Thinking Is Still Your Job

Socrates never wrote down any of his ideas. He thought writing would make people appear smarter while making them think less. He practiced philosophy through spoken dialogue and engaging others in active questioning. Most of what we know about his philosophy now exists thanks to his student Plato, who recorded these conversations after Socrates’ death. In Plato’s Phaedrus, he tells the story of Theuth, the god who presents writing to King Thamus as a gift that will improve memory and wisdom. Thamus is not impressed. When speaking about letters, Theuth said: “This branch of learning, my king, will make the Egyptians wiser and will improve their memory. The drug! for memory and wisdom has been discovered!” to which the king said, “For this will produce a forgetting in the souls of those who learn these letters as they fail to exercise their memory, because those who put trust in writing recollect from the outside with foreign signs, rather than themselves recollecting from within by themselves.” He argues that writing will create the “appearance of wisdom” rather than wisdom itself. The invention designed to make us wise, he feared, could replace thinking.

We’ve come a long way since Plato.

Every technology since has repeated a pattern.

They all take away our ability to use our heads. Of course, they are helpful. We do more, but at a cost to our minds. Each time, we told ourselves the tool freed us for higher thinking. Sometimes it did. But when the tool is doing all the reasoning for you, you should be careful. Right now Plato’s story is on a whole new level. The tool has changed, but the danger has not. Super thinking machines can do more than remember for us.

They’ve become authority figures.

Except they are prone to hallucinations. Their facts are not always true. Which is why thinking for ourselves is now more relevant than ever before. But thinking is expensive. It costs time, tolerance for ambiguity, and the willingness to be wrong in front of yourself before you’re right in front of anyone else. So people don’t trust themselves to do more of it. Daniel Kahneman spent a career studying why people avoid this cost. He proved that the mind defaults to fast, automatic judgment almost all the time. What he called system 1. And reserves slow, effortful reasoning for when it’s unavoidable. His research found that given the option, most people will choose cheap thinking almost every time.

“Laziness is built deep into our nature,” he says.

“Jumping to conclusions is efficient if the conclusions are likely to be correct and the costs of an occasional mistake acceptable, and if the jump saves much time and effort,” Kahneman writes in his book, Thinking, Fast and Slow. A machine that answers instantly exploits our desire for slow thinking. You no longer need patience. The machine has none of the impatience you’d otherwise feel. You no longer need to sit with yourself, thinking, pondering, reflecting. Engaging your brain. Why waste your time when one is available on demand, right? The danger is that being right isn’t the same as understanding or thinking for yourself. Richard Feynman once said, “I would rather have questions that can’t be answered than answers that can’t be questioned.” A generation raised on instant, confident (but sometimes false) answers is at risk of losing the taste for thinking (a productive discomfort that makes us wiser).

We are living through an accelerating decline in the habit of human thinking. Slow, effortful and sometimes painful thinking. What Kahneman says is “system 2” thinking. He wrote is deliberate thinking. “System 2″ is slower, more deliberative, and more logical. Using external tools reduces this mental effort. It helps to some extent. You do it every day. You don’t memorise phone numbers anymore. Who needs landmarks when you have GPS? You don’t do long division in your head. Some of that is fine. Some of it is even good. Einstein reportedly said, “My pencil is cleverer than I am.” Meaning his tools extended his thinking, not replaced it. A pencil doesn’t give you answers. It helps you find them. Writing develops thoughts beyond what the mind can hold alone. Intelligent tools are different. They give you the answers. And when the answer arrives, your brain gets a signal. “Job done as quickly as I needed it.” The productive struggle (uncomfortable work of forming a thought) never happens. The MIT researchers call this cognitive debt.

You borrow convenience today.

You pay in capacity tomorrow.

Your thinking skill gets compromised over the long-term.

Thamus was wrong about one thing. Writing didn’t destroy memory; it changed what memory was for. We adapted. We built new cognitive skills around the technology. But his caution is still valid. The same will likely be true in the future. The machines won’t destroy thinking. But they will show, clearly, who was thinking in the first place. For a while, the tool will make it hard to tell the difference between a mind still working and a mind that has retired. That confusion is temporary. It resolves the minute you are forced to think for yourself. Forced to make a decision with practical life consequences. A judgment nobody else can make for you. And the answer either comes from your head or it doesn’t come at all. Every external tool that reduces mental effort carries a risk.

The risk is that you’ll stop doing the work.

And if you stop doing the work long enough, you lose the ability to do it. It gets worse. The tools become so important to you that it makes thinking feel unnecessary. Another danger is the trust spiral. Michael Gerlich, who studies technology and cognition, warns that their ability to provide complete answers discourages critical evaluation. Why check the machine’s work when the machine is very confident? But thinking machines are not good at giving you the facts. They get it wrong a lot of the time. Most people never check. Trust is breeding laziness. They are too dependent to doubt. That dependency breeds a decline of thought.

There is tons of research that proves that people who are obsessed with super thinking tools give up on difficult problems.

They stop trying.

Alan Turing, the father of the thinking machine, famously wrote, “The original question, ‘can machines think?’ I believe to be too meaningless to deserve discussion.” Ten years from now, will you still think for yourself? Turing predicted something else that should make you uncomfortable. “Once the machine thinking method had started, it would not take long to outstrip our feeble powers.” And then: “There would be plenty to do in trying, say, to keep one’s intelligence up to the standard set by the machines.” He saw through it all. The machines would get better. The humans would have to work harder to keep up. And if they didn’t? They’d fall behind. And would have engineered their own fall.

The decline is not inevitable.

Here’s the thing. You can use thinking machines to find the facts. The answers you can’t think for yourself. The work has to be done anyway. If the tool can help, after you double-check, then go ahead. But be deliberate. Reduce cognitive offloading. Unguided use is the problem here. Some things will always require productive tools to save time. But do your own thinking when you must. If you offload the boring stuff and save your brain for the hard stuff, you get smarter. If you offload everything and let your brain be comfortable, you compromise your System 2 thinking process. The question is which path you’re on. No one can throw away their intelligence tools. These are here to stay. They’re useful. They can help you think through problems. But be more mindful.

The habit of thinking is fragile. You lose it if you stop using it. Use the machine. But use it like a pencil, not like a brain. Let it help you find the thought. Don’t let it have the thought for you.

Read the output.

Argue with it. Check its sources. Write your own thoughts down before you prompt. Sit with the discomfort of not knowing. Figure things out. Some people are using these tools the way a good carpenter uses a power drill. As an extension of skill. The drill never replaced a great carpenter. If you want the machines to give you options, fine. But argue with every one of them. Figure out what hasn’t been considered, then do the harder work of deciding what’s true. The tool should make you think.

Aristotle separated phronesis (practical experiential wisdom from mere technical knowledge, techne. A machine can give you techne instantly. The facts, drafts, summaries, and the code. It cannot give you phronesis. You have to earn it through the exercise of choosing, failing, and choosing again. You cannot outsource your thinking muscle and still expect it to be strong when you need it. The poverty of information/knowledge, true or false, is no longer a problem. The effort required to make it mean something to you is a skill you shouldn’t let go. Abundance of knowledge without meaningful connection guarantees scarcity of wisdom. Socrates feared writing would make us forgetful. He was partly right. But we adapted. We found new ways to think. We can adapt again. But right now it’s a choice. And you make it every time you decide whether to struggle with a problem or hand it to the machine.

The machines are getting smarter. The danger is that you’ll stop trying to keep up. Use your ability to think or lose it. Thinking is still your job.