I saw this video and realised that all mentioned Old English plurals sound pretty natural for me, even though I'm native Czech speaker. Also in German I think inflection seems to follow some universal model. For instance, female form you create by adding -in. But for example word Hündin ( [hʏndin] - female dog) is an exception which I think relates to the meaning much more than [hundin] would.

What is this caused by? Is the more open vowel in plural/female form common also for Asian or tribal languages or is it ingrained just in the Indo-European ones? Or am I just influenced by the life in Western culture (or used to the correct forms)?

Thank you.

  • 2
    Sound "natural" does not have any meaning in linguistics. Sound "naturally" is not even English. To say nothing of "conjungtion".
    – fdb
    Oct 10, 2015 at 10:32
  • @fdb I meant by "natural" that I would guess the form correctly.
    – Probably
    Oct 10, 2015 at 10:51

3 Answers 3


As the other answers have pointed out, "natural" is not a category that is used in linguistics. However, there is a cross-linguistic tendency for plurals to be formed by the addition of a morpheme to the singular - i.e. plurals tend to be longer than singulars. Apart from that, as a Czech speaker you might find some plural forms in other Indo-European languages familiar because they have similarities that go back to the earliest form of Indo-European (proto Indo-European). Also, the umlaut/ablaut pattern to form plurals is shared by (some) Germanic languages, i.e. 'Mann' ~ 'Männer', 'man' ~ 'men'.

  • 1
    Good point that length(plural(x)) >= length(sing(x)), basically universally. However I would not give too much weight to common Germanic or Indo-European. Consider Italian vs very similar neighbouring South Slavic plurals, or pluralisation with -s which occurs in Spanish, French, English, Dutch (half the time), German (in words from Northern dialect). Whereas Indo-European languages Armenian, Persian tend towards regular plural suffix with no stem change, similar to Turkic. So pluralisation patterns can spread by contact and be categorised as much by geography as by language family. Oct 11, 2015 at 9:20
  • I agree, contact can in this question be more important than common ancestry.
    – robert
    Oct 11, 2015 at 14:55
  • @AdamM.B. It's not always true that length(plural(x)) >= length(sing(x)) (altought it is in like 99% cases). There are some exceptions. For example Polish "dziecko" or (old Polish) "dziecię" (child) and "dzieci" (children).
    – Arsen
    Oct 11, 2015 at 15:31
  • 2
    @Arsen Thanks for the counterexample. There are some others (eg celovek vs ljudi). But, as you say, on average the rule is still true for Polish - by "universal" I mean it is true (on average) in all languages. I would be very interested in a counter example to that. (I can't think of one.) Oct 11, 2015 at 18:38
  • @AdamM.B. Yes, I think You are right. I just wanted to point that it's not a rule that will be right in 100%. From the other hand if we'd count 2nd Polish (again) plural form which in many cases is shorter than singular (in Polish and many other Slavic languages 1st plural form connects with unspecified number of objects and 2, 3, 4, 22, 23 and so on, while 2nd one with all the other numbers). E.g."1 koza (goat), 2 kozy, 5 kóz". But the 2nd form is actually a genitive case (just like we can say milion of people, in Polish it starts at "5 of people"), so it's not strictly a plural form.
    – Arsen
    Oct 11, 2015 at 20:33

That's just some misconception.

The arbitrariness of sound-meaning correspondence is one of the most uncontroversial things in linguistics. There isn't anything 'natural' about the form in the language, these are purely formal aspects of sound and its mapping to meaning. Nor is there anything universal about the (phonological) form of inflectional morphology in the languages of the world.

Indeed, there will be intra-family similarities. But whatever reasons are for your intuition that the sound of some suffix (or ablaut) 'makes sense' for its meaning, it is but illusory.

That said, there is also phonosemantics, which is about how certain phonological shapes have a certain 'feel' to them, but that's a different story.

  • There's also markedness. I get the impression that inflectional suffixes tend to have less-marked consonants than other lexical items, but I might have a biased sample to base that impression on. Oct 11, 2015 at 5:00
  • @sumelic: Hm, interesting point. Would you care to give an example of what you mean? I never thought that markedness can really be defined on phonological inventories. Oct 11, 2015 at 7:11
  • Hmm. Well, for example, maybe rare consonants like "th"-sounds are less common in inflectional suffixes. There's a list of frequencies for Arabic here. Frequencies like this probably depend a lot on the language, though. It seems notable to me that in English, the main inflectional endings are coronal, like /z/, /d/ and /t/. Then again, maybe "rare" sounds are rare in inflectional suffixes because they're rare in general. Sorry, it wasn't a very coherent thought. Oct 11, 2015 at 8:42

“Hund” ~ “Hündin” is like “Mann” ~ “Männer”, or, in English, “man” ~ “men”. These all illustrate what is called “Umlaut”, the fronting of the stem vowel before a suffix with a front vowel. There is nothing “natural” about it. It is however a wide-spread phenomenon in the Germanic branch of Indo-European.

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