Thinking About Thought (Part 2)

Here I continue to read through An Enquiry Concerning Human Understanding by David Hume.  You can read Part 1 here.  This will likely be a five-parter.  I have been finding this book to be quite interesting, and I hope you will agree.

Chapter IV – Sceptical Doubts Concerning the Operations of the Understanding

Part I

So far Hume has focussed on the different perceptions of the mind: impressions, which include things like hearing, seeing, feeling, willing, loving, etc, and ideas, which are copies (recollections) of impressions.  Then he dwelt on associations of ideas, and defined them as resemblance, contiguity, and cause and effect.  In this chapter, he turns to those objects that we actually think about.  He divides the chapter into a series of questions:

Question 1: What kinds of objects do we think about?

Answer: There are two:

1. Relations of ideas

2. Matters of fact

Relations of ideas are primarily mathematical, and are not the focus of the chapter.  An example of a relation of ideas is ‘two plus two equals four’.  Here there is a relationship between two sets of two, and a logical outcome.  The chief identifiers of relations of ideas are that they can be worked out by reason alone (even if nothing else existed, two plus two would still equal four – experience is not necessary for one to discover this), and contradictions are not logically possible (two plus two cannot equal anything but four).

Matters of fact are everything else, from the knowledge that the sun will rise tomorrow, to the fact that a billiard ball striking another billiard ball with such-and-such force at such-and-such an angle will cause the other billiard ball to move at a certain velocity in a particular direction.    The chief identifier of matters of fact are that a contradiction can be fathomed by the mind (we could easily conceive of a world in which the sun does not rise tomorrow, but we could never conceive of a world in which two plus two equals five).

Hume argues that, if ‘the sun will rise tomorrow’ and ‘the sun will not rise tomorrow’ can both be readily understood by the mind, then there is no way to disprove one statement or the other.

Yet this is completely foreign to how we live our lives.  We all know that the sun will rise tomorrow; in fact, we can calculate the exact time it will rise at different places on the planet.  But if the contradiction is logically possible, then how could we possibly know such a thing?  This is where things get interesting.

Question 2: How do we know that matters of fact are true?

Answer: Matters of fact are a product of cause and effect.  This allows us to extrapolate beyond what our senses or memory tells us, and can convince us that one statement is true over another.  A man finds a watch on a deserted island; he can then make the intellectual leap that humans had once been there.  Your friend goes to France.  How do you know he is there?  He told you he would; he is trustworthy; he sent you a letter.  You have no experience or sensation of his being in France, but you know it to be a matter of fact anyway, because you have the past experience of his being trustworthy.  Cause and effect.

The relationship between cause and effect can be categorized in a few ways:

1. Cause and effect can be near or remote (Hume doesn’t explain this, but I think he means that the cause and effect could be separated spatially or temporally by varying degrees)

2. Cause and effect can be direct or collateral (direct would be something like a lit match starting a fire; collateral would be the resultant heat and light of that fire).

Question 3: If all matters of fact are known through cause and effect, how do we know how the relationship between cause and effect works for each matter of fact?

Answer:  In a previous chapter, Hume categorized cause and effect as one of the ways that ideas are associated with one another.  Prior to that, he argued that ideas are only meaningful if they are based on experience.  Here we have the culmination of those arguments: if matters of fact are based on cause and effect, and cause and effect are associations of ideas, we can only know the nature of the cause/effect relationship through experience.

In other words, we can never understand a cause and effect relationship a priori.  ‘Let an object be presented to a man of ever so strong natural reason and abilities; if that object be entirely new to him, he will not be able, by the most accurate examination of its sensible qualities, to discover any of its causes or effects.’  The first time you encountered water, you could never have known that it would suffocate you.  You could only learn that through the unfortunate experience of others.  In the same way, you could never, by beholding any object for the first time, know of its cause or its effects simply by reflecting on what your senses tell you about it.

In brief, Hume’s proposition is that ‘causes and effects are discoverable, not by reason but by experience.’

Who could possibly suspect that magnets would stick together, or dynamite could explode, or that large stones and small stones would fall at the same rate, or that a hand could not pass through a door, without first experiencing those things or things that are similar?

But, we object, surely I know that a hand can’t pass through a door, without first trying it?  Ah, says Hume, but when you were first born, were you born with that knowledge?  Or, through habitual experience since birth, have you come to learn that two bodies of matter cannot occupy the same space?  Surely you had to learn that a hand cannot pass through a door, but it can pass through water.  All of this knowledge of cause and effect is based on experience, even if those experiences are so ingrained in us that we never thought of them as being experiences.  Such ingrained experience Hume refers to as custom. 

Question 4: Why is it impossible to reason out the cause and effect relationship?

Answer: It is because every effect is completely distinct from its cause.  To the mind, there is no reason why, when a billiard ball at rest is struck by another billiard ball, the resting ball should move with an equal and opposite force, or be jumped over by the moving ball, or cause both balls to come to a rest, or explode, or any other unlimited number of options.  The cause (a rolling billiard ball) is logically distinct from the event it produces, and only experience will show us what the true effect will be. 

Even if we see the outcome, there is no reason to believe that the same thing will happen the second time.  Again, the effect is distinct from the cause.  Only custom, the repeated experience of the same effect from the same cause, would ever give us confidence to believe in a matter of fact. 

But, we object, Hume was saying this in the 1700s, well before we understood how everything works.  Today we can predict the effects that a cause will have, just by looking at what the objects are made of.  For example, we have never had the experience of preventing a meteor from colliding with our planet, but we can come up with numerous causes that will have the effect of deflecting such a meteor, and those causes should work.

Ah, says Hume, but the reason you know it will work is because you have determined cause and effect relationships for other similar things.  And ultimately, even if we could manipulate matter to create a brand new object never before seen, and use a priori reasoning to determine how it would work, that a priori reasoning is itself based on a posteriori reasoning when it comes to how matter fundamentally works.  There is no reason for matter to behave the way it does; we had to observe/experience it in order to develop ideas about how it causes effects.

One interesting consequence of this is that we can never know the ultimate cause to everything that we see.  We can narrow every effect down to some general causes, such as gravity and other such physical laws.  But inspection of the natural world will never bring us to the first cause, because we have no experience of any such first cause.  ‘The most perfect philosophy of the natural kind only staves off our ignorance a little longer: as perhaps the most perfect philosophy of the moral or metaphysical kind serves only to discover larger portions of it.  Thus the observation of human blindness and weakness is the result of all philosophy.’

In other words, because we have no experience of the first cause that got the whole chain of causes going (or even if there was a first cause), science and philosophy (and therefore human knowledge) are necessarily limited.  Not even math can save us from this.  Although we can use equations to peer back in time, ultimately the validity of those equations rests on experience.  There is no logical reason for choosing one equation to describe reality over another, apart from experience, and our experience of the first cause is, as already mentioned, lacking.

Part II

Hume is not content to leave experience unmolested.  After proving that the cause and effect relationship is determined by experience, and not by reason, he goes on to ask:

Question 5: What is the foundation of all conclusions from experience?

Answer: Hume here is not concerned with providing a definite answer to this question, but rather proceeds to show that, whatever the foundation, it is not based on reasoning or understanding.  The larger question at play here is whether we can ever expect the future to conform to the past.  Cause and effect relationships are learned through experience, but the experience is of past cause and effect relationships – can we reasonably project that cause and effect relationship into the future?

Hume begins with an illustration: suppose, through experience, we learn that a slice of bread is good for nourishing the body.  We now have certain and definite information regarding that particular slice of bread that we ate.  There was absolutely nothing about the appearance of the bread itself that would indicate it would be nourishing; only experience could teach us that.  Now we have before us a second slice of bread, similar in appearance to the first.  Since the first slice of bread was nourishing, can we conclude that this second slice of bread will also be nourishing?  Hume says that the answer is ‘no’.  There is no logical connection between the appearance (i.e. ‘sensible qualities’) of the first slice of bread and its ‘secret’ nourishing powers.  Likewise, then, there can be no link between the sensible qualities of the second slice of bread and its ‘secret’ powers.  The properties of that slice, which are hidden from our senses, could very well be poisonous. 

Our minds, however, always do make the jump from the sensible qualities and secret powers of one object and the like sensible qualities of a second object, to the second object having like secret powers.  But whatever this intellectual leap is, which our minds always make, it is not based on reason or understanding.

In short, Hume says that our minds somehow connect two very different propositions and treat them as if they were the same:

Proposition one: ‘I have found that such an object has always been attended with such an effect.’

Proposition two: ‘I foresee, that other objects, which are, in appearance, similar, will be attended with similar effects.’

Our minds infer proposition two from proposition one, but this is not based on reasoning.

There are two types of reasoning:

1. Demonstrative reasoning – concerns relations of ideas (mathematics)

2. Moral reasoning – concerns matters of fact and existence

What Hume wants to know is if either of these types of reasoning is sufficient to provide us with certain knowledge about the future, given past experience.  Obviously it cannot be demonstrative reasoning, since the contradiction of any past experience can be held by the mind with ease (unlike two plus two equals five).  Could it be moral reasoning?  One possible model of moral reasoning could be to use probability: such-and-such happened in the past x times out of y, and therefore has probability z of happening in the future.  But, cries Hume, even such probability-based thinking is rooted in the belief that the probabilities of the past will be the same as the probabilities of the future!  So we end up with a circular argument.

Hume provides four additional reasons that predictions of the future based on experiences of the past cannot be founded on reason. 

1. The main lesson of experience is that ‘from causes which appear similar we expect similar effects.’  If the foundation of this conclusion was reason, we should expect it to always hold true, but it clearly does not.  Out of a dozen eggs, all of which have identical sensible qualities, there might be such variation in flavour that some are unpalatable to us.  In our scientific age, we might object that the chemical composition of those eggs must have differed, but this makes no difference to Hume’s argument, as the chemical composition is still a sensible quality.  Its ‘secret’ power of taste has little to do with its chemical makeup, for why should such a chemical taste egg-like?  And even if all twelve eggs were chemically identical, they still will not all taste the same, based on changes in our taste buds, overall health, other odours in the room, and other foods on our palate.  When twelve identical eggs have differing tastes, what process of reasoning do we use to link the next egg with egg-taste?

2. Reason only needs one example to form a firm conclusion.  Experience requires numerous examples.  No one would declare that all eggs are equally delicious after eating only one.  They must first try many.  But reason need only reflect on a single circle to come up with its conclusions, and those conclusions would not be altered by reflecting on additional circles.

3. What about inference?  After doing hundreds of identical experiments and finding very similar results, can we not infer that, if we do the same experiment in the future, we will get the same results?  This is certainly what we do, but it is still not founded on reason.  For all inference assumes that the future will conform to the past; if the course of nature were to drastically change tomorrow, then all inference based on experience becomes meaningless.  ‘Let the course of things be allowed hitherto ever so regular; that alone, without some new argument or inference, proves not that, for the future, it will continue so.  In vain do you pretend to have learned the nature of bodies from your past experience.  Their secret nature, and consequently all their effects and influence, may change, without any change in their sensible qualities.  This happens sometimes, and with regard to some objects: Why may it not happen always, and with regard to all objects?  What logic, what process of argument secures you against this supposition?’

4. Have you ever seen a toddler put their hand over the flame of a candle?  Their immediate reaction is to move their hand away.  They then associate all things that have the sensible qualities of a candle flame with the burning pain they felt in the past.  This is identical to our own process of experiential learning.  If you say that the foundation for this is reasoning and understanding, then you have given such faculties of philosophical speculation to the toddler, which is absurd. 

And so, concludes Hume, experience is founded on some principle that is not reason.  What, then, is that principle?

Chapter V: Sceptical Solution of These Doubts

Part I

After clearly and eloquently showing that experience is the foundation of all of our ideas and, therefore, our understanding of the world, Hume then seeks to determine the principle that stands apart from reason and that allows us to make any conclusions from experience.

But first he makes an interesting digression about scepticism.  Apparently, in the 1700s the sceptical school of thought, which constantly doubted and always tried to limit human understanding, was under attack.  Hume argues that it was despised because it was the most innocent.  People often begin with pre-conceived biases; they then use religion or philosophy to validate and indulge their bias.  Conclusions can then be overstated.  Sceptics seek to state only that which can be known; they doubt, and suspend judgment, and avoid hasty conclusions, and renounce ‘all speculations which lie not within the limits of common life and practice.’  I find this definition of scepticism remarkably helpful, as it outlines all of Hume’s biases, which he declares are innocent because they all revolve around truth.  Whether he is right in his assessment of his biases remains to be seen. 

Next Hume turns to the problem raised in the last chapter, in which he developed a negative philosophy against reason but did not provide a substitute.  In this chapter he identifies the principle that allows us to connect past experience with future prediction: custom, which he also calls habit.

Consider a man suddenly brought into this world with his full faculties of reason.  He is immediately confronted by objects, but beyond seeing how one object succeeds another or one event follows the next, there is nothing more he could discover.  He could never, at this stage, reason out cause and effect.  But he continues to live, and notices that one event is always conjoined with another.  This experience now teaches him to expect the second event if he sees the first, or to infer the first event if he sees the second.  He has now perceived a cause and effect relationship.  But why this relationship works as it does is never brought to his mind, nor does he need it to make predictions.  The only thing that he needs to survive in his world is, not reason, but custom: the habitual experience of the same event following the same cause.

‘Without the influence of custom, we should be entirely ignorant of every matter of fact beyond what is immediately present to the memory and senses.’  If we relied solely on reason, we would be constantly confronted with an infinite number of possible effects from any single cause; we would be left adrift in life, hopeless, unable to make any predictions beyond what we are currently experiencing and, therefore, completely incapable of action.  Custom, a completely reasonless process, ingrains in us that there is always the same effect from any one cause; despite the fact that this is disprovable, it is essential to our survival and allows us to make decisions and act on those decisions.

Reason is bound by contradiction; custom is bound by only one thing: that ‘some fact must always be present to the senses or memory.’  If we never find an ancient city in the desert, we can never know that there once was an ancient city in the desert.  But if we were to find such a city, custom tells us that such a city must have been built by people, and that they must have died out or abandoned their city long ago. 

What this means is that, if you have some conviction about truth, that conviction must be bound to a fact.  If I were to ask you about the truthfulness of that fact, it must be bound to another fact, and another, and another.  This cannot go on forever; at some point, the inquiry would terminate in a fact present to your senses or memory.  If such a fact does not exist, ‘you must allow that your belief is entirely without foundation.’

Any belief is founded on a chain of inferences, but since matters of fact have no foundation in reasoning, these links are only based on experience.  Thus any belief in which the chain does not terminate in a fact present to the senses or memory disintegrates under its own emptiness.

Here, then, is the result of Hume’s argument from experience: ‘all belief of matter of fact or real existence is derived merely from some object, present to the memory or senses, and a customary conjunction between that and some other object.’ 

For example, if experience teaches me that snow and cold are conjoined (that is, that snow is always cold), if I see from the window that it is snowing, custom operates such that I expect that this snow also is cold.  But custom doesn’t just cause me to expect that it is cold, but also to believe that it is cold, and without any further proof I would bundle up in winter clothes before going outdoors.  Upon going outside, I would discover that my belief was correct.  Such is the power of belief.  But my belief would have been unfounded if I hadn’t looked outside at all.

Part II

But what is the nature of belief, and the customary conjunctions that allow it?  To explore belief, Hume returns to his definition of ideas as copies of impressions.

The imagination, Hume already told us, seems unlimited it what it can conceive, but in fact it is limited by what we experience.  We can create any number of fictional worlds, so long as they have some connection to our own experiences; we cannot form ideas of anything we have not experienced.  Unicorns and all manner of monsters can prance about in our minds, but even they are based on experience.  Now, if our minds can develop all sorts of fantasies and fictions, how is it ever able to sort out fact from fantasy?  Although I can see a polka-dotted fire-breathing giant taco, why can I never, no matter how hard I try, persuade my mind to believe in its existence?  What is the difference, to the mind, between fiction and belief?

The difference cannot be an idea that is connected with belief but not with fiction; otherwise, our minds would easily be able to attach that idea to fiction to fool us into belief.  Instead, the difference must be one of feeling or sentiment which attaches itself to belief, but not to fiction, and which we have no control over.

If ‘the sun will rise tomorrow’ and ‘the sun will not rise tomorrow’ can both be imagined by the mind, and there was no sentiment attached to the former that could distinguish belief from fiction, then there would be no difference between those two statements and we would hold them both as equally valid.  But custom conjoins ‘sun’ with ‘morning’ and ‘rise’, such that when our alarm goes off at six in the morning, we believe that we will soon see the sun peek over the horizon, and not the alternative. 

But what is this sentiment?  It is difficult to define, like all sentiments are.  We call it the sentiment of belief, and that is good enough.  We could describe it as ‘nothing but a more vivid, lively, forcible, firm, steady conception of an object, than what the imagination alone is ever able to attain.’ 

Hume then goes off on what appears to be a tangent, but quickly brings it back to belief.  As a general law of the mind, Hume states that, if an object is presented to the senses or memory, and it moves the mind to an associated idea, that idea will appear to the mind in a steadier and stronger conception, than if the object had not appeared to the senses or memory in the first place.

To prove that this is a general rule, he goes through the three ways in which ideas are associated, which I discussed in my last post.  They were: resemblance, contiguity, and cause and effect.  The law will only be general if it holds for all three associations.

Let us say that we wish to think of a friend.  If we see his picture, the thought of our friend and all of the emotions that come along with it carry a greater force than if we saw a picture of a stranger, or saw no photo at all.  This proves the law for resemblance.  (Hume also discusses Catholicism and its use of images to bring us to a greater force of meditation on God).

In the same way, objects that are close to my home bring up stronger emotions and ideas of home, than objects that are thousands of kilometres away.  Objects thousands of kilometres away may bring to mind an idea of something close to home, and thus of home itself, but it will always be weaker than if I were to actually see the corner store that loomed over so much of my childhood.  So much for contiguity.

If my friend was to die, and I was to see his son, his son would bring to mind stronger ideas of my friend than me just thinking about my friend.  So much for cause and effect.

With this general rule proved, Hume argues belief plays a role.  In associations of ideas, seeing my neighbourhood will only bring to mind a stronger idea of my home if I actually believe that my home exists.  Belief is necessary for reinforcing the connection between ideas. 

And this is the way it needs to be.  The imagination is limited by experience, but can still develop all sorts of cause and effects never found in nature; reason alone can never allow us to decide between them.  But experience also fosters belief, which strengthens the connection between an object presented to our senses or memory, and an associated idea.  Fictional cause and effect relationships do not receive such strengthened bonds, and thus appear to the mind with less force than reality.  Thus we find harmony between nature and our ideas – we are able to enhance conjoined ideas that reflect how nature works.  This quickly becomes essential for our survival – we can conjoin exposure to cold with hypothermia or to pleasure; we can conjoin rotting meat with disease or to nourishment, but our beliefs allow us to filter through all of the logical possibilities, find those ideas that are conjoined by custom rather than fiction, and find what we need to survive.

But this is in no way based on reason.  Rotting meat will not always make us sick; the human constitution could miraculously change such that long-term exposure to the cold does not kill us.  There is no way to know, using reason, what the cause and effect relationship in the future will be.  But this is not a bad thing; in fact, argues Hume, it is very good that we use beliefs formed by custom, rather than reason, as reason is slow, prone to error, and is a product of maturity, whereas survival requires something that is quick, works most of the time, and can be found in even an infant.

Chapter VI – Probability

This is a short chapter, and will round off today’s summary. 

Suppose you had a six-sided die.  On four sides there was one dot, on the other two there were two dots.  When you roll the die, how does your mind operate to predict what will happen?  Hume relies on the discussion of belief in the previous chapter.  The mind approaches the die as if all six events (sides of the die) are equally probable.  However, four events are identical (one dot) and the other two are identical (two dots).  Belief, which is a sentiment that colours the idea with greater or lesser force, attaches greater strength to the idea of one dot than to the idea of two dots, because the mind turns to the idea of one more than to that of two.  Both are possible, but one is more probable, and we thus have greater assurance that the event will be ‘one’ instead of ‘two’.  If it were a thousand-sided die with 999 ‘one’ and 1 ‘two’, our belief in ‘one’ would have even greater force.  Such is the probability of chance.

There is also probability in causation.  Some causes never vary in the effects they produce – if a human were to breathe underwater, he would always drown – but some causes are not uniform.  Some medicines, for example, only work some of the time.  When we need to project a cause and effect relationship into the future, but past experience shows different outcomes for the same cause, what do we do?  We simply take the ratio of events in the past, and project it into the future, with the greater frequency of the same event being connected to the stronger belief.

Implications

This has turned into a rather lengthy article, so I will not belabour the implications this week.  But several certainly jump out at me:

1. Science is based solely on experience and not on reason.

The entire experimental method is based on observation and experience.  It is simply trying to determine the cause and effect relationships we see, and develop models to predict future events.  What makes science so difficult is that the causes are not always readily detectable, and there are always levels of causation that can be explored (such as the levels in an animal of gene, genome, cell, tissue, organ, organ system, organism, relatives, population, species, ecosystem).  Hume completely agrees that science is observation-based.  What he also does, though, is knock it down a peg.  Those who believe that science explains everything find, if not an enemy, at least a challenger in Hume.  First causes lie outside the realm of science; any hubris about the intellectual discoveries of science are deflated; science is merely a highly ordered form of what we all do every single day of our lives, and is not based on reason.

2. The importance of belief

Hume recognizes that we are incapable of believing anything we want.  There are fictions, and there are beliefs.  This is interesting to read in an age where belief, particularly religious belief, is equated with fiction.  The question for Christians is, what are the objects of experience to which religious belief is directed?  Hume says that all beliefs in matters of fact must terminate in an object present to the senses or memory.  If someone were to question a Christian about their faith, at what fact would their chain of inferences terminate? 

3. The importance of revelation

Hume has argued that human knowledge is necessarily limited when it comes to first causes, because we lack any experience of such a cause.  We cannot reason out the universe.  How much less, then, could we reason out God or His salvation history!  Hume has, I think, only strengthened the case for the importance of special revelation over natural revelation.

4. Natural selection

Hume writes well before natural selection was elucidated by Darwin and others, and yet his model of experiential knowledge fits very nicely with the evolution of behaviour.  Since we are descended from animals, it would be expected that we should still have animal-like instincts.  By giving experiential learning such priority over reason, this is exactly what Hume inadvertently suggests.  He rhapsodizes about custom over reason.  Reason would give us certainty if certainty could be known; custom lets us act despite our uncertainty, and even prevents us from realizing that we are uncertain!  Custom is not interested in truth, but rather about survival.  And it can change as experiences change, whereas reason is slow, ponderous, and unchanging.  When he rhapsodizes about custom all I could see was natural selection at work.