PY1006 (2009-10)

REASONING AND KNOWLEDGE

LECTURE TWLEVE:

Internalism and Externalism

8/2/2010

(student version)
Dr. Patrick Greenough

1
1. Recap.

• Last time, we looked at what it is to give

a good analytic definition.

• This time, we are going to apply that

wisdom to the question of how to define
‘knowledge’.

2
2. Is knowledge just belief?

Question: is knowledge just belief?

3
2. Is knowledge just belief?

• Put another way: is belief both necessary

and sufficient for knowledge that P?

• That is, do we have the following

analysis?

A subject S knows that P if and only if S

believes that P

4
2. Is knowledge just belief?

Question: But is belief necessary for

knowledge?

• YES.

Potential counterexample: …

5
2. Is knowledge just belief?

Question: But is belief sufficient for

knowledge?

• Surely knowledge is some kind of special

belief.

• Mere belief, while necessary, is not

sufficient for knowledge.

6
 3. Is knowledge just true belief?

Question: Is knowledge just true belief?

• That is, do we have the following?

S knows that P if and only if:

(i) S believes that P and

(ii) P is true.
7
3. Is knowledge just true belief?

Question: Is truth necessary for

knowledge knowledge?

Arguably: Yes – one cannot know a

proposition which is false.

Potential counterexample: …

8
3. Is knowledge just true belief?

Question: Is true belief sufficient for

knowledge?

9
3. Is knowledge just true belief?

Counterexample to this sufficiency claim:

I believe that the moon contains 3 billion
tonnes of lead. I formed this belief
without any evidence (e.g. via
guesswork). Nonetheless my belief turns
out to be true – but surely it is not
knowledge!

10
3. Is knowledge just true belief?

Basic Underlying Thought: a belief needs

to be a bit more special than that in order
to count as knowledge.

11
3. Is knowledge just true belief?

Underlying Thought One: we don’t want

beliefs that are accidentally true to count
as knowledge.

Underlying Thought Two: we don’t want

true beliefs that are not supported by
evidence to count as knowledge.

12
4. Is knowledge justified true belief?

Question: Is knowledge justified true

belief?

Put another way: Are truth, belief, and

justification not only individually
necessary conditions for knowledge, are
they also jointly sufficient for knowledge?

13
4. Is knowledge justified true belief?

• This is called The Tripartite Analysis of

Knowledge.

Alternative name: The JTB Analysis.

14
4. Is knowledge justified true belief?

JTB Analysis: a subject S knows that P if

and only if

(i) S believes that P,

(ii) P is true,

(iii) S is justified in believing that P.

15
4. Is knowledge justified true belief?

• That is, …

16
5. What do we mean by ‘justification’ here?

• The sort of justification in the JTB

analysis is fallibilist justification: one can
have a justified but false belief.

• …

17
5. What do we mean by ‘justification’ here?

• It's a so-called 'threshold concept'--you

need a certain amount of justification in
order for your true belief to count as
knowledge.

• A bit of evidence won't do--you need

sufficient evidence.

18
6. Is JTB enough for knowledge?

Question: But is a justified true belief

sufficient for knowledge?

19
7. A Gettier counterexample: Chisholm’s sheep case.

See the lecture …

20
8. A Gettier counterexample: Wimbledon final case.

(i) You fall asleep during the 2010

Wimbledon men’s final between Mr
Smash and Mr Volley.
(ii) Some time later you wake up and the
final still seems to be in progress.
(iii) In fact, you are watching the
complete 2009 final which was also
played by Mr Smash and Mr Volley: the
BBC has forgot to put the repeat emblem
in the corner of the screen.
21
8. A Gettier counterexample: Wimbledon final case.

(iv) Indeed, the 2010 final is over and Mr

Smash won.
(v) He also won in 2009.
(vi) On the basis of what the BBC is now
showing, you form the belief that Mr
Smash has won the 2010 title.
(vii) Your belief is true and indeed
justified – but it is not knowledge …

22
9. A Gettier counterexample: Russell’s stopped clock.

See the lecture …

23
10. What is a Gettier counterexample?

Definition: A (basic) Gettier

counterexample is a case where a
subject has a justified true belief, but
where this justified true belief is not
sufficient (not enough) for knowledge.

24
10. What is a Gettier counterexample?

• So, (basic) Gettier counterexamples are

counterexamples to the sufficiency claim
in the JTB analysis.

They are designed to show that a

justified true belief is not sufficient for
knowledge.

25
10. What is a Gettier counterexample?

• Note that …

26
11. How should we respond?

• Three basic strategies:

(1) Take ‘knowledge’ to be basic and

give up on the possibility of a reductive
definition.

(Seems a bit hasty ….)

27
11. How should we respond?

(2) Replace the justification condition

with something better.

(3) Beef-up the sufficiency conditions (by

adding an extra condition).

28
11. How should we respond?

• We are going to look at two ways in

which you might employ the second
strategy …

• …

29
12. Why do the Gettier cases arise? (Part I)

Diagnosis One: Because the justification

for believing that P is not in any way
connected the fact that P.

• That is, one is believing the right thing

but for the wrong reasons … so one
does not know …

30
13. The causal theory of knowledge.

• If we follow this diagnosis then one

natural thought is that one’s belief needs
to be properly ‘tethered’ to the facts if
one is to know.

• Enter the causal theory of knowledge.

31
13. The causal theory of knowledge.

Causal Analysis: a subject S knows that

P if and only if

(i) S believes that P,

(ii) P is true,

(iii) S’s belief is caused by the fact that P.

32
13. The causal theory of knowledge.

• Note that the justification condition has

been replaced by a causal condition.

• …

33
13. The causal theory of knowledge.

Thumbnail Causal View: One knows that

P just in case one’s belief that P is
(appropriately) hooked up to the fact that
P.

34
14. Internalism versus externalism.

• The causal theory represents a

revolution in epistemology … away from
internal factors being important to
making external factors important.

35
14. Internalism versus externalism.

Traditional Epistemology: one knows just

in case one has a true belief that is
backed by the right kind of evidence or
reasons.

• Here evidence is internal to the subject: it

is the kind of thing the subject can cite.

36
14. Internalism versus externalism.

Externalist Epistemology: one knows just

in case one’s belief bears the right kind
of external relation to the facts.

• The causal theory of knowledge is one

kind of externalist theory.

37
15. A Gettier counterexample for the causal theory.

• The causal theory has its own Gettier-

style counterexample: The Goldman-
Ginet Barn-Façade case.

See the lecture …

38
15. A Gettier counterexample for the causal theory.

• In such a case, one has a true belief that

is causally hooked up to the facts.

• Yet, intuitively, one (allegedly) lacks

knowledge ….

• So, bad news for the causal theory.

39
15. A Gettier counterexample for the causal theory.

Note: we have here a generic Gettier

counterexample: i.e. a case where the
subject has a true belief which also
meets whatever further sufficiency
conditions are proposed, yet is not a
case of knowledge.

40
15. A Gettier counterexample for the causal theory.

Upshot: the causal theory of knowledge

makes the wrong predictions.

• We need to go back to the why all of the

Gettier cases arise…

41
16. Why do the Gettier cases arise? (Part II).

Diagnosis Two: Because the subject’s

true belief that P is just too lucky to count
as knowledge.

• That is, where one’s belief gets to be true

by accident then one fails to know.

42
17. Reliabilism.

• So, why don’t we just define knowledge

as follows:

Reliabilist Account: Knowledge is non-

accidental true belief.

(See Dretske 1970, 1971.)

43
17. Reliabilism.

• Another way of expressing knowledge as

non-accidental true belief is to say that
for a true belief to count as knowledge,
the belief must be reliably connected to
the truth in some way.

44
17. Reliabilism.

Reliabilist Analysis: a subject S knows

that P if and only if

(i) S believes that P,

(ii) P is true,

(iii) S’s belief is non-accidentally true.

45
17. Reliabilism.

The father of reliabilism is Bertrand

Russell, who said:

People do not say that a barometer

‘knows’ when it is going to rain; but I
doubt if there is any essential difference
in this respect between the barometer
and the meteorologist who observes it.
(Russell 1923, Australasian Journal of
Philosophy).
46
17. Reliabilism.

• Five points of note:

(i) Again, here the justification condition

has been replaced

(ii) The truth condition has become

redundant.

47
17. Reliabilism.

(iii) It deals with the Goldman-Ginet Barn

Façade case.

• Why? Because the subject’s belief fails

the third condition on knowing.

• So, the theory makes the right

predictions.
48
17. Reliabilism.

(iv) It deals with all the other Gettier

cases we have encountered also.

(Or so it would seem. One really needs

to combine Reliabilism with the causal
theory to get the right result.)

49
17. Reliabilism.

(v) It’s a form of externalism.

• Whether or not one’s belief is knowledge

depends on external factors …

50
17. Reliabilism.

• Reliabilism allows that small children and

animals have knowledge.

• Why? Because one does not need

evidence (that one can cite) in order to
know, one just needs to have a belief
that is non-accidentally true.

51
18. The Problem of Bonjour’s clairvoyant.

• Suppose someone can reliably detect,

just by exercising some kind of
clairvoyant power, the presence of
President Obama.

• So, given Reliabilsm, this clairvoyant has

knowledge of Obama’s whereabouts.

52
18. The Problem of Bonjour’s clairvoyant.

• Such a case is supposed to challenge

the sufficiency of the reliability clauses.

• Just getting it blindly right (especially

when one has evidence that one should
not be able to get it right) is not sufficient
for knowledge.

53
18. The Problem of Bonjour’s clairvoyant.

Solution: add in an extra clause:

(iv) The subject must have evidence that

they can cite in favour of their beliefs.

(This leads to a kind of hybrid view of

internalism and externalism.)

54
19. The Problem of the chicken-sexers.

• Industrial chicken sexers can sex

chickens even though they have false
beliefs about how they do it.

• Some have argued that they lack

knowledge even though Reliabilism
predicts that they do have knowledge …

55
19. The Problem of the chicken-sexers.

• But it would seem way to much to

demand that one always be able to cite
explicit evidence in favour of one’s
beliefs.

• I know that the battle of Waterloo took

place in 1815 – even though I have
forgotten my sources …
56
20. The method of counterexamples.

• The method of counterexamples is a

useful tool in discovering some accurate
definition of some concept.

• But what if for every patch there is

always another puncture?

57
20. The method of counterexamples.

• Perhaps the best we can do in the end is

to say that knowledge is a true belief that
is believed for the right reasons.

• But then there is the danger of

circularity…

58
20. The method of counterexamples.

• Perhaps the best we can do in the end is

to say that knowledge is basic and
admits of no reductive definition….

• The project of defining knowledge on

such a view is quixotic…

59
21. Do you want to read more?

• Stephen Hetherington (1996):

Knowledge Puzzles, Chs. 5, 6, 7 (very
short chapters), Westview.

• J. Dancy (1985) : Introduction to

Contemporary Epistemology, Ch. 2, 3,
Blackwells.

60