( great importance1 9 :i* ! empirici/s 7 9 ! "r z aga9/ ! rationali/s4 ,no?+ c be kn[n to exi/ except by ! help ( exp]i5ce4 ,t is to say1 if we wi% to prove t some?+ ( :i* we h no direct exp]i5ce exi/s1 we mu/ h among our premisses ! exi/5ce ( "o or m ?+s ( :i* we h direct exp]i5ce4 ,our belief t ! ,emp]or ( ,*9a exi/s1 = example1 re/s ^u te/imony1 & te/imony consi/s1 9 ! la/ analysis1 ( s5se-data se5 or he>d 9 read+ or be+ spok5 to4 ,rationali/s believ$ t1 f g5]al consid]ation z to :at mu/ be1 !y c\ld d$uce ! exi/5ce ( ?is or t 9 ! actual _w4 ,9 ?is belief !y seem to h be5 mi/ak54 ,all ! k t we c acquire a priori conc]n+ exi/5ce seems to be hypo!tical3 x tells u t if "o ?+ exi/s1 ano!r mu/ exi/1 or1 m g5]ally1 t if "o proposition is true1 ano!r mu/ be true4 ,?is is exemplifi$ by ! pr9ciples we h already dealt )1 su* z 'if ?is is true1 & ?is implies t1 !n t is true'1 or 'if ?is & t h be5 repeat$ly f\nd connect$1 !y w probably be connect$ 9 ! next in/ance 9 :i* "o ( !m is f\nd'4 ,?us ! scope & p[] ( a priori pr9ciples is /rictly limit$4 ,all k t some?+ exi/s mu/ be 9 "p dep5d5t on exp]i5ce4 ,:5 any?+ is kn[n imm$iately1 its exi/5ce is kn[n by exp]i5ce alone2 :5 any?+ is prov$ to exi/1 )\t be+ kn[n imm$iately1 bo? exp]i5ce & a priori pr9ciples mu/ be requir$ 9 ! pro(4 ,k is call$ empirical :5 x re/s :olly or p>tly ^u exp]i5ce4 ,?us all k :i* ass]ts exi/5ce is empirical1 & ! only a priori k conc]n+ exi/5ce is hypo!tical1 giv+ connexions among ?+s t exi/ or may exi/1 b n giv+ actual exi/5ce4 ,a priori k is n all ( ! logical k9d we h be5 hi!rto consid]+4 ,p]haps ! mo/ important example ( non-logical a priori k is k z to e?ical value4 ,i am n speak+ ( judgem5ts z to :at is useful or z to :at is virtu\s1 = su* judgem5ts d require empirical premisses2 ,i am speak+ ( judgem5ts z to ! intr9sic desirability ( ?+s4 ,if some?+ is useful1 x mu/ be useful because x secures "s end2 ! end mu/1 if we h gone f> en\gh1 be valuable on its own acc\nt1 & n m]ely because x is useful = "s fur!r end4 ,?us all judgem5ts z to :at is useful dep5d ^u judgem5ts z to :at has value on its own acc\nt4 ,we judge1 = example1 t happ9ess is m desirable ?an mis]y1 k ?an ignorance1 goodwill ?an hatr$1 & s on4 ,su* judgem5ts mu/1 9 "p at lea/1 be imm$iate & a priori4 ,l our previ\s a priori judgem5ts1 !y may be elicit$ by exp]i5ce1 & inde$ !y mu/ be2 = x seems n possible to judge :e!r any?+ is intr9sically valuable unless we h exp]i5c$ some?+ ( ! same k9d4 ,b x is fairly obvi\s t !y _c be prov$ by exp]i5ce2 = ! fact t a ?+ exi/s or does n exi/ _c prove ei!r t x is good t x %\ld exi/ or t x is bad4 ,! pursuit ( ?is subject belongs to e?ics1 :]e ! impossibility ( d$uc+ :at ought to be f :at is has to be e/abli%$4 ,9 ! pres5t connexion1 x is only important to realize t k z to :at is intr9sically ( value is a priori 9 ! same s5se 9 :i* logic is a priori1 namely 9 ! s5se t ! tru? ( su* k c be nei!r prov$ nor disprov$ by exp]i5ce4 ,all pure ma!matics is a priori1 l logic4 ,?is 0 /r5u\sly d5i$ by ! empirical philosoph]s1 :o ma9ta9$ t exp]i5ce 0 z mu* ! s\rce ( our k ( ari?metic z ( our k ( geography4 ,!y ma9ta9$ t by ! repeat$ exp]i5ce ( see+ two ?+s & two o!r ?+s1 & f9d+ t altoge!r !y made f\r ?+s1 we 7 l$ by induction to ! conclusion t two ?+s & two o!r ?+s w\ld always make f\r ?+s altoge!r4 ,if1 h[ev]1 ?is 7 ! s\rce ( our k t two & two are f\r1 we %\ld proce$ diff]5tly1 9 p]suad+ ourselves ( its tru?1 f ! way 9 :i* we d actually proce$4 ,9 fact1 a c]ta9 numb] ( in/ances are ne$$ to make u ?9k ( two ab/ractly1 r ?an ( two co9s or two books or two p1 or two ( any o!r specifi$ k9d4 ,b z soon z we are able to dive/ our ?\ghts ( irrelevant p>ticul>ity1 we become able to see ! g5]al pr9ciple t two & two are f\r2 any "o in/ance is se5 to be typical1 & ! exam9ation ( o!r in/ances becomes unnecess>y47#a7 7#a7 ,cf4 ,a4 ,n4 ,:itehead1 ,introduction to ,ma!matics 7,home ,univ]sity ,libr>y74 ,! same ?+ is exemplifi$ 9 geometry4 ,if we want to prove "s prop]ty ( all triangles1 we draw "s "o triangle & reason ab\t x2 b we c avoid mak+ use ( any prop]ty :i* x does n %>e ) all o!r triangles1 & ?us1 f our p>ticul> case1 we obta9 a g5]al result4 ,we d n1 9 fact1 feel our c]ta9ty t two & two are f\r increas$ by fre% in/ances1 because1 z soon z we h se5 ! tru? ( ?is proposition1 our c]ta9ty becomes s great z to be incapable ( gr[+ great]4 ,moreov]1 we feel "s quality ( necessity ab\t ! proposition 'two & two are f\r'1 :i* is abs5t f ev5 ! be/ atte/$ empirical g5]alizations4 ,su* g5]alizations always rema9 m]e facts3 we feel t !re might be a _w 9 :i* !y 7 false1 ?\gh 9 ! actual _w !y happ5 to be true4 ,9 any possible _w1 on ! contr>y1 we feel t two & two w\ld be f\r3 ?is is n a m]e fact1 b a necessity to :i* ev]y?+ actual & possible mu/ con=m4 ,! case may be made cle>] by consid]+ a g5u9ely-empirical g5]alization1 su* z ',all m5 are mortal4' ,x is pla9 t we believe ?is proposition1 9 ! fir/ place1 because !re is no kn[n in/ance ( m5 liv+ beyond a c]ta9 age1 & 9 ! second place because !re seem to be physiological gr\nds = ?9k+ t an organism su* z a man's body mu/ soon] or lat] we> out4 ,neglect+ ! second gr\nd1 & consid]+ m]ely our exp]i5ce ( m5's mortality1 x is pla9 t we %\ld n be cont5t ) "o q cle>ly und]/ood in/ance ( a man dy+1 :]eas1 9 ! case ( 'two & two are f\r'1 "o in/ance does suffice1 :5 c>efully consid]$1 to p]suade u t ! same mu/ happ5 9 any o!r in/ance4 ,also we c be =c$ to admit1 on reflection1 t !re may be "s d\bt1 h[ev] slight1 z to :e!r all m5 are mortal4 ,?is may be made pla9 by ! attempt to imag9e two diff]5t worlds1 9 "o ( :i* !re are m5 :o are n mortal1 :ile 9 ! o!r two & two make five4 ,:5 ,swift invites u to consid] ! race ( ,/ruldbugs :o nev] die1 we are able to acquiesce 9 imag9ation4 ,b a _w :]e two & two make five seems q on a diff]5t level4 ,we feel t su* a _w1 if !re 7 "o1 w\ld upset ! :ole fabric ( our k & r$uce u to utt] d\bt4 ,! fact is t1 9 simple ma!matical judgem5ts su* z 'two & two are f\r'1 & also 9 _m judgem5ts ( logic1 we c "k ! g5]al proposition )\t inf]r+ x f in/ances1 al?\gh "s in/ance is usually necess>y to make cle> to u :at ! g5]al proposition means4 ,?is is :y !re is real utility 9 ! process ( d$uction1 :i* goes f ! g5]al to ! g5]al1 or f ! g5]al to ! p>ticul>1 z well z 9 ! process ( induction1 :i* goes f ! p>ticul> to ! p>ticul>1 or f ! p>ticul> to ! g5]al4 ,x is an old debate among philosoph]s :e!r d$uction ev] gives new k4 ,we c n[ see t 9 c]ta9 cases1 at lea/1 x does d s4 ,if we already "k t two & two always make f\r1 & we "k t ,br[n & ,jones are two1 & s are ,rob9son & ,smi?1 we c d$uce t ,br[n & ,jones & ,rob9son & ,smi? are f\r4 ,?is is new k1 n conta9$ 9 our premisses1 because ! g5]al proposition1 'two & two are f\r'1 nev] told u !re 7 su* p z ,br[n & ,jones & ,rob9son & ,smi?1 & ! p>ticul> premisses d n tell u t !re 7 f\r ( !m1 :]eas ! p>ticul> proposition d$uc$ does tell u bo? ^! ?+s4 ,b ! newness ( ! k is mu* less c]ta9 if we take ! /ock in/ance ( d$uction t is always giv5 9 books on logic1 namely1 ',all m5 are mortal2 ,socrates is a man1 !re=e ,socrates is mortal4' ,9 ?is case1 :at we really "k beyond reasonable d\bt is t c]ta9 m51 ,a1 ,b1 ,c1 7 mortal1 s9ce1 9 fact1 !y h di$4 ,if ,socrates is "o ( ^! m51 x is fooli% to g ! r\ndab\t way ?r\gh 'all m5 are mortal' to arrive at ! conclusion t probably ,socrates is mortal4 ,if ,socrates is n "o ( ! m5 on :om our induction is bas$1 we %all /ill d bett] to argue /raight f our ,a1 ,b1 ,c1 to ,socrates1 ?an to g r\nd by ! g5]al proposition1 'all m5 are mortal'4 ,= ! probability t ,socrates is mortal is great]1 on our data1 ?an ! probability t all m5 are mortal4 7,?is is obvi\s1 because if all m5 are mortal1 s is ,socrates2 b if ,socrates is mortal1 x does n foll[ t all m5 are mortal47 ,h5ce we %all rea* ! conclusion t ,socrates is mortal ) a great] approa* to c]ta9ty if we make our argum5t purely inductive ?an if we g by way ( 'all m5 are mortal' & !n use d$uction4 ,?is illu/rates ! diff]5ce betwe5 g5]al propositions kn[n a priori su* z 'two & two are f\r'1 & empirical g5]alizations su* z 'all m5 are mortal'4 ,9 reg>d to ! =m]1 d$uction is ! "r mode ( argum5t1 :]eas 9 reg>d to ! latt]1 induction is always !oretically pref]able1 & w>rants a great] confid5ce 9 ! tru? ( our conclusion1 because all empirical g5]alizations are m unc]ta9 ?an ! in/ances ( !m4 ,we h n[ se5 t !re are propositions kn[n a priori1 & t among !m are ! propositions ( logic & pure ma!matics1 z well z ! fundam5tal propositions ( e?ics4 ,! "q :i* mu/ next occupy u is ?is3 ,h[ is x possible t !re %\ld be su* k8 ,& m p>ticul>ly1 h[ c !re be k ( g5]al propositions 9 cases :]e we h n exam9$ all ! in/ances1 & inde$ nev] c exam9e !m all1 because _! numb] is inf9ite8 ,^! que/ions1 :i* 7 fir/ br\ght prom95tly =w>d by ! ,g]man philosoph] ,kant 7#agbd-ahjd71 are v difficult1 & hi/orically v important4 ,,*apt] ,,viii4 ,,h[ ,a ,,priori ,,k ,,is ,,possible ,immanuel ,kant is g5]ally reg>d$ z ! greate/ ( ! mod]n philosoph]s4 ,?\gh he liv$ ?r\gh ! ,sev5 ,ye>s ,w> & ! ,fr5* ,revolution1 he nev] int]rupt$ 8 tea*+ ( philosophy at ,knigsb]g 9 ,ea/ ,prussia4 ,8 mo/ di/9ctive contribution 0 ! inv5tion ( :at he call$ ! 'critical' philosophy1 :i*1 assum+ z a datum t !re is k ( v>i\s k9ds1 inquir$ h[ su* k comes to be possible1 & d$uc$1 f ! answ] to ?is inquiry1 _m metaphysical results z to ! nature ( ! _w4 ,:e!r ^! results 7 valid may well be d\bt$4 ,b ,kant und\bt$ly des]ves cr$it = two ?+s3 fir/1 = hav+ p]ceiv$ t we h a priori k :i* is n purely 'analytic'1 i4e4 su* t ! opposite w\ld be self-contradictory1 & secondly1 = hav+ made evid5t ! philosophical importance ( ! !ory ( k4 ,be=e ! "t ( ,kant1 x 0 g5]ally held t :atev] k 0 a priori mu/ be 'analytic'4 ,:at ?is ^w means w be be/ illu/rat$ by examples4 ,if ,i say1 ',a bald man is a man'1 ',a plane figure is a figure'1 ',a bad poet is a poet'1 ,i make a purely analytic judgem5t3 ! subject spok5 ab\t is giv5 z hav+ at lea/ two prop]ties1 ( :i* "o is s+l$ out to be ass]t$ ( x4 ,su* propositions z ! above are trivial1 & w\ld nev] be enunciat$ 9 real life except by an orator prep>+ ! way = a piece ( sophi/ry4 ,!y are call$ 'analytic' because ! pr$icate is obta9$ by m]ely analys+ ! subject4 ,be=e ! "t ( ,kant x 0 ?\ght t all judgem5ts ( :i* we c\ld be c]ta9 a priori 7 ( ?is k9d3 t 9 all ( !m !re 0 a pr$icate :i* 0 only "p ( ! subject ( :i* x 0 ass]t$4 ,if ?is 7 s1 we %\ld be involv$ 9 a def9ite contradiction if we attempt$ to d5y any?+ t c\ld be kn[n a priori4 ',a bald man is n bald' w\ld ass]t & d5y baldness ( ! same man1 & w\ld !re=e contradict itself4 ,?us accord+ to ! philosoph]s be=e ,kant1 ! law ( contradiction1 :i* ass]ts t no?+ c at ! same "t h & n h a c]ta9 prop]ty1 suffic$ to e/abli% ! tru? ( all a priori k4 ,hume 7#agaa-gf71 :o prec$$ ,kant1 accept+ ! usual view z to :at makes k a priori1 discov]$ t1 9 _m cases :i* _h previ\sly be5 suppos$ analytic1 & notably 9 ! case ( cause & effect1 ! connexion 0 really syn!tic4 ,be=e ,hume1 rationali/s at lea/ _h suppos$ t ! effect c\ld be logically d$uc$ f ! cause1 if only we _h suffici5t k4 ,hume argu$--correctly1 z w\ld n[ be g5]ally admitt$--t ?is c\ld n be done4 ,h5ce he inf]r$ ! f> m d\btful proposition t no?+ c\ld be kn[n a priori ab\t ! connexion ( cause & effect4 ,kant1 :o _h be5 educat$ 9 ! rationali/ tradition1 0 mu* p]turb$ by ,hume's scepticism1 & endeav\r$ to f9d an answ] to x4 ,he p]ceiv$ t n only ! connexion ( cause & effect1 b all ! propositions ( ari?metic & geometry1 are 'syn!tic'1 i4e4 n analytic3 9 all ^! propositions1 no analysis ( ! subject w reveal ! pr$icate4 ,8 /ock in/ance 0 ! proposition #g #e #ab4 ,he po9t$ out1 q truly1 t #g & #e h to be put toge!r to give #ab3 ! idea ( #ab is n conta9$ 9 !m1 nor ev5 9 ! idea ( add+ !m toge!r4 ,?us he 0 l$ to ! conclusion t all pure ma!matics1 ?\gh a priori1 is syn!tic2 & ?is conclusion rais$ a new problem ( :i* he endeav\r$ to f9d ! solution4 ,! "q :i* ,kant put at ! beg9n+ ( 8 philosophy1 namely ',h[ is pure ma!matics possible8' is an int]e/+ & difficult "o1 to :i* e philosophy :i* is n purely sceptical mu/ f9d "s answ]4 ,! answ] ( ! pure empirici/s1 t our ma!matical k is d]iv$ by induction f p>ticul> in/ances1 we h already se5 to be inadequate1 = two reasons3 fir/1 t ! validity ( ! inductive pr9ciple itself _c be prov$ by induction2 secondly1 t ! g5]al propositions ( ma!matics1 su* z 'two & two always make f\r'1 c obvi\sly be kn[n ) c]ta9ty by consid]ation ( a s+le in/ance1 & ga9 no?+ by enum]ation ( o!r cases 9 :i* !y h be5 f\nd to be true4 ,?us our k ( ! g5]al propositions ( ma!matics 7& ! same applies to logic7 mu/ be acc\nt$ = o!rwise ?an our 7m]ely probable7 k ( empirical g5]alizations su* z 'all m5 are mortal'4 ,! problem arises ?r\gh ! fact t su* k is g5]al1 :]eas all exp]i5ce is p>ticul>4 ,x seems /range t we %\ld app>5tly be able to "k "s tru?s 9 advance ab\t p>ticul> ?+s ( :i* we h z yet no exp]i5ce2 b x _c easily be d\bt$ t logic & ari?metic w apply to su* ?+s4 ,we d n "k :o w be ! inhabitants ( ,london a hundr$ ye>s h5ce2 b we "k t any two ( !m & any o!r two ( !m w make f\r ( !m4 ,?is app>5t p[] ( anticipat+ facts ab\t ?+s ( :i* we h no exp]i5ce is c]ta9ly surpris+4 ,kant's solution ( ! problem1 ?\gh n valid 9 my op9ion1 is int]e/+4 ,x is1 h[ev]1 v difficult1 & is diff]5tly und]/ood by diff]5t philosoph]s4 ,we c1 !re=e1 only give ! m]e/ outl9e ( x1 & ev5 t w be ?\ght mislead+ by _m expon5ts ( ,kant's sy/em4 ,:at ,kant ma9ta9$ 0 t 9 all our exp]i5ce !re are two elem5ts to be di/+ui%$1 ! "o due to ! object 7i4e4 to :at we h call$ ! 'physical object'71 ! o!r due to our own nature4 ,we saw1 9 discuss+ matt] & s5se-data1 t ! physical object is diff]5t f ! associat$ s5se-data1 & t ! s5se-data are to be reg>d$ z result+ f an int]action betwe5 ! physical object & ourselves4 ,s f>1 we are 9 agreem5t ) ,kant4 ,b :at is di/9ctive ( ,kant is ! way 9 :i* he apportions ! %>es ( ourselves & ! physical object respectively4 ,he consid]s t ! crude mat]ial giv5 9 s5sation--! col\r1 h>dness1 etc4--is due to ! object1 & t :at we supply is ! arrangem5t 9 space & "t1 & all ! relations betwe5 s5se-data :i* result f comp>ison or f consid]+ "o z ! cause ( ! o!r or 9 any o!r way4 ,8 *ief reason 9 fav\r ( ?is view is t we seem to h a priori k z to space & "t & causality & comp>ison1 b n z to ! actual crude mat]ial ( s5sation4 ,we c be sure1 he says1 t any?+ we %all ev] exp]i5ce mu/ %[ ! *>act]i/ics affirm$ ( x 9 our a priori k1 because ^! *>act]i/ics are due to our own nature1 & !re=e no?+ c ev] come into our exp]i5ce )\t acquir+ ^! *>act]i/ics4 ,! physical object1 :i* he calls ! '?+ 9 itself'17#a7 he reg>ds z ess5tially unkn[able2 :at c be kn[n is ! object z we h x 9 exp]i5ce1 :i* he calls ! 'ph5om5on'4 ,! ph5om5on1 be+ a jo9t product ( u & ! ?+ 9 itself1 is sure to h ^? *>act]i/ics :i* are due to u1 & is !re=e sure to con=m to our a priori k4 ,h5ce ?is k1 ?\gh true ( all actual & possible exp]i5ce1 mu/ n be suppos$ to apply outside exp]i5ce4 ,?us 9 spite ( ! exi/5ce ( a priori k1 we _c "k any?+ ab\t ! ?+ 9 itself or ab\t :at is n an actual or possible object ( exp]i5ce4 ,9 ?is way he tries to reconcile & h>monize ! cont5tions ( ! rationali/s ) ! argum5ts ( ! empirici/s4 7#a7 ,kant's '?+ 9 itself' is id5tical 9 def9ition ) ! physical object1 namely1 x is ! cause ( s5sations4 ,9 ! prop]ties d$uc$ f ! def9ition x is n id5tical1 s9ce ,kant held 79 spite ( "s inconsi/5cy z reg>ds cause7 t we c "k t none ( ! categories are applicable to ! '?+ 9 itself'4 ,ap>t f m9or gr\nds on :i* ,kant's philosophy may be criticiz$1 !re is "o ma9 objection :i* seems fatal to any attempt to deal ) ! problem ( a priori k by 8 me?od4 ,! ?+ to be acc\nt$ = is our c]ta9ty t ! facts mu/ always con=m to logic & ari?metic4 ,to say t logic & ari?metic are contribut$ by u does n acc\nt = ?is4 ,our nature is z mu* a fact ( ! exi/+ _w z any?+1 & !re c be no c]ta9ty t x w rema9 con/ant4 ,x might happ51 if ,kant is "r1 t to-morr[ our nature w\ld s *ange z to make two & two become five4 ,?is possibility seems nev] to h occurr$ to him1 yet x is "o :i* utt]ly de/roys ! c]ta9ty & univ]sality :i* he is anxi\s to v9dicate = ari?metical propositions4 ,x is true t ?is possibility1 =mally1 is inconsi/5t ) ! ,kantian view t "t itself is a =m impos$ by ! subject ^u ph5om5a1 s t our real ,self is n 9 "t & has no to-morr[4 ,b he w /ill h to suppose t ! "t-ord] ( ph5om5a is det]m9$ by *>act]i/ics ( :at is beh9d ph5om5a1 & ?is suffices = ! sub/ance ( our argum5t4 ,reflection1 moreov]1 seems to make x cle> t1 if !re is any tru? 9 our ari?metical beliefs1 !y mu/ apply to ?+s equally :e!r we ?9k ( !m or n4 ,two physical objects & two o!r physical objects mu/ make f\r physical objects1 ev5 if physical objects _c be exp]i5c$4 ,to ass]t ?is is c]ta9ly )9 ! scope ( :at we mean :5 we /ate t two & two are f\r4 ,its tru? is j z indubitable z ! tru? ( ! ass]tion t two ph5om5a & two o!r ph5om5a make f\r ph5om5a4 ,?us ,kant's solution unduly limits ! scope ( a priori propositions1 9 addition to fail+ 9 ! attempt at expla9+ _! c]ta9ty4 ,ap>t f ! special doctr9es advocat$ by ,kant1 x is v common among philosoph]s to reg>d :at is a priori z 9 "s s5se m5tal1 z conc]n$ r ) ! way we mu/ ?9k ?an ) any fact ( ! out] _w4 ,we not$ 9 ! prec$+ *apt] ! ?ree pr9ciples commonly call$ 'laws ( ?\ght'4 ,! view :i* l$ to _! be+ s nam$ is a natural "o1 b !re are /rong reasons = ?9k+ t x is errone\s4 ,let u take z an illu/ration ! law ( contradiction4 ,?is is commonly /at$ 9 ! =m ',no?+ c bo? be & n be'1 :i* is int5d$ to express ! fact t no?+ c at once h & n h a giv5 quality4 ,?us1 = example1 if a tree is a bee* x _c also be n a bee*2 if my table is rectangul> x _c also be n rectangul>1 & s on4 ,n[ :at makes x natural to call ?is pr9ciple a law ( ?\ght is t x is by ?\ght r ?an by outw>d obs]vation t we p]suade ourselves ( its necess>y tru?4 ,:5 we h se5 t a tree is a bee*1 we d n ne$ to look aga9 9 ord] to asc]ta9 :e!r x is also n a bee*2 ?\ght alone makes u "k t ?is is impossible4 ,b ! conclusion t ! law ( contradiction is a law ( ?\ght is nev]!less errone\s4 ,:at we believe1 :5 we believe ! law ( contradiction1 is n t ! m9d is s made t x mu/ believe ! law ( contradiction4 ,?is belief is a subsequ5t result ( psy*ological reflection1 :i* presupposes ! belief 9 ! law ( contradiction4 ,! belief 9 ! law ( contradiction is a belief ab\t ?+s1 n only ab\t ?\ghts4 ,x is n1 e4g41 ! belief t if we ?9k a c]ta9 tree is a bee*1 we _c at ! same "t ?9k t x is n a bee*2 x is ! belief t if ! tree is a bee*1 x _c at ! same "t be n a bee*4 ,?us ! law ( contradiction is ab\t ?+s1 & n m]ely ab\t ?\ghts2 & al?\gh belief 9 ! law ( contradiction is a ?\ght1 ! law ( contradiction itself is n a ?\ght1 b a fact conc]n+ ! ?+s 9 ! _w4 ,if ?is1 :i* we believe :5 we believe ! law ( contradiction1 7 n true ( ! ?+s 9 ! _w1 ! fact t we 7 compell$ to ?9k x true w\ld n save ! law ( contradiction f be+ false2 & ?is %[s t ! law is n a law ( ?\ght4 ,a simil> argum5t applies to any o!r a priori judgem5t4 ,:5 we judge t two & two are f\r1 we are n mak+ a judgem5t ab\t our ?\ghts1 b ab\t all actual or possible c\ples4 ,! fact t our m9ds are s con/itut$ z to believe t two & two are f\r1 ?\gh x is true1 is emphatically n :at we ass]t :5 we ass]t t two & two are f\r4 ,& no fact ab\t ! con/itution ( our m9ds c\ld make x true t two & two are f\r4 ,?us our a priori k1 if x is n errone\s1 is n m]ely k ab\t ! con/itution ( our m9ds1 b is applicable to :atev] ! _w may conta91 bo? :at is m5tal & :at is non-m5tal4 ,! fact seems to be t all our a priori k is conc]n$ ) entities :i* d n1 prop]ly speak+1 exi/1 ei!r 9 ! m5tal or 9 ! physical _w4 ,^! entities are su* z c be nam$ by p>ts ( spee* :i* are n sub/antives2 !y are su* entities z qualities & relations4 ,suppose1 = in/ance1 t ,i am 9 my room4 ,i exi/1 & my room exi/s2 b does '9' exi/8 ,yet obvi\sly ! ^w '9' has a mean+2 x d5otes a relation :i* holds betwe5 me & my room4 ,?is relation is some?+1 al?\gh we _c say t x exi/s 9 ! same s5se 9 :i* ,i & my room exi/4 ,! relation '9' is some?+ :i* we c ?9k ab\t & und]/&1 =1 if we c\ld n und]/& x1 we c\ld n und]/& ! s5t5ce ',i am 9 my room'4 ,_m philosoph]s1 foll[+ ,kant1 h ma9ta9$ t relations are ! "w ( ! m9d1 t ?+s 9 !mselves h no relations1 b t ! m9d br+s !m toge!r 9 "o act ( ?\ght & ?us produces ! relations :i* x judges !m to h4 ,?is view1 h[ev]1 seems op5 to objections simil> to ^? :i* we urg$ be=e aga9/ ,kant4 ,x seems pla9 t x is n ?\ght :i* produces ! tru? ( ! proposition ',i am 9 my room'4 ,x may be true t an e>wig is 9 my room1 ev5 if nei!r ,i nor ! e>wig nor any "o else is aw>e ( ?is tru?2 = ?is tru? conc]ns only ! e>wig & ! room1 & does n dep5d ^u any?+ else4 ,?us relations1 z we %all see m fully 9 ! next *apt]1 mu/ be plac$ 9 a _w :i* is nei!r m5tal nor physical4 ,?is _w is ( great importance to philosophy1 & 9 p>ticul> to ! problems ( a priori k4 ,9 ! next *apt] we %all proce$ to develop its nature & its be>+ ^u ! que/ions ) :i* we h be5 deal+4 ,,*apt] ,,ix4 ,,! ,,_w ,,( ,,univ]sals ,at ! end ( ! prec$+ *apt] we saw t su* entities z relations appe> to h a be+ :i* is 9 "s way diff]5t f t ( physical objects1 & also diff]5t f t ( m9ds & f t ( s5se-data4 ,9 ! pres5t *apt] we h to consid] :at is ! nature ( ?is k9d ( be+1 & also :at objects !re are t h ?is k9d ( be+4 ,we w beg9 ) ! latt] "q4 ,! problem ) :i* we are n[ conc]n$ is a v old "o1 s9ce x 0 br\ght into philosophy by ,plato4 ,plato's '!ory ( ideas' is an attempt to solve ?is v problem1 & 9 my op9ion x is "o ( ! mo/ successful attempts hi!rto made4 ,! !ory to be advocat$ 9 :at foll[s is l>gely ,plato's1 ) m]ely su* modifications z "t has %[n to be necess>y4 ,! way ! problem arose = ,plato 0 m or less z foll[s4 ,let u consid]1 say1 su* a notion z ju/ice4 ,if we ask ourselves :at ju/ice is1 x is natural to proce$ by consid]+ ?is1 t1 & ! o!r j act1 ) a view to discov]+ :at !y h 9 common4 ,!y mu/ all1 9 "s s5se1 p>take ( a common nature1 :i* w be f\nd 9 :atev] is j & 9 no?+ else4 ,?is common nature1 9 virtue ( :i* !y are all j1 w be ju/ice itself1 ! pure ess5ce ! admixture ( :i* ) facts ( ord9>y life produces ! multiplicity ( j acts4 ,simil>ly ) any o!r ^w :i* may be applicable to common facts1 su* z ':it5ess' = example4 ,! ^w w be applicable to a numb] ( p>ticul> ?+s because !y all p>ticipate 9 a common nature or ess5ce4 ,?is pure ess5ce is :at ,plato calls an 'idea' or '=m'4 7,x mu/ n be suppos$ t 'ideas'1 9 8 s5se1 exi/ 9 m9ds1 ?\gh !y may be appreh5d$ by m9ds47 ,! 'idea' ju/ice is n id5tical ) any?+ t is j3 x is some?+ o!r ?an p>ticul> ?+s1 :i* p>ticul> ?+s p>take (4 ,n be+ p>ticul>1 x _c itself exi/ 9 ! _w ( s5se4 ,moreov] x is n fleet+ or *angeable l ! ?+s ( s5se3 x is et]nally itself1 immutable & inde/ructible4 ,?us ,plato is l$ to a supra-s5sible _w1 m real ?an ! common _w ( s5se1 ! un*angeable _w ( ideas1 :i* alone gives to ! _w ( s5se :atev] pale reflection ( reality may belong to x4 ,! truly real _w1 = ,plato1 is ! _w ( ideas2 = :atev] we may attempt to say ab\t ?+s 9 ! _w ( s5se1 we c only succe$ 9 say+ t !y p>ticipate 9 su* & su* ideas1 :i*1 !re=e1 con/itute all _! "*4 ,h5ce x is easy to pass on into a my/icism4 ,we may hope1 9 a my/ic illum9ation1 to see ! ideas z we see objects ( s5se2 & we may imag9e t ! ideas exi/ 9 heav54 ,^! my/ical developm5ts are v natural1 b ! basis ( ! !ory is 9 logic1 & x is z bas$ 9 logic t we h to consid] x4 ,! ^w 'idea' has acquir$1 9 ! c\rse ( "t1 _m associations :i* are q mislead+ :5 appli$ to ,plato's 'ideas'4 ,we %all !re=e use ! ^w 'univ]sal' in/ead ( ! ^w 'idea'1 to describe :at ,plato meant4 ,! ess5ce ( ! sort ( entity t ,plato meant is t x is oppos$ to ! p>ticul> ?+s t are giv5 9 s5sation4 ,we speak ( :atev] is giv5 9 s5sation1 or is ( ! same nature z ?+s giv5 9 s5sation1 z a p>ticul>2 by opposition to ?is1 a univ]sal w be any?+ :i* may be %>$ by _m p>ticul>s1 & has ^? *>act]i/ics :i*1 z we saw1 di/+ui% ju/ice & :it5ess f j acts & :ite ?+s4 ,:5 we exam9e common words1 we f9d t1 broadly speak+1 prop] names /& = p>ticul>s1 :ile o!r sub/antives1 adjectives1 prepositions1 & v]bs /& = univ]sals4 ,pron\ns /& = p>ticul>s1 b are ambigu\s3 x is only by ! context or ! circum/ances t we "k :at p>ticul>s !y /& =4 ,! ^w 'n[' /&s = a p>ticul>1 namely ! pres5t mom5t2 b l pron\ns1 x /&s = an ambigu\s p>ticul>1 because ! pres5t is always *ang+4 ,x w be se5 t no s5t5ce c be made up )\t at lea/ "o ^w :i* d5otes a univ]sal4 ,! ne>e/ approa* w\ld be "s su* /atem5t z ',i l ?is'4 ,b ev5 "h ! ^w 'l' d5otes a univ]sal1 = ,i may l o!r ?+s1 & o!r p may l ?+s4 ,?us all tru?s involve univ]sals1 & all k ( tru?s involves acqua9tance ) univ]sals4 ,see+ t ne>ly all ! words to be f\nd 9 ! diction>y /& = univ]sals1 x is /range t h>dly anybody except /ud5ts ( philosophy ev] realizes t !re are su* entities z univ]sals4 ,we d n naturally dwell ^u ^? words 9 a s5t5ce :i* d n /& = p>ticul>s2 & if we are =c$ to dwell ^u a ^w :i* /&s = a univ]sal1 we naturally ?9k ( x z /&+ = "s "o ( ! p>ticul>s t come "u ! univ]sal4 ,:51 = example1 we he> ! s5t5ce1 ',*>les ,i's head 0 cut (f'1 we may naturally en\gh ?9k ( ,*>les ,i1 ( ,*>les ,i's head1 & ( ! op]ation ( cutt+ (f 8 head1 :i* are all p>ticul>s2 b we d n naturally dwell ^u :at is meant by ! ^w 'head' or ! ^w 'cut'1 :i* is a univ]sal3 ,we feel su* words to be incomplete & insub/antial2 !y seem to dem& a context be=e any?+ c be done ) !m4 ,h5ce we succe$ 9 avoid+ all notice ( univ]sals z su*1 until ! /udy ( philosophy =ces !m ^u our att5tion4 ,ev5 among philosoph]s1 we may say1 broadly1 t only ^? univ]sals :i* are nam$ by adjectives or sub/antives h be5 mu* or (t5 recogniz$1 :ile ^? nam$ by v]bs & prepositions h be5 usually ov]look$4 ,?is omission has _h a v great effect ^u philosophy2 x is h>dly too mu* to say t mo/ metaphysics1 s9ce ,sp9oza1 has be5 l>gely det]m9$ by x4 ,! way ?is has occurr$ is1 9 outl9e1 z foll[s3 ,speak+ g5]ally1 adjectives & common n\ns express qualities or prop]ties ( s+le ?+s1 :]eas prepositions & v]bs t5d to express relations betwe5 two or m ?+s4 ,?us ! neglect ( prepositions & v]bs l$ to ! belief t e proposition c be reg>d$ z attribut+ a prop]ty to a s+le ?+1 r ?an z express+ a relation betwe5 two or m ?+s4 ,h5ce x 0 suppos$ t1 ultimately1 !re c be no su* entities z relations betwe5 ?+s4 ,h5ce ei!r !re c be only "o ?+ 9 ! univ]se1 or1 if !re are _m ?+s1 !y _c possibly int]act 9 any way1 s9ce any int]action w\ld be a relation1 & relations are impossible4 ,! fir/ ( ^! views1 advocat$ by ,sp9oza & held 9 our own day by ,bradley & _m o!r philosoph]s1 is call$ monism2 ! second1 advocat$ by ,leibniz b n v common n[adays1 is call$ monadism1 because ea* ( ! isolat$ ?+s is call$ a monad4 ,bo? ^! oppos+ philosophies1 int]e/+ z !y are1 result1 9 my op9ion1 f an undue att5tion to "o sort ( univ]sals1 namely ! sort repres5t$ by adjectives & sub/antives r ?an by v]bs & prepositions4 ,z a matt] ( fact1 if any "o 7 anxi\s to d5y altoge!r t !re are su* ?+s z univ]sals1 we %\ld f9d t we _c /rictly prove t !re are su* entities z qualities1 i4e4 ! univ]sals repres5t$ by adjectives & sub/antives1 :]eas we c prove t !re mu/ be relations1 i4e4 ! sort ( univ]sals g5]ally repres5t$ by v]bs & prepositions4 ,let u take 9 illu/ration ! univ]sal :it5ess4 ,if we believe t !re is su* a univ]sal1 we %all say t ?+s are :ite because !y h ! quality ( :it5ess4 ,?is view1 h[ev]1 0 /r5u\sly d5i$ by ,b]keley & ,hume1 :o h be5 foll[$ 9 ?is by lat] empirici/s4 ,! =m :i* _! d5ial took 0 to d5y t !re are su* ?+s z 'ab/ract ideas '4 ,:5 we want to ?9k ( :it5ess1 !y said1 we =m an image ( "s p>ticul> :ite ?+1 & reason conc]n+ ?is p>ticul>1 tak+ c>e n to d$uce any?+ conc]n+ x :i* we _c see to be equally true ( any o!r :ite ?+4 ,z an acc\nt ( our actual m5tal processes1 ?is is no d\bt l>gely true4 ,9 geometry1 = example1 :5 we wi% to prove some?+ ab\t all triangles1 we draw a p>ticul> triangle & reason ab\t x1 tak+ c>e n to use any *>act]i/ic :i* x does n %>e ) o!r triangles4 ,! beg9n]1 9 ord] to avoid error1 (t5 f9ds x useful to draw sev]al triangles1 z unlike ea* o!r z possible1 9 ord] to make sure t 8 reason+ is equally applicable to all ( !m4 ,b a difficulty em]ges z soon z we ask ourselves h[ we "k t a ?+ is :ite or a triangle4 ,if we wi% to avoid ! univ]sals :it5ess & triangul>ity1 we %all *oose "s p>ticul> pat* ( :ite or "s p>ticul> triangle1 & say t any?+ is :ite or a triangle if x has ! "r sort ( resemblance to our *os5 p>ticul>4 ,b !n ! resemblance requir$ w h to be a univ]sal4 ,s9ce !re are _m :ite ?+s1 ! resemblance mu/ hold betwe5 _m pairs ( p>ticul> :ite ?+s2 & ?is is ! *>act]i/ic ( a univ]sal4 ,x w be useless to say t !re is a diff]5t resemblance = ea* pair1 = !n we %all h to say t ^! resemblances resemble ea* o!r1 & ?us at la/ we %all be =c$ to admit resemblance z a univ]sal4 ,! relation ( resemblance1 !re=e1 mu/ be a true univ]sal4 ,& hav+ be5 =c$ to admit ?is univ]sal1 we f9d t x is no long] wor? :ile to inv5t difficult & unplausible !ories to avoid ! admission ( su* univ]sals z :it5ess & triangul>ity4 ,b]keley & ,hume fail$ to p]ceive ?is refutation ( _! rejection ( 'ab/ract ideas'1 because1 l _! adv]s>ies1 !y only ?\ght ( qualities1 & altoge!r ignor$ relations z univ]sals4 ,we h !re=e "h ano!r respect 9 :i* ! rationali/s appe> to h be5 9 ! "r z aga9/ ! empirici/s1 al?\gh1 ow+ to ! neglect or d5ial ( relations1 ! d$uctions made by rationali/s 71 if any?+1 m apt to be mi/ak5 ?an ^? made by empirici/s4 ,hav+ n[ se5 t !re mu/ be su* entities z univ]sals1 ! next po9t to be prov$ is t _! be+ is n m]ely m5tal4 ,by ?is is meant t :atev] be+ belongs to !m is indep5d5t ( _! be+ ?\ght ( or 9 any way appreh5d$ by m9ds4 ,we h already t\*$ on ?is subject at ! end ( ! prec$+ *apt]1 b we mu/ n[ consid] m fully :at sort ( be+ x is t belongs to univ]sals4 ,consid] su* a proposition z ',ed9burgh is nor? ( ,london'4 ,"h we h a relation betwe5 two places1 & x seems pla9 t ! relation subsi/s indep5d5tly ( our k ( x4 ,:5 we come to "k t ,ed9burgh is nor? ( ,london1 we come to "k some?+ :i* has to d only ) ,ed9burgh & ,london3 we d n cause ! tru? ( ! proposition by com+ to "k x1 on ! contr>y we m]ely appreh5d a fact :i* 0 !re be=e we knew x4 ,! "p ( ! e>?'s surface :]e ,ed9burgh /&s w\ld be nor? ( ! "p :]e ,london /&s1 ev5 if !re 7 no human be+ to "k ab\t nor? & s\?1 & ev5 if !re 7 no m9ds at all 9 ! univ]se4 ,?is is1 ( c\rse1 d5i$ by _m philosoph]s1 ei!r = ,b]keley's reasons or = ,kant's4 ,b we h already consid]$ ^! reasons1 & decid$ t !y are inadequate4 ,we may !re=e n[ assume x to be true t no?+ m5tal is presuppos$ 9 ! fact t ,ed9burgh is nor? ( ,london4 ,b ?is fact involves ! relation 'nor? ('1 :i* is a univ]sal2 & x w\ld be impossible = ! :ole fact to involve no?+ m5tal if ! relation 'nor? ('1 :i* is a con/itu5t "p ( ! fact1 did involve any?+ m5tal4 ,h5ce we mu/ admit t ! relation1 l ! t]ms x relates1 is n dep5d5t ^u ?\ght1 b belongs to ! indep5d5t _w :i* ?\ght appreh5ds b does n create4 ,?is conclusion1 h[ev]1 is met by ! difficulty t ! relation 'nor? (' does n seem to exi/ 9 ! same s5se 9 :i* ,ed9burgh & ,london exi/4 ,if we ask ',:]e & :5 does ?is relation exi/8' ! answ] mu/ be ',n[h]e & n[h5'4 ,!re is no place or "t :]e we c f9d ! relation 'nor? ('4 ,x does n exi/ 9 ,ed9burgh any m ?an 9 ,london1 = x relates ! two & is neutral z betwe5 !m4 ,nor c we say t x exi/s at any p>ticul> "t4 ,n[ ev]y?+ t c be appreh5d$ by ! s5ses or by introspection exi/s at "s p>ticul> "t4 ,h5ce ! relation 'nor? (' is radically diff]5t f su* ?+s4 ,x is nei!r 9 space nor 9 "t1 nei!r mat]ial nor m5tal2 yet x is some?+4 ,x is l>gely ! v peculi> k9d ( be+ t belongs to univ]sals :i* has l$ _m p to suppose t !y are really m5tal4 ,we c ?9k ( a univ]sal1 & our ?9k+ !n exi/s 9 a p]fectly ord9>y s5se1 l any o!r m5tal act4 ,suppose1 = example1 t we are ?9k+ ( :it5ess4 ,!n 9 "o s5se x may be said t :it5ess is '9 our m9d'4 ,we h "h ! same ambiguity z we not$ 9 discuss+ ,b]keley 9 ,*apt] ,,iv4 ,9 ! /rict s5se1 x is n :it5ess t is 9 our m9d1 b ! act ( ?9k+ ( :it5ess4 ,! connect$ ambiguity 9 ! ^w 'idea'1 :i* we not$ at ! same "t1 also causes confusion "h4 ,9 "o s5se ( ?is ^w1 namely ! s5se 9 :i* x d5otes ! object ( an act ( ?\ght1 :it5ess is an 'idea'4 ,h5ce1 if ! ambiguity is n gu>d$ aga9/1 we may come to ?9k t :it5ess is an 'idea' 9 ! o!r s5se1 i4e4 an act ( ?\ght2 & ?us we come to ?9k t :it5ess is m5tal4 ,b 9 s ?9k+1 we rob x ( its ess5tial quality ( univ]sality4 ,"o man's act ( ?\ght is necess>ily a diff]5t ?+ f ano!r man's2 "o man's act ( ?\ght at "o "t is necess>ily a diff]5t ?+ f ! same man's act ( ?\ght at ano!r "t4 ,h5ce1 if :it5ess 7 ! ?\ght z oppos$ to its object1 no two diff]5t m5 c\ld ?9k ( x1 & no "o man c\ld ?9k ( x twice4 ,t :i* _m diff]5t ?\ghts ( :it5ess h 9 common is _! object1 & ?is object is diff]5t f all ( !m4 ,?us univ]sals are n ?\ghts1 ?\gh :5 kn[n !y are ! objects ( ?\ghts4 ,we %all f9d x conv5i5t only to speak ( ?+s exi/+ :5 !y are 9 "t1 t is to say1 :5 we c po9t to "s "t at :i* !y exi/ 7n exclud+ ! possibility ( _! exi/+ at all times74 ,?us ?\ghts & feel+s1 m9ds & physical objects exi/4 ,b univ]sals d n exi/ 9 ?is s5se2 we %all say t !y subsi/ or h be+1 :]e 'be+' is oppos$ to 'exi/5ce' z be+ timeless4 ,! _w ( univ]sals1 !re=e1 may also be describ$ z ! _w ( be+4 ,! _w ( be+ is un*angeable1 rigid1 exact1 delightful to ! ma!matician1 ! logician1 ! build] ( metaphysical sy/ems1 & all :o love p]fection m ?an life4 ,! _w ( exi/5ce is fleet+1 vague1 )\t %>p b\nd>ies1 )\t any cle> plan or arrangem5t1 b x conta9s all ?\ghts & feel+s1 all ! data ( s5se1 & all physical objects1 ev]y?+ t c d ei!r good or h>m1 ev]y?+ t makes any diff]5ce to ! value ( life & ! _w4 ,accord+ to our temp]am5ts1 we %all pref] ! contemplation ( ! "o or ( ! o!r4 ,! "o we d n pref] w probably seem to u a pale %ad[ ( ! "o we pref]1 & h>dly wor?y to be reg>d$ z 9 any s5se real4 ,b ! tru? is t bo? h ! same claim on our imp>tial att5tion1 bo? are real1 & bo? are important to ! metaphysician4 ,inde$ no soon] h we di/+ui%$ ! two worlds ?an x becomes necess>y to consid] _! relations4 ,b fir/ ( all we mu/ exam9e our k ( univ]sals4 ,?is consid]ation w occupy u 9 ! foll[+ *apt]1 :]e we %all f9d t x solves ! problem ( a priori k1 f :i* we 7 fir/ l$ to consid] univ]sals4 ,,*apt] ,x4 ,,on ,,our ,,k ,,( ,,univ]sals ,9 reg>d to "o man's k at a giv5 "t1 univ]sals1 l p>ticul>s1 may be divid$ into ^? kn[n by acqua9tance1 ^? kn[n only by description1 & ^? n kn[n ei!r by acqua9tance or by description4 ,let u consid] fir/ ! k ( univ]sals by acqua9tance4 ,x is obvi\s1 to beg9 )1 t we are acqua9t$ ) su* univ]sals z :ite1 r$1 black1 sweet1 s\r1 l\d1 h>d1 etc41 i4e4 ) qualities :i* are exemplifi$ 9 s5se-data4 ,:5 we see a :ite pat*1 we are acqua9t$1 9 ! fir/ in/ance1 ) ! p>ticul> pat*2 b by see+ _m :ite pat*es1 we easily le>n to ab/ract ! :it5ess :i* !y all h 9 common1 & 9 le>n+ to d ?is we are le>n+ to be acqua9t$ ) :it5ess4 ,a simil> process w make u acqua9t$ ) any o!r univ]sal ( ! same sort4 ,univ]sals ( ?is sort may be call$ 's5sible qualities'4 ,!y c be appreh5d$ ) less ef=t ( ab/raction ?an any o!rs1 & !y seem less remov$ f p>ticul>s ?an o!r univ]sals are4 ,we come next to relations4 ,! easie/ relations to appreh5d are ^? :i* hold betwe5 ! diff]5t p>ts ( a s+le complex s5se-datum4 ,= example1 ,i c see at a glance ! :ole ( ! page on :i* ,i am writ+2 ?us ! :ole page is includ$ 9 "o s5se-datum4 ,b ,i p]ceive t "s p>ts ( ! page are to ! left ( o!r p>ts1 & "s p>ts are above o!r p>ts4 ,! process ( ab/raction 9 ?is case seems to proce$ some:at z foll[s3 ,i see successively a numb] ( s5se-data 9 :i* "o "p is to ! left ( ano!r2 ,i p]ceive1 z 9 ! case ( diff]5t :ite pat*es1 t all ^! s5se-data h some?+ 9 common1 & by ab/raction ,i f9d t :at !y h 9 common is a c]ta9 relation betwe5 _! p>ts1 namely ! relation :i* ,i call 'be+ to ! left ('4 ,9 ?is way ,i become acqua9t$ ) ! univ]sal relation4 ,9 l mann] ,i become aw>e ( ! relation ( be=e & aft] 9 "t4 ,suppose ,i he> a *ime ( bells3 :5 ! la/ bell ( ! *ime s\nds1 ,i c reta9 ! :ole *ime be=e my m9d1 & ,i c p]ceive t ! e>li] bells came be=e ! lat] ones4 ,also 9 memory ,i p]ceive t :at ,i am rememb]+ came be=e ! pres5t "t4 ,f ei!r ( ^! s\rces ,i c ab/ract ! univ]sal relation ( be=e & aft]1 j z ,i ab/ract$ ! univ]sal relation 'be+ to ! left ('4 ,?us "t-relations1 l space-relations1 are among ^? ) :i* we are acqua9t$4 ,ano!r relation ) :i* we become acqua9t$ 9 mu* ! same way is resemblance4 ,if ,i see simultane\sly two %ades ( gre51 ,i c see t !y resemble ea* o!r2 if ,i also see a %ade ( r$3 at ! same "t1 ,i c see t ! two gre5s h m resemblance to ea* o!r ?an ei!r has to ! r$4 ,9 ?is way ,i become acqua9t$ ) ! univ]sal resemblance or simil>ity4 ,betwe5 univ]sals1 z betwe5 p>ticul>s1 !re are relations ( :i* we may be imm$iately aw>e4 ,we h j se5 t we c p]ceive t ! resemblance betwe5 two %ades ( gre5 is great] ?an ! resemblance betwe5 a %ade ( r$ & a %ade ( gre54 ,"h we are deal+ ) a relation1 namely 'great] ?an'1 betwe5 two relations4 ,our k ( su* relations1 ?\gh x requires m p[] ( ab/raction ?an is requir$ = p]ceiv+ ! qualities ( s5se-data1 appe>s to be equally imm$iate1 & 7at lea/ 9 "s cases7 equally indubitable4 ,?us !re is imm$iate k conc]n+ univ]sals z well z conc]n+ s5se-data4 ,return+ n[ to ! problem ( a priori k1 :i* we left unsolv$ :5 we began ! consid]ation ( univ]sals1 we f9d ourselves 9 a position to deal ) x 9 a mu* m satisfactory mann] ?an 0 possible be=e4 ,let u rev]t to ! proposition 'two & two are f\r'4 ,x is fairly obvi\s1 9 view ( :at has be5 said1 t ?is proposition /ates a relation betwe5 ! univ]sal 'two' & ! univ]sal 'f\r'4 ,?is sugge/s a proposition :i* we %all n[ endeav\r to e/abli%3 namely1 ,all a priori k deals exclusively ) ! relations ( univ]sals4 ,?is proposition is ( great importance1 & goes a long way t[>ds solv+ our previ\s difficulties conc]n+ a priori k4 ,! only case 9 :i* x might seem1 at fir/ sight1 z if our proposition 7 untrue1 is ! case 9 :i* an a priori proposition /ates t all ( "o class ( p>ticul>s belong to "s o!r class1 or 7:at comes to ! same ?+7 t all p>ticul>s hav+ "s "o prop]ty also h "s o!r4 ,9 ?is case x might seem z ?\gh we 7 deal+ ) ! p>ticul>s t h ! prop]ty r ?an ) ! prop]ty4 ,! proposition 'two & two are f\r' is really a case 9 po9t1 = ?is may be /at$ 9 ! =m 'any two & any o!r two are f\r'1 or 'any collection =m$ ( two twos is a collection ( f\r'4 ,if we c %[ t su* /atem5ts z ?is really deal only ) univ]sals1 our proposition may be reg>d$ z prov$4 ,"o way ( discov]+ :at a proposition deals ) is to ask ourselves :at words we mu/ und]/&--9 o!r words1 :at objects we mu/ be acqua9t$ )--9 ord] to see :at ! proposition means4 ,z soon z we see :at ! proposition means1 ev5 if we d n yet "k :e!r x is true or false1 x is evid5t t we mu/ h acqua9tance ) :atev] is really dealt ) by ! proposition4 ,by apply+ ?is te/1 x appe>s t _m propositions :i* might seem to be conc]n$ ) p>ticul>s are really conc]n$ only ) univ]sals4 ,9 ! special case ( 'two & two are f\r'1 ev5 :5 we int]pret x z mean+ 'any collection =m$ ( two twos is a collection ( f\r'1 x is pla9 t we c und]/& ! proposition1 i4e4 we c see :at x is t x ass]ts1 z soon z we "k :at is meant by 'collection' & 'two' & 'f\r'4 ,x is q unnecess>y to "k all ! c\ples 9 ! _w3 if x 7 necess>y1 obvi\sly we c\ld nev] und]/& ! proposition1 s9ce ! c\ples are inf9itely num]\s & !re=e _c all be kn[n to u4 ,?us al?\gh our g5]al /atem5t implies /atem5ts ab\t p>ticul> c\ples1 z soon z we "k t !re are su* p>ticul> c\ples1 yet x does n itself ass]t or imply t !re are su* p>ticul> c\ples1 & ?us fails to make any /atem5t :atev] ab\t any actual p>ticul> c\ple4 ,! /atem5t made is ab\t 'c\ple'1 ! univ]sal1 & n ab\t ?is or t c\ple4 ,?us ! /atem5t 'two & two are f\r' deals exclusively ) univ]sals1 & !re=e may be kn[n by anybody :o is acqua9t$ ) ! univ]sals conc]n$ & c p]ceive ! relation betwe5 !m :i* ! /atem5t ass]ts4 ,x mu/ be tak5 z a fact1 discov]$ by reflect+ ^u our k1 t we h ! p[] ( sometimes p]ceiv+ su* relations betwe5 univ]sals1 & !re=e ( sometimes kn[+ g5]al a priori propositions su* z ^? ( ari?metic & logic4 ,! ?+ t seem$ my/]i\s1 :5 we =m]ly consid]$ su* k1 0 t x seem$ to anticipate & control exp]i5ce4 ,?is1 h[ev]1 we c n[ see to h be5 an error4 ,no fact conc]n+ any?+ capable ( be+ exp]i5c$ c be kn[n indep5d5tly ( exp]i5ce4 ,we "k a priori t two ?+s & two o!r ?+s toge!r make f\r ?+s1 b we d n "k a priori t if ,br[n & ,jones are two1 & ,rob9son & ,smi? are two1 !n ,br[n & ,jones & ,rob9son & ,smi? are f\r4 ,! reason is t ?is proposition _c be und]/ood at all unless we "k t !re are su* p z ,br[n & ,jones & ,rob9son & ,smi?1 & ?is we c only "k by exp]i5ce4 ,h5ce1 al?\gh our g5]al proposition is a priori1 all its applications to actual p>ticul>s involve exp]i5ce & !re=e conta9 an empirical elem5t4 ,9 ?is way :at seem$ my/]i\s 9 our a priori k is se5 to h be5 bas$ ^u an error4 ,x w s]ve to make ! po9t cle>] if we contra/ our g5u9e a priori judgem5t ) an empirical g5]alization1 su* z 'all m5 are mortals'4 ,"h z be=e1 we c und]/& :at ! proposition means z soon z we und]/& ! univ]sals involv$1 namely man & mortal4 ,x is obvi\sly unnecess>y to h an individual acqua9tance ) ! :ole human race 9 ord] to und]/& :at our proposition means4 ,?us ! diff]5ce betwe5 an a priori g5]al proposition & an empirical g5]alization does n come 9 ! mean+ ( ! proposition2 x comes 9 ! nature ( ! evid5ce = x4 ,9 ! empirical case1 ! evid5ce consi/s 9 ! p>ticul> in/ances4 ,we believe t all m5 are mortal because we "k t !re are innum]able in/ances ( m5 dy+1 & no in/ances ( _! liv+ beyond a c]ta9 age4 ,we d n believe x because we see a connexion betwe5 ! univ]sal man & ! univ]sal mortal4 ,x is true t if physiology c prove1 assum+ ! g5]al laws t gov]n liv+ bodies1 t no liv+ organism c la/ = ev]1 t gives a connexion betwe5 man & mortality :i* w\ld enable u to ass]t our proposition )\t appeal+ to ! special evid5ce ( m5 dy+4 ,b t only means t our g5]alization has be5 subsum$ "u a wid] g5]alization1 = :i* ! evid5ce is /ill ( ! same k9d1 ?\gh m ext5sive4 ,! progress ( sci5ce is con/antly produc+ su* subsumptions1 & !re=e giv+ a con/antly wid] inductive basis = sci5tific g5]alizations4 ,b al?\gh ?is gives a great] degree ( c]ta9ty1 x does n give a diff]5t k9d3 ! ultimate gr\nd rema9s inductive1 i4e4 d]iv$ f in/ances1 & n an a priori connexion ( univ]sals su* z we h 9 logic & ari?metic4 ,two opposite po9ts are to be obs]v$ conc]n+ a priori g5]al propositions4 ,! fir/ is t1 if _m p>ticul> in/ances are kn[n1 our g5]al proposition may be arriv$ at 9 ! fir/ in/ance by induction1 & ! connexion ( univ]sals may be only subsequ5tly p]ceiv$4 ,= example1 x is kn[n t if we draw p]p5dicul>s to ! sides ( a triangle f ! opposite angles1 all ?ree p]p5dicul>s meet 9 a po9t4 ,x w\ld be q possible to be fir/ l$ to ?is proposition by actually draw+ p]p5dicul>s 9 _m cases1 & f9d+ t !y always met 9 a po9t2 ?is exp]i5ce might lead u to look = ! g5]al pro( & f9d x4 ,su* cases are common 9 ! exp]i5ce ( e ma!matician4 ,! o!r po9t is m int]e/+1 & ( m philosophical importance4 ,x is1 t we may sometimes "k a g5]al proposition 9 cases :]e we d n "k a s+le in/ance ( x4 ,take su* a case z ! foll[+3 ,we "k t any two numb]s c be multipli$ toge!r1 & w give a ?ird call$ _! product4 ,we "k t all pairs ( integ]s ! product ( :i* is less ?an #ajj h be5 actually multipli$ toge!r1 & ! value ( ! product record$ 9 ! multiplication table4 ,b we also "k t ! numb] ( integ]s is inf9ite1 & t only a f9ite numb] ( pairs ( integ]s ev] h be5 or ev] w be ?\ght ( by human be+s4 ,h5ce x foll[s t !re are pairs ( integ]s :i* nev] h be5 & nev] w be ?\ght ( by human be+s1 & t all ( !m deal ) integ]s ! product ( :i* is ov] #ajj4 ,h5ce we arrive at ! proposition3 ',all products ( two integ]s1 :i* nev] h be5 & nev] w be ?\ght ( by any human be+1 are ov] #ajj4' ,"h is a g5]al proposition ( :i* ! tru? is und5iable1 & yet1 f ! v nature ( ! case1 we c nev] give an in/ance2 because any two numb]s we may ?9k ( are exclud$ by ! t]ms ( ! proposition4 ,?is possibility1 ( k ( g5]al propositions ( :i* no in/ance c be giv51 is (t5 d5i$1 because x is n p]ceiv$ t ! k ( su* propositions only requires a k ( ! relations ( univ]sals1 & does n require any k ( in/ances ( ! univ]sals 9 "q4 ,yet ! k ( su* g5]al propositions is q vital to a great deal ( :at is g5]ally admitt$ to be kn[n4 ,= example1 we saw1 9 our e>ly *apt]s1 t k ( physical objects1 z oppos$ to s5se-data1 is only obta9$ by an inf]5ce1 & t !y are n ?+s ) :i* we are acqua9t$4 ,h5ce we c nev] "k any proposition ( ! =m '?is is a physical object'1 :]e '?is' is some?+ imm$iately kn[n4 ,x foll[s t all our k conc]n+ physical objects is su* t no actual in/ance c be giv54 ,we c give in/ances ( ! associat$ s5se-data1 b we _c give in/ances ( ! actual physical objects4 ,h5ce our k z to physical objects dep5ds ?r\gh\t ^u ?is possibility ( g5]al k :]e no in/ance c be giv54 ,& ! same applies to our k ( o!r p's m9ds1 or ( any o!r class ( ?+s ( :i* no in/ance is kn[n to u by acqua9tance4 ,we may n[ take a survey ( ! s\rces ( our k1 z !y h appe>$ 9 ! c\rse ( our analysis4 ,we h fir/ to di/+ui% k ( ?+s & k ( tru?s4 ,9 ea* !re are two k9ds1 "o imm$iate & "o d]ivative4 ,our imm$iate k ( ?+s1 :i* we call$ acqua9tance1 consi/s ( two sorts1 accord+ z ! ?+s kn[n are p>ticul>s or univ]sals4 ,among p>ticul>s1 we h acqua9tance ) s5se-data & 7probably7 ) ourselves4 ,among univ]sals1 !re seems to be no pr9ciple by :i* we c decide :i* c be kn[n by acqua9tance1 b x is cle> t among ^? t c be s kn[n are s5sible qualities1 relations ( space & "t1 simil>ity1 & c]ta9 ab/ract logical univ]sals4 ,our d]ivative k ( ?+s1 :i* we call k by description1 always involves bo? acqua9tance ) some?+ & k ( tru?s4 ,our imm$iate k ( tru?s may be call$ intuitive k1 & ! tru?s s kn[n may be call$ self-evid5t tru?s4 ,among su* tru?s are includ$ ^? :i* m]ely /ate :at is giv5 9 s5se1 & also c]ta9 ab/ract logical & ari?metical pr9ciples1 & 7?\gh ) less c]ta9ty7 "s e?ical propositions4 ,our d]ivative k ( tru?s consi/s ( ev]y?+ t we c d$uce f self-evid5t tru?s by ! use ( self-evid5t pr9ciples ( d$uction4 ,if ! above acc\nt is correct1 all our k ( tru?s dep5ds ^u our intuitive k4 ,x !re=e becomes important to consid] ! nature & scope ( intuitive k1 9 mu* ! same way z1 at an e>li] /age1 we consid]$ ! nature & scope ( k by acqua9tance4 ,b k ( tru?s raises a fur!r problem1 :i* does n arise 9 reg>d to k ( ?+s1 namely ! problem ( error4 ,"s ( our beliefs turn out to be errone\s1 & !re=e x becomes necess>y to consid] h[1 if at all1 we c di/+ui% k f error4 ,?is problem does n arise ) reg>d to k by acqua9tance1 =1 :atev] may be ! object ( acqua9tance1 ev5 9 dreams & halluc9ations1 !re is no error involv$ s long z we d n g beyond ! imm$iate object3 error c only arise :5 we reg>d ! imm$iate object1 i4e4 ! s5se-datum1 z ! m>k ( "s physical object4 ,?us ! problems connect$ ) k ( tru?s are m difficult ?an ^? connect$ ) k ( ?+s4 ,z ! fir/ ( ! problems connect$ ) k ( tru?s1 let u exam9e ! nature & scope ( our intuitive judgem5ts4 ,,*apt] ,,xi4 ,,on ,,intuitive ,,k ,!re is a common impression t ev]y?+ t we believe ought to be capable ( pro(1 or at lea/ ( be+ %[n to be highly probable4 ,x is felt by _m t a belief = :i* no reason c be giv5 is an unreasonable belief4 ,9 ! ma91 ?is view is j4 ,almo/ all our common beliefs are ei!r inf]r$1 or capable ( be+ inf]r$1 f o!r beliefs :i* may be reg>d$ z giv+ ! reason = !m4 ,z a rule1 ! reason has be5 =gott51 or has ev5 nev] be5 consci\sly pres5t to our m9ds4 ,few ( u ev] ask ourselves1 = example1 :at reason !re is to suppose ! food we are j go+ to eat w n turn out to be poison4 ,yet we feel1 :5 *all5g$1 t a p]fectly good reason c\ld be f\nd1 ev5 if we are n ready ) x at ! mom5t4 ,& 9 ?is belief we are usually ju/ifi$4 ,b let u imag9e "s insi/5t ,socrates1 :o1 :atev] reason we give him1 cont9ues to dem& a reason = ! reason4 ,we mu/ soon] or lat]1 & probably be=e v long1 be driv5 to a po9t :]e we _c f9d any fur!r reason1 & :]e x becomes almo/ c]ta9 t no fur!r reason is ev5 !oretically discov]able4 ,/>t+ ) ! common beliefs ( daily life1 we c be driv5 back f po9t to po9t1 until we come to "s g5]al pr9ciple1 or "s in/ance ( a g5]al pr9ciple1 :i* seems lum9\sly evid5t1 & is n itself capable ( be+ d$uc$ f any?+ m evid5t4 ,9 mo/ que/ions ( daily life1 su* z :e!r our food is likely to be n\ri%+ & n poison\s1 we %all be driv5 back to ! inductive pr9ciple1 :i* we discuss$ 9 ,*apt] ,,vi4 ,b beyond t1 !re seems to be no fur!r regress4 ,! pr9ciple itself is con/antly us$ 9 our reason+1 sometimes consci\sly1 sometimes unconsci\sly2 b !re is no reason+ :i*1 />t+ f "s simpl] self-evid5t pr9ciple1 leads u to ! pr9ciple ( induction z its conclusion4 ,& ! same holds = o!r logical pr9ciples4 ,_! tru? is evid5t to u1 & we employ !m 9 con/ruct+ demon/rations2 b !y !mselves1 or at lea/ "s ( !m1 are incapable ( demon/ration4 ,self-evid5ce1 h[ev]1 is n conf9$ to ^? among g5]al pr9ciples :i* are incapable ( pro(4 ,:5 a c]ta9 numb] ( logical pr9ciples h be5 admitt$1 ! re/ c be d$uc$ f !m2 b ! propositions d$uc$ are (t5 j z self-evid5t z ^? t 7 assum$ )\t pro(4 ,all ari?metic1 moreov]1 c be d$uc$ f ! g5]al pr9ciples ( logic1 yet ! simple propositions ( ari?metic1 su* z 'two & two are f\r'1 are j z self-evid5t z ! pr9ciples ( logic4 ,x w\ld seem1 also1 ?\gh ?is is m disputable1 t !re are "s self-evid5t e?ical pr9ciples1 su* z 'we ought to pursue :at is good'4 ,x %\ld be obs]v$ t1 9 all cases ( g5]al pr9ciples1 p>ticul> in/ances1 deal+ ) famili> ?+s1 are m evid5t ?an ! g5]al pr9ciple4 ,= example1 ! law ( contradiction /ates t no?+ c bo? h a c]ta9 prop]ty & n h x4 ,?is is evid5t z soon z x is und]/ood1 b x is n s evid5t z t a p>ticul> rose :i* we see _c be bo? r$ & n r$4 7,x is ( c\rse possible t p>ts ( ! rose may be r$ & p>ts n r$1 or t ! rose may be ( a %ade ( p9k :i* we h>dly "k :e!r to call r$ or n2 b 9 ! =m] case x is pla9 t ! rose z a :ole is n r$1 :ile 9 ! latt] case ! answ] is !oretically def9ite z soon z we h decid$ on a precise def9ition ( 'r$'47 ,x is usually ?r\gh p>ticul> in/ances t we come to be able to see ! g5]al pr9ciple4 ,only ^? :o are practis$ 9 deal+ ) ab/ractions c readily grasp a g5]al pr9ciple )\t ! help ( in/ances4 ,9 addition to g5]al pr9ciples1 ! o!r k9d ( self-evid5t tru?s are ^? imm$iately d]iv$ f s5sation4 ,we w call su* tru?s 'tru?s ( p]ception'1 & ! judgem5ts express+ !m we w call 'judgem5ts ( p]ception'4 ,b "h a c]ta9 am\nt ( c>e is requir$ 9 gett+ at ! precise nature ( ! tru?s t are self-evid5t4 ,! actual s5se-data are nei!r true nor false4 ,a p>ticul> pat* ( col\r :i* ,i see1 = example1 simply exi/s3 x is n ! sort ( ?+ t is true or false4 ,x is true t !re is su* a pat*1 true t x has a c]ta9 %ape & degree ( brightness1 true t x is surr\nd$ by c]ta9 o!r col\rs4 ,b ! pat* itself1 l ev]y?+ else 9 ! _w ( s5se1 is ( a radically diff]5t k9d f ! ?+s t are true or false1 & !re=e _c prop]ly be said to be true4 ,?us :atev] self-evid5t tru?s may be obta9$ f our s5ses mu/ be diff]5t f ! s5se-data f :i* !y are obta9$4 ,x w\ld seem t !re are two k9ds ( self-evid5t tru?s ( p]ception1 ?\gh p]haps 9 ! la/ analysis ! two k9ds may coalesce4 ,fir/1 !re is ! k9d :i* simply ass]ts ! exi/5ce ( ! s5se-datum1 )\t 9 any way analys+ x4 ,we see a pat* ( r$1 & we judge '!re is su*-&-su* a pat* ( r$'1 or m /rictly '!re is t'2 ?is is "o k9d ( intuitive judgem5t ( p]ception4 ,! o!r k9d arises :5 ! object ( s5se is complex1 & we subject x to "s degree ( analysis4 ,if1 = in/ance1 we see a r\nd pat* ( r$1 we may judge 't pat* ( r$ is r\nd'4 ,?is is aga9 a judgem5t ( p]ception1 b x diff]s f our previ\s k9d4 ,9 our pres5t k9d we h a s+le s5se-datum :i* has bo? col\r & %ape3 ! col\r is r$ & ! %ape is r\nd4 ,our judgem5t analyses ! datum into col\r & %ape1 & !n recomb9es !m by /at+ t ! r$ col\r is r\nd 9 %ape4 ,ano!r example ( ?is k9d ( judgem5t is '?is is to ! "r ( t'1 :]e '?is' & 't' are se5 simultane\sly4 ,9 ?is k9d ( judgem5t ! s5se-datum conta9s con/itu5ts :i* h "s relation to ea* o!r1 & ! judgem5t ass]ts t ^! con/itu5ts h ?is relation4 ,ano!r class ( intuitive judgem5ts1 analog\s to ^? ( s5se & yet q di/9ct f !m1 are judgem5ts ( memory4 ,!re is "s dang] ( confusion z to ! nature ( memory1 ow+ to ! fact t memory ( an object is apt to be accompani$ by an image ( ! object1 & yet ! image _c be :at con/itutes memory4 ,?is is easily se5 by m]ely notic+ t ! image is 9 ! pres5t1 :]eas :at is rememb]$ is kn[n to be 9 ! pa/4 ,moreov]1 we are c]ta9ly able to "s ext5t to comp>e our image ) ! object rememb]$1 s t we (t5 "k1 )9 some:at wide limits1 h[ f> our image is accurate2 b ?is w\ld be impossible1 unless ! object1 z oppos$ to ! image1 7 9 "s way be=e ! m9d4 ,?us ! ess5ce ( memory is n con/itut$ by ! image1 b by hav+ imm$iately be=e ! m9d an object :i* is recogniz$ z pa/4 ,b = ! fact ( memory 9 ?is s5se1 we %\ld n "k t !re ev] 0 a pa/ at all1 nor %\ld we be able to und]/& ! ^w 'pa/'1 any m ?an a man born bl9d c und]/& ! ^w 'light'4 ,?us !re mu/ be intuitive judgem5ts ( memory1 & x is ^u !m1 ultimately1 t all our k ( ! pa/ dep5ds4 ,! case ( memory1 h[ev]1 raises a difficulty1 = x is notori\sly fallaci\s1 & ?us ?r[s d\bt on ! tru/wor?9ess ( intuitive judgem5ts 9 g5]al4 ,?is difficulty is no light "o4 ,b let u fir/ n>r[ its scope z f> z possible4 ,broadly speak+1 memory is tru/wor?y 9 proportion to ! vividness ( ! exp]i5ce & to its ne>ness 9 "t4 ,if ! h\se next door 0 /ruck by lightn+ half a m9ute ago1 my memory ( :at ,i saw & he>d w be s reliable t x w\ld be prepo/]\s to d\bt :e!r !re _h be5 a fla% at all4 ,& ! same applies to less vivid exp]i5ces1 s long z !y are rec5t4 ,i am absolutely c]ta9 t half a m9ute ago ,i 0 sitt+ 9 ! same *air 9 :i* ,i am sitt+ n[4 ,go+ backw>d ov] ! day1 ,i f9d ?+s ( :i* ,i am q c]ta91 o!r ?+s ( :i* ,i am almo/ c]ta91 o!r ?+s ( :i* ,i c become c]ta9 by ?\ght & by call+ up att5dant circum/ances1 & "s ?+s ( :i* ,i am by no means c]ta94 ,i am q c]ta9 t ,i ate my breakfa/ ?is morn+1 b if ,i 7 z indiff]5t to my breakfa/ z a philosoph] %\ld be1 ,i %\ld be d\btful4 ,z to ! conv]sation at breakfa/1 ,i c recall "s ( x easily1 "s ) an ef=t1 "s only ) a l>ge elem5t ( d\bt1 & "s n at all4 ,?us !re is a cont9ual gradation 9 ! degree ( self-evid5ce ( :at ,i rememb]1 & a correspond+ gradation 9 ! tru/wor?9ess ( my memory4 ,?us ! fir/ answ] to ! difficulty ( fallaci\s memory is to say t memory has degrees ( self-evid5ce1 & t ^! correspond to ! degrees ( its tru/wor?9ess1 rea*+ a limit ( p]fect self-evid5ce & p]fect tru/wor?9ess 9 our memory ( ev5ts :i* are rec5t & vivid4 ,x w\ld seem1 h[ev]1 t !re are cases ( v firm belief 9 a memory :i* is :olly false4 ,x is probable t1 9 ^! cases1 :at is really rememb]$1 9 ! s5se ( be+ imm$iately be=e ! m9d1 is some?+ o!r ?an :at is falsely believ$ 91 ?\gh some?+ g5]ally associat$ ) x4 ,george ,,iv is said to h at la/ believ$ t he 0 at ! battle ( ,wat]loo1 because he _h s (t5 said t he 04 ,9 ?is case1 :at 0 imm$iately rememb]$ 0 8 repeat$ ass]tion2 ! belief 9 :at he 0 ass]t+ 7if x exi/$7 w\ld be produc$ by association ) ! rememb]$ ass]tion1 & w\ld !re=e n be a g5u9e case ( memory4 ,x w\ld seem t cases ( fallaci\s memory c probably all be dealt ) 9 ?is way1 i4e4 !y c be %[n to be n cases ( memory 9 ! /rict s5se at all4 ,"o important po9t ab\t self-evid5ce is made cle> by ! case ( memory1 & t is1 t self-evid5ce has degrees3 x is n a quality :i* is simply pres5t or abs5t1 b a quality :i* may be m or less pres5t1 9 gradations rang+ f absolute c]ta9ty d[n to an almo/ imp]ceptible fa9tness4 ,tru?s ( p]ception & "s ( ! pr9ciples ( logic h ! v highe/ degree ( self-evid5ce2 tru?s ( imm$iate memory h an almo/ equally high degree4 ,! inductive pr9ciple has less self-evid5ce ?an "s ( ! o!r pr9ciples ( logic1 su* z ':at foll[s f a true premiss mu/ be true'4 ,memories h a dim9i%+ self-evid5ce z !y become remot] & fa9t]2 ! tru?s ( logic & ma!matics h 7broadly speak+7 less self-evid5ce z !y become m complicat$4 ,judgem5ts ( intr9sic e?ical or ae/hetic value are apt to h "s self-evid5ce1 b n mu*4 ,degrees ( self-evid5ce are important 9 ! !ory ( k1 s9ce1 if propositions may 7z seems likely7 h "s degree ( self-evid5ce )\t be+ true1 x w n be necess>y to ab&on all connexion betwe5 self-evid5ce & tru?1 b m]ely to say t1 :]e !re is a conflict1 ! m self-evid5t proposition is to be reta9$ & ! less self-evid5t reject$4 ,x seems1 h[ev]1 highly probable t two diff]5t notions are comb9$ 9 'self-evid5ce' z above expla9$2 t "o ( !m1 :i* corresponds to ! highe/ degree ( self-evid5ce1 is really an infallible gu>antee ( tru?1 :ile ! o!r1 :i* corresponds to all ! o!r degrees1 does n give an infallible gu>antee1 b only a great] or less presumption4 ,?is1 h[ev]1 is only a sugge/ion1 :i* we _c z yet develop fur!r4 ,aft] we h dealt ) ! nature ( tru?1 we %all return to ! subject ( self-evid5ce1 9 connexion ) ! di/9ction betwe5 k & error4 ,,*apt] ,,xii4 ,,tru? ,,& ,,falsehood ,our k ( tru?s1 unlike our k ( ?+s1 has an opposite1 namely error4 ,s f> z ?+s are conc]n$1 we may "k !m or n "k !m1 b !re is no positive /ate ( m9d :i* c be describ$ z errone\s k ( ?+s1 s long1 at any rate1 z we conf9e ourselves to k by acqua9tance4 ,:atev] we are acqua9t$ ) mu/ be some?+2 we may draw wrong inf]5ces f our acqua9tance1 b ! acqua9tance itself _c be deceptive4 ,?us !re is no dualism z reg>ds acqua9tance4 ,b z reg>ds k ( tru?s1 !re is a dualism4 ,we may believe :at is false z well z :at is true4 ,we "k t on v _m subjects diff]5t p hold diff]5t & incompatible op9ions3 h5ce "s beliefs mu/ be errone\s4 ,s9ce errone\s beliefs are (t5 held j z /rongly z true beliefs1 x becomes a difficult "q h[ !y are to be di/+ui%$ f true beliefs4 ,h[ are we to "k1 9 a giv5 case1 t our belief is n errone\s8 ,?is is a "q ( ! v greate/ difficulty1 to :i* no completely satisfactory answ] is possible4 ,!re is1 h[ev]1 a prelim9>y "q :i* is r less difficult1 & t is3 ,:at d we mean by tru? & falsehood8 ,x is ?is prelim9>y "q :i* is to be consid]$ 9 ?is *apt]4 ,9 ?is *apt] we are n ask+ h[ we c "k :e!r a belief is true or false3 we are ask+ :at is meant by ! "q :e!r a belief is true or false4 ,x is to be hop$ t a cle> answ] to ?is "q may help u to obta9 an answ] to ! "q :at beliefs are true1 b = ! pres5t we ask only ',:at is tru?8' & ',:at is falsehood8' n ',:at beliefs are true8' & ',:at beliefs are false8' ,x is v important to keep ^! diff]5t que/ions entirely sep>ate1 s9ce any confusion betwe5 !m is sure to produce an answ] :i* is n really applicable to ei!r4 ,!re are ?ree po9ts to obs]ve 9 ! attempt to discov] ! nature ( tru?1 ?ree requisites :i* any !ory mu/ fulfil4 7#a7 ,our !ory ( tru? mu/ be su* z to admit ( its opposite1 falsehood4 ,a good _m philosoph]s h fail$ adequately to satisfy ?is condition3 !y h con/ruct$ !ories accord+ to :i* all our ?9k+ ought to h be5 true1 & h !n _h ! greate/ difficulty 9 f9d+ a place = falsehood4 ,9 ?is respect our !ory ( belief mu/ diff] f our !ory ( acqua9tance1 s9ce 9 ! case ( acqua9tance x 0 n necess>y to take acc\nt ( any opposite4 7#b7 ,x seems fairly evid5t t if !re 7 no beliefs !re c\ld be no falsehood1 & no tru? ei!r1 9 ! s5se 9 :i* tru? is correlative to falsehood4 ,if we imag9e a _w ( m]e matt]1 !re w\ld be no room = falsehood 9 su* a _w1 & al?\gh x w\ld conta9 :at may be call$ 'facts'1 x w\ld n conta9 any tru?s1 9 ! s5se 9 :i* tru?s are ?+s ( ! same k9d z falsehoods4 ,9 fact1 tru? & falsehood are prop]ties ( beliefs & /atem5ts3 h5ce a _w ( m]e matt]1 s9ce x w\ld conta9 no beliefs or /atem5ts1 w\ld also conta9 no tru? or falsehood4 7#c7 ,b1 z aga9/ :at we h j said1 x is to be obs]v$ t ! tru? or falsehood ( a belief always dep5ds ^u some?+ :i* lies outside ! belief itself4 ,if ,i believe t ,*>les ,i di$ on ! scaffold1 ,i believe truly1 n because ( any intr9sic quality ( my belief1 :i* c\ld be discov]$ by m]ely exam9+ ! belief1 b because ( an hi/orical ev5t :i* happ5$ two & a half c5turies ago4 ,if ,i believe t ,*>les ,i di$ 9 8 b$1 ,i believe falsely3 no degree ( vividness 9 my belief1 or ( c>e 9 arriv+ at x1 prev5ts x f be+ false1 aga9 because ( :at happ5$ long ago1 & n because ( any intr9sic prop]ty ( my belief4 ,h5ce1 al?\gh tru? & falsehood are prop]ties ( beliefs1 !y are prop]ties dep5d5t ^u ! relations ( ! beliefs to o!r ?+s1 n ^u any int]nal quality ( ! beliefs4 ,! ?ird ( ! above requisites leads u to adopt ! view--:i* has on ! :ole be5 commone/ among philosoph]s--t tru? consi/s 9 "s =m ( correspond5ce betwe5 belief & fact4 ,x is1 h[ev]1 by no means an easy matt] to discov] a =m ( correspond5ce to :i* !re are no irrefutable objections4 ,by ?is p>tly--& p>tly by ! feel+ t1 if tru? consi/s 9 a correspond5ce ( ?\ght ) some?+ outside ?\ght1 ?\ght c nev] "k :5 tru? has be5 atta9$--_m philosoph]s h be5 l$ to try to f9d "s def9ition ( tru? :i* %all n consi/ 9 relation to some?+ :olly outside belief4 ,! mo/ important attempt at a def9ition ( ?is sort is ! !ory t tru? consi/s 9 coh]5ce4 ,x is said t ! m>k ( falsehood is failure to coh]e 9 ! body ( our beliefs1 & t x is ! ess5ce ( a tru? to =m "p ( ! completely r\nd$ sy/em :i* is ,! ,tru?4 ,!re is1 h[ev]1 a great difficulty 9 ?is view1 or r two great difficulties4 ,! fir/ is t !re is no reason to suppose t only "o coh]5t body ( beliefs is possible4 ,x may be t1 ) suffici5t imag9ation1 a noveli/ might inv5t a pa/ = ! _w t w\ld p]fectly fit on to :at we "k1 & yet be q diff]5t f ! real pa/4 ,9 m sci5tific matt]s1 x is c]ta9 t !re are (t5 two or m hypo!ses :i* acc\nt = all ! kn[n facts on "s subject1 & al?\gh1 9 su* cases1 m5 ( sci5ce endeav\r to f9d facts :i* w rule out all ! hypo!ses except "o1 !re is no reason :y !y %\ld always succe$4 ,9 philosophy1 aga91 x seems n uncommon = two rival hypo!ses to be bo? able to acc\nt = all ! facts4 ,?us1 = example1 x is possible t life is "o long dream1 & t ! out] _w has only t degree ( reality t ! objects ( dreams h2 b al?\gh su* a view does n seem inconsi/5t ) kn[n facts1 !re is no reason to pref] x to ! common-s5se view1 accord+ to :i* o!r p & ?+s d really exi/4 ,?us coh]5ce z ! def9ition ( tru? fails because !re is no pro( t !re c be only "o coh]5t sy/em4 ,! o!r objection to ?is def9ition ( tru? is t x assumes ! mean+ ( 'coh]5ce' kn[n1 :]eas1 9 fact1 'coh]5ce' presupposes ! tru? ( ! laws ( logic4 ,two propositions are coh]5t :5 bo? may be true1 & are incoh]5t :5 "o at lea/ mu/ be false4 ,n[ 9 ord] to "k :e!r two propositions c bo? be true1 we mu/ "k su* tru?s z ! law ( contradiction4 ,= example1 ! two propositions1 '?is tree is a bee*' & '?is tree is n a bee*'1 are n coh]5t1 because ( ! law ( contradiction4 ,b if ! law ( contradiction itself 7 subject$ to ! te/ ( coh]5ce1 we %\ld f9d t1 if we *oose to suppose x false1 no?+ w any long] be incoh]5t ) any?+ else4 ,?us ! laws ( logic supply ! skeleton or framework )9 :i* ! te/ ( coh]5ce applies1 & !y !mselves _c be e/abli%$ by ?is te/4 ,= ! above two reasons1 coh]5ce _c be accept$ z giv+ ! mean+ ( tru?1 ?\gh x is (t5 a mo/ important te/ ( tru? aft] a c]ta9 am\nt ( tru? has become kn[n4 ,h5ce we are driv5 back to correspond5ce ) fact z con/itut+ ! nature ( tru?4 ,x rema9s to def9e precisely :at we mean by 'fact'1 & :at is ! nature ( ! correspond5ce :i* mu/ subsi/ betwe5 belief & fact1 9 ord] t belief may be true4 ,9 accordance ) our ?ree requisites1 we h to seek a !ory ( tru? :i* 7#a7 all[s tru? to h an opposite1 namely falsehood1 7#b7 makes tru? a prop]ty ( beliefs1 b 7#c7 makes x a prop]ty :olly dep5d5t ^u ! relation ( ! beliefs to outside ?+s4 ,! necessity ( all[+ = falsehood makes x impossible to reg>d belief z a relation ( ! m9d to a s+le object1 :i* c\ld be said to be :at is believ$4 ,if belief 7 s reg>d$1 we %\ld f9d t1 l acqua9tance1 x w\ld n admit ( ! opposition ( tru? & falsehood1 b w\ld h to be always true4 ,?is may be made cle> by examples4 ,o!llo believes falsely t ,desdemona loves ,cassio4 ,we _c say t ?is belief consi/s 9 a relation to a s+le object1 ',desdemona's love = ,cassio'1 = if !re 7 su* an object1 ! belief w\ld be true4 ,!re is 9 fact no su* object1 & !re=e ,o!llo _c h any relation to su* an object4 ,h5ce 8 belief _c possibly consi/ 9 a relation to ?is object4 ,x might be said t 8 belief is a relation to a diff]5t object1 namely 't ,desdemona loves ,cassio'2 b x is almo/ z difficult to suppose t !re is su* an object z ?is1 :5 ,desdemona does n love ,cassio1 z x 0 to suppose t !re is ',desdemona's love = ,cassio'4 ,h5ce x w be bett] to seek = a !ory ( belief :i* does n make x consi/ 9 a relation ( ! m9d to a s+le object4 ,x is common to ?9k ( relations z ?\gh !y always held betwe5 two t]ms1 b 9 fact ?is is n always ! case4 ,"s relations dem& ?ree t]ms1 "s f\r1 & s on4 ,take1 = in/ance1 ! relation 'betwe5'4 ,s long z only two t]ms come 91 ! relation 'betwe5' is impossible3 ?ree t]ms are ! smalle/ numb] t r5d] x possible4 ,york is betwe5 ,london & ,ed9burgh2 b if ,london & ,ed9burgh 7 ! only places 9 ! _w1 !re c\ld be no?+ :i* 0 betwe5 "o place & ano!r4 ,simil>ly jeal\sy requires ?ree p3 !re c be no su* relation t does n involve ?ree at lea/4 ,su* a proposition z ',a wi%es ,b to promote ,c's m>riage ) ,d' involves a relation ( f\r t]ms2 t is to say1 ,a & ,b & ,c & ,d all come 91 & ! relation involv$ _c be express$ o!rwise ?an 9 a =m involv+ all f\r4 ,in/ances might be multipli$ indef9itely1 b en\gh has be5 said to %[ t !re are relations :i* require m ?an two t]ms be=e !y c occur4 ,! relation involv$ 9 judg+ or believ+ mu/1 if falsehood is to be duly all[$ =1 be tak5 to be a relation betwe5 sev]al t]ms1 n betwe5 two4 ,:5 ,o!llo believes t ,desdemona loves ,cassio1 he mu/ n h be=e 8 m9d a s+le object1 ',desdemona's love = ,cassio'1 or 't ,desdemona loves ,cassio '1 = t w\ld require t !re %\ld be objective falsehoods1 :i* subsi/ indep5d5tly ( any m9ds2 & ?is1 ?\gh n logically refutable1 is a !ory to be avoid$ if possible4 ,?us x is easi] to acc\nt = falsehood if we take judgem5t to be a relation 9 :i* ! m9d & ! v>i\s objects conc]n$ all occur sev]ally2 t is to say1 ,desdemona & lov+ & ,cassio mu/ all be t]ms 9 ! relation :i* subsi/s :5 ,o!llo believes t ,desdemona loves ,cassio4 ,?is relation1 !re=e1 is a relation ( f\r t]ms1 s9ce ,o!llo also is "o ( ! t]ms ( ! relation4 ,:5 we say t x is a relation ( f\r t]ms1 we d n mean t ,o!llo has a c]ta9 relation to ,desdemona1 & has ! same relation to lov+ & also to ,cassio4 ,?is may be true ( "s o!r relation ?an believ+2 b believ+1 pla9ly1 is n a relation :i* ,o!llo has to ea* ( ! ?ree t]ms conc]n$1 b to all ( !m toge!r3 !re is only "o example ( ! relation ( believ+ involv$1 b ?is "o example knits toge!r f\r t]ms4 ,?us ! actual occurr5ce1 at ! mom5t :5 ,o!llo is ent]ta9+ 8 belief1 is t ! relation call$ 'believ+' is knitt+ toge!r into "o complex :ole ! f\r t]ms ,o!llo1 ,desdemona1 lov+1 & ,cassio4 ,:at is call$ belief or judgem5t is no?+ b ?is relation ( believ+ or judg+1 :i* relates a m9d to sev]al ?+s o!r ?an itself4 ,an act ( belief or ( judgem5t is ! occurr5ce betwe5 c]ta9 t]ms at "s p>ticul> "t1 ( ! relation ( believ+ or judg+4 ,we are n[ 9 a position to und]/& :at x is t di/+ui%es a true judgem5t f a false "o4 ,= ?is purpose we w adopt c]ta9 def9itions4 ,9 e act ( judgem5t !re is a m9d :i* judges1 & !re are t]ms conc]n+ :i* x judges4 ,we w call ! m9d ! subject 9 ! judgem5t1 & ! rema9+ t]ms ! objects4 ,?us1 :5 ,o!llo judges t ,desdemona loves ,cassio1 ,o!llo is ! subject1 :ile ! objects are ,desdemona & lov+ & ,cassio4 ,! subject & ! objects toge!r are call$ ! con/itu5ts ( ! judgem5t4 ,x w be obs]v$ t ! relation ( judg+ has :at is call$ a 's5se' or 'direction'4 ,we may say1 metaphorically1 t x puts its objects 9 a c]ta9 ord]1 :i* we may indicate by means ( ! ord] ( ! words 9 ! s5t5ce4 7,9 an inflect$ language1 ! same ?+ w be indicat$ by inflections1 e4g4 by ! diff]5ce betwe5 nom9ative & accusative47 ,o!llo's judgem5t t ,cassio loves ,desdemona diff]s f 8 judgem5t t ,desdemona loves ,cassio1 9 spite ( ! fact t x consi/s ( ! same con/itu5ts1 because ! relation ( judg+ places ! con/itu5ts 9 a diff]5t ord] 9 ! two cases4 ,simil>ly1 if ,cassio judges t ,desdemona loves ,o!llo1 ! con/itu5ts ( ! judgem5t are /ill ! same1 b _! ord] is diff]5t4 ,?is prop]ty ( hav+ a 's5se' or 'direction' is "o :i* ! relation ( judg+ %>es ) all o!r relations4 ,! 's5se' ( relations is ! ultimate s\rce ( ord] & s]ies & a ho/ ( ma!matical concepts2 b we ne$ n conc]n ourselves fur!r ) ?is aspect4 ,we spoke ( ! relation call$ 'judg+' or 'believ+' z knitt+ toge!r into "o complex :ole ! subject & ! objects4 ,9 ?is respect1 judg+ is exactly l e o!r relation4 ,:5ev] a relation holds betwe5 two or m t]ms1 x unites ! t]ms into a complex :ole4 ,if ,o!llo loves ,desdemona1 !re is su* a complex :ole z ',o!llo's love = ,desdemona'4 ,! t]ms unit$ by ! relation may be !mselves complex1 or may be simple1 b ! :ole :i* results f _! be+ unit$ mu/ be complex4 ,:]ev] !re is a relation :i* relates c]ta9 t]ms1 !re is a complex object =m$ ( ! union ( ^? t]ms2 & conv]sely1 :]ev] !re is a complex object1 !re is a relation :i* relates its con/itu5ts4 ,:5 an act ( believ+ occurs1 !re is a complex1 9 :i* 'believ+' is ! unit+ relation1 & subject & objects are arrang$ 9 a c]ta9 ord] by ! 's5se' ( ! relation ( believ+4 ,among ! objects1 z we saw 9 consid]+ ',o!llo believes t ,desdemona loves ,cassio'1 "o mu/ be a relation--9 ?is in/ance1 ! relation 'lov+'4 ,b ?is relation1 z x occurs 9 ! act ( believ+1 is n ! relation :i* creates ! unity ( ! complex :ole consi/+ ( ! subject & ! objects4 ,! relation 'lov+'1 z x occurs 9 ! act ( believ+1 is "o ( ! objects--x is a brick 9 ! /ructure1 n ! cem5t4 ,! cem5t is ! relation 'believ+'4 ,:5 ! belief is true1 !re is ano!r complex unity1 9 :i* ! relation :i* 0 "o ( ! objects ( ! belief relates ! o!r objects4 ,?us1 e4g41 if ,o!llo believes truly t ,desdemona loves ,cassio1 !n !re is a complex unity1 ',desdemona's love = ,cassio'1 :i* is compos$ exclusively ( ! objects ( ! belief1 9 ! same ord] z !y _h 9 ! belief1 ) ! relation :i* 0 "o ( ! objects occurr+ n[ z ! cem5t t b9ds toge!r ! o!r objects ( ! belief4 ,on ! o!r h&1 :5 a belief is false1 !re is no su* complex unity compos$ only ( ! objects ( ! belief4 ,if ,o!llo believes falsely t ,desdemona loves ,cassio1 !n !re is no su* complex unity z ',desdemona's love = ,cassio'4 ,?us a belief is true :5 x corresponds to a c]ta9 associat$ complex1 & false :5 x does n4 ,assum+1 = ! sake ( def9it5ess1 t ! objects ( ! belief are two t]ms & a relation1 ! t]ms be+ put 9 a c]ta9 ord] by ! 's5se' ( ! believ+1 !n if ! two t]ms 9 t ord] are unit$ by ! relation into a complex1 ! belief is true2 if n1 x is false4 ,?is con/itutes ! def9ition ( tru? & falsehood t we 7 9 se>* (4 ,judg+ or believ+ is a c]ta9 complex unity ( :i* a m9d is a con/itu5t2 if ! rema9+ con/itu5ts1 tak5 9 ! ord] :i* !y h 9 ! belief1 =m a complex unity1 !n ! belief is true2 if n1 x is false4 ,?us al?\gh tru? & falsehood are prop]ties ( beliefs1 yet !y are 9 a s5se extr9sic prop]ties1 = ! condition ( ! tru? ( a belief is some?+ n involv+ beliefs1 or 79 g5]al7 any m9d at all1 b only ! objects ( ! belief4 ,a m9d1 :i* believes1 believes truly :5 !re is a correspond+ complex n involv+ ! m9d1 b only its objects4 ,?is correspond5ce ensures tru?1 & its abs5ce entails falsehood4 ,h5ce we acc\nt simultane\sly = ! two facts t beliefs 7a7 dep5d on m9ds = _! exi/5ce1 7b7 d n dep5d on m9ds = _! tru?4 ,we may re/ate our !ory z foll[s3 ,if we take su* a belief z ',o!llo believes t ,desdemona loves ,cassio'1 we w call ,desdemona & ,cassio ! object-t]ms1 & lov+ ! object-relation4 ,if !re is a complex unity ',desdemona's love = ,cassio'1 consi/+ ( ! object-t]ms relat$ by ! object-relation 9 ! same ord] z !y h 9 ! belief1 !n ?is complex unity is call$ ! fact correspond+ to ! belief4 ,?us a belief is true :5 !re is a correspond+ fact1 & is false :5 !re is no correspond+ fact4 ,x w be se5 t m9ds d n create tru? or falsehood4 ,!y create beliefs1 b :5 once ! beliefs are creat$1 ! m9d _c make !m true or false1 except 9 ! special case :]e !y conc]n future ?+s :i* are )9 ! p[] ( ! p]son believ+1 su* z cat*+ tra9s4 ,:at makes a belief true is a fact1 & ?is fact does n 7except 9 exceptional cases7 9 any way involve ! m9d ( ! p]son :o has ! belief4 ,hav+ n[ decid$ :at we mean by tru? & falsehood1 we h next to consid] :at ways !re are ( kn[+ :e!r ?is or t belief is true or false4 ,?is consid]ation w occupy ! next *apt]4 ,,*apt] ,,xiii4 ,,k1 ,,error1 ,,& ,,probable ,,op9ion ,! "q z to :at we mean by tru? & falsehood1 :i* we consid]$ 9 ! prec$+ *apt]1 is ( mu* less int]e/ ?an ! "q z to h[ we c "k :at is true & :at is false4 ,?is "q w occupy u 9 ! pres5t *apt]4 ,!re c be no d\bt t "s ( our beliefs are errone\s2 ?us we are l$ to inquire :at c]ta9ty we c ev] h t su* & su* a belief is n errone\s4 ,9 o!r words1 c we ev] "k any?+ at all1 or d we m]ely sometimes by good luck believe :at is true8 ,be=e we c attack ?is "q1 we mu/1 h[ev]1 fir/ decide :at we mean by 'kn[+'1 & ?is "q is n s easy z might be suppos$4 ,at fir/ sight we might imag9e t k c\ld be def9$ z 'true belief'4 ,:5 :at we believe is true1 x might be suppos$ t we _h a*iev$ a k ( :at we believe4 ,b ?is w\ld n accord ) ! way 9 :i* ! ^w is commonly us$4 ,to take a v trivial in/ance3 ,if a man believes t ! late ,prime ,m9i/]'s la/ "n began ) a ,b1 he believes :at is true1 s9ce ! late ,prime ,m9i/] 0 ,sir ,h5ry ,campbell ,bann]man4 ,b if he believes t ,mr4 ,balf\r 0 ! late ,prime ,m9i/]1 he w /ill believe t ! late ,prime ,m9i/]'s la/ "n began ) a ,b1 yet ?is belief1 ?\gh true1 w\ld n be ?\ght to con/itute k4 ,if a newspap]1 by an intellig5t anticipation1 ann\nces ! result ( a battle be=e any telegram giv+ ! result has be5 receiv$1 x may by good =tune ann\nce :at aft]w>ds turns out to be ! "r result1 & x may produce belief 9 "s ( its less exp]i5c$ read]s4 ,b 9 spite ( ! tru? ( _! belief1 !y _c be said to h k4 ,?us x is cle> t a true belief is n k :5 x is d$uc$ f a false belief4 ,9 l mann]1 a true belief _c be call$ k :5 x is d$uc$ by a fallaci\s process ( reason+1 ev5 if ! premisses f :i* x is d$uc$ are true4 ,if ,i "k t all ,greeks are m5 & t ,socrates 0 a man1 & ,i inf] t ,socrates 0 a ,greek1 ,i _c be said to "k t ,socrates 0 a ,greek1 because1 al?\gh my premisses & my conclusion are true1 ! conclusion does n foll[ f ! premisses4 ,b are we to say t no?+ is k except :at is validly d$uc$ f true premisses8 ,obvi\sly we _c say ?is4 ,su* a def9ition is at once too wide & too n>r[4 ,9 ! fir/ place1 x is too wide1 because x is n en\gh t our premisses %\ld be true1 !y mu/ also be kn[n4 ,! man :o believes t ,mr4 ,balf\r 0 ! late ,prime ,m9i/] may proce$ to draw valid d$uctions f ! true premiss t ! late ,prime ,m9i/]'s "n began ) a ,b1 b he _c be said to "k ! conclusions rea*$ by ^! d$uctions4 ,?us we %all h to am5d our def9ition by say+ t k is :at is validly d$uc$ f kn[n premisses4 ,?is1 h[ev]1 is a circul> def9ition3 x assumes t we already "k :at is meant by 'kn[n premisses'4 ,x c1 !re=e1 at be/ def9e "o sort ( k1 ! sort we call d]ivative1 z oppos$ to intuitive k4 ,we may say3 ',d]ivative k is :at is validly d$uc$ f premisses kn[n intuitively'4 ,9 ?is /atem5t !re is no =mal defect1 b x leaves ! def9ition ( intuitive k /ill to seek4 ,leav+ on "o side1 = ! mom5t1 ! "q ( intuitive k1 let u consid] ! above sugge/$ def9ition ( d]ivative k4 ,! *ief objection to x is t x unduly limits k4 ,x con/antly happ5s t p ent]ta9 a true belief1 :i* has gr[n up 9 !m because ( "s piece ( intuitive k f :i* x is capable ( be+ validly inf]r$1 b f :i* x has n1 z a matt] ( fact1 be5 inf]r$ by any logical process4 ,take1 = example1 ! beliefs produc$ by read+4 ,if ! newspap]s ann\nce ! dea? ( ! ,k+1 we are fairly well ju/ifi$ 9 believ+ t ! ,k+ is dead1 s9ce ?is is ! sort ( ann\ncem5t :i* w\ld n be made if x 7 false4 ,& we are q amply ju/ifi$ 9 believ+ t ! newspap] ass]ts t ! ,k+ is dead4 ,b "h ! intuitive k ^u :i* our belief is bas$ is k ( ! exi/5ce ( s5se-data d]iv$ f look+ at ! pr9t :i* gives ! news4 ,?is k sc>cely rises into consci\sness1 except 9 a p]son :o _c read easily4 ,a *ild may be aw>e ( ! %apes ( ! lett]s1 & pass gradually & pa9fully to a realization ( _! mean+4 ,b anybody accu/om$ to read+ passes at once to :at ! lett]s mean1 & is n aw>e1 except on reflection1 t he has d]iv$ ?is k f ! s5se-data call$ see+ ! pr9t$ lett]s4 ,?us al?\gh a valid inf]5ce f !-lett]s to _! mean+ is possible1 & c\ld be p]=m$ by ! read]1 x is n 9 fact p]=m$1 s9ce he does n 9 fact p]=m any op]ation :i* c be call$ logical inf]5ce4 ,yet x w\ld be absurd to say t ! read] does n "k t ! newspap] ann\nces ! ,k+'s dea?4 ,we mu/1 !re=e1 admit z d]ivative k :atev] is ! result ( intuitive k ev5 if by m]e association1 provid$ !re is a valid logical connexion1 & ! p]son 9 "q c\ld become aw>e ( ?is connexion by reflection4 ,!re are 9 fact _m ways1 besides logical inf]5ce1 by :i* we pass f "o belief to ano!r3 ! passage f ! pr9t to its mean+ illu/rates ^! ways4 ,^! ways may be call$ 'psy*ological inf]5ce'4 ,we %all1 !n1 admit su* psy*ological inf]5ce z a means ( obta9+ d]ivative k1 provid$ !re is a discov]able logical inf]5ce :i* runs p>allel to ! psy*ological inf]5ce4 ,?is r5d]s our def9ition ( d]ivative k less precise ?an we c\ld wi%1 s9ce ! ^w 'discov]able' is vague3 x does n tell u h[ mu* reflection may be ne$$ 9 ord] to make ! discov]y4 ,b 9 fact 'k' is n a precise conception3 x m]ges into 'probable op9ion'1 z we %all see m fully 9 ! c\rse ( ! pres5t *apt]4 ,a v precise def9ition1 !re=e1 %\ld n be s\ght1 s9ce any su* def9ition mu/ be m or less mislead+4 ,! *ief difficulty 9 reg>d to k1 h[ev]1 does n arise ov] d]ivative k1 b ov] intuitive k4 ,s long z we are deal+ ) d]ivative k1 we h ! te/ ( intuitive k to fall back ^u4 ,b 9 reg>d to intuitive beliefs1 x is by no means easy to discov] any crit]ion by :i* to di/+ui% "s z true & o!rs z errone\s4 ,9 ?is "q x is sc>cely possible to rea* any v precise result3 all our k ( tru?s is infect$ ) "s degree ( d\bt1 & a !ory :i* ignor$ ?is fact w\ld be pla9ly wrong4 ,some?+ may be done1 h[ev]1 to mitigate ! difficulties ( ! "q4 ,our !ory ( tru?1 to beg9 )1 supplies ! possibility ( di/+ui%+ c]ta9 tru?s z self-evid5t 9 a s5se :i* ensures infallibility4 ,:5 a belief is true1 we said1 !re is a correspond+ fact1 9 :i* ! sev]al objects ( ! belief =m a s+le complex4 ,! belief is said to con/itute k ( ?is fact1 provid$ x fulfils ^? fur!r some:at vague conditions :i* we h be5 consid]+ 9 ! pres5t *apt]4 ,b 9 reg>d to any fact1 besides ! k con/itut$ by belief1 we may also h ! k9d ( k con/itut$ by p]ception 7tak+ ?is ^w 9 its wide/ possible s5se74 ,= example1 if y "k ! h\r ( ! sunset1 y c at t h\r "k ! fact t ! sun is sett+3 ?is is k ( ! fact by way ( k ( tru?s2 b y c also1 if ! wea!r is f9e1 look to ! we/ & actually see ! sett+ sun3 y !n "k ! same fact by ! way ( k ( ?+s4 ,?us 9 reg>d to any complex fact1 !re are1 !oretically1 two ways 9 :i* x may be kn[n3 7#a7 by means ( a judgem5t1 9 :i* its sev]al p>ts are judg$ to be relat$ z !y are 9 fact relat$2 7#b7 by means ( acqua9tance ) ! complex fact itself1 :i* may 79 a l>ge s5se7 be call$ p]ception1 ?\gh x is by no means conf9$ to objects ( ! s5ses4 ,n[ x w be obs]v$ t ! second way ( kn[+ a complex fact1 ! way ( acqua9tance1 is only possible :5 !re really is su* a fact1 :ile ! fir/ way1 l all judgem5t1 is liable to error4 ,! second way gives u ! complex :ole1 & is !re=e only possible :5 its p>ts d actually h t relation :i* makes !m comb9e to =m su* a complex4 ,! fir/ way1 on ! contr>y1 gives u ! p>ts & ! relation sev]ally1 & dem&s only ! reality ( ! p>ts & ! relation3 ! relation may n relate ^? p>ts 9 t way1 & yet ! judgem5t may occur4 ,x w be rememb]$ t at ! end ( ,*apt] ,,xi we sugge/$ t !re might be two k9ds ( self-evid5ce1 "o giv+ an absolute gu>antee ( tru?1 ! o!r only a p>tial gu>antee4 ,^! two k9ds c n[ be di/+ui%$4 ,we may say t a tru? is self-evid5t1 9 ! fir/ & mo/ absolute s5se1 :5 we h acqua9tance ) ! fact :i* corresponds to ! tru?4 ,:5 ,o!llo believes t ,desdemona loves ,cassio1 ! correspond+ fact1 if 8 belief 7 true1 w\ld be ',desdemona's love = ,cassio'4 ,?is w\ld be a fact ) :i* no "o c\ld h acqua9tance except ,desdemona2 h5ce 9 ! s5se ( self-evid5ce t we are consid]+1 ! tru? t ,desdemona loves ,cassio 7if x 7 a tru?7 c\ld only be self-evid5t to ,desdemona4 ,all m5tal facts1 & all facts conc]n+ s5se-data1 h ?is same privacy3 !re is only "o p]son to :om !y c be self-evid5t 9 our pres5t s5se1 s9ce !re is only "o p]son :o c be acqua9t$ ) ! m5tal ?+s or ! s5se-data conc]n$4 ,?us no fact ab\t any p>ticul> exi/+ ?+ c be self-evid5t to m ?an "o p]son4 ,on ! o!r h&1 facts ab\t univ]sals d n h ?is privacy4 ,_m m9ds may be acqua9t$ ) ! same univ]sals2 h5ce a relation betwe5 univ]sals may be kn[n by acqua9tance to _m diff]5t p4 ,9 all cases :]e we "k by acqua9tance a complex fact consi/+ ( c]ta9 t]ms 9 a c]ta9 relation1 we say t ! tru? t ^! t]ms are s relat$ has ! fir/ or absolute k9d ( self-evid5ce1 & 9 ^! cases ! judgem5t t ! t]ms are s relat$ mu/ be true4 ,?us ?is sort ( self-evid5ce is an absolute gu>antee ( tru?4 ,b al?\gh ?is sort ( self-evid5ce is an absolute gu>antee ( tru?1 x does n enable u to be absolutely c]ta91 9 ! case ( any giv5 judgem5t1 t ! judgem5t 9 "q is true4 ,suppose we fir/ p]ceive ! sun %9+1 :i* is a complex fact1 & !nce proce$ to make ! judgem5t '! sun is %9+'4 ,9 pass+ f ! p]ception to ! judgem5t1 x is necess>y to analyse ! giv5 complex fact3 we h to sep>ate out '! sun' & '%9+' z con/itu5ts ( ! fact4 ,9 ?is process x is possible to commit an error2 h5ce ev5 :]e a fact has ! fir/ or absolute k9d ( self-evid5ce1 a judgem5t believ$ to correspond to ! fact is n absolutely infallible1 because x may n really correspond to ! fact4 ,b if x does correspond 79 ! s5se expla9$ 9 ! prec$+ *apt]71 !n x mu/ be true4 ,! second sort ( self-evid5ce w be t :i* belongs to judgem5ts 9 ! fir/ in/ance1 & is n d]iv$ f direct p]ception ( a fact z a s+le complex :ole4 ,?is second k9d ( self-evid5ce w h degrees1 f ! v highe/ degree d[n to a b>e incl9ation 9 fav\r ( ! belief4 ,take1 = example1 ! case ( a horse trott+ away f u along a h>d road4 ,at fir/ our c]ta9ty t we he> ! ho(s is complete2 gradually1 if we li/5 int5tly1 !re comes a mom5t :5 we ?9k p]haps x 0 imag9ation or ! bl9d up/airs or our own he>tbeats2 at la/ we become d\btful :e!r !re 0 any noise at all2 !n we ?9k we no long] he> any?+1 & at la/ we "k we no long] he> any?+4 ,9 ?is process1 !re is a cont9ual gradation ( self-evid5ce1 f ! highe/ degree to ! lea/1 n 9 ! s5se-data !mselves1 b 9 ! judgem5ts bas$ on !m4 ,or aga93 ,suppose we are comp>+ two %ades ( col\r1 "o blue & "o gre54 ,we c be q sure !y are diff]5t %ades ( col\r2 b if ! gre5 col\r is gradually alt]$ to be m & m l ! blue1 becom+ fir/ a blue-gre51 !n a gre5y-blue1 !n blue1 !re w come a mom5t :5 we are d\btful :e!r we c see any diff]5ce1 & !n a mom5t :5 we "k t we _c see any diff]5ce4 ,! same ?+ happ5s 9 tun+ a musical in/rum5t1 or 9 any o!r case :]e !re is a cont9u\s gradation4 ,?us self-evid5ce ( ?is sort is a matt] ( degree2 & x seems pla9 t ! high] degrees are m to be tru/$ ?an ! l[] degrees4 ,9 d]ivative k our ultimate premisses mu/ h "s degree ( self-evid5ce1 & s mu/ _! connexion ) ! conclusions d$uc$ f !m4 ,take = example a piece ( reason+ 9 geometry4 ,x is n en\gh t ! axioms f :i* we />t %\ld be self-evid5t3 x is necess>y also t1 at ea* /ep 9 ! reason+1 ! connexion ( premiss & conclusion %\ld be self-evid5t4 ,9 difficult reason+1 ?is connexion has (t5 only a v small degree ( self-evid5ce2 h5ce errors ( reason+ are n improbable :]e ! difficulty is great4 ,f :at has be5 said x is evid5t t1 bo? z reg>ds intuitive k & z reg>ds d]ivative k1 if we assume t intuitive k is tru/wor?y 9 proportion to ! degree ( its self-evid5ce1 !re w be a gradation 9 tru/wor?9ess1 f ! exi/5ce ( notewor?y s5se-data & ! simpl] tru?s ( logic & ari?metic1 :i* may be tak5 z q c]ta91 d[n to judgem5ts :i* seem only j m probable ?an _! opposites4 ,:at we firmly believe1 if x is true1 is call$ k1 provid$ x is ei!r intuitive or inf]r$ 7logically or psy*ologically7 f intuitive k f :i* x foll[s logically4 ,:at we firmly believe1 if x is n true1 is call$ error4 ,:at we firmly believe1 if x is nei!r k nor error1 & also :at we believe hesitat+ly1 because x is1 or is d]iv$ f1 some?+ :i* has n ! highe/ degree ( self-evid5ce1 may be call$ probable op9ion4 ,?us ! great] "p ( :at w\ld commonly pass z k is m or less probable op9ion4 ,9 reg>d to probable op9ion1 we c d]ive great assi/ance f coh]5ce1 :i* we reject$ z ! def9ition ( tru?1 b may (t5 use z a crit]ion4 ,a body ( individually probable op9ions1 if !y are mutually coh]5t1 become m probable ?an any "o ( !m w\ld be individually4 ,x is 9 ?is way t _m sci5tific hypo!ses acquire _! probability4 ,!y fit into a coh]5t sy/em ( probable op9ions1 & ?us become m probable ?an !y w\ld be 9 isolation4 ,! same ?+ applies to g5]al philosophical hypo!ses4 ,(t5 9 a s+le case su* hypo!ses may seem highly d\btful1 :ile yet1 :5 we consid] ! ord] & coh]5ce :i* !y introduce into a mass ( probable op9ion1 !y become pretty ne>ly c]ta94 ,?is applies1 9 p>ticul>1 to su* matt]s z ! di/9ction betwe5 dreams & wak+ life4 ,if our dreams1 night aft] night1 7 z coh]5t "o ) ano!r z our days1 we %\ld h>dly "k :e!r to believe ! dreams or ! wak+ life4 ,z x is1 ! te/ ( coh]5ce condemns ! dreams & confirms ! wak+ life4 ,b ?is te/1 ?\gh x increases probability :]e x is successful1 nev] gives absolute c]ta9ty1 unless !re is c]ta9ty already at "s po9t 9 ! coh]5t sy/em4 ,?us ! m]e organization ( probable op9ion w nev]1 by itself1 trans=m x into indubitable k4 ,,*apt] ,,xiv4 ,,! ,,limits ,,( ,,philosophical ,,k ,9 all t we h said hi!rto conc]n+ philosophy1 we h sc>cely t\*$ on _m matt]s t occupy a great space 9 ! writ+s ( mo/ philosoph]s4 ,mo/ philosoph]s--or1 at any rate1 v _m--pr(ess to be able to prove1 by a priori metaphysical reason+1 su* ?+s z ! fundam5tal dogmas ( religion1 ! ess5tial rationality ( ! univ]se1 ! illusor9ess ( matt]1 ! unreality ( all evil1 & s on4 ,!re c be no d\bt t ! hope ( f9d+ reason to believe su* !ses z ^! has be5 ! *ief inspiration ( _m life-long /ud5ts ( philosophy4 ,?is hope1 ,i believe1 is va94 ,x w\ld seem t k conc]n+ ! univ]se z a :ole is n to be obta9$ by metaphysics1 & t ! propos$ pro(s t1 9 virtue ( ! laws ( logic su* & su* ?+s mu/ exi/ & su* & su* o!rs _c1 are n capable ( surviv+ a critical scrut9y4 ,9 ?is *apt] we %all briefly consid] ! k9d ( way 9 :i* su* reason+ is attempt$1 ) a view to discov]+ :e!r we c hope t x may be valid4 ,! great repres5tative1 9 mod]n times1 ( ! k9d ( view :i* we wi% to exam9e1 0 ,hegel 7#aggj-ahca74 ,hegel's philosophy is v difficult1 & comm5tators diff] z to ! true int]pretation ( x4 ,accord+ to ! int]pretation ,i %all adopt1 :i* is t ( _m1 if n mo/1 ( ! comm5tators & has ! m]it ( giv+ an int]e/+ & important type ( philosophy1 8 ma9 !sis is t ev]y?+ %ort ( ! ,:ole is obvi\sly fragm5t>y1 & obvi\sly incapable ( exi/+ )\t ! complem5t suppli$ by ! re/ ( ! _w4 ,j z a comp>ative anatomi/1 f a s+le bone1 sees :at k9d ( animal ! :ole mu/ h be51 s ! metaphysician1 accord+ to ,hegel1 sees1 f any "o piece ( reality1 :at ! :ole ( reality mu/ be--at lea/ 9 its l>ge outl9es4 ,e app>5tly sep>ate piece ( reality has1 z x 71 hooks :i* grapple x to ! next piece2 ! next piece1 9 turn1 has fre% hooks1 & s on1 until ! :ole univ]se is recon/ruct$4 ,?is ess5tial incomplet5ess appe>s1 accord+ to ,hegel1 equally 9 ! _w ( ?\ght & 9 ! _w ( ?+s4 ,9 ! _w ( ?\ght1 if we take any idea :i* is ab/ract or incomplete1 we f9d1 on exam9ation1 t if we =get its incomplet5ess1 we become involv$ 9 contradictions2 ^! contradictions turn ! idea 9 "q into its opposite1 or anti!sis2 & 9 ord] to escape1 we h to f9d a new1 less incomplete idea1 :i* is ! syn!sis ( our orig9al idea & its anti!sis4 ,?is new idea1 ?\gh less incomplete ?an ! idea we />t$ )1 w be f\nd1 nev]!less1 to be /ill n :olly complete1 b to pass into its anti!sis1 ) :i* x mu/ be comb9$ 9 a new syn!sis4 ,9 ?is way ,hegel advances until he rea*es ! ',absolute ,idea'1 :i*1 accord+ to him1 has no incomplet5ess1 no opposite1 & no ne$ ( fur!r developm5t4 ,! ,absolute ,idea1 !re=e1 is adequate to describe ,absolute ,reality2 b all l[] ideas only describe reality z x appe>s to a p>tial view1 n z x is to "o :o simultane\sly surveys ! ,:ole4 ,?us ,hegel rea*es ! conclusion t ,absolute ,reality =ms "o s+le h>moni\s sy/em1 n 9 space or "t1 n 9 any degree evil1 :olly rational1 & :olly spiritual4 ,any appe>ance to ! contr>y1 9 ! _w we "k1 c be prov$ logically--s he believes--to be entirely due to our fragm5t>y piecemeal view ( ! univ]se4 ,if we saw ! univ]se :ole1 z we may suppose ,god sees x1 space & "t & matt] & evil & all /riv+ & /ruggl+ w\ld disappe>1 & we %\ld see in/ead an et]nal p]fect un*ang+ spiritual unity4 ,9 ?is conception1 !re is und5iably some?+ sublime1 some?+ to :i* we c\ld wi% to yield ass5t4 ,nev]!less1 :5 ! argum5ts 9 support ( x are c>efully exam9$1 !y appe> to involve mu* confusion & _m unw>rantable assumptions4 ,! fundam5tal t5et ^u :i* ! sy/em is built up is t :at is incomplete mu/ be n self-subsi/5t1 b mu/ ne$ ! support ( o!r ?+s be=e x c exi/4 ,x is held t :atev] has relations to ?+s outside itself mu/ conta9 "s ref]5ce to ^? outside ?+s 9 its own nature1 & c\ld n1 !re=e1 be :at x is if ^? outside ?+s did n exi/4 ,a man's nature1 = example1 is con/itut$ by 8 memories & ! re/ ( 8 k1 by 8 loves & hatr$s1 & s on2 ?us1 b = ! objects :i* he kn[s or loves or hates1 he c\ld n be :at he is4 ,he is ess5tially & obvi\sly a fragm5t3 tak5 z ! sum-total ( reality he w\ld be self-contradictory4 ,?is :ole po9t ( view1 h[ev]1 turns ^u ! notion ( ! 'nature' ( a ?+1 :i* seems to mean 'all ! tru?s ab\t ! ?+'4 ,x is ( c\rse ! case t a tru? :i* connects "o ?+ ) ano!r ?+ c\ld n subsi/ if ! o!r ?+ did n subsi/4 ,b a tru? ab\t a ?+ is n "p ( ! ?+ itself1 al?\gh x mu/1 accord+ to ! above usage1 be "p ( ! 'nature' ( ! ?+4 ,if we mean by a ?+'s 'nature' all ! tru?s ab\t ! ?+1 !n pla9ly we _c "k a ?+'s 'nature' unless we "k all ! ?+'s relations to all ! o!r ?+s 9 ! univ]se4 ,b if ! ^w 'nature' is us$ 9 ?is s5se1 we %all h to hold t ! ?+ may be kn[n :5 its 'nature' is n kn[n1 or at any rate is n kn[n completely4 ,!re is a confusion1 :5 ?is use ( ! ^w 'nature' is employ$1 betwe5 k ( ?+s & k ( tru?s4 ,we may h k ( a ?+ by acqua9tance ev5 if we "k v few propositions ab\t x--!oretically we ne$ n "k any propositions ab\t x4 ,?us1 acqua9tance ) a ?+ does n involve k ( its 'nature' 9 ! above s5se4 ,& al?\gh acqua9tance ) a ?+ is involv$ 9 our kn[+ any "o proposition ab\t a ?+1 k ( its 'nature'1 9 ! above s5se1 is n involv$4 ,h5ce1 7#a7 acqua9tance ) a ?+ does n logically involve a k ( its relations1 & 7#b7 a k ( "s ( its relations does n involve a k ( all ( its relations nor a k ( its 'nature' 9 ! above s5se4 ,i may be acqua9t$1 = example1 ) my too?a*e1 & ?is k may be z complete z k by acqua9tance ev] c be1 )\t kn[+ all t ! d5ti/ 7:o is n acqua9t$ ) x7 c tell me ab\t its cause1 & )\t !re=e kn[+ its 'nature' 9 ! above s5se4 ,?us ! fact t a ?+ has relations does n prove t its relations are logically necess>y4 ,t is to say1 f ! m]e fact t x is ! ?+ x is we _c d$uce t x mu/ h ! v>i\s relations :i* 9 fact x has4 ,?is only seems to foll[ because we "k x already4 ,x foll[s t we _c prove t ! univ]se z a :ole =ms a s+le h>moni\s sy/em su* z ,hegel believes t x =ms4 ,& if we _c prove ?is1 we also _c prove ! unreality ( space & "t & matt] & evil1 = ?is is d$uc$ by ,hegel f ! fragm5t>y & relational "* ( ^! ?+s4 ,?us we are left to ! piecemeal inve/igation ( ! _w1 & are unable to "k ! *>act]s ( ^? p>ts ( ! univ]se t are remote f our exp]i5ce4 ,?is result1 disappo9t+ z x is to ^? ^: hopes h be5 rais$ by ! sy/ems ( philosoph]s1 is 9 h>mony ) ! inductive & sci5tific temp] ( our age1 & is borne out by ! :ole exam9ation ( human k :i* has occupi$ our previ\s *apt]s4 ,mo/ ( ! great ambiti\s attempts ( metaphysicians h proce$$ by ! attempt to prove t su* & su* app>5t features ( ! actual _w 7 self-contradictory1 & !re=e c\ld n be real4 ,! :ole t5d5cy ( mod]n ?\ght1 h[ev]1 is m & m 9 ! direction ( %[+ t ! suppos$ contradictions 7 illusory1 & t v little c be prov$ a priori f consid]ations ( :at mu/ be4 ,a good illu/ration ( ?is is af=d$ by space & "t4 ,space & "t appe> to be inf9ite 9 ext5t1 & inf9itely divisible4 ,if we travel along a /raight l9e 9 ei!r direction1 x is difficult to believe t we %all f9ally rea* a la/ po9t1 beyond :i* !re is no?+1 n ev5 empty space4 ,simil>ly1 if 9 imag9ation we travel backw>ds or =w>ds 9 "t1 x is difficult to believe t we %all rea* a fir/ or la/ "t1 ) n ev5 empty "t beyond x4 ,?us space & "t appe> to be inf9ite 9 ext5t4 ,aga91 if we take any two po9ts on a l9e1 x seems evid5t t !re mu/ be o!r po9ts betwe5 !m h[ev] small ! di/ance betwe5 !m may be3 e di/ance c be halv$1 & ! halves c be halv$ aga91 & s on ad inf9itum4 ,9 "t1 simil>ly1 h[ev] little "t may elapse betwe5 two mom5ts1 x seems evid5t t !re w be o!r mom5ts betwe5 !m4 ,?us space & "t appe> to be inf9itely divisible4 ,b z aga9/ ^! app>5t facts--inf9ite ext5t & inf9ite divisibility--philosoph]s h advanc$ argum5ts t5d+ to %[ t !re c\ld be no inf9ite collections ( ?+s1 & t !re=e ! numb] ( po9ts 9 space1 or ( in/ants 9 "t1 mu/ be f9ite4 ,?us a contradiction em]g$ betwe5 ! app>5t nature ( space & "t & ! suppos$ impossibility ( inf9ite collections4 ,kant1 :o fir/ emphasiz$ ?is contradiction1 d$uc$ ! impossibility ( space & "t1 :i* he decl>$ to be m]ely subjective2 & s9ce 8 "t v _m philosoph]s h believ$ t space & "t are m]e appe>ance1 n *>act]i/ic ( ! _w z x really is4 ,n[1 h[ev]1 ow+ to ! lab\rs ( ! ma!maticians1 notably ,georg ,cantor1 x has appe>$ t ! impossibility ( inf9ite collections 0 a mi/ake4 ,!y are n 9 fact self-contradictory1 b only contradictory ( c]ta9 r ob/9ate m5tal prejudices4 ,h5ce ! reasons = reg>d+ space & "t z unreal h become inop]ative1 & "o ( ! great s\rces ( metaphysical con/ructions is dri$ up4 ,! ma!maticians1 h[ev]1 h n be5 cont5t ) %[+ t space z x is commonly suppos$ to be is possible2 !y h %[n also t _m o!r =ms ( space are equally possible1 s f> z logic c %[4 ,"s ( ,euclid's axioms1 :i* appe> to common s5se to be necess>y1 & 7 =m]ly suppos$ to be necess>y by philosoph]s1 are n[ kn[n to d]ive _! appe>ance ( necessity f our m]e famili>ity ) actual space1 & n f any a priori logical f\ndation4 ,by imag9+ worlds 9 :i* ^! axioms are false1 ! ma!maticians h us$ logic to loos5 ! prejudices ( common s5se1 & to %[ ! possibility ( spaces diff]+--"s m1 "s less--f t 9 :i* we live4 ,& "s ( ^! spaces diff] s little f ,euclidean space1 :]e di/ances su* z we c measure are conc]n$1 t x is impossible to discov] by obs]vation :e!r our actual space is /rictly ,euclidean or ( "o ( ^! o!r k9ds4 ,?us ! position is completely rev]s$4 ,=m]ly x appe>$ t exp]i5ce left only "o k9d ( space to logic1 & logic %[$ ?is "o k9d to be impossible4 ,n[1 logic pres5ts _m k9ds ( space z possible ap>t f exp]i5ce1 & exp]i5ce only p>tially decides betwe5 !m4 ,?us1 :ile our k ( :at is has become less ?an x 0 =m]ly suppos$ to be1 our k ( :at may be is enorm\sly increas$4 ,in/ead ( be+ %ut 9 )9 n>r[ walls1 ( :i* e nook & cranny c\ld be explor$1 we f9d ourselves 9 an op5 _w ( free possibilities1 :]e mu* rema9s unkn[n because !re is s mu* to "k4 ,:at has happ5$ 9 ! case ( space & "t has happ5$1 to "s ext5t1 9 o!r directions z well4 ,! attempt to prescribe to ! univ]se by means ( a priori pr9ciples has brok5 d[n2 logic1 in/ead ( be+1 z =m]ly1 ! b> to possibilities1 has become ! great lib]ator ( ! imag9ation1 pres5t+ innum]able alt]natives :i* are clos$ to unreflective common s5se1 & leav+ to exp]i5ce ! task ( decid+1 :]e decision is possible1 betwe5 ! _m worlds :i* logic (f]s = our *oice4 ,?us k z to :at exi/s becomes limit$ to :at we c le>n f exp]i5ce--n to :at we c actually exp]i5ce1 =1 z we h se51 !re is mu* k by description conc]n+ ?+s ( :i* we h no direct exp]i5ce4 ,b 9 all cases ( k by description1 we ne$ "s connexion ( univ]sals1 enabl+ u1 f su* & su* a datum1 to inf] an object ( a c]ta9 sort z impli$ by our datum4 ,?us 9 reg>d to physical objects1 = example1 ! pr9ciple t s5se-data are signs ( physical objects is itself a connexion ( univ]sals2 & x is only 9 virtue ( ?is pr9ciple t exp]i5ce enables u to acquire k conc]n+ physical objects4 ,! same applies to ! law ( causality1 or1 to desc5d to :at is less g5]al1 to su* pr9ciples z ! law ( gravitation4 ,pr9ciples su* z ! law ( gravitation are prov$1 or r are r5d]$ highly probable1 by a comb9ation ( exp]i5ce ) "s :olly a priori pr9ciple1 su* z ! pr9ciple ( induction4 ,?us our intuitive k1 :i* is ! s\rce ( all our o!r k ( tru?s1 is ( two sorts3 pure empirical k1 :i* tells u ( ! exi/5ce & "s ( ! prop]ties ( p>ticul> ?+s ) :i* we are acqua9t$1 & pure a priori k1 :i* gives u connexions betwe5 univ]sals1 & enables u to draw inf]5ces f ! p>ticul> facts giv5 9 empirical k4 ,our d]ivative k always dep5ds ^u "s pure a priori k & usually also dep5ds ^u "s pure empirical k4 ,philosophical k1 if :at has be5 said above is true1 does n diff] ess5tially f sci5tific k2 !re is no special s\rce ( wisdom :i* is op5 to philosophy b n to sci5ce1 & ! results obta9$ by philosophy are n radically diff]5t f ^? obta9$ f sci5ce4 ,! ess5tial *>act]i/ic ( philosophy1 :i* makes x a /udy di/9ct f sci5ce1 is criticism4 ,x exam9es critically ! pr9ciples employ$ 9 sci5ce & 9 daily life2 x se>*es out any inconsi/5cies !re may be 9 ^! pr9ciples1 & x only accepts !m :51 z ! result ( a critical inquiry1 no reason = reject+ !m has appe>$4 ,if1 z _m philosoph]s h believ$1 ! pr9ciples und]ly+ ! sci5ces 7 capable1 :5 dis5gag$ f irrelevant detail1 ( giv+ u k conc]n+ ! univ]se z a :ole1 su* k w\ld h ! same claim on our belief z sci5tific k has2 b our inquiry has n reveal$ any su* k1 & !re=e1 z reg>ds ! special doctr9es ( ! bold] metaphysicians1 has _h a ma9ly negative result4 ,b z reg>ds :at w\ld be commonly accept$ z k1 our result is 9 ! ma9 positive3 we h seldom f\nd reason to reject su* k z ! result ( our criticism1 & we h se5 no reason to suppose man incapable ( ! k9d ( k :i* he is g5]ally believ$ to possess4 ,:51 h[ev]1 we speak ( philosophy z a criticism ( k1 x is necess>y to impose a c]ta9 limitation4 ,if we adopt ! attitude ( ! complete sceptic1 plac+ ourselves :olly outside all k1 & ask+1 f ?is outside position1 to be compell$ to return )9 ! circle ( k1 we are dem&+ :at is impossible1 & our scepticism c nev] be refut$4 ,= all refutation mu/ beg9 ) "s piece ( k :i* ! disputants %>e2 f blank d\bt1 no argum5t c beg94 ,h5ce ! criticism ( k :i* philosophy employs mu/ n be ( ?is de/ructive k9d1 if any result is to be a*iev$4 ,aga9/ ?is absolute scepticism1 no logical argum5t c be advanc$4 ,b x is n difficult to see t scepticism ( ?is k9d is unreasonable4 ,desc>tes' 'me?odical d\bt'1 ) :i* mod]n philosophy began1 is n ( ?is k9d1 b is r ! k9d ( criticism :i* we are ass]t+ to be ! ess5ce ( philosophy4 ,8 'me?odical d\bt' consi/$ 9 d\bt+ :atev] seem$ d\btful2 9 paus+1 ) ea* app>5t piece ( k1 to ask himself :e!r1 on reflection1 he c\ld feel c]ta9 t he really knew x4 ,?is is ! k9d ( criticism :i* con/itutes philosophy4 ,"s k1 su* z k ( ! exi/5ce ( our s5se-data1 appe>s q indubitable1 h[ev] calmly & ?or\ghly we reflect ^u x4 ,9 reg>d to su* k1 philosophical criticism does n require t we %\ld ab/a9 f belief4 ,b !re are beliefs--su*1 = example1 z ! belief t physical objects exactly resemble our s5se-data--:i* are ent]ta9$ until we beg9 to reflect1 b are f\nd to melt away :5 subject$ to a close inquiry4 ,su* beliefs philosophy w bid u reject1 unless "s new l9e ( argum5t is f\nd to support !m4 ,b to reject ! beliefs :i* d n appe> op5 to any objections1 h[ev] closely we exam9e !m1 is n reasonable1 & is n :at philosophy advocates4 ,! criticism aim$ at1 9 a ^w1 is n t :i*1 )\t reason1 det]m9es to reject1 b t :i* consid]s ea* piece ( app>5t k on its m]its1 & reta9s :atev] /ill appe>s to be k :5 ?is consid]ation is complet$4 ,t "s risk ( error rema9s mu/ be admitt$1 s9ce human be+s are fallible4 ,philosophy may claim ju/ly t x dim9i%es ! risk ( error1 & t 9 "s cases x r5d]s ! risk s small z to be practically negligible4 ,to d m ?an ?is is n possible 9 a _w :]e mi/akes mu/ occur2 & m ?an ?is no prud5t advocate ( philosophy w\ld claim to h p]=m$4 ,,*apt] ,,xv4 ,,! ,,value ,,( ,,philosophy ,hav+ n[ come to ! end ( our brief & v incomplete review ( ! problems ( philosophy1 x w be well to consid]1 9 conclusion1 :at is ! value ( philosophy & :y x ought to be /udi$4 ,x is ! m necess>y to consid] ?is "q1 9 view ( ! fact t _m m51 "u ! influ5ce ( sci5ce or ( practical affairs1 are incl9$ to d\bt :e!r philosophy is any?+ bett] ?an innoc5t b useless trifl+1 hair-splitt+ di/9ctions1 & controv]sies on matt]s conc]n+ :i* k is impossible4 ,?is view ( philosophy appe>s to result1 p>tly f a wrong conception ( ! ends ( life1 p>tly f a wrong conception ( ! k9d ( goods :i* philosophy /rives to a*ieve4 ,physical sci5ce1 ?r\gh ! m$ium ( inv5tions1 is useful to innum]able p :o are :olly ignorant ( x2 ?us ! /udy ( physical sci5ce is to be recomm5d$1 n only1 or prim>ily1 because ( ! effect on ! /ud5t1 b r because ( ! effect on mank9d 9 g5]al4 ,?us utility does n belong to philosophy4 ,if ! /udy ( philosophy has any value at all = o!rs ?an /ud5ts ( philosophy1 x mu/ be only indirectly1 ?r\gh its effects ^u ! lives ( ^? :o /udy x4 ,x is 9 ^! effects1 !re=e1 if any:]e1 t ! value ( philosophy mu/ be prim>ily s\ght4 ,b fur!r1 if we are n to fail 9 our endeav\r to det]m9e ! value ( philosophy1 we mu/ fir/ free our m9ds f ! prejudices ( :at are wrongly call$ 'practical' m54 ,! 'practical' man1 z ?is ^w is (t5 us$1 is "o :o recognizes only mat]ial ne$s1 :o realizes t m5 mu/ h food = ! body1 b is oblivi\s ( ! necessity ( provid+ food = ! m9d4 ,if all m5 7 well (f1 if pov]ty & disease _h be5 r$uc$ to _! l[e/ possible po9t1 !re w\ld /ill rema9 mu* to be done to produce a valuable society2 & ev5 9 ! exi/+ _w ! goods ( ! m9d are at lea/ z important z ! goods ( ! body4 ,x is exclusively among ! goods ( ! m9d t ! value ( philosophy is to be f\nd2 & only ^? :o are n indiff]5t to ^! goods c be p]suad$ t ! /udy ( philosophy is n a wa/e ( "t4 ,philosophy1 l all o!r /udies1 aims prim>ily at k4 ,! k x aims at is ! k9d ( k :i* gives unity & sy/em to ! body ( ! sci5ces1 & ! k9d :i* results f a critical exam9ation ( ! gr\nds ( our convictions1 prejudices1 & beliefs4 ,b x _c be ma9ta9$ t philosophy has _h any v great measure ( success 9 its attempts to provide def9ite answ]s to its que/ions4 ,if y ask a ma!matician1 a m9]alogi/1 a hi/orian1 or any o!r man ( le>n+1 :at def9ite body ( tru?s has be5 asc]ta9$ by 8 sci5ce1 8 answ] w la/ z long z y are will+ to li/54 ,b if y put ! same "q to a philosoph]1 he w1 if he is c&id1 h to confess t 8 /udy has n a*iev$ positive results su* z h be5 a*iev$ by o!r sci5ces4 ,x is true t ?is is p>tly acc\nt$ = by ! fact t1 z soon z def9ite k conc]n+ any subject becomes possible1 ?is subject ceases to be call$ philosophy1 & becomes a sep>ate sci5ce4 ,! :ole /udy ( ! heav5s1 :i* n[ belongs to a/ronomy1 0 once includ$ 9 philosophy2 ,newton's great "w 0 call$ '! ma!matical pr9ciples ( natural philosophy'4 ,simil>ly1 ! /udy ( ! human m9d1 :i* 0 a "p ( philosophy1 has n[ be5 sep>at$ f philosophy & has become ! sci5ce ( psy*ology4 ,?us1 to a great ext5t1 ! unc]ta9ty ( philosophy is m app>5t ?an real3 ^? que/ions :i* are already capable ( def9ite answ]s are plac$ 9 ! sci5ces1 :ile ^? only to :i*1 at pres5t1 no def9ite answ] c be giv51 rema9 to =m ! residue :i* is call$ philosophy4 ,?is is1 h[ev]1 only a "p ( ! tru? conc]n+ ! unc]ta9ty ( philosophy4 ,!re are _m que/ions--& among !m ^? t are ( ! pr(\nde/ int]e/ to our spiritual life--:i*1 s f> z we c see1 mu/ rema9 insoluble to ! human intellect unless its p[]s become ( q a diff]5t ord] f :at !y are n[4 ,has ! univ]se any unity ( plan or purpose1 or is x a =tuit\s conc\rse ( atoms8 ,is consci\sness a p]man5t "p ( ! univ]se1 giv+ hope ( indef9ite gr[? 9 wisdom1 or is x a transitory accid5t on a small planet on :i* life mu/ ultimately become impossible8 ,are good & evil ( importance to ! univ]se or only to man8 ,su* que/ions are ask$ by philosophy1 & v>i\sly answ]$ by v>i\s philosoph]s4 ,b x w\ld seem t1 :e!r answ]s be o!rwise discov]able or n1 ! answ]s sugge/$ by philosophy are none ( !m demon/rably true4 ,yet1 h[ev] slight may be ! hope ( discov]+ an answ]1 x is "p ( ! bus9ess ( philosophy to cont9ue ! consid]ation ( su* que/ions1 to make u aw>e ( _! importance1 to exam9e all ! approa*es to !m1 & to keep alive t speculative int]e/ 9 ! univ]se :i* is apt to be kill$ by conf9+ ourselves to def9itely asc]ta9able k4 ,_m philosoph]s1 x is true1 h held t philosophy c\ld e/abli% ! tru? ( c]ta9 answ]s to su* fundam5tal que/ions4 ,!y h suppos$ t :at is ( mo/ importance 9 religi\s beliefs c\ld be prov$ by /rict demon/ration to be true4 ,9 ord] to judge ( su* attempts1 x is necess>y to take a survey ( human k1 & to =m an op9ion z to its me?ods & its limitations4 ,on su* a subject x w\ld be unwise to pron\nce dogmatically2 b if ! inve/igations ( our previ\s *apt]s h n l$ u a/ray1 we %all be compell$ to r5\nce ! hope ( f9d+ philosophical pro(s ( religi\s beliefs4 ,we _c1 !re=e1 include z "p ( ! value ( philosophy any def9ite set ( answ]s to su* que/ions4 ,h5ce1 once m1 ! value ( philosophy mu/ n dep5d ^u any suppos$ body ( def9itely asc]ta9able k to be acquir$ by ^? :o /udy x4 ,! value ( philosophy is1 9 fact1 to be s\ght l>gely 9 its v unc]ta9ty4 ,! man :o has no t9cture ( philosophy goes ?r\gh life imprison$ 9 ! prejudices d]iv$ f common s5se1 f ! habitual beliefs ( 8 age or 8 nation1 & f convictions :i* h gr[n up 9 8 m9d )\t ! co-op]ation or cons5t ( 8 delib]ate reason4 ,to su* a man ! _w t5ds to become def9ite1 f9ite1 obvi\s2 common objects r\se no que/ions1 & unfamili> possibilities are contemptu\sly reject$4 ,z soon z we beg9 to philosophize1 on ! contr>y1 we f9d1 z we saw 9 our op5+ *apt]s1 t ev5 ! mo/ ev]yday ?+s lead to problems to :i* only v incomplete answ]s c be giv54 ,philosophy1 ?\gh unable to tell u ) c]ta9ty :at is ! true answ] to ! d\bts :i* x raises1 is able to sugge/ _m possibilities :i* enl>ge our ?\ghts & free !m f ! tyranny ( cu/om4 ,?us1 :ile dim9i%+ our feel+ ( c]ta9ty z to :at ?+s are1 x greatly increases our k z to :at !y may be2 x removes ! some:at arrogant dogmatism ( ^? :o h nev] travell$ into ! region ( lib]at+ d\bt1 & x keeps alive our s5se ( wond] by %[+ famili> ?+s 9 an unfamili> aspect4 ,ap>t f its utility 9 %[+ unsuspect$ possibilities1 philosophy has a value--p]haps its *ief value--?r\gh ! greatness ( ! objects :i* x contemplates1 & ! fre$om f n>r[ & p]sonal aims result+ f ?is contemplation4 ,! life ( ! in/9ctive man is %ut up )9 ! circle ( 8 private int]e/s3 family & fri5ds may be includ$1 b ! out] _w is n reg>d$ except z x may help or h9d] :at comes )9 ! circle ( in/9ctive wi%es4 ,9 su* a life !re is some?+ fev]i% & conf9$1 9 comp>ison ) :i* ! philosophic life is calm & free4 ,! private _w ( in/9ctive int]e/s is a small "o1 set 9 ! mid/ ( a great & p[]ful _w :i* mu/1 soon] or lat]1 lay our private _w 9 ru9s4 ,unless we c s enl>ge our int]e/s z to include ! :ole out] _w1 we rema9 l a g>rison 9 a beleagur$ =tress1 kn[+ t ! enemy prev5ts escape & t ultimate surr5d] is inevitable4 ,9 su* a life !re is no peace1 b a con/ant /rife betwe5 ! insi/5ce ( desire & ! p[]lessness ( w4 ,9 "o way or ano!r1 if our life is to be great & free1 we mu/ escape ?is prison & ?is /rife4 ,"o way ( escape is by philosophic contemplation4 ,philosophic contemplation does n1 9 its wide/ survey1 divide ! univ]se into two ho/ile camps--fri5ds & foes1 helpful & ho/ile1 good & bad--x views ! :ole imp>tially4 ,philosophic contemplation1 :5 x is unalloy$1 does n aim at prov+ t ! re/ ( ! univ]se is ak9 to man4 ,all acquisition ( k is an enl>gem5t ( ! ,self1 b ?is enl>gem5t is be/ atta9$ :5 x is n directly s\ght4 ,x is obta9$ :5 ! desire = k is alone op]ative1 by a /udy :i* does n wi% 9 advance t its objects %\ld h ?is or t "*1 b adapts ! ,self to ! *>act]s :i* x f9ds 9 its objects4 ,?is enl>gem5t ( ,self is n obta9$ :51 tak+ ! ,self z x is1 we try to %[ t ! _w is s simil> to ?is ,self t k ( x is possible )\t any admission ( :at seems ali54 ,! desire to prove ?is is a =m ( self-ass]tion &1 l all self-ass]tion1 x is an ob/acle to ! gr[? ( ,self :i* x desires1 & ( :i* ! ,self kn[s t x is capable4 ,self-ass]tion1 9 philosophic speculation z else:]e1 views ! _w z a means to its own ends2 ?us x makes ! _w ( less acc\nt ?an ,self1 & ! ,self sets b\nds to ! greatness ( its goods4 ,9 contemplation1 on ! contr>y1 we />t f ! n-,self1 & ?r\gh its greatness ! b\nd>ies ( ,self are enl>g$2 ?r\gh ! inf9ity ( ! univ]se ! m9d :i* contemplates x a*ieves "s %>e 9 inf9ity4 ,= ?is reason greatness ( s\l is n fo/]$ by ^? philosophies :i* assimilate ! univ]se to ,man4 ,k is a =m ( union ( ,self & n-,self2 l all union1 x is impair$ by dom9ion1 & !re=e by any attempt to =ce ! univ]se into con=mity ) :at we f9d 9 ourselves4 ,!re is a widespread philosophical t5d5cy t[>ds ! view :i* tells u t ,man is ! measure ( all ?+s1 t tru? is man-made1 t space & "t & ! _w ( univ]sals are prop]ties ( ! m9d1 & t1 if !re be any?+ n creat$ by ! m9d1 x is unkn[able & ( no acc\nt = u4 ,?is view1 if our previ\s discussions 7 correct1 is untrue2 b 9 addition to be+ untrue1 x has ! effect ( robb+ philosophic contemplation ( all t gives x value1 s9ce x fett]s contemplation to ,self4 ,:at x calls k is n a union ) ! n-,self1 b a set ( prejudices1 habits1 & desires1 mak+ an imp5etrable veil betwe5 u & ! _w beyond4 ,! man :o f9ds pleasure 9 su* a !ory ( k is l ! man :o nev] leaves ! dome/ic circle = fe> 8 ^w might n be law4 ,! true philosophic contemplation1 on ! contr>y1 f9ds its satisfaction 9 e enl>gem5t ( ! n-,self1 9 ev]y?+ t magnifies ! objects contemplat$1 & !reby ! subject contemplat+4 ,ev]y?+1 9 contemplation1 t is p]sonal or private1 ev]y?+ t dep5ds ^u habit1 self-int]e/1 or desire1 di/orts ! object1 & h5ce impairs ! union :i* ! intellect seeks4 ,by ?us mak+ a b>ri] betwe5 subject & object1 su* p]sonal & private ?+s become a prison to ! intellect4 ,! free intellect w see z ,god might see1 )\t a "h & n[1 )\t hopes & fe>s1 )\t ! trammels ( cu/om>y beliefs & traditional prejudices1 calmly1 dispassionately1 9 ! sole & exclusive desire ( k--k z imp]sonal1 z purely contemplative1 z x is possible = man to atta94 ,h5ce also ! free intellect w value m ! ab/ract & univ]sal k into :i* ! accid5ts ( private hi/ory d n ent]1 ?an ! k br\ght by ! s5ses1 & dep5d5t1 z su* k mu/ be1 ^u an exclusive & p]sonal po9t ( view & a body ^: s5se-organs di/ort z mu* z !y reveal4 ,! m9d :i* has become accu/om$ to ! fre$om & imp>tiality ( philosophic contemplation w pres]ve some?+ ( ! same fre$om & imp>tiality 9 ! _w ( action & emotion4 ,x w view its purposes & desires z p>ts ( ! :ole1 ) ! abs5ce ( insi/5ce t results f see+ !m z inf9itesimal fragm5ts 9 a _w ( :i* all ! re/ is unaffect$ by any "o man's de$s4 ,! imp>tiality :i*1 9 contemplation1 is ! unalloy$ desire = tru?1 is ! v same quality ( m9d :i*1 9 action1 is ju/ice1 & 9 emotion is t univ]sal love :i* c be giv5 to all1 & n only to ^? :o are judg$ useful or admirable4 ,?us contemplation enl>ges n only ! objects ( our ?\ghts1 b also ! objects ( our actions & our affections3 x makes u citiz5s ( ! univ]se1 n only ( "o wall$ city at w> ) all ! re/4 ,9 ?is citiz5%ip ( ! univ]se consi/s man's true fre$om1 & 8 lib]ation f ! ?raldom ( n>r[ hopes & fe>s4 ,?us1 to sum up our discussion ( ! value ( philosophy2 ,philosophy is to be /udi$1 n = ! sake ( any def9ite answ]s to its que/ions1 s9ce no def9ite answ]s c1 z a rule1 be kn[n to be true1 b r = ! sake ( ! que/ions !mselves2 because ^! que/ions enl>ge our conception ( :at is possible1 enri* our intellectual imag9ation & dim9i% ! dogmatic assurance :i* closes ! m9d aga9/ speculation2 b above all because1 ?r\gh ! greatness ( ! univ]se :i* philosophy contemplates1 ! m9d also is r5d]$ great1 & becomes capable ( t union ) ! univ]se :i* con/itutes its highe/ good4 ,,bibliographical ,,note ,! /ud5t :o wi%es to acquire an elem5t>y k ( philosophy w f9d x bo? easi] & m pr(itable to read "s ( ! works ( ! great philosoph]s ?an to attempt to d]ive an all-r\nd view f h&books4 ,! foll[+ are specially recomm5d$3 ,plato3 ,republic1 especially ,books ,,vi & ,,vii4 ,desc>tes3 ,m$itations4 ,sp9oza3 ,e?ics4 ,leibniz3 ,! ,monadology4 ,b]keley3 ,?ree ,dialogues betwe5 ,hylas & ,philon\s4 ,hume3 ,enquiry conc]n+ ,human ,und]/&+4 ,kant3 ,prolegom5a to any ,future ,metaphysic4