Does English contain accented letters? Is there a standard list of accented letters that are used in the English language?

In my application I'm currently allowing the characters a-z and áéóúíý äëöüïÿ '-. The reason I'm allowing '- as well is because some people names contain them.

  • 1
    Are English names the only inputs you are considering? Jun 26 '14 at 16:13
  • 6
    This question appears to be off-topic because it is about accents/diacritics, which aren't really a feature of English at all. So it's entirely subjective which particular symbols some people might accept as "valid in English" (imho, none). Jun 26 '14 at 16:28
  • 1
    @FumbleFingers What about cliché? coöperate? Or, since OP asks about names, Noël? Renée? The diacritics are not essential, but they are used here and there - and we generally concede nominees the right to spell their names in any way they choose. Jun 26 '14 at 18:13
  • 1
    @FumbleFingers I think that at worst, this question is on topic indirectly and at best, it is on topic.. well... indirectly. Because people who read English will read words that have diacritics. It has become a small world, after all. The OP properly selected the answer that indicates English doesn't have accented letters which is basically your point. And the two answers are good. Note: I agreed with you to close at first, but then retracted after reading the answers. Jun 26 '14 at 20:12
  • 2
    Off topic here - but what will you do if somebody types a character that's not on your list? Will you tell them their name is invalid? Aug 9 '14 at 1:15

You could count accents on imported words (café, resumé, piñata). Some names also include accents (Zoë, Brontë, Beyoncé). Then there's the rare case of using accents and umlauts over certain vowels (naïve, coöperation) - this is a non-standard convention not used by the majority of English speakers.

It's rare to see accents or diacritics in modern English as anything other than a stylistic typographical choice. A good example of this is Häagen-Dazs ice cream, whose Bronx, NY creators put a superfluous umlaut in the name to add a European flair.

Many other languages with Latin alphabets use umlauts and other diacritics. Examples from the original question are missing ñ, ø, vowels with macron (top bar ā, ī etc) or circumflex (â,ê etc) as well as many others.

That gets to the key point of the question, which is that you should be asking about the Extended Latin alphabet - not "English."

Assuming by "application" you mean software, you should look into how Unicode characters work, and what localization options are provided by the platform you're developing for. This is a complicated topic that goes well beyond just the language, and is better suited to asking at StackOverflow.

  • Thanks for saving me. I was ready to answer, no, until I read your excellent examples of English words with accents. A brief enrichment - in the past, accents were omitted because typewriters didn't include those keys. Word processors made the inclusion of accentuated letters much easier, so accented letters are coming back into vogue.
    – JimM
    Nov 15 '14 at 21:13

English traditionally uses very few diacritical marks compared to other European languages.

English employs a huge number of loanwords, however, so to provide enough characters to represent all commonly used English words, you may as well use the entire Latin-1 character set. We would not want, for example, to deprive people of proper orthography for their

  • ångströms
  • arrastãos
  • daïses
  • entrepôts
  • façades
  • fêtes
  • Māori
  • piñatas

Certain romanization systems require still further characters. For example, the capital of North Korea should technically be rendered as P'yŏngyang and the lingua franca of Malawi is technically Chicheŵa.

As you can see, where diacritics remain in use, they are mostly retained from their original languages, and are especially retained where they help avoid confusion with another word. For example, façade is fading in favor of facade, but maté and résumé are still going strong.

Traditionally, the diaresis (¨) is used to indicate that two adjacent vowels should be pronounced separately instead of as a dipthong, as in naïve. This usage has been fading fast, with The New Yorker among the few famously clinging to the old convention.

Otherwise, diacritics are largely found in poetry as scansion markers. Most commonly, the grave accent (`) indicates that a vowel should be pronounced where it is ordinarily suppressed; for example, blessèd and learnèd have two syllables whereas blessed and learned have one, but such examples are only really found in poetry. Similarly, the acute accent (´) indicates which syllable should receive stress, the breve (˘) shortens a syllable, and the macron (¯) lengthens it, but these are almost never encountered elsewhere outside of loanwords.

  • I don't deny people the right to use words like ångströms if they're writing in their native language, but I don't endorse it if they're writing in English. Jun 28 '14 at 14:52

In English and a couple of other languages, when two vowels occur in adjacent syllables, with no intervening consonant, you may place two dots over the second to indicate that this is not a diphthong. So you may write coöperate, daïs, preëmptive, reëlect and a handful of others. My favourite is skiïng.

This is English. This is nothing to do with whether a word is introduced from a foreign language. So it is entirely incorrect to say that English has no accented letters. It is, of course, optional; but one publication for which this is the "house style" is The New Yorker.

Of course, if you include introduced words, then there are dozens, possibly hundreds. Café, fiancé, hāngi, naïve and ångström are the obvious ones for me.

  • Adopting diacritics as a stylistic convention (an imported practice itself) is very different from them being a fundamental part of the language. In other languages, 'ö' and 'o' are separate characters. Among the 26 letters of modern English none have accents. Google Ngrams for all examples above show 100% without umlauts - the other versions aren't even found. While "you may," nobody does. All this aside, the question was fundamentally a misunderstanding about the extended Latin alphabet used in a programming context. It really didn't belong on this site.
    – mc01
    Aug 10 '14 at 0:29
  • @mc01 - it's not true that "nobody does", which is why I cited The New Yorker as an example. I have seen all of the words that I listed, written as I wrote them here. It is a feature of English, unless you want to limit the word "English" to mean the particular variety that you yourself write. So your argument in the comment here is as incorrect as your answer. Aug 10 '14 at 1:35
  • Edited my answer above. I concede this exists & I've seen it too, but this remains a rare, non-standard, stylistic choice. I amend my comment to say "practically nobody" has used this convention for several decades. Even the New Yorker admits to being an obstinate, holdout edge case. My goal was to address the intent of the OP's question (how to account for accents & diacritics in software) without getting bogged down in minutiae that doesn't help solve the problem. You are correct, but I didn't see value in burying the key point among a laundry list of qualifying fringe examples. Cheers ~
    – mc01
    Aug 12 '14 at 0:05

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .