I know that the umlaut vowels were also written as ae oe and ue, and this orthography shows the process of assimilation with a high vowel. But were these vowels ever actually pronounced as a diphthong, before they became fronted? For example schön being pronounced as schoen, with both vowels being pronounced separately as a diphthong? I think there must've been some sort of intermediate stage of pronunciation between Old High German sconi and Modern German schön.

  • By the way, in many dialects (and Yiddish) there is Entrundung: ö is pronounced as e and ü as i. It may be some clue to the intermediate stage. Commented Mar 1, 2019 at 6:20

2 Answers 2


Umlaut itself—as in the process, not the dots—was a sort of vowel harmony that was productive for a long time in Germanic. The thing you're asking about specifically is called i-umlaut; there was also a-umlaut and u-umlaut.

The way it's generally understood (in Proto-Germanic), when there was a back vowel in one syllable, and an */i/ or */j/ in the next syllable, the back vowel shifted forward in the mouth. Phonetically, this is quite reasonable: these sorts of harmony processes (where vowels shift forward or back to match the syllables around them) are common, and you also see them in e.g. modern Turkish and Finnish.

The back vowels in Proto-Germanic were */a o u/, which would have had front allophones *[æ ø y]. In the earliest written records, this allophony isn't written: it was fully predictable from the context, and *[æ ø y] didn't show up anywhere else, so there was no ambiguity. The front allophones were used if */i j/ were in the next syllable, the back allophones otherwise. Nice and straightforward.

But then, sound changes happened, and certain suffixes started disappearing. For example, the plural of *mann "man" used to be *manniz, with a nice predictable [æ]. But when that suffix disappeared, the early Germanics were left with *mann and *mænn. The umlaut had become phonemic.

So now, how to write this? The Latin alphabet didn't have letters for these sounds! So different languages improvised in different ways. Old English used æ, oe, y for /æ ø y/. The Germans used ae, oe, ue, which developed into ä, ö, ü through abbreviation. In Norway, they used æ, ø, y. But by comparing the developments in different languages, it seems clear that these all represented the same phonemes, and that they were monophthongs, not diphthongs.

(P.S. English ended up losing all these fronted vowels, merging them into other phonemes. You can see some of their descendants in man~men, foot~feet, mouse~mice. There was also a vowel length distinction that I'm ignoring here for simplicity: for full details, Wikipedia has a nice chart.)

(P.P.S. Old English was originally written in the Futhorc alphabet, which did have specific runes for /æ ø y/. But then the Latin alphabet took over, so they had to make do with ligatures and digraphs.)

  • Thanks for the detailed answer, but I already understood the concept of Umlaut. My question was about the original pronunciation of those vowels that shifted. From what you've said there was no intermediate stage between the pronunciation of the vowels,and they were already pronounced as their fronted allophones. But now that I think about it,in the dutch word schoon this phenomenon didn't happen. This means that originally the vowels weren't pronounced as their fronted allophones. .I think that the vowels shifted their quality gradually.
    – X30Marco
    Commented Feb 28, 2019 at 16:38
  • 1
    It think it is because the umlaut you see in schön seems to only have happened after and only in the old high German branch and in the old Saxon branch. The proto-Germanic word from where schoon and schön came is *skauniz, I am not 100% sure, but I think that i-umlaut didn't affect dipthongs (based on what I saw from *skauniz descendants). That's why the old English equivalent is sċīene, that later evolved to "modern" sheen. I'll do a better research on this and, if someone doesn't answer it, I will bring here what I found! If I am wrong, please some one correct me Commented Feb 28, 2019 at 17:02
  • 2
    @X30Marco I think this answer does cover the "original" pronunciation of these vowels: specifically, it says they were fronted allophones of /a/, /o/ and /u/ occurring due to a vowel harmony process; so there was no intermediate pronunciation as diphthongs, which the answer mentions. As to Dutch schoon, Wikipedia says western Dutch dialects were largely unaffected by i-umlaut, while it looks like the word was pronounced with umlaut in eastern dialects.
    – LjL
    Commented Feb 28, 2019 at 19:12
  • @LjL Eastern dialects in the Netherlands do indeed have Umlaut here, somehwat more in the "Franconian" south than the "Saxon" North. Old Frisian had derounded e: (from an ø like sound, probably) namely "skēn", which became modern West Frisian "skjin" (via "skien" (with falling diphthong) and modern WF breaking). Commented Mar 3, 2019 at 23:58
  • @X30Marco Dutch is a bad example because it had a lot of umlauts but got rid of umlaut-connected irregularities in paradigmata by analogy with the exception of stad-steden. Commented Nov 26, 2019 at 13:50

The answer to your question is no. The German umlauted vowels were never diphthongs. In early New High German they were written as a, o, and u with a small superscript e. Later, this “e” was reduced to two dots. The spellings with ae, oe, and ue are merely typographical attempts to deal with the miniature “e”.

  • I think in Old High German they were diphthongs, still preserved in some Swiss German dialects. Commented Mar 4, 2019 at 0:00
  • @HennoBrandsma: there is no evidence that this was the case in OHG.
    – fdb
    Commented Mar 4, 2019 at 11:28
  • You don't think the [uo] in "bluot" was a diphthong (and its umlauted version too)? Commented Mar 4, 2019 at 18:07
  • @HennoBrandsma Your statement was wrong and your followup question misses the point spectacularly. Of course /uo/ was a diphthong but that doesn't have anything to do with diphthongs. And of course it's umlauted version was a diphthong. But that is completely beside the point because it is an umlaut of a diphthong like äu < öü < ou which is also preserved in Standard German. Commented Nov 26, 2019 at 13:43
  • I have my doubts about your statement that "ae, oe, and ue are merely typographical attempts to deal with" aͤ oͤ uͤ as to my knowledge dipgraphs like ae predate them. At first the small superscript e came up as a variant like uo was spelled as uͦ (or ů) as a variant. Commented Nov 26, 2019 at 13:47

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.