I've noticed that on Wiktionary for Esperanto words that include hyphenation, polymorphemic words (usually ones with derivational suffixes) are not hyphenated as I would expect.

For example, the word pafilo is listed as being hyphenated as paf‧i‧lo rather than pa‧fi‧lo. I would have expected the latter since I would have thought that the /f/ of the root paf- should occur in the onset and usually when we break words up through hyphenation we pay attention to syllable boundaries and not so much morphological boundaries.

I suspect that this is not actually a rule of hyphenation in Esperanto but an unfortunate side effect of inattentive coding since these hyphenations and IPA transcriptions are produced automatically by a template.

Am I right in this or is this actually a quirk of Esperanto orthographical conventions?

P.S. This is also mirrored in the IPA transcription such that pafilo is transcribed /pafˈilo/, which implies a syllable boundary directly after /f/, rather than /paˈfilo/. It seems more obvious to me that this is an oversight by whoever coded the IPA transcription template.

3 Answers 3


I worked on that script on the English Wiktionary and it's not perfect. It's actually really hard to automatically syllabize words with a script.

I would syllabize that word as "pa‧fi‧lo" and the script was supposed to syllabize it like that. Somebody overrode the script. I corrected it.

  • Good to know! Unfortunately, it seems that the overriding of the correct syllabification by the script has occurred on many, many pages.
    – Miztli
    Commented Apr 26, 2019 at 9:13

There are no obligatory rules nor is there an overwhelming tendency in practice for hyphenation in Esperanto. In a Lingva Respondo ("Linguistic Answer") from 1893*, Zamenhof stated that morphological hyphenation would be most logical, but that the question in fact is not important and you can divide words as you like.

It is my impression, however, that the most common, if at all, kind of hyphenation today is according to syllables. The syllable structure of Esperanto is partly flexible nevertheless, you can get an impression from the §§2-3 in the Fundamenta Ekzercaro.

(*) La Esperantisto, 1893, p. 32

Transportante la vortojn el unu linio en la sekvantan, ni ordinare dividas ilin per iliaj partoj gramatikaj, ĉar ĉiu parto gramatika en nia lingvo prezentas apartan vorton. Tiel ni ekzemple dividas: «Esper-anto», «ricev-ita» k.t.p. Sed tio ĉi tute ne estas deviga regulo; ni faras ĝin nur por ne rompi subite kun la kutimoj de aliaj lingvoj: efektive tiu ĉi maniero havas nenian celon kaj signifon, ĉar la transportado de la vortoj estas afero pure papera, havanta nenion komunan kun la leĝoj de la lingvo; ni konsilas al vi per nenio vin ĝeni en la dividado de la vortoj kaj fari ĝin tute tiel, kiel en la donita okazo estos al vi pli oportune. Eĉ se vi dividos ekzemple «aparteni-s», ni vidus en tio ĉi nenion malregulan, kvankam la aliaj lingvoj (tute sen ia logika kaŭzo) ne permesas tian dividadon.


I would disagree with your criticism of the IPA transcription: IMHO the /paf'ilo/ one is correct, as the pronunciation mirrors the morpheme structure, not just the syllables. With /pa'filo/ it almost sounds like you're talking about something related to filo, rather a gun. The phonology here supports the correct reading of the morphological structure.

As nothing is mentioned about hyphenation in the 16 grammatical rules, I would think it's not standardised and thus arbitrary. My own preference would always be to follow morpheme boundaries where possible.

  • Except that would go against a very, very strong tendency in natural languages to prefer to, all other things being equal, syllabify such words as CV.CV rather than CVC.V. That doesn't mean that that can't happen in Esperanto, it would just make it exceptionally odd in typological terms.
    – Miztli
    Commented Apr 26, 2019 at 10:30
  • 1
    As for hyphenation, usually in languages of the world, hyphenation follows the language's internal rules of syllabifcation, though I would understand why Esperanto might deviate from this and hyphenate according to morpheme boundaries first and syllable boundaries second (assuming what I say above about syllabifcation is true, for the sake of argument, at least).
    – Miztli
    Commented Apr 26, 2019 at 10:30
  • Hyphenation is not always the same as syllabification: in German, for example, the word for a high school degree is Abitur. According to German syllabic rules it would be hyphenated Abi-tur, but in fact the correct hyphenation is Ab-itur, as it is derived from Latin ab-ire. So again the (historical) morphology trumps the syllable structure when it comes to hyphenation. Commented Apr 26, 2019 at 12:11
  • 1
    (It might be worth adding that syllables are related to spoken language, whereas hyphenation is only used in writing. They do not have to coincide, though they often do. But nothing is language is ever without exceptions) Commented Apr 26, 2019 at 12:14
  • 2
    I strongly disagree with the claim that there is difference in pronounciation between a word like "koleg/o" and "kol/eg/o" (except for emphasis). In fact it's a feature of affixoids that they can't trigger syllable boundaries. So "pafilo" is clearly [pa.'fi.lᴐ], and a "natural" hyphanation would be either "pa-fi-lo" or "paf-il-o". German "Ab-itur" is a real exception, there is also "Abi-tur", as German hyphenation usually follows the syllable strcuture. Commented Apr 28, 2019 at 11:26

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.