Papex
Explore
← Blog

Note16 min read

Pauli's neutrino letter and other hypotheses that waited for evidence

From Pauli's neutrino letter to the Higgs boson: how famous hypotheses were shared, how many years the evidence took and how each is credited today.

Papex

Wolfgang Pauli proposed the neutrino on 4 December 1930, in an open letter that began "Dear radioactive ladies and gentlemen". It took 26 years before Clyde Cowan, Frederick Reines and their colleagues detected the particle. Many famous scientific hypotheses followed the same path: shared first in a letter, a talk or a short paper, then tested years or decades later, usually by other people. The cases suggest that a date alone rarely settled anything. What helped was a precise statement, a way to test it and a record that others could find and cite.

Below are seven such hypotheses, from Pauli's letter to the Higgs boson: how each was shared, how long the evidence took, how it is credited today and what the cases suggest about sharing a hypothesis well. Dates and quotations come from the original papers, archive copies, Nobel Prize pages and scholarly histories, all checked on 13 October 2026.

Seven hypotheses that waited for evidence

HypothesisWhoHow it was sharedSharedEvidence or acceptanceWhenGap
The neutrinoWolfgang PauliOpen letter to a meeting in Tübingen1930Reactor neutrinos detected by Cowan, Reines and colleagues195626 years
Eka-aluminium, eka-boron, eka-siliconDmitri MendeleevGaps in a textbook table (1869), described in an 1871 article1871Gallium, scandium and germanium discovered1875–18864 to 15 years
Gravitational wavesAlbert EinsteinPaper to the Prussian Academy of Sciences1916Binary pulsar (indirect, reported 1978); LIGO (direct)201599 years
Continental driftAlfred WegenerTalk in Frankfurt and articles (1912), then a book (1915)1912Seafloor spreading; plate tectonics, with a different mechanism1960sabout 50 years
Chemiosmotic couplingPeter MitchellTalk in Stockholm (1960), then a paper in Nature1961Experiments in many laboratories; Nobel Prize197817 years
The Higgs mechanism and bosonEnglert and Brout; Higgs; Guralnik, Hagen and KibbleThree short papers in Physical Review Letters1964New particle observed by ATLAS and CMS at CERN201248 years
The positronPaul DiracPaper in Proceedings of the Royal Society1931Positron tracks recorded by Carl Anderson19321 year

What did Pauli's neutrino letter say?

Pauli proposed that beta decay sends out a small, electrically neutral particle along with the electron, carrying off the energy that seemed to go missing. The letter is dated Zürich, 4 December 1930, and headed as an open letter to the radioactivity group at a meeting in Tübingen. CERN Archives, which holds a typed copy, gives its opening as "Dear radioactive ladies and gentlemen" ("Liebe Radioaktive Damen und Herren" in the copy itself).

He called the particles "neutrons", and he was specific. He gave them spin 1/2 and a mass that, in the English that Reines quoted in his Nobel lecture, "should be of the same order of magnitude as that of the electron and in any event no greater than 0.01 of the proton mass". The Nobel Prize's account says Pauli offered what he called a "desperate solution". He was candid about the odds: "I admit that my expedient may seem rather improbable from the first, because if neutrons existed they would have been discovered long since." In the German original he wrote that he did not yet dare to publish the idea, and asked the experimenters how such a particle might be detected. He did not attend: Reines recalled that he "had to go out to a dance".

Reines's lecture cites the letter as "unpublished", alongside Pauli's remarks in the proceedings of the 1933 Solvay conference, printed in 1934. Enrico Fermi built the particle into a theory of beta decay in 1934, and after James Chadwick found the neutron, Fermi renamed Pauli's particle the "neutrino". Detection looked hopeless: Hans Bethe and Rudolf Peierls concluded in 1934 that "there is no practically possible way of observing the neutrino".

Reines and Cowan used a nuclear reactor instead, and in June 1956 cabled Pauli that they had "definitely detected neutrinos". Their paper, "Detection of the Free Neutrino: a Confirmation", appeared in Science on 20 July 1956. (Strictly, what they caught was the electron antineutrino.) Pauli's reply, which never arrived, read: "Thanks for the message. Everything comes to him who knows how to wait." Reines shared the 1995 Nobel Prize in Physics "for the detection of the neutrino"; Cowan had died before the award.

How did Mendeleev predict gallium and germanium?

Mendeleev left gaps in his periodic table and, by 1871, described the elements that should fill three of them. His first table, printed in his 1869 textbook, had question marks where no element was known. According to the American Institute of Physics, "At first Mendeleev did not think of those placeholders as explicit predictions of unknown elements." In what Philip Stewart calls his major article of 1871, Mendeleev "devoted several pages to discussing the properties to be expected of eka-aluminium, eka-boron and eka-silicon", with expected atomic weights of 68, 44 and 72.

The elements turned up within fifteen years. Paul-Émile Lecoq de Boisbaudran found gallium in Paris in 1875; the Royal Society of Chemistry notes that "What Boisbaudran didn't realise was that its existence, and properties, had been predicted by Mendeleev." Lars Nilson found scandium in 1879 and Clemens Winkler found germanium in 1886, with an atomic weight of 72.63. The discoverers are credited with the elements and Mendeleev with the prediction: two kinds of credit, both on the record.

When did Einstein predict gravitational waves?

Einstein predicted gravitational waves in a paper presented to the Prussian Academy of Sciences on 22 June 1916, and LIGO first detected them directly on 14 September 2015, 99 years later. The paper, "Näherungsweise Integration der Feldgleichungen der Gravitation", has a section on plane gravitational waves.

Einstein doubted his own prediction. The 2017 Nobel press release notes that he "was convinced it would never be possible to measure them". In 1936 he told Max Born that he and a young collaborator had found "that gravitational waves do not exist". When the Physical Review sent him a critical referee's report on the paper, written with Nathan Rosen, he withdrew it in anger. The version printed in the Journal of the Franklin Institute in 1937 appeared, Daniel Kennefick writes in Physics Today, "with radically altered conclusions". Before he had found a way out, he told a Princeton audience: "If you ask me whether there are gravitational waves or not, I must answer that I do not know. But it is a highly interesting problem."

The first evidence was indirect. Russell Hulse and Joseph Taylor found a binary pulsar in 1974 whose orbit shrinks as it loses energy "in accordance with what Einstein in 1916 predicted"; the first report of the effect came at the end of 1978. LIGO's two detectors recorded the direct signal at 09:50:45 UTC on 14 September 2015, published on 11 February 2016 as "the first direct detection of gravitational waves". Its first reference is Einstein's 1916 paper, page 688. Rainer Weiss, Barry Barish and Kip Thorne received the 2017 Nobel Prize in Physics.

Why was continental drift rejected at first?

Continental drift was rejected for decades mainly because Alfred Wegener could not say what force moved the continents. He presented the idea to the Geologische Vereinigung in Frankfurt on 6 January 1912 and published it that April, despite advice from his father-in-law not to interfere in geological matters as a meteorologist. His book, Die Entstehung der Kontinente und Ozeane, followed in 1915 and reached a fourth revised edition in 1929. His evidence was the fit of the coastlines, matching fossils across the Atlantic and signs of past climates in the wrong places.

The US Geological Survey calls this "a fatal weakness": his critics asked what forces could move "such large masses of solid rock over such great distances". He suggested that continents plough through the ocean floor, and Harold Jeffreys "argued correctly" that this was physically impossible. The theory "never gained acceptance in his lifetime"; Wegener died on the Greenland ice cap in 1930.

The answer came from the sea floor: from 1961, scientists began to suggest that it splits apart along mid-ocean ridges. In 1963, Frederick Vine and Drummond Matthews published "Magnetic Anomalies Over Oceanic Ridges" in Nature. As Henry Frankel wrote in 1988, it "attracted little immediate interest" but became one of the seminal contributions to the theory of plate tectonics. Credit today is careful: the USGS calls continental drift "the forerunner to the theory of plate tectonics". The continents do move, but not as Wegener proposed, and the USGS notes the irony that what drives the plates is still among the field's chief outstanding questions.

What was Peter Mitchell's chemiosmotic hypothesis?

Peter Mitchell proposed in 1961 that cells make ATP using a gradient of protons across a membrane, not through a direct chemical link between enzymes. He first outlined the idea at a meeting in Stockholm in 1960 and set it out in Nature in July 1961: electron flow drives protons across the membrane, and their flow back drives ATP synthesis.

It went against the field. According to the Nobel Prize's account, most rival hypotheses postulated a direct chemical interaction, yet "no experimental evidence could be obtained for any of these hypotheses". Mitchell's theory "was first received with scepticism", until work in his and many other laboratories over 15 years showed "that the basic postulates of his theory are correct". He received the 1978 Nobel Prize in Chemistry alone, and his Nobel lecture is unusually frank: "Of course, I might have been wrong."

Who predicted the Higgs boson in 1964?

Three groups published the mechanism in Physical Review Letters in 1964, and Peter Higgs's paper drew attention to the boson that now bears his name. François Englert and Robert Brout's paper was received on 26 June 1964, Higgs's paper on 31 August and Gerald Guralnik, Carl Hagen and Tom Kibble's paper on 12 October.

The University of Edinburgh records that Higgs first sent his paper to Physics Letters, "but it was rejected on the grounds that it did not warrant rapid publication". He revised it for Physical Review Letters, and the revised paper "drew attention to the possibility of a massive spin-zero boson in its final paragraph". The referee, Yoichiro Nambu, asked him to relate it to Englert and Brout's paper, which Higgs had not seen because "the Brussels group did not send preprints to Edinburgh".

On 4 July 2012, CERN announced that its experiments had observed a "particle consistent with long-sought Higgs boson", at a mass around 125–126 GeV, cautioning that "Positive identification of the new particle's characteristics will take considerable time and data." The 2013 Nobel Prize went to Englert and Higgs, who "proposed the theory independently of each other (Englert together with his now deceased colleague Robert Brout)". Guralnik, Hagen and Kibble were not included. The Nobel statutes allow no more than three people per prize and do not consider work by someone who has died. The American Physical Society's 2010 Sakurai Prize went to all six.

Do predictions always take decades?

No: some are confirmed within a year, and some are still waiting. Paul Dirac predicted an antiparticle of the electron in a 1931 paper, and Carl Anderson photographed its tracks in August 1932. Physics Today notes that "Anderson had not set out to hunt antimatter" and only later realised his particle matched Dirac's. His 1933 paper named them: "These particles will be called positrons." Anderson received the 1936 Nobel Prize "for his discovery of the positron".

At the other end, the Riemann hypothesis, formulated in Riemann's 1859 paper, is still a Clay Millennium Prize Problem: checked for the first ten trillion cases, but not proved.

Do most hypotheses turn out to be right?

No. Most turn out to be wrong, and that is how science normally works. A list like this one is survivorship bias in action: it is made of the hypotheses that survived. Even Mendeleev's record is mixed. Eric Scerri's scorecard of his predictions, as Stewart reports it, concludes: "A success rate of half is clearly not outstanding by any stretch of the imagination." The rival hypotheses in Mitchell's field found no experimental evidence despite intensive research, and Einstein was wrong in 1936 about his own prediction. A hypothesis is worth sharing because it can be tested, not because it is likely to be right.

What these cases suggest about sharing a hypothesis well

None of these people had a registry. What helped their ideas get tested and credited were habits anyone can copy.

  • Precise. Pauli gave a spin and an upper bound for the mass; Mendeleev gave atomic weights and properties. A precise claim can be checked, and it can fail.
  • Testable. Pauli asked, in the letter itself, how the particle could be detected. Saying what result would show you wrong tells others what to build or measure.
  • Dated. Pauli's letter carries its place and date, and "received" dates put the three 1964 papers in order. A date does not settle credit on its own, as "independently of each other" shows, but without one there is little to discuss. Versions matter too: Higgs revised his paper after a rejection, and Einstein reversed his 1936 conclusion.
  • Addressable. Reines had to cite Pauli's letter as "unpublished", and Higgs missed a paper that was never sent to Edinburgh. The LIGO paper, a century on, could cite Einstein by journal and page. A stable address, such as a DOI or a permanent link, lets others find and cite the exact version you meant.
  • Attributable. Prizes are limited; records need not be. Naming everyone who contributed, as the Sakurai Prize did, keeps each part visible.

For the practical side, from writing the claim to choosing where to put it, see How to publish a hypothesis. If you are worried about being scooped before you share, see how to protect a research idea.

Recording a hypothesis on Papex

Papex is one place to keep such a record. A Hypothesis record has fields for the claim, why it matters, what would show it wrong, the expected outcomes and how it could be tested, with references and "builds on" links to exact versions of other records. It starts as a private draft. When you sign, the signature links you to that exact version; a later edit becomes a new version with its own signature and date, and earlier versions stay as they were. At signing only a salted fingerprint leaves Papex, and it is timestamped with OpenTimestamps and anchored in Bitcoin; the proof checks with any OpenTimestamps tool, without Papex.

You then choose to keep the version private, share it with people you invite or publish it. A published version has a permanent address of the form papex.org/id/PXH-…/v/1, one per version, and its page gives a citation to copy, with BibTeX, RIS, CFF and JSON-LD under Other formats. Other signed-in people can add supporting evidence or counter-evidence with sources, or raise challenges, shown on the record as activity rather than a verdict.

The limits matter. Papex mints no DOI; if you need one, use Zenodo, OSF or a journal. It is not a journal and does no peer review. A signed, timestamped record shows who signed which version and when; it does not show that the hypothesis is right, original or first, or that anyone owns the idea. What a Bitcoin timestamp proves explains the mechanism, and you can start a private draft at papex.org.

Sources

Keep reading

How to publish a hypothesis without data: journals, preprints and DOIs

Which journals and preprint servers accept a hypothesis without data, from Medical Hypotheses to Octopus, how to get a DOI, and how to cite it.

Guide

How to cite ChatGPT, Claude AI and Gemini in APA, MLA and Chicago

APA 7, MLA 9, Chicago, AMA, IEEE and Harvard formats for ChatGPT, Claude and Gemini, from each guide's current advice, and when to disclose instead of cite.

Guide

How to write an AI disclosure statement for a research paper

AI disclosure statement examples and templates for research papers: what to include, where it goes, if grammar help counts, and 11 journal policies compared.

Guide