Philosophy Weeks 9 and 10
October 28, 2013
Nozick’s tracking theory
S know that P if and only if
(1) P is true
(2) S believes that P
(3) If P were false, S wouldn’t believe that P (Tracking condition, “justification”, sufficient evidence)
(4) If P were true, S would believe that P
What makes a thermometer reliable?
For a thermometer to be reliable, its reading must be true (if it’s 25C in here, the thermometer should read
We need to rule out the possibility that the thermometer is stuck at 25C. You need to know what happens
when the temperature changes.
Counterfactuals are conditional statement: if x were true, then y is true.
What motivates condition (3)?
Nozick says that having a reliable mind is like having a reliable thermometer; it tracks the way things
change. The mind’s ability to track the way things are is indicative of whether or not it knows that P.
Example: Russell’s stuck clock case. This is a test case for any theory of knowledge.
If it weren’t 12:08, would you still believe that it is 12:08?
Yes! So condition (3) fails. S does not know that it is 12:08, even though it’s true and S believes it.
Suppose you have a fair lottery ticket. Like most tickets, it is a losing ticket. You believe that it is a losing
ticket. If it were a winning ticket (if P were false), you would still believe that it is a losing ticket.
Therefore, you don’t know if it is a losing ticket.
(3) Helps elucidate the idea of tracking
Acounterfactional conditional is stating what would be if some condition in an event were different.
Ex. “If Harper hadn’t won, an NDP member would’ve been prime minister” – true
“If Harper hadn’t won, a green party member would’ve been prime minister.”
Nozick’s theory does not necessarily satisfy the skeptic: it does not rule out the possibility that the mind is
being deceived. However, it does support the fact that if the mind is able to track the way things are, it
knows about many things.
- Find out why Nozick has the fourth condition.
Knowledge vs Justified Belief
Dismissive reaction to knowledge skepticism. - Fall back on justified belief, compare the degree to which some beliefs are more supported by
Some beliefs seem more supported than others by evidence.
Hume’s skepticism is about inductive inference.
An inductive argument
Ex. If I survey 1000 people and 56% say that they will vote for Harper in the next election, I may
conclude that Harper will win the next election.
We have a stronger argument whose strength depends on the representativeness of the sample.
The sun has risen in the past: Therefore, the sun will rise tomorrow. This is based on PUN
Hume’s problem of induction (this is a deductive argument)
(1)All inductive arguments presuppose PUN (Principle of Uniformity of Nature)
(2) To justify induction, we must justify PUN
(3) To justify PUN, we must use either an inductive or deductive argument.
(4) We can’t justify PUN deductively because we can’t deduce claims about the future from plans about
the past. We can’t deduce uniformity from what we observe(d).
(5) We can’t justify PUN inductively (it would be circular, since all inductive arguments presuppose
PUN. We can’t use PUN to justify PUN)
(6) Hence, we can’t justify PUN
(7) Hence, we can’t justify induction
(8) Hence, all of our beliefs based on inductions and generalization are unjustified. D:
Deductive argument: the premises guarantee the truth of the conclusions
What we want when we talk about justifying induction/PUN is that we want to justify the predictions and
generalizations we make that go beyond what we have observed.
Objection to premise (5) by Max Black
You can justify induction inductively in the following way
(P) Induction has been reliable in the past/ PUN has held true in the past
(C) Induction will (probably) be reliable if I use it now or in the future
Further, the premise is true. Secondly, it is not circular. Why not?An argument is circular if the
conclusion appears as one of the premises. The conclusion is not a premise, so Black’s argument is not
Objection to Black’s Argument
Counterinduction: IfAs have be B in the past, assumeAs won’t be B in the future.
Ex. If gravity has been x in the past, assume gravity won’t work that way in the future.
We want to justify counterinduction as a bad rule to prove that we can somehow show that induction is a
better/more justified rule than counterinduction. (P) Counterinduction has been unreliable in the past.
(C) Counterinduction will be reliable if I use it now or in the future.
There’s more to an argument than the premise and conclusion, there are also the rules of inference that
you use to justify your logic.
Problem: if the inductive justification for induction is not circular, then neither is the counterinductive
justification for counterinduction.
In these arguments, the method of inference is at stake. These arguments are circular in a broader sense
because they use the very method of inference at stake to prove their truth.
Level of knowledge Kind of belief Examples
3 Predictions and generalizations The sun will rise tomorrow.
The sun will keep rising.
2 Descartes’Problem ^Beliefs about the external ^There is a white board up