Old English has several glyphs that are obsolete today: For example þ, replaced by "th".

Old Japanese had several kana not used today, for example ゑ.

However, how about German? Did German used to have anything other than a-z, Umlaute, and ß?


5 Answers 5


In all the following cases, it’s up to you what you consider a separate letter:

  1. The long s (ſ) was a variant of the letter s with orthographic rules where to use it. In an extreme (contrived) example, its use could affect the meaning of the word:

    • Wachſtube = Wach-stube = guardroom
    • Wachstube = Wachs-tube = wax tube
  2. The r rotunda (ꝛ) was a variant of the letter used after certain letters. The rules for typesetting affecting it were mostly typographical (based on position) and not orthographical (based on content). It survived longest in the abbrevation ꝛc. (etc.).

  3. All the umlauts were preceded by variants where a small e was placed above the vowel, e.g., .

  4. In very early prints, you had further similar combinations, in particular (u with a small o above; similar to ů).

  5. Distinguishing between i and j is a rather new thing. Thus, from some point of view, a letter comprising i and j is gone. Similarly, u, v, and w all originate from the same letter.

  6. In older handwriting it was common to use a macron accent over a single m or n to indicate doubling, e.g. writing Don̄erstim̄e instead of Donnerstimme.

  7. Before moveable type, there was a plethora of shorthands and abbreviatures merging several letters or even entire words into one symbol.

Here is an example for Cases 2–5:

Example of ꝛ, uͦ, and uͤ in use.

  • What a magnificent example!
    – HalvarF
    Sep 6, 2022 at 14:02
  • By “rotunda” sehe ich sechs Balken. Ist das Absicht oder ein verunglückter Font?
    – gnasher729
    Sep 12, 2022 at 21:22
  • @gnasher729: Hört sich nach letzterem an. Im Zweifelsfall geben das Bild unten und der Wikipedia-Artikel Beispiele, wie es aussehen soll.
    – Wrzlprmft
    Sep 12, 2022 at 21:56
  • The "long s", the Umlauts and the overbar for doubling "m" are specific to German. The "ij" and "uvw" ambiguity is relatively common in all European languages, originally from Latin.
    – tofro
    Dec 26, 2023 at 9:53
  • 1
    @tofro: The "long s", the Umlauts […] are specific to German. – Not really. The long s was used by many other languages and I have seen in old Norwegian prints (IIRC).
    – Wrzlprmft
    Dec 26, 2023 at 10:18

The only one I can think of: in older German scripts like Fraktur or Sütterlin there was an alternative form of the letter s, the "long s", see for example https://de.wiktionary.org/wiki/%C5%BF (in German.)

  • 2
    The long s existed in English as well, starting to disappear around 1800. It still exists in German as part of 'ß'. Both English and German were written with runes at one time, and thorn was carried over from that into Old English as well as several other languages. The long s is a very different case, being adopted from Roman cursive,
    – RDBury
    Sep 4, 2022 at 19:38

In German Fraktur typesetting, there were five ligatures that behaved as single glyphs: ch, ck, ſt, ſz, and tz. When letterspacing was increased, these ligatures stayed glued together as single glyphs. Increased letterspacing was relatively common. It was the usual method for highlighting words, much as italics or boldface is in Roman type. A word like „ſchwitzen“, for instance, would look like „ ſ ch w i tz e n “ when letterspacing was increased.

The Fraktur ligature glyph ſz was carried over to Roman type in the 1876/1901 German spelling reform. This is how the modern Roman type letter Eszett ß originated – it was not used in Roman type before the 1876/1901 spelling reform. The other four Fraktur ligature glyphs ch, ck, ſt, and tz were not carried over to Roman type. Therefore, they can be considered obsolete old German letters.

  • I'm afraid you are mixing up the terms letter and glyph just like the OP. The ligature glyph ch is still the letters "c" and "k".
    – tofro
    Sep 6, 2022 at 6:52
  • @tofro: Thanks for pointing that out. I am well aware of the distinction between characters and glyphs, but I have avoided the issue. In this particular case, the distinction is problematic, since the ligature glyph ſz became the character ß when it was carried from Fraktur over to Roman type. To my knowledge, the term letter is not as well defined. I wonder whether a ligature that behaves like a single glyph with respect to glyph processing tasks like increased letterspacing might be called a letter.
    – mach
    Sep 6, 2022 at 11:13

There is another which became Ö: Œ œ -> Ö ö


  • 2
    That's not quite right. The article states that "ö" stems from an "o" with an "e" on top, which clearly isn't "œ".
    – tofro
    Sep 5, 2022 at 13:40
  • Danke für deinen Kommentar. Ich habe es mir erlaubt, eine Generation zu überspringen. hahaha Das O mit Umlaut wurde erst als OE geschrieben. Später entwickelte sich daraus ein O mit einem kleingeschriebenen E darüber (Oͤ/oͤ). In der Kurrentschreibschrift wurde das e in zwei kurzen, senkrechten Strichen dargestellt, welche mit der Zeit zu zwei Punkten stilisiert wurden. de.wikipedia.org/wiki/%C3%96
    – Hans
    Sep 5, 2022 at 19:05

As @TilmanSchmidt and @mach stated the "long s" comes from Fraktur. It was used to set apart an "s" in the middle of a syllable from one at the end. This way one could distinguish i.e. between "kreischen" ("krei-schen", to scream) and "Kreischen" ("Kreis-chen", circlet).

Where these two came together within the same syllable they where replaced by a "ß", basically a ligature consisting of a "long s" (looks like a "f" without the horizontal line) and a "letter s".

This is the rationale behind the original set of rules covering the use of "ß" (the "Adelungsche Regel"). One writes "ss" in a "Silbengelenk" (also known as "ambisyllabischer Konsonant") , but "ß" otherwise: "Wasser" but "wäßrig", "müssen" but "muß".

(Notice, though, that "Silbengelenk" is a term coined by Peter Eisenberg, a modern german linguist, Adelung himself gave different rules but they lead effectively to the same result.)

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.