The French alphabet has 5 diacritics and 2 orthographic ligatures, to make 16 extra letters. In Latin scripts, letters with diacritics like ä, å or à, ñ, ö, and ü can be transcribed as ae, aa, gn, oe, and ue.

However, in telegraphy, there are no codes for Ââ, Êê, Ëë, Îî, Ïï, Ôô, Ùù, Ûû, Ÿÿ in Continental Morse. In Telex, all letters with diacritics including Çç, Èè are totally gone but in European variant of original Baudot code where only Éé existed.

Can modern French do without diacritic, or only with Éé at most, in telecommunication especially?

  • 2
    Curious... Is <ñ> actually transcribed as <gn>? I know the latter represent the same sound in French and Italian as the former does in Spanish, but I'd have guessed Spanish speakers would not in general see it as a straightforward replacement.
    – LjL
    Oct 10, 2019 at 15:59
  • 1
    @LjL Ñ is the shorthand of nn (as in old Spanish), which nh, ny, nh and gn in other languages sound like.
    – Schezuk
    Oct 10, 2019 at 16:19
  • 1
    @Schezuk I know that, but that wasn't my question. My question was whether it's common in Spanish to transcribe <ñ> as <gn> when the former is not available (such as in telegraphy as mentioned by the OP). I have a feeling it's not actually common. Colloquially on the internet, in my experience, Spanish speakers write <n> when <ñ> is not available.
    – LjL
    Oct 10, 2019 at 18:25
  • 2
    @LjL, The transcription was an Italian proposal to the 1890 International Telegraph Conference in Paris, which was declined by a Spanish delegate. Ñ's mnemonic in Morse code was assigned as TY.
    – Schezuk
    Oct 11, 2019 at 3:47
  • ha, that explains it given Italian orthography! I'm not too surprised that the Spanish iñored it.
    – LjL
    Oct 11, 2019 at 22:02

3 Answers 3


There is no common conventions in French for replacing letters with diacritics by digraphs. In contexts where the diacritics are not available, the usage is just to omit them. This is still common on uppercase letters, and was very common in the early days of Internet when programs were not 8-bit clean and accented letters were causing problems.

The result is usually understandable, but cases for which ambiguities arise are known (DUPOND-MORETTI A LA BARRE is a case which happened and is really better with an accent although there is little risk of misunderstanding, at least for those who know who Dupond-Moretti is; UN INTERNE TUE didn't, as far as I know, occur in the wild but it is impossible to choose between the four possibilities without the context).

  • It's important to note that if the sender is aware that there is no diacritics, they see and try to avoid ambiguities by rephrasing. I don't think anyone would write a telegraph with "UN INTERNE TUE" and think it will be understood unambiguously. So in practice it's pretty rare to have real ambiguity problems because of the absence of accents. Oct 11, 2019 at 15:39
  • There have also been imfamous (or possibly apocryphal) cases of telegrams whose meaning changed critically due to omission of a comma.
    – Barmar
    Oct 11, 2019 at 18:52

Absolutely, jst lke Englsh cn wrk jst fne wth all the vwls in the mddle of wrds rmvd cmpltly. If you're a native English-speaker, you probably read that sentence without much difficulty, even though I removed a whole bunch of information from it.

Languages, as a rule, include a whole lot of redundancy. Speaking is a pretty noisy channel, and a lot of information can get lost. So spoken language has evolved to compensate, and written language gets those benefits as a useful side effect. Sometimes written language ends up with even more redundancy for historical reasons; this happened in both French and English (e.g. writing lets us distinguish "bare" from "bear", while speech doesn't).

In French, diacritics do carry information—a "has" (from avoir) vs à "to", du "of" vs "had to do it" (from devoir), ca "approximately" (abbreviation of circa) vs ça "it". If you remove those diacritics, you lose that information. But there's enough redundancy that you can almost always make do without them, and you can add even more if necessary: if confusion between ca and ça is causing problems, you can use circa and ce instead. This is what's happening when pilots say "affirmative" and "negative" instead of "yes" and "no"—it adds more redundancy, since radio is an even noisier channel than normal speech, and confusing the two could be disastrous.

  • 3
    Cn y rd ths? D y wnt t lrn Hbrw? Hbrw s wrttn wtht mst vwls. T's fn n hlps bld brn clls. Oct 11, 2019 at 13:34
  • 2
    I can read it, and I don't even need to be a native speaker like Draconis seems to require! And honestly most of my brain cells are fried, so it must be pretty easy.
    – LjL
    Oct 11, 2019 at 14:23
  • @LjL Entirely valid! I just wanted to hedge slightly in case OP is a non-native speaker who runs into trouble with my example.
    – Draconis
    Oct 11, 2019 at 15:43
  • m t. bt thn gn, y s, n r l s nt snk. we could spawn a new area51 puzzling site from this. or does it exist already?
    – dlatikay
    Oct 11, 2019 at 18:43
  • @dlatikay I must declare defeat at n r l s.
    – LjL
    Oct 11, 2019 at 21:33

French with accents and diacritics is confortable to read, as it provides all necessary information. "Bare" French requires some guesswork. Middle French used not to have accents, but there was quite a lot of mute consonants, that were used to clarify neighboring vowels.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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