The Classical Analysis of Knowledge
Steve drives to work and at the lights he sees his colleague, Ann, in her car.
He smiles and waves at her, but she looks at him as if she has never seen him
before. This upsets Steve and he keeps thinking for the rest of the day why Ann
has acted that way.
The scenario above fits the Classical Analysis of
Knowledge as Justified True Belief:
S knows that p
if and only if
(i) p is true;
(ii) S believes that p;
(iii) S is justified in believing that p (Ichikawa).
Steve forms a belief that he sees Ann, he is
justified in his belief - he sees her in the car next to his, this is a
standard morning situation, they usually meet at the lights on their way to
work. And Steve is right, Ann is in the car. All conditions are met, so Steve’s
belief can count as knowledge.
Gettier
cases
There is a twist in the story above.
Steve
doesn’t know that the woman behind the wheel is not Ann, but her identical twin
sister, driving Ann to work in Ann’s car that morning. Ann is sitting next to
her, but at the moment when the cars meet, she is bending down trying to pick
her lipstick from the car floor. In the light of these facts, does Steve really
have knowledge?
The scenario is an example of a
Gettier-style case, which shows that it is possible to lack knowledge even though one’s judgement is both true and
supported by evidence, therefore justified. Apparently, the JTB is insufficient
for knowledge. The problem was first acknowledged in the Middle Ages by Dharmottara, but
only after Edmund Gettier’s famous paper "Is Justified True Belief
Knowledge?” (1963),
it spurred epistemologists to rethink the concept of knowledge and to fix what
was missing from the tripartite analysis (Ichikawa). Gettier cases show the
following pattern:
(i)
S believes that p
(ii)
p is true
(iii)
S has good evidence for the p
(iv)
there is a hidden fact S is not aware of which causes him to make a false
assumption regarding the belief.
Gettier
“knowers” arrive at justified true beliefs unaware of hidden facts. There are three
major factors that undermine the JTB: false assumptions, fallible
justification, and the presence of luck. All Gettier “knowers” form beliefs on
false premises, and despite good evidence they end up with justified false
beliefs. Only thanks to luck the false beliefs turn out to be true.
Strengthening justification -
adding a fourth condition (JTB + X)
In the light of the above conclusion,
the solution seems simple – let’s strengthen the justification condition by
eliminating false beliefs and hidden facts that destroy the justification. That
was the idea behind M. Clark’s No false lemmas approach, followed by K. Lehre
and T. Paxson’s No defeater approach, both adding a fourth condition to the JTB
(Ichikawa).
According to No false lemmas,
S knows that p if and only if
(iv) It is on true grounds that S believes that p (Clark 447).
The problem is the condition excludes good cases of knowledge. In real life,
one false belief does not undermine the weight of solid evidence supporting a
true belief. It would not be reasonable to deny knowledge every time a false
premise slips in between a tonne of true ones. Investigators working on crime
cases would not be able to claim they have made justified judgements against
the suspect, if just one witness gave a false testimony (Nagel 49). If
knowledge requires forming a belief on entirely true grounds, then this one
false testimony ruins everything. Another problem here is that the approach
fits only cases where the knowers form beliefs on the basis of inference, and
does not apply to instances where knowers do not rely on an inference.
Lehre/Paxson’s
No defeater proposes a different fourth condition. In Gettier cases the
justification the “knowers” have to support having knowledge is destroyed by
a defeater – a hidden fact they are not aware of (Ichikawa). The analysis of
knowledge should be then Undefeated Justified True Belief and the fourth
condition is:
(iv) there are no justification destroying facts (defeaters).
The problem here, however, is that one can have knowledge despite hidden
defeaters, as long as one defeater cancels out the other one (e.g.
Lehre/Paxson’s Tom Grabit case) (Sudduth). But what if the defeaters cancel one another, and the knower’s
belief is still “gettiered”? For example, Sue knows she has $5 bill in her
wallet – that is how much change she got after paying for coffee. What she doesn’t
know is that the bill slipped out when she was putting it in the wallet. Later
that day, her husband who owed her $5, puts a $5 bill in her wallet,
unbeknownst to her. She knows that she has $5 in the wallet and she is right.
Does this count as knowledge given the circumstances?
Dropping Justification : The
Causal Theory of Knowledge, Reliabilism, and Tracking Theory
Adding a
fourth condition to the JTB does not help. Maybe the problem lies in the JTB
itself? In 1967, A. Goldman suggested dropping justification in favour of causal connection (Nagel 51). The
problem with justification is that it does not always lead to truth, it can
support false beliefs and mislead the knower. What is more, sometimes we cannot
justify how we came to know a fact, which does not mean that we cannot claim
knowledge. Knowing a fact is caused by a belief about that fact (Goldman 451). Therefore
S knows that p if and only if
(iii) p is causally connected in an appropriate way with S’s believing p.
Sense perception, testimony, and inference – all count as appropriate causal
connection. As long as the knower’s belief is causally connected to the fact
through any of above ways, he can claim knowledge. Gettier “knowers” are
causally disconnected from the facts about which they form a beliefs, for
example the traveller in the desert believing he sees water is not connected
through sense perception to said water. It is his mind playing tricks on him.
So appropriate causal connection might fix the problem, and also save the
instances of knowledge for which the knower does not have a proper
justification on hand.
The Causal Theory
is not free from objections. What exactly is meant by “appropriate causal
chain”? Or can we really quit the justification condition? And what about cases
of knowledge which do not depend on causal connection but some environmental
factors? Goldman’s Fake Barn scenario illustrates a case in which the knower is
causally connected to a fact and does not have knowledge: Henry visits
a county where 99% of the barns he drives by are barn facades, of which Henry
does not know. When he points at one of the barns, showing it to his son, it is
the only genuine barn (Nagel 52). However, with all the fakes around, can we
say that he knows he is looking at a real barn? The causal connection here is
intact, and yet it is questionable whether Henry knows. Given the environment,
Henry is forming his belief in an unreliable way. Gettier-style luck and
accidentally true belief play a major role here.
This led Goldman
to replace the Causal Theory with Reliabilism:
S knows that p if and only if
(iii) S has a true belief formed by a reliable belief-forming process (Nagel 53).
Reliable belief-forming process (e.g. through sense
perception) has to be likely to deliver the truth in the knower’s environment.
And this is what causes objections here: if reliable means ‘likely to produce a
true belief’, then how much of “likely”
will be enough to count as “producing true belief” and counting as knowledge? How
do we identify the reliable belief-forming mechanism? Depending on the
mechanism we choose, we either end up having knowledge or not (Nagel 53). And
even when our belief-forming mechanism is a reliable one, e.g. judging the
probability of how likely it is that you have a winning lottery ticket, you
cannot really know until the results are announced (Nagel 53).
To avoid
the problems posed by Reliabilism, another approach to knowledge was proposed, R.
Nozik’s Tracking Theory, which introduced two counterfactual conditions (tracking
conditions):
S knows that p if and only if
(iii) If p were not true and S used
belief-forming method M, then S would not
believe p;
(iv) If p were true and S used belief
forming method M, then S would believe that p
(Adams 19).
To know a fact it is not enough to be right about the state
of affairs. Our beliefs must track the truth given the methods by which we form
them. The tracking theory solves certain problems caused by the Causal Theory
and Reliabilism, but generates new ones, e.g. it is incompatible with the
closure principle.
Conclusions
Fixing the problem pointed out by
Gettier seems unsolvable. All the alternative analyses prove untenable, being either
too strict and ruling out cases of knowledge, or exposing further problems. In real life,
knowers have no access to all the necessary facts that contribute to whether
they have knowledge or not. To construct an analysis that would be immune to
objections, it would need to contain
a condition that the knower is omniscient. Otherwise, there is no way of
escaping false assumptions, hidden facts, ugly surprises, false testimonies, and
lucky circumstances. Perhaps this is why the analysis of knowledge is ineffective.
Word count: 1556
Works cited
Adams, Fred. Tracking Theories of
Knowledge. Veritas – Revista de Filosofia da
Pucrs 50 (4):1-35. Pontificia
Universidade Catolica, 2005
Clark, Michael.
“Knowledge and Grounds: A Comment on Mr. Gettier’s Paper.”
in M. Huemer (ed), Epistemology
- Contemporary Readings, London and New York, Routledge, pp. 447-9
Collins, John M. “Epistemic Closure Principles.” Internet Encyclopedia of
Philosophy,
iep.utm.edu/epis-clo/#SH3b. Accessed 26 Mar. 2021
Goldman, Alvin. ”A Causal Theory of Knowing.” in M. Huemer (ed), Epistemology
- Contemporary Readings, London
and New York, Routledge, pp. 450-63
Ichikawa, Jonathan
Jenkins and Matthias Steup. "The Analysis of Knowledge",
The Stanford Encyclopedia of
Philosophy (Summer 2018 Edition), Edward N.
Zalta (ed.), plato.stanford.edu/archives/sum2018/entries/knowledge-
analysis/. Accessed 20 Mar. 2021
Nagel, Jennifer. Knowledge: A Very Short Introduction. Kindle ed., OUP
Oxford,
2014
Sudduth, Michael, “Defeaters in Epistemology.” Internet Encyclopedia of
Philosophy,
iep.utm.edu/ep-defea/. Accessed 26 Mar. 2021
Bibliography
Gettier, E. L. “Is
Justified True Belief Knowledge?” in M. Huemer (ed.), Epistemology
- Contemporary Readings, London
and New York, Routledge, pp. 444–6
Lehrer, Keith and Thomas Paxson. “Knowledge: Undefeated Justified True Belief.”
in M. Huemer (ed.), Epistemology - Contemporary
Readings, London and New
York, Routledge, pp. 464-74
Pritchard, Duncan. What
Is This Thing Called Knowledge? 4th ed., Kindle ed.,
Routledge, 2018
No comments:
Post a Comment