V. Humean Skepticism
Some info about Hume:
David Hume (1711-1776)
- British philosopher and historian.
- Considered the greatest philosopher to write in the English language
- Greatest Philosophical Work: A Treatise of Human Nature (1739) (began when he was 23 years old)
- Famous Doctrines: empiricism; skepticism (about causation, induction, the external world, the self); "no necessary connections between distinct existences"; "you can't derive ought from is"
- Immanuel Kant said that reading Hume "awoke me from my dogmatic slumber."
- Adam Smith wrote of Hume that "upon the whole, I have always considered him, both in his life-time and since his death, as approaching as nearly to the idea of a perfectly wise and virtuous man, as perhaps the nature of human frailty will admit."
A. Two Kinds of Skepticism
1. about Knowledge
Descartes was not a skeptic. True, he dug us into a skeptical hole, but he thought he had arguments (concerning God) to save us. If you're convinced by the arguments that got us into the skeptical hole, but not convinced by those that are supposed to get us out, Descartes may turn you into a skeptic. In any event, this is skepticism about knowledge. Skepticism about knowledge is the view that we very little knowledge.
If knowledge requires absolute certainty, certainly beyond any shadow of a doubt, then maybe it's not so surprising and not such a big deal that we don't have much of it. So long as we can have some justification for our beliefs, and we can still say that some beliefs that some people have are more reasonable than other beliefs other people have, then who cares if none of it is ever properly called knowledge?
2. about Justified Belief
Hume is here to shatter our hope that we can even have many reasonable beliefs. So Hume isn't just a skeptic about knowledge. He is a skeptic about justified belief. He thinks we have it a lot less that we thought we did. His argument for this skepticism comes in the form of his so-called Problem of Induction.
B. Some Distinctions Among Propositions
1. relations of ideas vs. matters of fact
(a.k.a.: analytic vs. synthetic statements)
relations of ideas - statements that are made true simply in virtue of the concepts contained in the statement
(relations of ideas are also called 'analytic' statements)
In his glossary (p. 569), Sober defines an analytic sentence as "one whose truth or falsehood is deductively entailed by definitions."
Some example of relations of ideas / analytic truths:
'All bachelors are unmarried.'
'All sisters are female.'
'All triangles have three sides'.
'The internal angles of a triangle add up to 180 degrees.'
'Either it is raining or it is not raining.'
These statement are relations of ideas because they are made true simply in virtue of the concepts contained them. You could deduce that they are true simply by the definitions of the words in the sentences.
matters of fact - statements that are not relations of ideas
(matter of fact are also called 'synthetic' statements)
Sober defines an synthetic sentence as one that is not analytic.
Some example of matters of fact / synthetic truths:
'The earth revolves around the sun.'
'Either it is raining or it is snowing.'
'All bachelors live in messy apartments.'
'Every human being will die someday.'
'If you throw a brick at a window, the window will break'.
2. a priori vs. a posteriori
This is an epistemological distinction. (That means that it has to do with knowledge.)
a priori statements - statements one can know, or be justified in believing, without needing to appeal to any particular experience to serve as evidence for the statement.
From Sober's glossary (p. 569):
"a priori - A proposition that can be known or justified independent of sense experience. An a priori proposition can be known of justified by reason alone (once you grasp the constituent concepts). Truths of mathematics and definitions are often thought to be a priori."
a posteriori statements - statements that can be known, or justified, only by experiencing the world.
From Sober's glossary:
"a posteriori - A proposition that can be know of justified only by experience."
A Humean Thesis: a statement is a priori only if it is a relation of ideas.
(Notice this thesis is an 'only if' and not an 'if and only if'. This is because presumably there are some relations of ideas (e.g., of mathematics) that are too complex for us even to understand, let alone to know.)
C. Hume's Problem of Induction
Hume's Main Thesis:
we are not at all justified in using induction; we have no reason
to believe the conclusion of any inductive argument.
Examples of Inductive Arguments
I've observed many emeralds, and each has been green.
Therefore, all emeralds are green.
Every time I have eaten bread it has nourished me.
The next time I eat bread it will nourish me.
Every time I have pressed the brake pedal in my car, my car has stopped.
The next time I press my brake pedal, my car will stop.
So Hume's Main Thesis is saying that we have no reason at all to believe the conclusion of arguments like these. If Hume is right, then it would, for instance, be just as rational to press the radio dial in your car in order to stop it as it would to press the brake pedal.
Hume's Main Thesis implies that we have no reason to believe any of the following sorts of statements (since we believe such things using inductive arguments):
- any matter of fact about the future
- any "general" matter of fact (such as that all emeralds are green)
Hume is not merely saying that we can never know such things. His claim is much more radical. It is that we have absolutely no reason at all to believe them. It would be just as reasonable to believe the opposite.
1. The Principle of the Uniformity of Nature (PUN)
PUN: the future will resemble the past.
Hume claims that all inductive arguments make use of PUN. He will argue that we have no reason at all to believe in PUN. It follows from this that we have no reason at all to believe the conclusion of any argument that makes use of PUN.
2. Hume's Argument that There Is No Reason to Believe in PUN
Here is my formulation of the argument. I say it differently from Sober, but the arguments amount to the same thing.
(1) If there is any reason to believe in PUN, then our justification for PUN is either a priori or a posteriori.
This premise is true because these two options -- a priori and a posteriori -- are, by definition, exhaustive.
(2) Our justification for PUN is not a priori.
Because if it were a priori then, given "A Humean Thesis" above, PUN would be a relation of ideas. But is not a relation of ideas -- we can coherently imagine PUN being false. There is no contradiction is supposing that the future is utterly unlike the past in the way nature operates. It is at least coherent to suppose that, e.g., all of a sudden brake pedals no longer stop cars or bread no longer nourishes.
(3) Our justification for PUN is not a posteriori.
An a posteriori justification for PUN would have to come in the form of an argument with premises about our experience and with a conclusion that is PUN itself. Such an argument must be either a deductive or an inductive argument. That is, our reasons for believing PUN would either have to deductively entail that PUN is true, or would have to inductively support PUN (i.e., make PUN more likely).
Deduction won't work. Because none of our experiences, which of course are only of the past and the present, deductively imply anything about the future. (See Hume quote, p. 255).
Induction won't work. Well, we usually use inductive arguments to draw conclusions about the future based on our experience of the past. And this is exactly what we're trying to do here -- we're trying to infer PUN (a claim about the future) from our experiences of the past. But we can't use an inductive argument to establish PUN, because all inductive arguments for conclusions about the future presuppose PUN. They need to use PUN as a premise. So they certainly can't establish it as a conclusion. That would make the argument circular.
[See Sober, p. 186 for more on this.]
(4) Therefore, there is no reason to believe in PUN.
[follows validly from (1), (2), and (3)]
(5) If there is no reason to believe in PUN, then there is no reason to believe the conclusion of any inductive argument.
Because any inductive argument implicitly uses PUN as a premise.
(6) Therefore, there is no reason to believe the conclusion of any inductive argument.
[follows validly from (4) and (5)]
This -- (6) -- is an extremely radical conclusion. It means that not only do you not know the following, you in fact have no reason to believe any of the following:
'It is always colder during the winter.'
'The next time I press my brake pedal, my car will stop'
'The next time I drink water, it will quench my thirst instead of burning my throat like acid'
'My next breath of air will not kill me.'
Hume's conclusion shows that the beliefs above are no more reasonable than these:
'Next winter it will be sunny and 85 everyday.'
'The next time I press my brake pedal, my car will explode'
'The next time I drink water, it will burn my throat like acid'
'My next breath of air will kill me.'
D. Hume's Anti-Rationalism
Hume's point is not that we should stop trusting experience and stop using induction. He knows we will continue to use induction. Rather, his point is to show that this very basic form of reasoning is not rationally justifiable. All we can say about it is that we in fact do use it, not that we rationally ought to. It is just part of our nature to reason this way. Hume writes (p. 254):
"Even after we have experience of the operations of cause and effect, our conclusions from that experience are not founded on reasoning of any process of the understanding."