Knowledge Creates Resources
Treat scarcity as a knowledge problem, not a fixed physical limit.
- Difficulty
- Moderate
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 68%
Deutsch endorses a definition of wealth as the set of physical transformations you can effect. An individual's capacity scales somewhat with the resources available to them and far more with the knowledge available to them, which makes knowledge the multiplier. Resources are then whatever gets combined with knowledge to create wealth — a definition that makes the category mutable. New knowledge promotes previously useless matter into resources and lets you discard the ones you were running out of: wood to coal to oil to nuclear. The doom argument only works if knowledge growth stops. Deutsch will not claim the knowledge will be created — we are fallible and may miss a solution right in front of us — but he does argue we have everything it takes, so failure would trace to bad choices and anti-rational memes rather than to constraints imposed by the planet.
Origin
Deutsch's account of wealth, resources and optimism from The Beginning of Infinity, drawn out here by Naval in the context of natalism and sustainability.
Core principles
- 01Wealth is the set of physical transformations you can effect, not a pile of stuff.
- 02A resource is whatever you can combine with knowledge to produce those transformations.
- 03New knowledge promotes non-resources into resources and demotes the ones running out.
- 04People are on balance productive — two hands, one mouth — so more of them is upside.
- 05We have what it takes; failure would come from bad choices, not from planetary limits.
- 06Demanding that future knowledge be named in advance is a bad argument, not a standard.
How to run it
- 1
Restate the shortage as a knowledge claim
Convert 'we are running out of X' into 'with current knowledge, X is the only input that performs this transformation'. The scarcity is then a property of what we know, not of the universe.
- 2
Identify the transformation, not the material
Ask what physical transformation you actually need — heat, motion, computation, nutrition. Materials are interchangeable candidates for delivering it; the transformation is the requirement.
Pro tip Naming the transformation immediately widens the candidate list.
- 3
Check the substitution history
Look at how the same need has been met before. Energy went from wood to coal to oil to nuclear, each step turning something previously inert into a resource and retiring the one people feared losing.
Watch out Every generation has a reason why the substitutions stop with them.
- 4
Refuse the name-it-now demand
Opponents demand that you specify the future knowledge before it counts as real. Deutsch calls this a bad argument: it is a catch-22 that would have invalidated every past substitution at the moment before it happened.
Watch out Refusing the demand is not the same as promising the discovery — stay honest about fallibility.
- 5
Separate capability from outcome
Argue that we have what it takes, not that we will succeed. If we fail, the cause will be bad choices — anti-rational memes that restrict knowledge growth, or well-intentioned errors nobody could see.
Pro tip This is the honest and the stronger position: it cannot be refuted by pointing at a failure.
- 6
Audit the proposed remedy
Check whether the recommended fix works mainly by restricting people and knowledge creation. Under this model that suppresses the only mechanism that has ever produced new resources.
In the wild
Deutsch and Naval run through energy history as the standard case. Each transition took something that was not previously a resource and made it one, while the input people feared exhausting became less relevant. At every stage the objection recurs — now we are out of ideas, now we have to freeze the frame and conserve what we have. The counter is that new knowledge will produce new resources, so the old ones matter less than the panic assumes. The rebuttal to that is the demand to name the new resource in advance, which Deutsch classes as a bad argument rather than a real test.
→ The scarcity in each era turned out to be a fact about the knowledge of that era, not about the planet.
As a graduate student visiting Texas, Deutsch met libertarians whose immigration slogan was two hands, one mouth. He treats it as a compact statement of what human beings are: they consume and they produce, and on balance they produce more than they consume. Virtually all humans, mass murderers aside, create more wealth than they destroy, so other things being equal we should want more of them. He allows moral exceptions — bringing someone into a war zone may be unfair on them — while noting that on cold economic grounds it is probably still positive.
→ Population reframes from a load on fixed resources to additional capacity for creating the knowledge that makes new ones.
Common mistakes
Promising the breakthrough
Claiming the knowledge will definitely be created overreaches and is easy to refute. The defensible claim is that we have what it takes and that failure would be a choice.
Accepting the name-it-now test
Agreeing to specify unknown future knowledge before it counts concedes an unmeetable standard that no past discovery could have satisfied either.
Using the model to dismiss real constraints
The argument is about long-run substitutability, not about today's operations. Fallibility is part of the model — the solution right under our nose can still be missed.
Is it for you?
Best for
Evaluating sustainability, energy, and population arguments, or any long-horizon forecast built on the assumption that current inputs are fixed.
Not ideal for
Immediate operational shortages where the substitute knowledge does not yet exist and the constraint has to be managed today.
From the transcript
“wealth as a set of physical Transformations that we can affect”
“new knowledge allows you to use new things as resources”
“we have what it takes we have everything that it takes to achieve that”
From the episode
David Deutsch: Knowledge Creation and The Human Race, Part 1