I know, that English "t" and German "s" may be a cognate

it -> es
out -> aus
what -> was
that -> das?

Why do "house" and "mouse" have "s" on the end?

  • 3
    Actually, they have "e" on the end. – Greg Lee Aug 7 at 21:48
  • 2
    @GregLee True, but I assume he meant "/s/ on the end". – Draconis Aug 7 at 23:06

Because they go back to Protogermanic words already containing *s, *hūsą in the case of house and *mūs in the case of mouse.

Note that sound shifts work in a forward direction, and mergers can occur as in the case of High German aus, Maus, and Haus—they are now rhyming but they weren't at an earlier stage of of the language. This means that you cannot undo sound shifts without further knowledge (e.g., attested older stages of words or cognates in other languages), and that you cannot transfer words, say from German to English, by just applying sound shifts.


Jknappen's answer is entirely correct, but just to add some more examples:

  • The Proto-Germanic word for "out" was *ūt
  • The Proto-Germanic word for "what" was *hwat
  • The Proto-Germanic word for "house" was *hūsą
  • The Proto-Germanic word for "mouse" was *mūs

(Note: all forms marked with stars are reconstructions that aren't directly attested anywhere.)

You can see that, in German, Proto-Germanic *-t turned into -s, and Proto-Germanic *-s remained as -s.

In English, on the other hand, Proto-Germanic *-t stayed as -t, and Proto-Germanic *-s stayed as -s.

This lets us divide these words into two groups: words which end in -s in both languages probably had an -s in Proto-Germanic, while words which end in -t in English but -s in German probably had a -t in Proto-Germanic.

(This is, in fact, the basis of the comparative method: looking at German alone wouldn't let us distinguish these two classes; we need the data from English too, and only by comparing them can we reconstruct the proto-form. A more thorough analysis would look at a whole lot more words, and a whole lot more languages, in order to come up with detailed reconstructions like the ones I listed above.)

  • In fact both mouse and house reflect PIE *s. For house it's been suggested that it's an s-extension of *(s)kew-. For mouse I'm not aware of any hypothesis, a full-grade *múh₂s is reconstructed. It is not straight forward to compare *mey- viz "mini-" (e.g. mini) and it can be assumed that some early folk etymology is involved (rat is even less certain; also cp e.g. mite). The nature of the s-extensions is, in my humble opinion, what is interesting about this Question – vectory Aug 9 at 8:52
  • 1
    In modern West Frisian we still say “mûs” and “hûs” (long [u] sounds). Such forms also occur in Limburgian varieties in the Netherlands and some Low Saxon ones too. Islands of conservatism in this respect. In most Germanic varieties from the Netherlands and Belgium diphthongisation and later secondary monophthongisation kicked in, leaving many ways to pronounce these words. Interestingly, the isoglosses for these two words do not always run together, mouse tends sometimes to preserve older pronunciations. Kloeke, in his book on the sound change ascribed it to “taboo”, mouse is less frequent. – Henno Brandsma Aug 9 at 22:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.