There is no paradox: once you admit how little we know about each factor, an empty galaxy is the expected result, not a surprise

The Fermi question feels like a paradox because people multiply seven best guesses together and get a galaxy with thousands of civilisations, then look up and see none. But a best guess is the wrong thing to multiply. Several of the factors, above all the chance that life starts on a suitable planet, are uncertain not by a factor of a few but by tens or even hundreds of powers of ten, and when you run the equation with honest ranges instead of single best guesses, a large share of the outcomes have the galaxy empty and a good share have the whole observable universe empty. On this view there was never anything to explain: the silence is what you should expect from what we actually know, and the surprise was an artefact of false precision.

The argument is a point about arithmetic, made by three researchers at Oxford's Future of Humanity Institute in 2018. The Drake Equation multiplies seven numbers. If each is uncertain, the right way to combine them is not to pick a middle value for each and multiply, but to let each vary across its plausible range, draw one value from each range many thousands of times, multiply each time, and look at the spread of results. They did it two ways. Resampling the estimates published in the literature, the middle outcome was about a hundred civilisations per galaxy, yet there was still about a 30% chance that we are alone in the Milky Way. With ranges set from their own reading of current knowledge, the middle outcome was a fraction of a civilisation per galaxy (0.32), the chance that we are alone in the Milky Way came out at 52%, and the chance that we are alone in the entire observable universe came out at 38%.

The reason is that one or two of the factors dominate. The chance that life starts on a suitable planet could, on what is known, be anywhere from nearly certain to smaller than anyone can name: the paper spreads it as a bell curve in powers of ten, fifty powers of ten wide and centred on one start per planet. When a factor can be that small, the product can be that small, and no amount of optimism about stars and planets can rescue it. A single best guess hides this by quietly choosing a value in the middle of the range, which is where the odds are least likely to lie. Multiplying middles gives a confident-looking number that no individual step justifies.

The paper's second move was to turn the argument around. If the silence is what one expects, then the silence is also weak evidence about the factors, and the authors updated their ranges on the observation that nobody is visible. The result was the same in shape: the probability that we are alone went up, and the bulk of the uncertainty moved further toward the early steps, the origin of life and of intelligence. In the catalogue's terms this explanation lives with the "never there" family, because its conclusion is that the factors before the lifetime are probably small, but it reaches that conclusion by accounting rather than by any claim about biology.

This site's headline numbers follow the paper's method: every factor carries a range, the equation is run many thousands of times, and what goes on the home page is the spread. That is the practical reason the explanation gets a page of its own.

The main objection is that "dissolving" the paradox depends entirely on the ranges chosen, and the wide lower end for the origin of life is doing almost all the work. A natural objection is that the long low end of that spread is not an estimate anyone defends but a formal stand-in for ignorance, and that putting it into the range on equal terms with the rest nearly guarantees the conclusion; the paper itself says its conclusion can be driven by that one factor alone. The authors agree that the ranges are the argument, and reply that anyone who wants a crowded galaxy has to say which range is wrong and why, which is a better conversation than the one about paradoxes. A second objection is that the first step of the argument answers Drake's question (how many are detectable now) and not Hart's (why has nobody in ten billion years spread across the galaxy). The authors take up Hart's question in a second step, where the absence of visible settlement counts as evidence; that step raises the chance that we are alone to between 53% and 99.6% for the galaxy and between 39% and 85% for the observable universe, but it depends on how fast settlement is assumed to spread, for which the paper cites estimates from under a million to ten billion years.

What it says about the equation

Every explanation is a claim about one or more of the seven pieces that are multiplied to give the number of detectable civilisations. This one bears on 2 pieces:

f_l · The share of planets that could support life on which life actually starts piece 4 of 7 says this piece is smaller than it looks
f_i · The share of planets with life on which it becomes intelligent piece 5 of 7 says this piece is smaller than it looks

The test: what would strengthen or weaken it

Would strengthen it: A published argument or measurement that narrows the lower end of the origin-of-life range downward, confirming that very small values are plausible; continued absence of detections in surveys of the atmospheres of planets around other stars consistent with an empty neighbourhood.

Would weaken it: Any discovery of life beyond Earth, which would cut the origin-of-life range at the bottom and re-crowd the galaxy in the computation; a persuasive case that the lower bounds used were placeholders rather than estimates and should be dropped.

These two lines are what the record is read against when a new paper or discovery is logged. A result of the kind named here, published with its evidence, is what would settle it; a press claim on its own settles nothing.

The last reading

Once a week the records are read against the news, each against the test above. Open a line to see what that week held. Every reading ever made of this record is on the reading page.

2026-10-16 Logged, no change By hand

Two arguments about the lower ends of the ranges: Baak and Snoek (2026-03-02) treat humanity as one count and derive that at least 0.051 communicating civilisations in the observable universe is allowed at 95% confidence, making solitude in the observable universe disfavoured at 97.6%, while for the galaxy the bound says almost nothing; Bertolami (2025-10-08) proposes replacing the lifetime term. Neither moves the origin-of-life range's low end up or down. Logged both.

Baak, 2026 Bertolami, 2025 source 1 source 2

How seriously it is taken

Widely read and influential; its method is now the standard way to present the equation, and this site uses it. The strength of its conclusion is disputed, mostly over how wide a range to allow for the origin of life.

The papers that bear on it

Implications of the Pessimistic Lower Limit on the Drake Equation 2 March 2026 Max Baak, Hella Snoek · logged; moved no range
A small polymerase ribozyme that can synthesize itself and its complementary strand 13 February 2026 Edoardo Gianni, James Attwater, Philipp Holliger · logged; moved no range
On Bayesian inference considerations and other issues concerning Drake's equation of Astrosociobiology 8 October 2025 Orfeu Bertolami · logged; moved no range

Books, films and other work that take it up

Nothing in the library illustrates this explanation yet.

Notes

Corrected 2026-10-16, against the paper itself (Sandberg, Drexler and Ord, 'Dissolving the Fermi Paradox', arXiv:1806.02404, read in full). Confirmed: median of 0.32 civilisations in the galaxy, 52% chance we are alone in the galaxy, 38% alone in the observable universe (sections 3.2 and the 'No update' row of Table 2), and 53%-99.6% and 39%-85% after the silence is counted (section 4). These agree with data/papers/sandberg-2018.toml and data/factors/life.toml. Wrong: (1) the claim that the lowest value used for the chance life starts was one in 10 to the 36. The paper has no such bound; it uses a bell curve in powers of ten, 50 powers of ten wide (standard deviation), centred on one start per planet, giving a mean chance of 0.5 and a median of 0.63 (section 3.1.1). The summary, body and objection text were rewritten to say that. (2) The body said the 0.32 median came from ranges drawn from the published literature; it came from the authors' own ranges (their Table 1). Resampling the literature gave a median of about 100, 30% alone in the galaxy and 10% alone in the observable universe (section 2). (3) The summary said the chance life starts was uncertain by 'a factor of one followed by a hundred zeros'; the paper says tens or hundreds of orders of magnitude (it takes 200 as a reference value), so the wording now says tens or even hundreds of powers of ten. (4) The Hart-versus-Drake objection: no published source for it was found, and the paper itself does address Hart's question in its section 4 (absence of settlement as evidence), so the sentence was reworded as an objection to the first step only and the 'critics argue' phrasing was dropped.

Think this page gets the explanation wrong, or misses an objection? Say so, or read how disagreement gets folded in. The record itself is data/explanations/no-paradox-just-wide-ranges.toml in the public repository. This page's address will never change; if the explanation is ever retired, the address will say what replaced it.