Is Justified True Belief Knowledge?: Edmund Gettier
Is Justified True Belief Knowledge?: Edmund Gettier
Is Justified True Belief Knowledge?: Edmund Gettier
EDMUND GETTIER
Edmund Gettier is Professor Emeritus at the University of Massachusetts, Amherst. This short piece, published in 19 !, seemed to many decisively to refute an other"ise attractive analysis of #no"led$e. %t stimulated a rene"ed effort, still on$oin$, to clarify e&actly "hat #no"led$e comprises. '''''''''''''''''''' (rom Analysis. vol. )! *+ 9 ,. -opyri$ht . by Edmund Gettier. /eprinted by permission of the author. 0arious attempts have been made in recent years to state necessary and sufficient conditions for someone1s #no"in$ a $iven proposition. The attempts have often been such that they can be stated in a form similar to the fol1o"in$21 *a, 3 #no"s that P %(( *i.e., if and only if, *i, P is true, *ii, 3 believes that P, and *iii, 3 is 4ustified in believin$ that P. (or e&le, -hisholm has held that the follo"in$ $ives the necessary and sufficient conditions for #no"led$e2) *b, 3 #no"s that P %(( *i.e., if and only if, *i, 3 accepts P, *ii, 3 has ade5uate evidence for P, and *iii, P is true. Ayer has stated the necessary and sufficient conditions for #no"led$e as follo"s2! *c, 3 #no"s that P %(( *i, P is true, *ii, 3 is sure that P is true, and *iii, 3 has the ri$ht to be sure that P is true. % shall ar$ue that *a, is false in that the conditions stated therein do not constitute a sufficient condition for the truth of the proposition that 3 #no"s that P. The same ar$ument "ill sho" that *b, and *c, fail if 6has ade5uate evidence for6 or 6has the ri$ht to be sure that6 is substituted for 6is 4ustified in believin$ that6 throu$hout. % shal1 be$in by notin$ t"o points. (irst, in that sense of 64ustified6 in "hich 31s bein$ 4ustified in believin$ P is a necessary condition of 31s #no"in$ that P, it is possible for a person to be 4ustified in believin$ a proposition that is in fact false. 3econdly, for any proposition P, if 3 is 4ustified in believin$ P, and P entails 7, and 3 deduces 7 from P and accepts 7 as a result of this deduction, then 3 is 4ustified in believin$ 7. 8eepin$ these t"o points in mind, % shal1 no" present t"o cases in "hich the conditions stated in *a, are true for some proposition, thou$h it is at the same time false that the person in 5uestion #no"s that proposition.
CASE I
3uppose that 3mith and 9ones have applied for a certain 4ob. And suppose that 3mith has stron$ evidence for the fol1o"in$ con4unctive proposition2 *d, 9ones is the man "ho "ill $et the 4ob, and 9ones has ten coins in his poc#et. 3mith1s evidence for *d, mi$ht be that the president of the company assured him that 9ones "ould in the end be selected, and that he, 3mith, had counted the coins in 9ones1s poc#et ten minutes a$o. Proposition *d, entails2 *e, The man "ho "ill $et the 4ob has ten coins in his poc#et. :et us suppose that 3mith sees the entailment from *d, to *e,, and accepts *e, on the $rounds of *d,, for "hich he has stron$ evidence. %n this case, 3mith is clearly 4ustified in believin$ that *e, is true. ;ut ima$ine, further, that un#no"n to 3mith, he himself, not 9ones, "ill $et the 4ob. And, also, un#no"n to 3mith, he himself has ten coins in his poc#et. Proposition *e, is then true, thou$h proposition *d,, from "hich 3mith inferred *e,, is false. %n our e&le, then, all of the follo"in$ are true2 *i, *e, is true, *ii, 3mith believes that *e, is true, and *iii, 3mith is 4ustified in believin$ that *e, is true. ;ut it is e5ually clear that 3mith does not 8<=> that *e, is true? for *e, is true in virtue of the number of coins in 3mith1s poc#et, "hile 3mith does not #no" ho" many coins are in 3mith1s poc#et, and bases his belief in *e, on a count of the coins in 9ones1s poc#et, "hom he falsely believes to be the man "ho "ill $et the 4ob.
CASE II
:et us suppose that 3mith has stron$ evidence for the follo"in$ proposition2 *f, 9ones o"ns a (ord. 3mith1s evidence mi$ht be that 9ones has at all times in the past "ithin 3mith1s memory o"ned a car, and al"ays a (ord, and that 9ones has 4ust offered 3mith a ride "hile drivin$ a (ord. :et us ima$ine, no", that 3mith has another friend, ;ro"n, of "hose "hereabouts he is totally i$norant. 3mith selects three place names 5uite at random and constructs the follo"in$ three propositions2 *$, Either 9ones o"ns a (ord, or ;ro"n is in ;oston. *h, Either 9ones o"ns a (ord, or ;ro"n is in ;arcelona. *i, Either 9ones o"ns a (ord, or ;ro"n is in ;rest@:itovs#. Each of these propositions is entailed by *f,. %ma$ine that 3mith realiAes the entailment of each of these propositions he has constructed by *B, and proceeds to accept *$,, *h,, and *i, on the basis of *f,. 3mith has correctly inferred *$,, *h,, and *i, from a proposition for "hich he has stron$ evidence. 3mith is therefore completely 4ustified in believin$ each of these three propositions. 3mith, of course, has no idea "here ;ro"n is. ;ut ima$ine no" that t"o further conditions hold. (irst, 9ones does not o"n a (ord, but is at present drivin$ a rented car. And secondly, by the sheerest coincidence, and entirely un#no"n to 3mith, the place mentioned in proposition *h, happens really to be the place "here ;ro"n is. %f these t"o conditions hold, then 3mith does not 8<=> that *h, is true, even thou$h *i, *h, is true, *ii, 3mith does believe that *h, is true, and *iii, 3mith is 4ustified in believin$ that *h, is true. These t"o e&les sho" that definition *a, does not state a sufficient
condition for someone1s #no"in$ a $iven proposition. The same cases, "ith appropriate chan$es, "ill suffice to sho" that neither definition *b, nor definition *c, do so either.
NOTES
1. Plato seems to be considerin$ some such definition at Theaetetus )B %, and perhaps acceptin$ one at Meno 9C. ). /oderic# M. -hisholm, Perceivin$2 A Philosophical 3tudy. *%thaca, <.D., 19EF,, 1 . !. A. 9. Ayer. The Problem of 8no"led$e *:ondon, 19E ,.