Zenk 1 Benjamin Zenk Dr. Ron Bontekoe PHIL 445 5 December 2011 The Problem of the !

onnection" #cco$nt of !on%itional& in Robert 'talnaker(& # Theor) of !on%itional&" 'talnaker &tart& o$t hi& e&&a)* # Theor) of !on%itional&*" b) claimin+ that a con%itional ,ro,o&ition i& a f$nction* b$t not a tr$th-f$nction* of t.o other ,ro,o&ition&.1 De&,ite the .i%e&,rea% $&e of the tr$th-f$nctional anal)&i& of con%itional&* thi& i& not an e/tremel) contro0er&ial 0ie.. Ho.e0er* 'talnaker &ee& three ,roblem& a& ,re&entin+ them&el0e& in an) %e&cri,tion of the con%itional f$nction* tr$th-f$nctional or not1 the lo+ical ,roblem of con%itional&* the ,ra+matic ,roblem of co$nterfact$al&* an% the e,i&temolo+ical ,roblem of co$nterfact$al&. The) are ,roblem&* he claim&* for t.o &tan%ar% .a)& of e/,licatin+ the con%itional 2 the tr$th-f$nctional acco$nt an% the a,,eal to a connection* lo+ical or ca$&al* bet.een the antece%ent an% con&e3$ent 2 b$t the) are not ,roblem& for hi& o.n acco$nt of con%itional&.2 Hi& acco$nt take& for +rante% that con%itional&* &$bj$ncti0e or other.i&e* .ork. The) tell $& &omethin+ abo$t the .orl% an% .e &ho$l% be able to +i0e a coherent e/,lanation a& to .h) that i& the ca&e. I tr) to &ho.* thro$+h an e/,o&ition of &ome of 'talnaker(& main ,oint&* that in a0oi%in+ the connection" acco$nt of the con%itional f$nction he ha& act$all) m$%%ie% the .ater&. The three ,roblem& aro$n% .hich hi& e&&a) i& frame% are clo&el) relate%. The logical problem of conditionals refer& to the %iffic$ltie& that ari&e .hen one $n%ertake& to %e&cribe the formal ,ro,ertie& of the conditional function."4 Thi& i& the main ,roblem to
1

'talnaker* Robert. A Theory of Conditionals in Conditionals* 5rank 6ack&on* e%. 78e. 9ork1 1::1;* ,. 2<. 2 Ibid.* ,. 40. 4 Ibid.* ,. 2<.

Zenk 2 .hich 'talnaker %irect& hi& &ol$tion* beca$&e it* like hi& ,a,er* a,,lie& to con%itional& +enerall). The latter t.o ,roblem& are ,roblem& ari&in+ o$t of a con&i%eration of co$nterfact$al con%itional&. The pragmatic problem of counterfactuals i& that a %e&cri,tion of the formal ,ro,ertie& of con%itional& ,l$& the fact& &$rro$n%in+ them alone are not &$fficient to %etermine a %efiniti0e tr$th 0al$e for a con%itional. #%%itionall)* &ome criteria are nee%e% to %efen% certain inter,retation& o0er other&.4 The empirical problem of counterfactuals refer& to the ob&c$rit) latent in &tatement& that on the one han% &eem to re3$ire ob&er0able e0i%ence in or%er to be e0al$ate% an% on the other han% refer to circ$m&tance& that are them&el0e& co$nterfact$al* or co$nter-to-fact* an% hence $nob&er0able. Beca$&e the latter t.o ,roblem& ari&e o$t of the lo+ical ,roblem of con%itional&* it i& not &$r,ri&in+ that if a formal acco$nt of the con%itional f$nction rai&e& lo+ical %iffic$ltie&* it .ill likel) ha0e ,ra+matic an% e,i&temolo+ical i&&$e& .hen attem,tin+ to e0al$ate a co$nterfact$al* too. Take the tr$th-f$nctional acco$nt* for e/am,le. The tr$th-f$nctional acco$nt of the con%itional ren%er& a con%itional tr$e .hen either the antece%ent i& fal&e or the con&e3$ent i& tr$e. =b0io$& co$ntere/am,le& a&i%e* 'talnaker object&* The fal&it) of the antece%ent i& ne0er &$fficient rea&on to affirm a con%itional* e0en an in%icati0e con%itional."5 #l&o* +i0en that the antece%ent to a co$nterfact$al i& not it&elf fact$al an% ha& no tr$th-0al$e on it& o.n* the &tan%ar% tr$th-f$nctional acco$nt i& $nable to acco$nt for either the ,ra+matic or the e,i&temolo+ical ,roblem of co$nterfact$al&. 'talnaker ,oint& o$t that in recon&i%erin+ con%itional &tatement&* critic& are often le% to e/,lain con%itional& in term& of a lo+ical or ca$&al connection bet.een t.o
4 5

Ibid.* ,. 2:. Ibid.* ,. 41.

Zenk 4 ,ro,o&ition.> The tr$th of a con%itional* on thi& acco$nt* %e,en%& on &$ch a connection bein+ ,re&ent bet.een the antece%ent an% the con&e3$ent. B$t* 'talnaker claim&* there are in&tance& in .hich .e .o$l% acce,t a con%itional a& tr$e %e&,ite %en)in+ a connection.? I &hall no. tr) to &ho. that hi& claim i& mi&lea%in+. The e/am,le he $&e& to ill$&trate hi& ,oint i& the follo.in+1 If the !hine&e enter the @ietnam conflict* the Anite% 'tate& .ill $&e n$clear .ea,on&.< He think& that if .e .o$l% affirm the con&e3$ent in thi& e/am,le* then .e .o$l% be likel) to affirm the con%itional. #t fir&t +lance* I a+ree .ith 'talnaker on thi& ,oint. Bhen a&ke% to +i0e a tr$e or fal&e e0al$ation of thi& con%itional* the .ei+ht &eem& carrie% b) the tr$th-0al$e of the con&e3$ent. A,on f$rther con&i%eration* ho.e0er* there i& more to the &tor). Bhat .e act$all) &eem to mean b) affirmin+ thi& con%itional i& the follo.in+1 Re+ar%le&& of .hether or not the !hine&e enter the @ietnam conflict* the Anite% 'tate& .ill $&e n$clear .ea,on&. Bitho$t an) lo&& of meanin+* thi& can be ren%ere% tr$th-f$nctionall)1 Cither the !hine&e .ill enter the @ietnam conflict or the !hine&e .ill not enter the @ietnam conflict* an% the Anite% 'tate& .ill $&e n$clear .ea,on&. In thi& ca&e* 'talnaker i& ri+ht* ,re&ence of a connection i& not nece&&ar). #& I ha0e j$&t trie% to ,oint o$t* ho.e0er* thi& i& beca$&e the &tatement i& not reall) a con%itional. =r if it i&* it i& a meanin+le&& one. 'ince it i& the conj$nction of a ta$tolo+) an% a tr$e or fal&e ,ro,o&ition* the tr$th of &$ch a &tatement ob0io$&l) %e,en%& on the tr$th of the &econ%
> ?

Ibid.* ,. 41. Ibid.* ,. 41. < Ibid.* ,. 40.

Zenk 4 conj$nct. =r* if .e $&e lan+$a+e loo&el) eno$+h* it %e,en%& on the tr$th or fal&it) of the con&e3$ent. To clarif)* con&i%er another &entence of the &ame con%itional &tr$ct$re* b$t .ith a ne. antece%ent1 If Robert 'talnaker(& article i& ,$bli&he%* the Anite% 'tate& .ill $&e n$clear .ea,on&. C0en if .e &tron+l) felt that the con&e3$ent .ere tr$e* .e .o$l% not affirm thi& con%itional. Thi& i& beca$&e there i& no connection bet.een the antece%ent an% the con&e3$ent* at lea&t no a,,arent one. #& &oon a& a connection i& intelli+ible* ho.e0er* .e are more incline% to either affirm or %en) the con%itional. That i&* .e are more incline% to belie0e it make& an) &en&e a& a con%itional. In%ee%* .hat incline& $& to affirm the tr$th of the like-&tr$ct$re% con%itional in 'talnaker(& e/am,le i& that a 0a+$e connection i& reco+niDe% bet.een the antece%ent an% the con%itional. Bracketin+ thi& objection to 'talnaker(& co$ntere/am,le for the moment* let(& con&i%er hi& o.n acco$nt* .hich i& a &tren+thenin+ of an acco$nt of con%itional& taken from 5. P. Ram&e). Thi& acco$nt hol%&* +enerall)* that the tr$th-0al$e a&&i+ne% to a con%itional i& the &ame a& the h),othetical 0al$e of the con&e3$ent .hen )o$ h),otheticall) a%% the antece%ent to )o$r &tock of belief& an% then con&i%er the h),othetical tr$th of the con&e3$ent.: =n thi& acco$nt* if )o$ belie0e there i& a connection bet.een the antece%ent an% the con&e3$ent* )o$ .ill ob0io$&l) acce,t the con%itional. If )o$ %o not belie0e &$ch a connection hol%&* 'talnaker ar+$e& that )o$ .ill nonethele&& acce,t the con%itional if )o$ alrea%) belie0e the con&e3$ent i& tr$e. The ,roblem of loo&e lan+$a+e i& reenterin+ hi& acco$nt here an%* in%e,en%ent of a connection* if .e acce,t the con&e3$ent of If m) com,$ter i& a Eac* then Barbara 'trei&an% .ill make a comeback
:

Ibid.* ,. 42.

Zenk 5 to$r*" .e are bo$n% to acce,t the con%itional. #lrea%)* 'talnaker(& acco$nt i& %ee,l) ,roblematic. In ca&e& .here .e belie0e the antece%ent to be fal&e* .e cannot .itho$t contra%iction a%% &$ch a belief to o$r &tock of belief&. Therefore* .e m$&t minimall) b$t &$fficientl) a%j$&t o$r h),othetical belief& to remo0e &$ch contra%iction. In makin+ the tran&ition from belief con%ition& to tr$th con%ition&* 'talnaker $tiliDe& the conce,t of a ,o&&ible .orl%* the ontolo+ical analo+$e of a &tock of h),othetical belief&."10 He then %e0elo,& a mo%al &tr$ct$re in .hich the relation&hi,& amon+ ,o&&ible .orl%& are %etermine% b) in%$cti0e relation& an% ca$&al connection&. Po&&ible .orl%&* in t$rn* %etermine the tr$th-0al$e of con%itional&11 .itho$t the nee% for ,ara,hra&e. 'talnaker characteriDe& the ,ara,hra&in+ of con%itional& to ,re&er0e connection a& a bal%l) ad hoc mane$0er" an% &ee& 0irt$e in a0oi%in+ &$ch 3$alification.12 That i&* 'talnaker &ee& a%0anta+e& in treatin+ the con%itional a& a $ni0ocal conce,t.14 In maintainin+ an in%irect rather than %irect relation&hi, bet.een connection an% con%itional&* he claim& to +i0e a $nifie% acco$nt of con%itional& .hich e/,lain& the 0ariation& in their beha0io$r in %ifferent conte/t&."14 Rather than &ti,$latin+ or makin+ recommen%ation& re+ar%in+ or%inar) lan+$a+e* hi& acco$nt inten%& to e/,lain the conce,t a& commonl) $&e%.15 Thi& a,,roach &tan%& a,,ro,riatel) in contra&t to the tr$thf$nctional acco$nt* .hich* in re,lacin+ an $nclear conce,t .ith a clear one* cea&e% to
10 11

Ibid.* ,. 44. Ibid.* ,. 40. 12 Ibid.* ,. 4:. 14 Ibid.* ,. 41. 14 Ibid.* ,. 40. 15 Ibid.* ,. 40.

Zenk > re&emble the conce,t it inten%e% to clarif). 9et* a& m) ar+$ment abo0e in%icate&* there i& &ome tro$ble the incl$&i0e a,,roach to %e&cribin+ con%itional&. It r$n& the ri&k of bein+ $nnece&&aril) a,olo+etic for &lo,,) $&a+e. In 'talnaker(& ca&e* thi& a,,roach i& +$ilt) of it. Be&i%e& thi&* it %oe& not e/,lain a.a) the ob0io$& fla.& of the tr$th-f$nctional acco$nt* b$t &$b&$me& the tr$th-f$nctional acco$nt $n%er it. In a0oi%in+ the thoro$+h+oin+ commitment to connection in con%itional &tatement&* 'talnaker commit& him&elf to acce,tin+ man) of the &ame a.k.ar% &tatement& that .ere initiall) acce,te% b) the tr$th-f$nctional acco$nt.