Umlaut marks are usually used to show that an inflection has occurred, that is the umlaut is a synchronic change. But some words with an umlauted character, such as Ärger, Hälfte, Käse, Lärm, März, Rätsel, Träne are not inflected forms of something else, so why do they have umlaut spellings?

Duden says that Bär was originally spelt with an e:

mittelhochdeutsch ber, althochdeutsch bero, eigentlich = der Braune, verhüllend Bezeichnung

but I don't think the "eigentlich = der Braune" explains the ä.

Ärger comes from ärgern

mittelhochdeutsch ergern, argern, althochdeutsch argerōn, gebildet zum Komparativ von arg und eigentlich = ärger, schlechter machen

Here the ä can be traced back to an inflected form (in this case a comparative) but it not clear when the spelling was changed from e.

So, even when umlauts are not indicating inflections in Modern German, it is clear that they do not all have the same explanation. So my question is:

Looking at umlauts that are not simply inflections, what are the various explanations, and when did these changes occur?

  • 4
    Umlaut marks are usually used to show that an indirection has occurred, ... Is this your observation or do you have a source for this rule (of thumbs)? If so, would you please provide a link?
    – Arsak
    Commented Apr 10, 2019 at 6:19
  • 11
    »Umlaut marks are usually used to show that an inflection has occurred« This simply is wrong. If it was true my last name would not contain an umlaut. Commented Apr 10, 2019 at 6:43
  • 2
    Because your premise is wrong. Umlauts are not evidence of plural forms any more than the letter "e" is, despite the many plural endings involving "e". Commented Apr 10, 2019 at 7:14
  • 1
    @HubertSchölnast: Schonlast hat aber auch einen gewissen Reiz bzw. Sinn. Commented Apr 10, 2019 at 9:22
  • @userunknown: Mein Name ist der Name eines Musikinstruments: Ein Ast, auf dem Schellen (kleine Kugelglocken) hängen: Schellen-Ast (verwandt mit dem Schellenbaum), Mein Urgroßvater hieß auch tatsächlich noch Schellnast Allerdings hat das im Südosten der Steiermark, an der Grenze zum Südburgenland, niemand so ausgesprochen. Und prompt hat der Standesbeamte, der den Namen meines Großvaters niederschrieb, den Namen so geschrieben wie er ausgesprochen wurde; Schölnast (langes Ö statt kurzes E). Commented Apr 11, 2019 at 6:36

5 Answers 5


Umlaut marks are usually used to show that an inflection has occurred, that is the umlaut is a synchronic change.

Umlauted vowels are less common than unumlauted vowels, but I think it's going too far to say that they are "usually" used to show inflection. The front rounded vowel sounds /yː/, /ʏ/, /øː/, /œ/ are regularly written with ü and ö in German, and they don't only occur in inflected forms of words with a base form in u or o. For example, the prepositions über and für are not inflected forms, and the verb stören doesn't have o in any of its inflected forms. Historically, these sounds derive in most cases from the phonetic process called umlaut, which seems to have been active at some point before the "Middle High German" period. You can find more information about the evolution of German vowels over time on the following web page: German vowels over time (by Helmut Richter). Indo-European ablaut (the source of most vowel alternations in the conjugation of "strong" verbs) is a separate and much older type of vowel alternation.

In fact, all of your examples are with ä. The situation with ä is more complicated because, unlike ü and ö, it doesn't always represent a distinct sound of its own in the spelling of German words. It represents the same sound as the letter e in at least some cases. (The details vary depending on the word and the speaker.) So it is true in general to say that ä specifically is only used in modern spelling in words that are related in some way to words spelled with a. But the relationship does not have to be a matter of inflection: ä is also used in the spelling of some derived words. More details can be found in the answers to this Linguistics SE question: What was the original pronunciation of 'ä' in German?

  • In some cases, ä is even used to distinguish two words that would otherwise be homographs at least in some inflected forms, e. g. Lärche/Lerche and, arguably, Bären/Beeren.
    – Crissov
    Commented Apr 12, 2019 at 10:20
  • In diachrony über may well be an inflection; Ahd. ubar (8. Jh.), ubari Adv. (8. Jh.), mhd. über, (md.) uber, ober, asächs. oƀar, oƀer, ofer, mnd. mnl. ōver, nl. over, aengl. ofer, engl. over, […] eine Bildung mit komparativischem r-Suffix zu ie. *upo, *up, *eup ‘unten an etw. heran’, dann ‘von unten hinauf, über’ (s. auf, ob).*" ... It is comparative, stands next to oben, ober, auf. There's also a PIE *Hep-, (Gk. *apo, etc). Der Rest des Arguments erübrigt sich.
    – vectory
    Commented Jun 7, 2019 at 5:05

In most of your examples the vowel ä is an actual Umlaut, i.e. a changed vowel a or ā:

  1. Ärger relates to arg, https://www.dwds.de/wb/Ärger
  2. Hälfte relates to halb, https://www.dwds.de/wb/Hälfte
  3. Träne stems from Middle High German trān, Germanic *trahnu-, https://www.dwds.de/wb/Träne
  4. Käse stems from Latin cāseus, https://www.dwds.de/wb/Käse
  5. Lärm stems from Alarm, Italian all’arme ("to the weapons"), https://www.dwds.de/wb/Lärm

So your hypothesis that the vowel ä indicates inflection, is just a part of the truth. It seems to indicate that the word stem has been a in earlier stages of the word.

The exception Bär remains though. A hypothesis: Maybe the Umlaut indicates that the vowel has been long in Middle High German and is still long?

At least, other words with the morpheme ber- seem to have the short vowel nowadays:

  • Berkelium
  • Berlin
  • Bermuda
  • Berlocke
  • Berme
  • Bern
  • Berserker
  • bersten
  • Bertha
  • 2
    Eigennamen (egal ob Orte oder Personen) gehorchen aber nicht zwingend den allgemeinen Regeln, daher sind die Beispiele Berlin, Bermuda, Bern und Bertha mit Vorsicht zu genießen - und auch das Berkelium, das nach der kalifornischen Stadt Berkeley benannt wurde. Commented Apr 11, 2019 at 11:27
  • @VolkerLandgraf Ja, stimmt, danke für den Hinweis!
    – Jonathan Herrera
    Commented Apr 11, 2019 at 13:32
  • So oder so, ber- wird mit kurzem e gelesen. Deshalb Beere. Zudem wird ä auch ae umschrieben, das einen diphtong nahelegt. Die alt-englische Runenschrift hatte sogar ein eigenes Zeichen dafür, vgl. etwa Äsir, En Aesir. Zwar hat Englisch bear ...
    – vectory
    Commented Jun 7, 2019 at 5:24

First of all, let's distinguish between the linguistic phenomenon (Germanic) umlaut (or i-mutation) and the grapheme umlaut diacritic. In German, these two concepts are very closely related, because letters with umlaut diacritics (ä, ö, ü) are almost always used to spell vowels resulting from i-mutations and rarely used for other words. All modern Germanic languages have been affected by i-mutations and many languages (Germanic and non-Germanic) use vowels with umlaut diacritics, but only German has this close correspondence.

OP seems to have noticed a pattern of umlaut diacritics appearing in inflected words (presumably plural nouns for instance) and assumed there is a rule for this. The reason we see this pattern is that the i-mutation occurs in words with an /i/, /iː/, or /j/, and only certain inflections of a word had this sound when the i-mutation took place. From the Wikipedia article on Germanic umlaut:

Although umlaut was not a grammatical process, umlauted vowels often serve to distinguish grammatical forms (and thus show similarities to ablaut when viewed synchronically), as can be seen in the English word man. In ancient Germanic, it and some other words had the plural suffix -iz, with the same vowel as the singular. As it contained an i, this suffix caused fronting of the vowel, and when the suffix later disappeared, the mutated vowel remained as the only plural marker: men.

In this quote they use the English forms man/men, but it of course also applies to the German forms Mann/Männer.

The Wikipedia article also gives a few examples of loan words spelled with umlaut diacritic in German even though they have not been affected by i-mutaion (examples bolded by me):

Conversely, some foreign words have umlaut diacritics that do not mark a vowel produced by the sound change of umlaut. Notable examples are Känguru from English kangaroo, and Büro from French bureau. Here the diacritic is a purely phonological marker, indicating that the English and French sounds (or at least, the approximation of them used in German) are identical to the native German umlauted sounds. Similarly, Big Mac was originally spelt Big Mäc in German. In borrowings from Latin and Greek, Latin ae, oe, or Greek ai, oi, are rendered in German as ä and ö respectively (Ägypten, "Egypt", or Ökonomie, "economy"). However, Latin/Greek y is written y in German instead of ü (Psychologie).


I seriously doubt your premise of Umlaut marks. These two dots have been strokes before, which is simply Kurrent handwriting of the letter e. Older printed texts have an actual tiny e printed on top of a, u, o. And this is common practice even today: if no Umlaute are available on your computer, you should write ae, oe, ue instead.

These are just five additional vowels German has (short ä is identical to short e). Or diphthongs, whatever you like.

The sound changes in inflection and conjugation aren't special to the Umlaute, it's a general pattern and called Ablaut. This was already present in Proto-Indo-European.

You can see this in German where no Umlaute are involved:

rufen, er ruft, er rief

saufen, er säuft, er soff

bieten, er bietet, er bot

  • 1
    This seems to be a comment. As in English there are several Ablaute, as shown in sprechen, spricht, sprach, gesprochen. But one of these Ablaute is the set of sound changes that is typically associated with plurals, comparatives and a few other things. So whereas most of your examples are spelt phonetically, säuft is spelt using "Stammschreibung", that is, the au is retained with an extra mark (originally e, then ◌ͤ, then ¨). My point is that Bär does appear to come from *Bar or anything like it and Ärger relates to no modern word with a. This is what I am asking about. Commented Apr 10, 2019 at 0:25
  • The adjective arg is used in modern German, but not in all dialects and codes.
    – Janka
    Commented Apr 10, 2019 at 0:51
  • Bär is special because the ä-r is pronounced as Schwa-Tiefschwa.
    – Janka
    Commented Apr 10, 2019 at 0:54
  • @Janka I doubt that there is any German noun that has a Schwa as its only vowel.
    – sgf
    Commented Apr 10, 2019 at 14:00

To keep it simple, I think we should distinguish between

  • [ä], [ö], [ü] as altered (umgelautete) forms of [a], [o], [u] where this has some systemic function - such as, typically, indicating plural (but usually together with some suffix: Mann --> Männer, Maus --> Mäuse.
  • [ä], [ö], [ü] as separate, full-grown vowels in the phonological system of vowels of German language, testable via minimal pairs (Mus vs. Mass - mind that we are speaking here of phonems, not graphems, so the double-s is irrelevant; Bar vs. Bär).

Under this perspective, your questions appears being based on a wrong assumption. [ä], [ö], [ü] (i.e. the phonems expressed through these letters) are simple there, they are not necessarily derived from umgelautete other vowels. Or in other words: there are words that use those phonems (which then gets represented by ä, ö, ü in writing). You have to accept this as a given. Similarly, you probably wouldn't ask: "Why are there words in German that have a vowel [i]?" Or if you did, the answer would be: [i] is one of the vowel phonems of German. If i wasn't used, it wouldn't be counted part of the phonem system.

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.