How much do you know
⏳ Can you survive the Doomsday Argument?
Eight questions on birth rank, self-sampling, self-indication and the reference class — the places where reasoning about your own position goes wrong.
Lukasz Szramuk ·
Question 1 of 8
What single fact does the Doomsday Argument reason from?
Every question, and every answer
The whole quiz, written out — so you can read it without playing, and so a search engine can too.
What single fact does the Doomsday Argument reason from?
Answer: Your approximate birth rank among all humans who will ever live
That is what makes it unsettling. The argument uses no information about climate, war, asteroids or technology — only where you sit in a queue. If that alone shifts a forecast, either self-locating information is worth far more than it looks, or something in the reasoning is broken.
Read the essay →In Gott's Copernican version at 95% confidence, the future fraction of the human run lies between:
Answer: 1/39 and 39 times the fraction already past
The bound is (1−C)/(1+C) to (1+C)/(1−C). At C = 0.95 that is 1/39 to 39. The width is the honest result: fed 117 billion people and 130 million births a year, it gives a window from a couple of decades to tens of thousands of years, not a date.
Read the essay →The Self-Sampling Assumption says you should reason as if you are:
Answer: A random member of your reference class of observers
SSA is the step the Doomsday Argument needs. Treat your birth rank as a draw from the whole human run and a rank of 117 billion caps the plausible total. Everything then turns on what the reference class is, which SSA does not tell you.
Read the essay →How does the Self-Indication Assumption blunt the argument?
Answer: Your existence is itself evidence for a larger population, which offsets the update
SIA says a future containing trillions of people makes your existing at all more likely in the first place. That consideration pushes in exactly the opposite direction to the doomsday update, and in the simplest setups it cancels it precisely. Whether it cancels neatly or overcorrects is unsettled.
Read the essay →What is the reference class problem?
Answer: Nothing in the argument says who counts as one of 'us' for the draw
Homo sapiens, any hominid, any observer capable of running the argument — the answer changes the numbers by orders of magnitude, and the argument supplies no principle for choosing. This is the objection that most often persuades people the reasoning is not yet well posed.
Read the essay →Who is the Carter catastrophe named after, and what is his other famous contribution?
Answer: Brandon Carter, who also named the anthropic principle
Carter raised the argument in the early 1980s without publishing it as a prediction, and he named the anthropic principle at Kraków in 1973. They are the same move applied to time rather than to space: reason from the fact that you are an observer to a conclusion about the world.
Read the essay →A Boltzmann brain is a problem for anthropic reasoning because:
Answer: If you assume you are typical, you should conclude you are one
In many models, fluctuation-assembled observers with false memories vastly outnumber the ordinary kind. Applied consistently, typicality then says you are almost certainly one. Nobody accepts that, so a Boltzmann-brain-dominated model is treated as ruled out — which is also an admission that 'typical' needs supervision.
Read the essay →What is the most defensible thing to take from the argument?
Answer: That reasoning about your own position needs an explicit rule, and the rules disagree
The argument is not a prophecy and its authors did not offer it as one. What it reliably demonstrates is that self-locating belief is not optional in cosmology and not settled either: SSA and SIA are both defensible, they cannot both be right, and which you pick changes the answer.
Read the essay →