Our second philosopher is David Hume, an 18th-century Scottish thinker. A survey given a few years ago asked professors of philosophy which nonliving philosopher they most identified with — whose voice and ideas felt closest to their own — and Hume came in at number one. That is a measure of his enduring popularity among professional philosophers.
Kant, who will be our next thinker, came in at number two. Nietzsche ranked 17th, and our previous philosopher, Descartes, came in at number 18 — which is still not bad company to be in.
We will be focusing primarily on An Enquiry Concerning Human Understanding, published in 1748. There is good news and bad news about this book. The good news is that there is really only one argument in it — if you grasp that argument, you have grasped the whole work. The bad news is that it is a large, complicated, multi-part, and deeply counterintuitive argument that takes an entire book to unfold.
The first part of the book prepares for the argument: setting up the pieces, defining the terms, and explaining the necessary background. The middle section contains the argument itself, and the remainder explores its ramifications and consequences. That structure will guide our approach, which we can divide into three phases: the preparation, which is somewhat dry; the argument, which is strange; and the conclusions, which are genuinely fascinating. Unfortunately, you have to work through the first two to arrive at the third.
What makes the effort worthwhile is that this argument has the capacity — and has had precisely this effect on many generations of readers — to fundamentally change the way you think about yourself and about human nature.
The book is titled An Inquiry Concerning Human Understanding. It is about understanding, and therefore, like Descartes, it is concerned with epistemology. But notice the qualifying word: this is an inquiry concerning human understanding. Hume wanted that in the title. Why? It might seem gratuitous — what other kind of understanding could there be? Why emphasize that this is human understanding?
We do need to emphasize it, and the reason becomes clear when we look back at Descartes. Descartes was not interested in human understanding. He pared us down to our essence, which is a thinking thing. Our humanity, our embodied-ness, our living in a culture, a society, a history — none of that was relevant to thinking for Descartes. In fact, it was a distraction. Being in a body, as many philosophers since Plato have felt, is simply interference; it gets in the way of thought.
For Descartes, a title like this would simply have been An Inquiry Concerning Understanding — an inquiry concerning the understanding of a thinking thing, whether that thinking thing is a human, a computer, an angel, or a porpoise. It does not matter. The goal is to get down to the essence. Hume, however, says at the very beginning of the book — on page four of the Hackett edition — "Be a philosopher, but amidst all your philosophy, be still a man."
Hume thinks it is crucial that we maintain our humanity within our philosophy. The notion that we can strip that away and still have anything left that is real, significant, important, or truly us is, in his view, one of the professional deformations of philosophy. The only kind of understanding we can use, the only kind that matters for us to think about, is our understanding. We have no idea what angelic or divine understanding is, no idea what an alien understanding would be. The only understanding we know is human understanding, and for Hume, this has a profound effect on what thinking itself can be.
Consider the case of vision. If we were conducting an inquiry into vision, or visibility, we would quickly realize that human vision is fundamentally different from an owl's vision, a bat's vision, or a bee's vision. Each species' anatomy — the way their eyes are structured, the way their brains are structured — determines what they can see, how they see it, and how they decode it. The visual world that each species inhabits is fundamentally shaped by its biology.
Humans can only see within a range of approximately 380 to 720 nanometers — the visible spectrum. Anything below 380 or above 720 nanometers, whether ultraviolet or infrared, is simply invisible to us. No matter what we believe, no matter how hard we try, we cannot see beyond those limits.
Hume's insight is that the brain is an organ — a biological object in our skulls that has its own capabilities and limitations, just as our eyes do. Just as there are things we can see and things we cannot, there are things we can think and things we simply cannot. Because of the way our brains are structured, certain things fall within our understanding and certain things lie beyond our ken. No matter how hard we try, no matter how hopeful we are, we cannot think everything.
The problem is that if you are working in science and you make a fundamental error, the world will correct you. If you build a bridge based on the wrong principles of physics, it will fall down. You will get feedback. You will find out about your mistakes.
But if you do metaphysics — if you speculate about the nature of divinity or the fundamental structure of the universe — you never get corrected. Nothing will tell you that you are wrong. And so, for all of history, there have been geniuses, brilliant people who could have benefited humanity in countless ways, who instead spent their lives pouring their thoughts down the drain of metaphysics. They wasted their time thinking they could think thoughts that are, in fact, unthinkable — like straining your eyes to see infrared. They believed themselves capable of achieving genuine metaphysical insight when, according to Hume, such insight is fundamentally inaccessible to human beings.
But how can we prove this? If we do not get feedback the way we do from a bridge falling down, how could we ever find out? Metaphysics doesn't build anything, so nothing will break. That is precisely why so much intellectual firepower has been squandered on the futile pursuit of metaphysical truth — a truth that is simply not within our grasp.
Hume has an answer to this problem. We are going to inquire concerning human understanding in the same way that we would investigate the visible spectrum. To find out the visible spectrum, what would we do? We would study eyeballs. We would cut them open, examine their structure, and look at the relevant portions of the brain. This is precisely what Hume proposes to do for human understanding: he is going to discern the anatomy of the mind.
Not the brain — that was impossible at the time. But he could map the details of the mind itself. He undertakes what he calls a mental geography, a map of the mind. He is going to identify both the contents of the mind and its structure — what is in there, and what we do with what is in there. If we can figure that out, we can know how far we are able to take it.
What this project will do is draw a line — a limit — and a limit does two things simultaneously. It tells us both what is inside and what is outside; both what we can do and what we cannot. If we can establish this limit once and for all, then we can concentrate all of our effort as a species, all of our collective intellectual power, solely on what falls within those limits — solely on the questions we can actually answer — and simply set aside everything that lies outside that circle. This is what Hume wants to achieve with this book.
This is what Kant will call critical philosophy. Kant wrote three books, each of which he called a critique. As we will see in the next lecture, Kant was profoundly influenced by Hume and had enormous respect for him. He thought Hume had not gotten it quite right, but that Hume had the right problem — the right question, even if not the right answer.
Hume did not have a name for this approach, so we will use Kant's term: critical philosophy. Critical philosophy is the project of understanding the structure and nature of the mind in order to draw its limits. Those limits tell us both what we can know and understand, and what we cannot — so that we may give up the latter and focus entirely on the former. This is precisely what Hume's inquiry is meant to produce for us.
Hume begins his examination of the mind by looking at its contents — what is actually inside our heads. Like Descartes and many early modern philosophers of the period from 1600 to 1800, he uses the word idea to describe what we carry in our minds. But he argues that everything in the mind falls into one of two fundamental categories, a kind of dualism. The two major types are what he calls impressions and ideas.
An impression is an immediate, present experience of something. Picking up a pen and feeling it in your hand — that is an impression. Now set the pen down and recollect the feeling of holding it: that recollection is an idea. You can also anticipate picking it up again, imagining how it will feel — that anticipation is likewise an idea. The moment you pick it up once more, you are back to having an impression.
The impression of the pen and the idea of the pen share the same content, and yet they are fundamentally different. For Hume, the difference lies in what he calls relative force and vivacity. The impression is vivid, immediate, and commanding — it is right there, impossible to ignore. The idea carries the same content but has lost its grip; it is weak, like a faded copy of the original.
Think of what it is like to be intensely hungry. The hunger commands your attention; there is no denying it and no getting away from it. Now think of remembering that hunger when you are no longer hungry — you have the same content in mind, but it exerts no force on you whatsoever. That contrast between the gripping immediacy of the experience and the pale recollection of it is precisely the distinction Hume draws between impressions and ideas.
It is worth noting that Hume was a remarkable prose stylist, though his writing can feel ornate and dated to modern readers. He was among the first writers to make a living entirely from his own work — a genuinely new phenomenon at the time. His multi-volume History of England was a bestseller for decades, and he lived off its proceeds. By contrast, Descartes wrote an obsequious letter at the opening of the Meditations begging wealthy patrons to support his habit of staying in bed until noon to think; eventually Queen Christina of Sweden obliged, though that arrangement did not end well.
Hume takes this division and makes a very important claim: every idea you have in your head came from an impression. There is no idea in your mind that did not originate from an impression. This is considered the founding theory of what is known as empiricism.
Empiricism is the school of philosophy that holds empirical data to be the key to knowledge. This stands in contrast to rationalism, which, as the name suggests, places reason at the center of inquiry. Descartes was the founder of early modern rationalism, because he argued that thinking, reasoning, and a priori knowledge were far more important than experience. He acknowledged that we need sensory data in order to know the world, but insisted that we must translate that data into numbers so that the mind can grasp it and reason about it, rather than depending on the senses directly. For Descartes, reason takes priority over sensory experience.
Hume argues the opposite: that the senses are far more important than reason. This debate between rationalism and empiricism is one of the defining disputes of the early modern period, roughly 1600 to 1800. Hume was not the founder of empiricism — that distinction is generally credited to Locke — but he is the thinker who pushed empiricism to its logical conclusion. Anticipations of this debate can be traced all the way back to Plato and Aristotle, with Plato serving as a proto-rationalist and Aristotle as a proto-empiricist, though the parallel is not perfect. Still, while this tension has a long history, it truly took center stage during the early modern period.
Hume emphasizes that experience is the source of everything. He is inspired by John Locke, who had a very famous phrase: "The human mind is a tabula rasa." You may have heard this before — a blank slate. Locke said the human mind starts off as a blank piece of paper, and experience writes on it. If a baby were placed in a sensory deprivation tank from birth — kept alive intravenously but never given any empirical experience — it would never have a single thought. You need input from the senses to have any content in your mind at all. Without that, there is nothing, and you would remain a blank slate your entire life.
Hume offers two arguments in support of this view. In philosophy, you can make all kinds of claims, but you must have an argument to back them up. The first argument is straightforward: if you have never had the impression, you will not have the idea. His example is pineapples. Pineapples were a novelty at the time — explorers had brought them back to England, and people were tasting them for the first time. Before you tasted a pineapple, you simply did not have the idea of what one tasted like; that taste was not in your head.
You could look at a pineapple, examine it closely, analyze it under a microscope, determine its chemical composition, and learn everything about it — and still you could not know what it tasted like until you put a piece in your mouth and received the actual impression. The second argument is more complicated.
An easy response to Hume is: "I have all kinds of ideas in my head that I've never had an impression of. I have the idea of a unicorn, and I've never seen one." This is where Hume moves into the second aspect of his mental geography. The first aspect concerns the contents of our minds — the ideas themselves. The second concerns what we can do with those ideas: the powers or faculties of the mind.
Hume argues that once we have an idea inside our heads, we have some power over it — we can take these ideas and play with them. We can, for instance, combine them. If you've seen a horse, you have the idea of a horse. If you've seen a horn, you have the idea of a horn. The mind can then fuse those two together, and you arrive at the idea of a unicorn. You didn't have a direct impression of a unicorn, but you had a direct impression of a horse and a direct impression of a horn, and your mind combined them. So while it's true that you had no impression of the specific composite thing, you must have had impressions of its parts.
The mind can do a great deal with ideas once it has them. Take that idea of a unicorn and make it a hundred feet tall — all you've done is stretch it, like Silly Putty. You can make it purple, make it tiny, make it speak. But you can only do these things because you already have the idea of tall through an impression, the idea of purple through an impression. If you had never seen purple, you could not make that unicorn purple. You take the unicorn, you take purple, and you put them together.
All kinds of highly abstract ideas that one might try to use as counterexamples are, on Hume's account, simply constructions out of smaller pieces — pieces that must originally have come from impressions. Once we have those pieces, we can do all manner of things with them inside the playpen of our minds. But everything must ultimately be traced back to impressions. This is one of the fundamental tenets of empiricism, and it stands in direct contrast to Descartes, who held that we possess a priori ideas — such as those of mathematics — already present in the mind. Which raises the question: what does Hume have to say about mathematics?
This brings us to the second part of Hume's mental geography, which continues his account of what we do with our ideas. Having explained what is in our heads and where it came from — we have ideas, and they all came from the senses — the question now becomes: how do we think with those ideas? What is the nature of inference and reasoning? This is the second major piece of preparation before we arrive at the central argument. The first piece concerned ideas and impressions and their relationship: impressions have greater force and vivacity, and all ideas are derived from impressions. The second and final piece of preparation concerns the kinds of reasoning available to us.
Hume accepts mathematics. He even accepts what Descartes says about it: math is absolutely certain. But his reason for that certitude is very different from Descartes'. Consider our clear and distinct perception that two and two equals four. We know this with certainty; it is a necessary truth for Hume just as it is for Descartes. But the reason is this: any other possible answer would be a contradiction. If you said the answer is five, you would get five equals four, and that is a contradiction because five does not equal four. Whether you say the answer is three, eight, or the color blue, any other answer is not merely accidentally wrong — the way it happens to be wrong to misidentify the president or the color of someone's shirt — it is logically contradictory.
We know a priori that no other answer can be right. You do not have to consult the world or run experiments adding things together to verify that two and two equals four, because nothing else could equal four except four. One and only one answer is non-contradictory, and therefore we can know these truths a priori. To that extent, Descartes was right.
But the reason we can know this a priori is precisely because it is empty — because it says nothing. What does "two and two equals four" actually tell us? It tells us that four is four. It does not tell us whether there are four of anything in the world. It does not tell us whether there are two things and two other things. It simply tells us that the idea four is identical to itself. Hume calls these relations of ideas, because all we are doing is taking ideas and arranging them into relationships — in this case, relationships of identity.
Relations of ideas are absolutely certain — we will grant Descartes that — but that certainty is purchased at the price of saying nothing. It is purchased at the price of being completely uninformative. We know it is true, yet we learn nothing from it; it gives us no knowledge of the world. For Hume, it is a pyrrhic epistemological victory.
The other kind of knowledge, what Hume calls matters of fact, is completely different. A matter of fact is informative — it tells us something about the world. But the only way something can be genuinely informative is if more than one answer is possible. If only one answer is possible, then telling you that answer is true conveys nothing, because nothing else could have been the case. Something is informative precisely because it rules out other things that were actually possible.
Consider the claim that my shirt is blue. This is true, and the reason it tells you something is that it could have been red, polka-dotted, white, or any number of other colors. All of those alternatives are possible — not actual, not true, they didn't happen, but they could have. Every one of them is logically consistent and coherent. This is unlike three equaling four, which is not merely false but impossible. My wearing a white shirt, by contrast, is entirely possible; it simply didn't happen. There is no problem with it — it is completely intelligible and could easily have been the case.
And it has to be this way. The relationship between these properties and these two types of knowledge is internal. Relations of ideas achieve certainty — a priori certainty — precisely because they are uninformative, because nothing else could be the case. That is the reason you can know them without consulting the world, and the reason you can know them for certain: nothing else could possibly be true. Matters of fact, on the other hand, are informative only because other things could be true — but that also means other things could be true, so you have to go and find out. You have to check the world, which means they are empirical.
Relations of ideas are a priori. I do not need to look to confirm that four equals four; I do not need to run surveys or conduct experiments, because it could not be anything else. I can know it in the safety and comfort of my own mind. But I cannot know what shirt you are wearing without going and looking, because you could be wearing any number of different shirts. If you have four shirts, I know you have four shirts without inspecting them — but what color they are I cannot know without seeing them.
Relations of ideas are absolutely necessary, which is why I can know them a priori, and which is also why they tell me nothing about the world. Matters of fact are empirical because they possess a very important property: they are contingent. To be contingent means that things could have been otherwise. I did wear a blue shirt, but I could have worn many others. Contingency and the need for empirical verification are necessarily intertwined.
Consider a simple example: I cannot determine by pure reasoning alone what color shirt you are wearing, because it could be blue, black, white, or any number of other colors. That is a contingent fact about the world. By contrast, if you have four shirts, I can infer purely in my mind that you have four shirts — no observation required. This is the distinction between a priori knowledge and empirical knowledge. For Descartes, the a priori beliefs — the mathematical ones — were the crucial ones, because they possessed absolute certainty. They resisted doubt, and that resistance was precisely what he was after.
Hume's response is that Descartes paid an awfully high price for that certainty. The things that matter, the things that tell us what the world is actually like and that guide our actions, are all matters of fact. Science consists entirely of matters of fact. Our everyday beliefs — what is safe to eat, where we live, whom we will see — are all matters of fact. Matters of fact vastly outnumber relations of ideas, and they are vastly more important.
They are also philosophically stranger. There is no real mystery about how we know relations of ideas: they cannot be otherwise, and they are governed by the law of non-contradiction and the law of identity. Four equals four — it is pure identity. But the question Hume is pressing is how we know matters of fact. These are the beliefs that make up science and guide our every action. His answer, it turns out, is deeply surprising.