You are what you know

Thinking, memory and knowing

This is a long essay - 10,000 words - so do feel free to take your time. This is everything I believe to be true about what knowledge is and its relationship to memory and thought.

---

Now, what I want is, Facts. Teach these boys and girls nothing but Facts. Facts alone are wanted in life. Plant nothing else, and root out everything else. You can only form the minds of reasoning animals upon Facts: nothing else will ever be of any service to them. This is the principle on which I bring up my own children, and this is the principle on which I bring up these children. Stick to Facts, Sir!’

Charles Dickens, Hard Times

Gradgrind was, as I’m sure you know, a fictional character that Dickens invented as a caricature of what was doubtless some fairly awful teaching in Victorian England. But he isn’t real. If they ever did, no one thinks like that today. Or if they do, I’ve yet to encounter them.

Knowledge is not the same thing as facts. Facts are just one part of a much greater whole. Philosophers have been trying to work out what knowledge is for millennia. When Greece was still ancient, Aristotle broke it into three components which he called episteme, techne and phronesis. For Aristotle, episteme was timeless universal knowledge, such as the quality of the base angles of an isosceles triangle, while techne was knowledge that was dependent on time and place. Phronesis, sometimes translated as practical wisdom, is less concerned with what it true and more concerned with what is right.

We can think of episteme (propositional knowledge) as what we know, whereas techne (procedural knowledge) is know-how and is basically synonymous with skill. Phronesis can perhaps best be thought of as tacit knowledge and is made up of those things we’re unable to articulate and don’t necessarily know we know.

From episteme we get epistemology – the study of knowledge. Philosophers tend to think about knowledge as justified true belief. Getting to grips with this would involve recapping some drawn out, tangled philosophical debates. We’re not going to do that here.¹ Instead we’re going to think about knowledge from the perspective of cognitive science – structured collections of information acquired through perception or reasoning. This doesn’t have to be justified, or true, or even necessarily believed; it just has to be stored in the repository of our long-term memory. Our brains are as full of misconceptions, confusions and falsehoods as they are anything else.²

My contention is that you are what you know. Knowledge is all there is. All knowledge is biological – stored in the organic substance of our brains – and everything stored biologically is knowledge. This might sound a little extreme and is sometimes dismissed as unduly ‘cognitivist’, so I’ll try to explain what I mean. Psychologists used to believe that people could be entirely understood by observing their behaviour and that behaviour could be explained in terms of conditioning. Importantly, behaviourists believed that the mind was a blank slate, or tabula rasa, on which any instructions could be written. One of the founders of behaviourism, John B. Watson, even went so far as to claim:

Give me a dozen healthy infants, well-formed, and my own specified world to bring them up in and I’ll guarantee to take any one at random and train him to become any type of specialist I might select – doctor, lawyer, artist, merchant-chief and, yes, even beggar-man and thief, regardless of his talents, penchants, tendencies, abilities, vocations, and race of his ancestors. I am going beyond my facts and I admit it, but so have the advocates of the contrary and they have been doing it for many thousands of years.

Behaviourism, p. 82.

He was going beyond his facts and it’s now widely accepted that all human traits are in some part influenced by genes. As discussed here, there is good reason to believe that our genes have co-evolved with human culture to acquire adaptive capacities that make it easy for us to learn some forms of knowledge and ways of behaving, and that our genes, in part, select our environments.

Behaviourism fell from grace and was replaced by cognitivism. Arguably, the so-called ‘cognitivist revolution’ began with Noam Chomsky’s observation that human language has universal features, and that, given the vast complexity of these features, if they really were blank slates, 3-year-olds should not be able to deduce the rules of language given the limited exposure they typically have by that age. He argued that we must therefore have an innate capacity for learning language. This was revolutionary thinking indeed, and from there psychologists began to focus on our inner mental activities.

Cognitivism – and, in particular, the computational theory of mind³ – suggests that in order to understand ourselves we need to peer into the black box of the human mind wherein mental processes (thinking, memory and knowing) can be explored. In this view, knowledge, memory and thought are interchangeable, and learning is defined as a change in schema. I’m not subscribing wholesale to this paradigm, but that’s certainly where my sympathies lie. Everything we are – our personality, experiences, preferences, thoughts and feelings – are all stored in memory. There is nothing outside of these biological processes; mind and body are not distinct and there is no ‘little man’ or spooky stuff required to explain how and why we do what we do.⁴

When we think about who we are and what makes us unique, it’s remarkably tempting to consider ourselves as made of something special, something beyond the merely biological. We want to believe we have some kind of immortal, intangible essence, but such beliefs are wishful thinking.⁵ Even if we don’t want to believe something so numinous, we’re still tempted to suppose that there must be more to who we are than merely what we know. We work hard to come up with different labels for this stuff (skills, understanding, etc.) but these can just as easily be seen as synonyms for knowledge.

But what of the common-sense observation that knowledge and understanding are different? That wisdom is distinct from skill? Surely, it makes sense to create separate categories for each of these things that I’m grouping together as knowledge? Well, yes, of course it does. I can group Cheddar, Roquefort and mozzarella together as cheese but they each look and taste distinctively different. There are times when it makes sense to specify and times when it makes sense to aggregate. So, we can accept that factual knowledge, skill, wisdom and anything else you might care to speculate about are – at the same time – all knowledge and all different. Giving something a different name doesn’t alter its reality. Knowledge is all there is.

Thinking about

My position can be summarised in these three propositions:

1. Knowledge is both what we think with and about.

2. We cannot think with or about something we don’t know.

3. The more we know about something, the more sophisticated our thinking.

Nobody can think about something they don’t know. Try it for a moment. The best you can do is to ask, “What don’t I know?” but even then, you’re limited to what you know you don’t know. What we think about are concepts, ideas, experiences and facts – “nothing but facts”. We can think about the capital of China (and know that it’s Beijing, despite it being called Peking in earlier times). We can think about the length of the Nile (but this is subject to change with the seasons and frequently altering water courses). We can think about our favourite colour (no longer the pink we liked as a child) or what we’d like for our birthday (as long as it’s not the same as what we received last year). Some of what seem to us to be immutable facts are temporary ways of holding the world in mind. As our thinking changes, so does what to hold to be true. Knowledge changes us.

We can think about anything we know at least something about, but this can be a shallow, unfulfilling experience. The more things we know, the more detail we possess, the more links and connections we can make. Seeing these links is insight; making these connections is creative. Knowledge attracts knowledge; new stuff sticks to what we already know and our sphere of knowledge expands until, at a certain point, its growth is exponential.

Facts or propositions are what we think about. Propositional knowledge exists in the realm of conscious thought; when we are aware of a thing we are thinking about it. Propositional knowledge, declarative memory and crystallised intelligence are, if not the same thing, then broadly synonymous. Like most good ideas, I’m certainly not the first to have thought this. The psychologist John Anderson developed a theory of cognition he called Adaptive Control of Thought (ACT) in which he stated,

All that there is to intelligence is the simple accrual and tuning of many small units of knowledge that in total produce complex cognition. The whole is no more than the sum of its parts, but it has a lot of parts. [my emphasis]

ACT: A Simple Theory of Complex Cognition

This is something of an understatement. There are many more parts than we will ever be aware of, and the factual, propositional knowledge we possess is just the small part we are able to bring deliberately to mind.

Some examples of propositional knowledge:

• A horse is a quadruped.

• 85% of the people of the world live in ‘developed’ countries.

• The language Ewoks speak in Return of the Jedi is based on ancient Tibetan.

• Brain scans clearly show that we use most of our brain most of the time, even when we’re sleeping.

• Cleopatra was born 2,500 years after the Great Pyramid of Giza was built, yet only 2,000 years before the first lunar landing was achieved.

• Ostriches do not bury their heads in the sand to avoid danger.

Are they reasonable? Are they verifiably true? Do you believe them?

Think for a moment about how you felt on reading the list. The first item probably reminded you of the quote from Hard Times at the start of this post. Did that alter the way you thought about it? Maybe some of the propositions surprised you. Maybe you believe ostriches really do stick their heads in the sand at the sign of danger.⁶ Maybe you still do and you think I must be wrong, but how would they avoid suffocation? And, for that matter, how would they avoid predators? Moving on, can it really be true that 85% of the world’s population live in developed countries? Everyone knows that most of the world’s population live in poverty, right?⁷

Perhaps you have no strong feelings about ostriches or where people live and are inclined to believe these new facts. Perhaps you will verify them later. We change our beliefs all the time and as they change, so do we. It’s only rarely that such changes are noticeable, but many small changes have an incremental effect. Slowly, or quickly, knowledge alters who we are.

All of these propositions are formed from a variety of different facts. Think about the one about Cleopatra. You not only have to know who she was, you also have to know what the Great Pyramid is, where Giza is, how these relate to Cleopatra, something about the moon landings and the scientific progress that led to that moment in human history. You probably also have to know something about time; the fact that 2,000 years is a huge chunk of human history and that the civilisation of ancient Egypt is almost unimaginably old.

While they rely on lots of unstated knowledge, all these propositions are expressed in language. Thinking about these propositions can have the quality of an internal conversation between different perspectives. And while we catch glimpses of thought untethered to language, we can only really think about what we can put into words. That said, the idea that we think in English (or any other language) may be an illusion. The language of thought – call it ‘mentalese’ – is far more efficient and direct than spoken language, with all the trappings of grammar needed to communicate an idea from one mind to another, but this is something of which we are unaware.⁸ Of course, we can have thoughts we’re unable to express, but we struggle to think about anything without being able to say it. This is something we can test: try thinking about something you cannot say. Hard, isn’t it?

Words, and their meanings, are all facts. But, although it can difficult to discuss concepts for which you have no words, propositional knowledge is much more than a mere list of unconnected facts. Speaking about his knowledge of birds, everyone’s favourite physicist, Richard Feynman, observed:

You can know the name of a bird in all the languages of the world, but when you’re finished, you’ll know absolutely nothing whatever about the bird … So let’s look at the bird and see what it’s doing – that’s what counts. I learned very early the difference between knowing the name of something and knowing something.

1973 TV interview

It’s easy to belittle the idea that knowing what things are called is useful. Of course, simply knowing the name of something is quite different from apprehending the ‘thing itself’. Knowing that Einstein came up with the special theory of relativity tells you ‘absolutely nothing whatever’ about what the theory means. That second kind of knowing – understanding, if you prefer – requires more facts. The more you know about a thing, the better your understanding. The more you understand it, the better you’ll know it.

This is the distinction between inflexible and flexible knowledge. Knowledge is flexible when it is not tied to superficial features and can be applied to a wide range of contexts. For instance, knowing that Henry II was a medieval king is inflexible. It tells you absolutely nothing whatsoever about who Henry was or what it meant to be a medieval monarch. Knowing what medieval kingship entails is flexible and can be deployed to think more widely about a period of history. What we ultimately want is for children to have a flexible understanding that can be applied to a wide variety of new situations, but this is unlikely to occur spontaneously.

Cognitive scientist, Daniel Willingham puts it like this:

Inflexible knowledge is meaningful, but narrow; it’s narrow in that it is tied to the concept’s surface structure, and the deep structure of the concept is not easily accessed. ‘Deep structure’ refers to a principle that transcends specific examples; ‘surface structure’ refers to the particulars of an example meant to illustrate deep structure.

Inflexible Knowledge: The First Step to Expertise

The obvious solution would be to encourage children to think about what we want them to know in deeper, more abstract terms, so they will be better able to generalise what they learn to new contexts. Sadly, this doesn’t work.⁹ In practice, children need to fix their knowledge, however inflexible, in order to incrementally arrive at greater understanding. Knowing inflexibly is a necessary stepping-stone to more flexible knowledge.

Even though knowing the name of something may be an inflexible form of knowledge, does that mean learning names isn’t worth the effort? Not at all, says Feynman, “knowing the names of things is useful if you want to talk to somebody else – so you can tell them what you’re talking about”. And this is the point: we use names and labels for reasons of expediency. While I can say, “The theory which determined that the laws of physics are the same for all non-accelerating observers, which shows that the speed of light within a vacuum is the same no matter the speed at which an observer travels,” it’s quicker and easier to chunk this information as “Einstein’s special theory of relativity”. Likewise, I can say, “You can use a piece of punctuation that looks like a dot floating above a comma to connect two bits of a sentence if both bits make sense on their own and are also closely related,” but this is confusing and time consuming. How much more efficient it would be if I had the schematic knowledge to be able to say, “A semicolon connects two closely related independent clauses.” Not only is this more straightforward, it’s less ambiguous and consumes less working memory. The act of chunking facts together helps to make them more flexible, and when knowledge becomes flexible we can think with it as well as about it.

Inflexible ideas can be learned by rote. I can memorise a fact such as “commas are used to separate clauses in a sentence”, but just knowing this won’t necessarily help me to write a better punctuated sentence. On its own it’s too abstract to be useful. For this reason, children learn proxies like “put a comma where you take a breath”. This sounds plausible and contains knowledge that can be applied, but what if you’ve just been for a run? What if you have asthma? The limits of inflexible knowledge rapidly become clear. The antidote is more knowledge. I have to show you how to use a comma in a wide variety of contexts and then get you to practise writing correctly punctuated sentences.

Similarly, I can learn the names of all the European capital cities or memorise the times tables, but that doesn’t suggest I will be able to apply this information in any situation other than being directly asked, “What is the capital of Latvia?” or “What is 9 × 7?” This has resulted in inflexible knowledge being dismissed as only good for pub quizzes. The truth is that inflexibility is a necessary foundation upon which more flexible structures can be built. Let’s imagine you have learned all the square numbers up to 10: 12 = 1, 22 = 4, 32 = 9, etc. Let’s further imagine you have learned Pythagoras’ theorem for finding the area of a right-angled triangle: a2 + b2 = c2. Now, if you encounter a problem like the one below, you should, with minimal prompting, be able to produce the correct answer.

When we apply inflexible knowledge, we make new schematic connections. We don’t just remember facts, we remember all the ways we’ve thought about and used those facts before.

The more we apply propositional knowledge, the ‘chunkier’ and increasingly flexible it becomes. Instead of looking at a problem and trying to hold all the steps in mind, we are able to think with the whole schema. Eventually, these schemas can be applied to examples which are less obviously similar because we are able to ignore the superficial differences and concentrate on the similarities. The journey from inflexibility to flexibility produces a positive feedback loop – changes are amplified and enhanced and the new, more expert, way of thinking becomes permanent. In a very real sense, this is what it means to be clever. Eventually, knowledge can be mastered to the point where we no longer have to think about it; our working memory is free to handle chunks of propositional knowledge and arrive at solutions and insights far faster than someone who merely has high fluid intelligence ever could. The general rule is that expert knowledge always trumps raw ability.

So much for Gradgrind and his thirst for facts. All this propositional knowledge is just the tip of an unimaginably enormous iceberg. What, then, of skill?