Written by David Deutsch.
See also:
Links to this note
-
How to Detect and Eliminate Errors Is the Most Important Knowledge
For new knowledge to be created, there needs to be an error correcting mechanism. This makes it the most important knowledge for progress and innovation.
-
The real world and it’s behaviors are extraordinarily complex. To theorize and create good explanations necessarily requires some encapsulation of ideas through abstractions. It is possible to understand a phenomena by understanding abstractions and similarly, it is possible to create new explanations by building on top of them.
-
It’s a mistake when confusing an abstract process with a real-world process of the same name. An example from The Beginning of Infinity is the study of decision making. The assumptions that are built into their models are a simplified version of actual real-world decision making because there is not a good explanation of how it works in reality.
-
Extrapolating from past data points is not an explanation. Building your confidence that something that will happen—like Bayes Theorem—is useful for descrete, observable problems, but fails to reveal the truth. It’s the equivalent of saying “because it’s always been that way” which is a flawed way of reasoning about the world.
-
The Quest for Good Explanations Is Error Correcting
The process of seeking out good explanations is error correcting. It is tolerant of dissent with a healthy dose of skepticism and distrust of authority. It means that explanations are rejected when they are contradicted by better explanations.
-
Objective Knowledge Begins as Conjecture and Then Corrected With Criticism
Objective knowledge is possible and it comes from within. Communication is lossy so even precise books and teaching can not transfer objective knowledge. It starts with conjecture and then in repeated cycles of criticism it is corrected. Through this process objective knowledge can be found, but also shared.
-
The way I write more is by doing it every day. I write first thing in the morning (journaling and note taking) and publishing my notes (like this one). For work, I write product briefs to clarify the situation, my interpretation of the facts, and what we should do about it. I write memos for the team for anything important. I write investor updates. I do it without thinking—even when drafting a tricky email I’ll write it out to understand what I’m trying to do.
-
In a thought experiment from The Beginning of Infinity, the author introduces a universe traveling device. During a set interval, you hold the button you go from universe 1 to universe 2 for one minute then on to universe 3 for 30 seconds and so on until you release it and are taken back to universe 1. By the time two minutes is up, you will have traversed the infinite set of universes. If the instrument could take readings along the way, you have a way to measure infinity.
-
The Turing-Test Is an Empiricist Mistake
The Turing-test is rooted in the idea that a human can judge whether something is an Artificial Intelligence merely by the behaviors it exhibits during the test. In reality, a judgment of whether or not it’s a genuine AI requires an explanation of how it works.
-
Humans Transform Inhospitable Environments into Support Systems for Themselves
A popular view of the environment, “Spaceship Earth”, is that the planet provides just the right biosphere to support human life. That is misleading because humans are actually ill suited to living in most places. Take for example living in New York—you would freeze to death come winter if not for shelter, clothing, access to clean water, and food. This is technology that humans created to transform inhospitable environments into systems that support human life.
-
The Larger an Object Gets the Less Behavior Is Affected by Interference
Quantum interference does not affect the behavior of larger more complex objects (like human beings) the way it does for individual elementary particles due to entanglement. That’s why you don’t observe the concepts/theories in quantum mechanics like superposition and the multiverse in everyday life. The Beginning of Infinity, however, describes some scenarios where it could.
-
Multiple Explanations at Different Levels of Emergence Are Not Inconsistent
A reductionist argument against an explanation might be that it is incorrect because there are multiple explanations of the same phenomena. If good explanations are hard to vary, how could there be multiple explanations? This argument doesn’t take into account that multiple explanations can exist at different levels of emergence and this is not altogether inconsistent.
-
I remember when I first started working at Stripe I felt like the dumbest person in the room. I was amazed at how smart everyone seemed and the writing…gosh, the writing! If I wanted to be like that too, something needed to change.
-
A universe that collapses to a single point due to baryon annihilation required to produce energy needed for colonization. The universe is not infinite and even though we observe the universe expansion accelerating, that doesn’t mean it will continue to accelerate forever. This is like the counterpoint of the Big Bang—the “Big Crunch”.
-
All problems are soluble with the right knowledge. That doesn’t mean we know the solution already, but that the pursuit of good explanations will lead to progress towards one. It also doesn’t mean that all problems are solvable—there exists undecidable problems too (like a mathematical proof that proves something is undecidable) but even that provides something soluble (the absence of a solution) and new knowledge can be created (what if this undecidable theorem were true?).
-
Utopians Are Defined by Their Pessimism
Visionaries pushing some form of utopia are defined by their pessimism. Their belief that utopia can be achieved means they don’t believe things could ever get better, nor their idea of what is best. Unless new knowledge stops being created and all relevant knowledge is known, utopians exhibit the same parochial error as someone who believes the sun revolves around the Earth.
-
Imagine If Human Knowledge Received Half the Enthusiasm as AI
It’s striking how captivated the world is by LLMs and the latest techniques in AI. What if we were just as excited about progressing human knowledge and learning? Education doesn’t receive nearly the attention and hype it should yet we are far more capable.
-
Good Explanations Are Hard to Vary
A good explanation can not be modified or molded to fit when new information contradicts it. It predicts situations that are both known and unknown. The domain of its meaning and applicability is not yours to specify.
-
Predictions About the Future Don’t Account for New Knowledge
One of the reasons making broad, sweeping predictions about the future tend to be wrong is that it does not account for the creation of new knowledge. Trends are not explanations and without an explanatory model of how knowledge will change (i.e. creativity) predictions such as the end of the world are just another example of a Malthusian catastrophe.
-
Problems Are Conflicts Between Ideas
The essence of problems is when two ideas come into conflict with each other. Sometimes that’s the conflict between an explanation and a better explanation. Other times it’s the conflict between the current state of being and what you want. Without conflict, there is no problem.
-
Trying to Know the Unknowable Leads to Pessimism
We do not yet know what we have not discovered and trying to know the unknowable (prophesy) leads to pessimism. A Malthusian catastrophe ends up being wrong because it does not predict knowledge that resulted in efficiency of food production. Similarly the pessimism of energy economics is error laden because it can not predict what new discoveries we will make in social and political systems or new defenses.
-
Reductionists View High-Level Behavior as Consisting of Lower-Level Behavior Only
The reductionist view of science is that all high-level behavior consists of the underlying lower-level behavior and should be analyzed into components to fully understand. However, good explanations can be self-contained and sufficient without needing an explanation of every low-level detail. For example, you can have a theory of how water boils that doesn’t need to predict movement of individual atoms.
-
Parochial Errors Happen When You Have a Narrow View
A parochial error happens when you falsely believe that something in your narrow view of the world applies more broadly than it does. For example, thinking the seasons everywhere around the earth in the same way as your home town because that’s what you personally experience.
-
The Anthropic Principle Says the Universe Is the Way That it Is Because We Exist
The fact that humans exist with the capacity to observe and theorize about the universe is the explanation for the universe being the way that it is. The physics that creates the preconditions for intelligent life means that intelligent life could only ever observe this kind of universe. If other universes don’t support intelligent life, they will have different physics from our universe and, potentially, our physics is universal to intelligent life.
-
Outline for my annual essay about things I learned and reflections for the year.
-
Human Knowledge Is Meme Replication
Despite common ancestry between humans and apes, humans have the ability to grow collective knowledge through the replication of memes.
-
The Quantum Suicide Argument and Subjective Certainty
In an infinite multiverse, where every possibility allowed by physics is certain, the argument goes that one could be absolutely sure of winning the lottery using the following hack. Buy a lottery ticket and set up a machine that will automatically kill you in your sleep if you lose. Supposing you don’t care about any history in which you are not a lottery winner, this is way of winning the lottery with certainty…at least in universes where you are a lottery winner.
-
The jump to universality has two phases. Before universality, one needs to create specialized objects. For example, Roman numerals would need to add a new symbol to raise the maximum value or adding a pictogram to a language to represent a new word. After universality, one need only customize or configure a universal object. For example, printing a different book on a movable type printing press, sending an email to someone new, or inventing a new word with the same alphabet.
-
Justificationism Secures Ideas Against Change
One way to answer “how do we know…?” is to justify one’s belief by reference to an authoritative source or cornerstone of knowledge. This is, in effect, saying “by what authority do we claim…?” which seeks endorsement in order to have certainty. Justificationism as a theory of knowledge therefore resists change (or at least delays in a form of path dependence).
-
Techno-optimists believe technology can solve the world’s most pressing problems. With the right knowledge, we can find solutions to climate change like abundant clean energy. Can we acquire the knowledge to build nuclear fusion reactors? Can we do it in time?
-
Elegance Is a Heuristic Guide to Truth
It seems that the best explanations are often the most elegant. Sometimes it’s in the simplicity, sometimes it’s in the obviousness. Looking at some of the biggest discoveries in math and science (e.g. relativity, DNA, calculus, evolution, computing) and thinking, “how could it not have been so?”