Empiricism, Skepticism, and Hume

From empiricism to skepticism

Locke’s empiricist position that knowledge can only come from experience opens a gap between experience and reality.

Berkeley exploits this gap to draw idealist conclusions.

The Scottish philosopher David Hume (1711-1776) goes from empiricist principles to skeptical conclusions.

Hume and induction

Our focus is just one of Hume’s contributions to philosophy: the problem of induction.

Note Hume didn’t use the word ‘induction.’ He called them ‘arguments from experience.’

We’ll focus on the basic type of such arguments, called inductive generalizations, simple inductions, or induction by enumeration.

These are arguments of the form:

Simple Induction

‘All observed A’s are X.’ ‘Therefore, all A’s are X.’

Examples:

No dog I’ve ever seen or heard of flies.

Therefore, dogs don’t fly.

Examples of simple inductions

(These are just the conclusions.) Fires are hot. Water boils at 100°C. Gravity = mm′ / d² The villain of an action movie doesn’t win. The sun will rise tomorrow. Professors don’t teach class naked.

What’s the problem?

The ‘problem’ of induction is that it is not a logically valid argument.

It is always logically possible for the premises of a simple induction to be true, while the conclusion turns out to be false.

Indeed, such counter-examples are common.

Example of bad simple induction

‘Every swan I or anyone I know has ever seen or heard about has been white.

Therefore, all swans are white.’

Naked professor shocks students

A Chinese professor shocked his students by stripping naked during an art class. Prof. Mo Xiaoxin, 56, was trying to emphasize the “power” of the body and “challenge taboos.”

Mo, 56, arranged for four other models, including a man and woman in their 70s or 80s, to strip naked in front of the class while he lectured.

He also invited students to take off their clothes.

“Professor Mo appeared emotionally excited at the time,” a student said.

More bad inductions

Everyone in my town speaks English. So, everyone speaks English.

My aunt, mother, and grandmother are bad drivers.

So, all Asian women are bad drivers.

Still more bad inductions

‘All moments I have experienced have been before 2016.

Therefore, all moments will be before 2016.’

‘All moments I have experienced, I have been alive.

Therefore, all moments will be moments when I will be alive.’

The inductive chicken

‘Every day I have been on this farm, I have woken up, scratched around, and been fed. Therefore, every day I will be on this farm, I will wake up and be fed.’

The inductive chicken

Bertrand Russell: “these crude expectations of uniformity are liable to be misleading. The man who has fed the chicken every day throughout its life at last wrings its neck instead, showing that more refined views as to the uniformity of nature would have been useful to the chicken.”

Inductive errors in science

The ancient Greeks thought the stars never changed.

But – very rarely in human timespans – stars can explode.

Alchemists and chemists thought it impossible to make organic chemicals from inorganic ones.

Then, in 1828, F. Wöhler made urea from ammonia chloride and potassium cyanate.

This is clearly bad…

Every time I or anyone else has ever looked into my refrigerator, the light is on.

Therefore, the light is always on in my refrigerator.

…so isn’t this bad too?

‘Every time I or anyone I know has ever seen a tree fall, it makes a sound.

Therefore, anytime a tree falls it makes a sound.’

Hume: there is no difference

Every time I or anyone else has ever looked into my refrigerator, the light is on.

Therefore, the light is always on in my refrigerator.

Every time I or anyone else has ever seen a tree fall, it makes a sound.

Therefore, anytime a tree falls, it makes a sound.

The problem of induction So let’s sum up. We use simple induction

all the time. We use it all the time in common sense

(we drink water because it has relieved thirst in the past).

We also use it all the time in science (electrons are always negatively charged).

But it isn’t good reasoning: it often leads to false conclusions!

Hume vs. induction

Hume says if induction was valid reasoning, you’d have to assume that:

‘The A’s you haven’t seen are like the A’s you have seen.’

Hume: “Instances, of which we have had no experience, must resemble those, of which we have had experience, and that the course of nature continues always uniformly the same.”

In Hume’s words:

“all inferences from experience suppose, as their foundation, that the future will resemble the past … . If there be any suspicion, that the course of nature may change, and that the past may be no rule for the future, all experience becomes useless, and can give rise to no conclusion.”

Is induction logical?

As our many examples of bad inductions show, we cannot generally assume that the future will be like the past, or that nature is uniform.

So: what reason do we have to conclude that nature is uniform? (How is thinking this way logical?)

Hume’s argument

If it is logical to believe in the uniformity of nature, then it should be based in either valid proofs from self-evident tautologies (deduction), or probability (induction).

It’s not tautological: “nature is uniform” is not like “A=A” or “stallions are male horses”; it could be false.

So it isn’t deduction: there is no valid argument deducing that things we haven’t seen must be like things we have seen, not without appealing to experience.

Justifying induction inductively?

But if it’s based in experience, that’s more induction. – ‘It works!’– ‘It’s always been that way!’

To appeal to past successes is to make the assumption all over again (and that is begging the question).

It’s not probable, either, since we only measure probability by past experience.

Hume’s conclusion

If we have no logical reason to believe in the uniformity of nature, then we have no logical reason to believe inferences from experience.

And that is what Hume concludes: we have no logical reason to believe induction.

The implications of Hume’s position

This is an unbelievable conclusion.Hume is saying we have no reason to

think the sun will rise, or that fire is hot, or that trees make sound when they fall.

You may think: “this is how crazy philosophy makes you.”

Hume would agree!

Hume’s words (Treatise I, vii):

‘Fortunately it happens that, since reason is incapable of dispelling these clouds, nature herself cures me of this philosophical melancholy, either by relaxing this bent of mind, or by some lively impression of my senses which obliterate all these chimeras. I dine, I play a game of back-gammon, I converse and I am merry with my friends; and when after three or four hours’ amusement, I would return to these speculations, they appear so cold and strain’d, and ridiculous, that I cannot find in my heart to enter into them any further.’

Hume’s skeptical conclusions

We can’t help believing simple inductions. But simple inductions aren’t logical. Animals seem to make simple inductions. Animals aren’t logical. They learn by habit and custom. So Hume concludes we are like animals: Simple

induction is not reason, but habit.

Hume’s skepticism about causality

Hume also applied this same skepticism about the unobserved to cause and effect.

The idea of cause and effect is not something we see, it is something we infer based on constant conjunctions of ideas.

So we have no knowledge of cause and effect, only habits which make us expect there to be causality.

Hume on science Scientific (like all empirical) thinking is

founded not on logic, but on an animal assumption that the unobserved will resemble the observed, for which no logical reason can be given.

But this doesn’t mean that all inductive thinking is equally good (or bad). The wise thing to do, Hume thinks, is to reason consistently with this assumption, since it is natural and irresistible.

Hume’s overall philosophy

We do not have knowledge about the world beyond what we are seeing now: there is no logical justification for our beliefs to be true, and no certainty outside of the tautologies of math.

But: we do have nature, which makes some beliefs follow by habit and some morals follow by feeling.

Hume’s skeptical naturalism

It is not in our power to control our beliefs strictly by reason.

Fortunately, nature gives us beliefs in experience and causality.

As long as we keep our beliefs within the habits of nature, we don’t need knowledge.

Beliefs that go beyond nature we should suspend judgment about.

Example: Miracles

Why believe a miracle report?– Because I have experience that reports of

witnesses tend to be true. My belief is based on inductive generalization.

Why not believe a miracle report?– But the inductive evidence against any

miracle will be far stronger than the inductive evidence for it. I have lots more experience of people being wrong, misled, tricked …

Man’s place in nature

Man is not “Made in God’s Image”– Our thinking is a natural faculty. There’s no

basis for thinking of humans as privileged; instead, we should be viewed as part of the natural world.

Man is a subject of empirical study– Human beings, like the natural world, can be

understood only by observation, experiment, conjecture, and generalization.

Hume’s practical outcome

Furthermore, once we stop worrying about finding an impossible justification for beliefs and morals outside or beyond human nature, then we will realize that beliefs and morals are the result of human nature, and nothing more.

We’ll understand ourselves. This, Hume thinks, will bring us peace from the worries of skepticism.

A solution to Hume’s problem: the law of large numbers

Many philosophers have proposed solutions to Hume’s skepticism about induction. Here’s a promising one (from Donald Williams and David Stove): The vast majority of large samples are

representative of the population. So, given a large enough sample, the

population will be approximately the same.

Generalizations

For example:Every dog that I’ve ever seen (except for one) has been able to bark.

So almost all dogs bark.

Or:52% of likely voters surveyed said they will vote for Clinton.

So, 52% (+/-2%) of voters will vote for Clinton.

The law of large numbers is a deductive truth about probability The claim “the vast majority of large

samples are representative of the population” is not known from experience.

It is a mathematical principle known as the law of large numbers: roughly, that if the probability of something occurring is X percent, then over the long run the percentage of times that happens will be close to X percent.