I'm an elementary teacher and my students are learning syllable division. We noticed that before the suffix -tion, the "i" is always a short sound, despite being an open syllable. All the other vowels are mostly pronounced as long sounds, as expected in open syllables. Why is this?

Example -ition words:

  • position, competition, tuition, edition

Other -tion words:

  • (ation) station, nation, fixation, isolation (exception: ration)
  • (otion) notion, emotion, lotion
  • (ution) distribution, evolution
  • (etion) secretion, completion, depletion (exception: discretion)
  • 3
    I doubt your students are learning syllable division in the linguistic sense off the term. I think they're learning breaking rules for words that straddle printed lines. Never the twain shall meet. Syllabification would be pretty hard for elementary students and of vitally no benefit to any of them! Dec 13, 2022 at 18:40
  • @Araucaria-Nothereanymore. I really don't agree with you as concerns phonetic syllabification; good practices should be introduced early and learning how to interpret syllabification as used in a good dictionary, which is not such a hard task (if the students go about it progressively), will help to preserve an accent. It is clear that different systems of syllabification make for different accents. The syllabification for RP can't be that for General American.
    – LPH
    Dec 13, 2022 at 20:58
  • @LPH You can only learn the syllabification from the phonetic facts--if, that is, you agree on the theory in the first place. It is not predictable from the writing. It is not agreed upon by the dictionaries. It is not agreed upon by the linguists. It is as often as not impossible to get any idea of where a syllable boundary lies from the speech signal itself--and that is not to do with a paucity of information about the speech signal. Dec 13, 2022 at 22:41
  • @Araucaria-Nothereanymore. This might be true for another syllabification than RP; I know only RP, but in RP, once you know the stress pattern for a word a few principles allow you to determine the syllabification in a great many cases. An agreement by linguists is a matter of accent. Try and make the difference here: Bre /'stren ju əs/ AmE /'stre nju əs/ (syllabification according to my ear). Not taken into account in OALD, (1/2)
    – LPH
    Dec 13, 2022 at 23:09
  • 1
    @LPH There is no syllabification "according to my ear" in such cases. What phonetic facts are your picking up on in strenuous, for example? In a panel of RP experts consisting of Ashby, Windsor Lewis, Collins, Setter, their unanimous verdict-vociferously seconded by all other famous phoneticians who spoke in the audience-was that LPD (i.e. Wells) was by far and away the gold standard for phonetic analysis for English and in particular RP. Bar none. Dec 13, 2022 at 23:39

2 Answers 2


This is a general rule about the pronunciation of the letter "I" when it is followed by a single consonant letter and then two vowel letters, the first of which is "e" or "i". "I" is pronounced "short" in this context, while other vowel letters are pronounced "long". You see the same contrast between i and other vowel letters in words with other endings of this form:

  • vicious vs. tenacious
  • artificial vs. facial
  • efficient vs. patient
  • initiate vs. ingratiate

It applies to the letter I both when it occurs in the root of a word (as in vicious, which is related to vice) and when it occurs in a suffix.

This spelling rule is related to past sound changes in the English language, but the explanation for why i developed differently from other vowels is probably fairly technical and therefore I don't know whether there is a satisfying explanation for students. A previous answer I wrote with some examples and exceptions is Why do we pronounce a long second vowel in "decide", but a short second vowel in "decision"?

A possible technical explanation

The only attempt at an explanation that I have seen is in the dissertation "The Architecture of the English Lexicon", by Jonathan B. Alcántara (1998), which in §5.2.2 "Compensatory lengthening within the stem" gives a detailed theoretical hypothesis for why ultimately

the only instance in which CiV lengthening can take place is when a stressed open syllable with a low or mid vowel is followed by the weak member of the foot, a syllable with the nucleus /i/.

(page 5-182)

Here is my own summary of what I understood of the explanation:

  • Let's suppose that sequences like -ious, -eous, -ial start at an "underlying level" with the vowel sound /i/

  • However, we see that at the "surface level", this vowel is often not pronounced as its own syllable: it may be replaced with a phonetic glide, as in a disyllabic pronunciation of "genius" [ˈdʒinjəs], or it may fuse with the preceding consonant by palatalizing it, as in palatial /pəˈleʃəl/. One interpretation is that the reduced, non-syllabic pronunciations—as /j/ or palatalization—result from unstressed /i/ in this position losing a "mora" (a "mora" in linguistics is a unit of length). (A wrinkle is that a syllabic /i/ sometimes can appear in this context in present-day English, as in memorial—Alcántara proposes that even if it can be pronounced in a way that sounds syllabic, the /ri/ in this word might function as "the onset of the third surface syllable", giving the division "s(mem)-s(momm)-s(riaml)" (page 5-177).)

  • There is a concept in linguistics known as "compensatory lengthening": the idea is that when one sound in a word is shortened or lost (e.g. by losing a mora), another nearby sound may be lengthened as part of the same process. This can be thought of as a process of a mora "shifting" from one part of the word to another. For example, the spelling phenomenon of "silent e" is the result of a process of compensatory lengthening that occurred in Middle English: in stressed syllables, the vowels a, e, o were lengthened when followed by a single consonant if an unstressed /ə/ sound, spelled e, was lost from the following word-final syllable. We can explain this as being caused by a process that shifted a mora from the original final /ə/ to the vowel in the preceding syllable.

  • Compensatory lengthening may affect the vowels a, e, o differently from other vowels. In most languages, including in Middle English, the letters a, e, o represent what is called "non-high" vowels: these are vowels that are pronounced with the mouth relatively open. These vowels can be viewed as having greater "sonority" than the "high vowels" /i/ and /u/. Roughly speaking, languages are supposed to prefer having more sonorous sounds in the middle of syllables, and less sonorous sounds at the edge of syllables.

  • Combining all of the above ideas, it may be the case that a, e, o were lengthened in English before sequences like -ious, -eous, -ial as a form of compensatory lengthening when the following vowel /i/ lost a mora (resulting in some cases in it being phoneticlly reduced to the non-syllabic glide /j/). But because i is less sonorous than a, e, o, the vowel i could not be lengthened when it occurred in the same position. My assumption is that this lengthening process first occurred in Middle English, as with lengthening in words ending with -e; Alcántara implies that these rules remain part of the sound system of present-day English, but the extent to which this is true this seems arguable.

I didn't mention U because the vowel letter u is a special case: since Middle English, the letter u in the English spelling system has been used in open syllables to represent a diphthong /ju/ (an adaptation of the French u sound, /y/) rather than a simple vowel. So from a historical perspective, there may be separate explanations for the long a/e/o in -ation/-etion/-otion and the long u in -ution.

Another example of how the letter U is special is that it is regularly pronounced long before a single consonant letter + vowel letter even when followed by two or more syllables; other vowel letters are regularly short in this context (compare short a, e, i, o in sanity, extremity, stability, velocity but long u in community). The short vowel in these words is explained in some theories by a rule of vowel-shortening called "trisyllabic laxing"; Alcántara however disagrees with explaining these alternations in terms of a process of shortening, arguing that "so-called 'vowel shortening' is actually indicative of cases where the stem vowel fails to lengthen due to a different stem structure" (page 1-29).

If we consider "trisyllabic laxing" to exist and to apply to words like ignition, it would seem to lead to the awkward conclusion that words like fixation, emotion, completion are also in the right environment to undergo trisyllabic laxing (or shortening). Does that mean words like this have an underlyingly long vowel that is shortened and then lengthened again, as the result of the trisyllabic shortening rule being reversed by a countervailing a/e/o lengthening rule? That scenario seems a bit convoluted, so I'm reluctant to apply the concept of "trisyllabic laxing" to explain the pronunciation of words ending in -ition.

But all of this seems much too complicated to use in instruction aimed at just teaching students about English spelling.

  • Thank you for your explanation and I'm sorry for taking so long to reply! I posted about this last year and completely forgot about it!! I just read your link that included a list of Latin-origin words with exceptions!! Super helpful thank you so much!!!
    – alpackie
    Oct 21, 2023 at 15:47

This a particularity of the suffix "-ition" and not "-tion". First the i receives primary stress and second, it is pronounced short. This can be verified in Longman Pronunciation dictionary.

-ition ˈɪʃ ən opposition ˌɒp ə ˈzɪʃ ən

Note The syllable in which ɪ is found is "zɪʃ ", a closed phonetic syllable, and it must not be confused with an orthographic syllable. Furthermore, "Note that [open and closed syllables] have nothing to do with open and close vowels, but are defined according to the phoneme that ends the syllable: a vowel (open syllable) or a consonant (closed syllable)" (ref.); thus the syllable in "though" is open. One must be careful also about the particular phonetic syllabification that is being used. There are at least two.

  • 2
    You've missed trisyllabic laxing: compete/competition, compose/composition, define/definition, contrite/contrition, depose/deposition, erudite/erudition, extradite/extradition, ignite/ignition, oppose/opposition, propose/proposition, repeat/repetition, require/requisition, unite/unition.
    – tchrist
    Dec 12, 2022 at 19:03
  • 2
    @tchrist This is not the explanation of the pronunciation of the suffix; for instance in "repeat/repetition" the laxing concerns "ea/e", not "ea/i"; moreover, this is rather a topic in specialist knowledge about phonology and it is excessive to impose that much erudition on mere users of the language, in my opinion. (Wikipedia)
    – LPH
    Dec 12, 2022 at 20:40
  • It is interesting that not a single example in the trisyllabic laxing Wikipedia article involves an -ition word. They note divide→division, derive→derivative, and pronounce→pronunciation, but that's as close as they get to -ition. (And in fact, the use of depose→deposit as an example rather than depose→deposition feels especially significant.)
    – FeRD
    Dec 13, 2022 at 4:34
  • @FeRD It is even worse than that: stress is being displaced every time, and the actual phenomenon is a reduction of an unstressed middle vowel, not the changing of a stressed long vowel into a stressed short vowel.
    – LPH
    Dec 13, 2022 at 12:42
  • 1
    There is no ition suffix in English; There is: -tion AND -sion
    – Lambie
    Dec 13, 2022 at 20:07

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.