I first noticed in this answer that there is something sneaky going on with the word photon: its ‹t› is the stressed allophone of /t/, a fully aspirated [tʰ]. It does not reduce to [t] or [ɾ] the way it does in words like voting. Other words with the same issue include proton and lepton.

The only way I can explain this would be if the second syllable in such words bears secondary stress, so [ˈfoʊˌtʰɑn].

Even so, the question remains: why does this happen? Is it because these are all “new” words? Or does Greek somehow enter into it?

I notice other new scientific terms have the related issue of an unreduced vowel in the syllable without primary stress, such as in hadron, quasar, protein, baryon, genome.

So the ‹t› in proton and photon works more like it does in Motown or cow town, but those are at morphemic boundaries. The only “old” word I could think of where it might work the same way (remain aspirated) might be in canton. But there the vowel in the unstressed syllable does seem a bit reduced, just not all the way.

What is really going on here?

Edit: The mid-word aspirated ‹t› also occurs in futon and wonton (as in wonton soup), but not in Briton, Milton, tartan, titan, wanton (as in a wanton woman).

  • 3
    Perhaps because the 'endings' in those second syllables, although bound, have lexical rather than merely grammatical force? Compare proton/protean or photon/photic. – StoneyB Jan 16 '13 at 13:19
  • 4
    I suspect it's the secondary stress. Everybody says /proʊtɑn/ and not /proʊtən/. If the second syllable didn't have secondary stress, you could reduce the vowel. Compare crouton, which also has secondary stress on the on. (And listening to some pronunciations on Forvo, Avon seems to in the U.S. but not the U.K.) – Peter Shor Jan 16 '13 at 13:46
  • 5
    When I read this question initially I thought someone had accidentally posted a physics.SE question here! – Marcus_33 Jan 16 '13 at 14:43
  • 4
    For a rather drastically different scheme for naming particles, you might be interested in Uncleftish Beholding. The words it uses for proton and photon are firstbit and lightbit. :) – starwed Jan 16 '13 at 17:40
  • 5
    Is this behavior all that anomalous? Don't the words protein and eighteen do the same thing? Their second syllable has secondary stress (at least in AmE), which means the vowel doesn't get reduced and the t is aspirated. – Peter Shor Jan 21 '13 at 12:58

As Bruce Hayes (UCLA) puts it,

Word-medial voiceless stops are aspirated provided they are in the onset of a stressed syllable and are not preceded by a strident (Hayes 1995: 13).


proton [ˈpʰɹouˌtʰɑn] in AmE.

There's no secondary stress in "voting" in AmE.

See a note on aspiration in the Longman Pronunciation Dictionary (3rd ed., Wells 2008: 45) - those rules are rather comprehensive; at least, that should be enough for someone who's not a professional phonetician.

The most important thing to remember is that aspiration is not binary, aspirated or unaspirated, but rather gradual. See fig. 6.6 in Ladefoged 2006, A Course in English Phonetics.


Actually, it is the suffix -on that causes the secondary stress to happen in most situations, like when it is added after a consonant.

This phenomenon is very common in AmE. But not the case in BrE: OBED and many other British dictionaries show otherwise (no secondary stress). [ Though many British physicists still stress the word in a way that gives secondary stress on the second syllable.]

OE suggests that "on" made its first appearance in the early 19th century in atomic physics:

ion (n.) 1834 [interestedly I've heard people pronounce it two ways: ˈaɪən and ˈaɪˌɑn]

introduced by English physicist and chemist Michael Faraday (suggested by the Rev. William Whewell, English polymath), coined from Greek ion, neuter prp. of ienai "go," from PIE root *ei- "to go, to walk".

So called because ions move toward the electrode of opposite charge.

It was not until the late 19th century that it is used as a suffix.

In 1894, Stoney coined the term electron to describe these elementary charges, saying, "... an estimate was made of the actual amount of this most remarkable fundamental unit of electricity, for which I have since ventured to suggest the name electron". The word electron is a combination of the word electric(icity) and the Greek suffix "tron", meaning roughly 'the means by which it is done'. The suffix -on which is now used to designate other subatomic particles, such as a proton or neutron, is in turn derived from electron.

In the 20th century, it becomes a fashion to use it for coining new terms not only in physics:

proton is coined in 1920 {it was used earlier in embryology (1893) at a translation of German anlage},

neutron [ˈn(j)uˌtrɑn] is coined in 1921,

photon (a particle-like package of light) is coined in 1926,

positron [ˈpɑzəˌtrɑn] (the antiparticle of electron) is coined in 1933,

negatron [ˈnɛɡəˌtrɑn] (the antiparticle of proton) is coined around 1933,

fermion [ˈfɜrmiˌɑn] (any particle that follows the Pauli exclusion principle) and

boson [ˈboʊsˌɑn] (any particles that don't follow the exclusion principle) are coined in the mid 20th century

[A famous example of boson would be the higgs boson which may help us to gain a
deeper insight into dark matter and dark energy, recently discovered by CERN (though it is not exactly the higgs boson predicted by the Standard Model)]

lepton (any particle that does not undergo strong interactions but follows the exclusion principle E.g. electron) is coined in 1948

muon [ˈmjuˌɑn] (an unstable, negatively charged lepton) is coined in the mid 20th century

gluon [ˈgluˌɑn'] (an exchange particle responsible for strong interaction) is coined in the mid 20th century

but also in other branches of science:

interferon [ˌɪntərˈfirˌɑn] Biochemistry

a protein released by animal cells, usually in response to the entry of a virus, that has >the property of inhibiting virus replication.

1957, coined in English, so called because it "interferes" with the reduplication of viruses. From interfere + subatomic particle suffix -on.

codon [ˈkoʊˌdɑn] Biochemistry

a sequence of three nucleotides that together form a unit of genetic code in a DNA or RNA >molecule.

1962, from code (n.) + -on.

operon [ˈɑpəˌrɑn] Biology

a unit made up of linked genes that is thought to regulate other genes responsible for >protein synthesis.

1960s: from French opérer ‘to effect, work’ + -on.

radon [ˈreɪˌdɑn] Chemistry

the chemical element of atomic number 86, a rare radioactive gas belonging to the noble gas series.

*1918, from radium (q.v.) + -on suffix of inert gases. *

A very famous physicist, Richard Feynman, even used the prefix -on to make up his own word, "parton" when he explained a seemingly-complex concept. And thus you can tell at that time the prefix -on was pretty well-used in the science community.

Notes: Not all 19th & 20th century scientific terms ending with "on" make use of the prefix "-on". One example would be ion as mentioned above, and here is another example

neuron [ˈn(j)ʊˌrɑn]

"a nerve cell with appendages," 1891, from German Neuron, from Greek neuron (see neuro-). Used earlier (1884) for "the spinal cord and brain."

Both the suffix -on and terms that exhibit secondary stress, such as neuron, ion, come from Greeks words.

So you are right, the Greeks are behind this.

  • You don't address why the popular British pronunciation does not stress the last syllable. If the Greeks were behind it, then I think the Brits would have followed suit. See Brit.: /ˈnjʊərɒn/ vs. U.S.: /ˈn(j)ʊˌrɑn/. I think you are closer with the French roots, which thusly drove the dialect per my answer. – New Alexandria Jan 20 '13 at 20:43
  • 1
    @NewAlexandria You have to be careful there. Some people analyze any English syllable with an unreduced vowel as having at least some sort of stress, since unstressed syllables are supposed to reduce in English. If it were a schwa, then I would agree than it is unstressed, but as it has the CLOTH vowel, I wonder. Perhaps a better marker is the aspiration on the t at the onset of the “unstressed” syllables, which is indeed why I chose the examples I did. – tchrist Jan 20 '13 at 20:48
  • 1
    @tchrist yeah. I've watched many lectures (on Internent) and the professors always pronounce the particles with the secondary stress. – user19341 Jan 20 '13 at 21:03
  • 1
    @Arch I think it behoves us to find one. That would settle the matter. Surely there must be something by Dawkins we can listen to. – tchrist Jan 20 '13 at 21:36
  • 1
    @Arch Good find: I like Brian Cox. And you’re right, he definitely sounds like he is stressing the word proton in a way that gives secondary stress on the second syllable. It is not reducing, and it is aspirated. Strikes me as pretty clear confirmation. – tchrist Jan 21 '13 at 8:42

You are correct in your guess about Greek. The first "-ton" (in relation to particle physics) was the proton. Ernest Rutherford named the proton from the Greek πρῶτον, meaning "first".

Incidentally, Rutherford also had William Prout in mind when he named the proton and even suggested that one of two names be used: protons or proutons. Prout was the first person to theorize a fundamental particle of matter, which he referred to as a protyle.

As particle physics evolved, further discovered particles were named in a similar fashion. Photons are the elementary particles of light, and were named "φῶς + ton". Most other elementary particles are named either for Greek words + "on" or "ton", or else after various physicists who had a hand in their discovery/description with "on" tacked on the end to indicate that it is a subatomic particle.

I also think your speculation about them being new words likely has something to do with their pronunciation. Proton began appearing in science literature in the 1920's. Every other subatomic particle name has been coined since then - these words are all less than 100 years old. They all originated after the invention of audio recordings, and none are used commonly enough to be changed by regional dialects.

  • 2
    Wilczek supposedly named the axion after a brand of bleach, on the basis that it sounded like a particle name. :) (I've heard him mention this at a talk, and I think you can check this article for confirmation.) – starwed Jan 16 '13 at 17:34
  • 4
    Great answer but I have to point out that ion (1834) is actually the word that influenced other physicists to name new discovered particles with the suffix -on [instead of "ton"]. Electron (1894) was later coined. After that then proton (1920) was coined. I study quantum mechanics and particle physics so I'm pretty sensitive to this :) – user19341 Jan 20 '13 at 20:20
  • 1
    This answer only seems to cover US English. In UK English, there is no stress on the second syllable, and words like proton behave exactly as expected. – Rory Alsop Jan 20 '13 at 20:51
  • 1
    That makes no sense - what would the alternative be? Of course it is aspirated - it is a strongly defined t sound, as is the initial one. – Rory Alsop Jan 20 '13 at 22:08
  • 1
    Note also that the suffix/ending is -on, not -ton : the stem of Attic phôs "light" is phôt-, as can be seen in photograph, photovoltaic, photosynthesis.... Note that the stem was pha(e)- in other dialects than Attic (cf. Greek pharos, Phaethôn); it is related to Greek pha(i)n- "radiate, appear", as in English phaenomenon, phantastic... – Cerberus Jan 20 '13 at 23:23

The OED acknowledges that the pronunciations differ between US and British usage:

  • Brit. /ˈprəʊtɒn/
  • U.S. /ˈproʊˌtɑn/

I will now speculate generously, but precisely ~~~

Words like proton rose in prevalence within the US populace during the race into the 'Atomic Age'. Texas, due to the wealth of oil tycoons and the geological coincidence of petroleum and heavy metals, was the hotbed of nuclear developments in the US.

Texas are known for their particular drawl. The drawl is of French origin, and in the US became know as Cajun. This French-speaking influence is the basis for much of what is known as the "Southern Dialect". Particularly, this dialect is known for emphasizing the second syllable:

"Cajuns are quite distinct. They tend to place emphasis on the second syllable of a word, when possible. They use French words and phrases frequently in everyday speech. The names of Cajun cities and people often end in the very French suffixes '-ieux' and '-eaux', pronounced like a short 'o' sound"

Those familiar with the French style of speaking know that such words mentioned above, though 'short' in the Cajun dialect, are often emphasized.

Emphasis is often used for words of increased value or meaning.

As such, it is no stretch to conclude that the word "proton" would have garnered this attention by Southern speakers in the region, many of whom were staking business interest in booming nuclear operations - whose core science was based around the activities of Protons and other subatomic particles.

  • Could you please show phonetic transcriptions, not phonemic ones? Otherwise the distinction is lost. – tchrist Jan 20 '13 at 21:40
  • @tchrist For how the drawl pronounces 'photon' and the like? – New Alexandria Jan 21 '13 at 1:39
  • No, I mean for how you think the US and UK pronunciations are. I am looking for aspiration, remember. – tchrist Jan 21 '13 at 3:48
  • Notice how in the link that Arch posted, it certainly sounds like Professor Brian Cox is saying proton with secondary stress on the second syllable. – tchrist Jan 21 '13 at 8:59
  • I'll look for the same. Someone in my family was born and raised in England + various part of the EU, has PhD science training - and doesn't pronounce words like 'proton' with a AmE stress on the last syllable. – New Alexandria Jan 21 '13 at 18:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.