1

I recently saw this clip where Karl Pilkington visits a Vanuatu tribe, in which it is said that every word of the Ninde language begins with the letter 'n'. I soon called BS, and as the wiki-page I linked to confirms, it was. My reason for doing so is that I'd think that if there was a language where every word began with an 'n', the speakers wouldn't be able to infer any information from it, and it would surely be dropped.

My question is whether or not this gut feeling is actually true: Has there ever been observed a (natural) language where some rule that in principle carried no information was still used?

  • 2
    It would contain information about where words start and stop. People don't speak with spaces. I know what you're trying to say, but it's hard for me to think of a rule that would convey absolutely zero information. A lot of agreement is redundant, but usually there are also some cases where it can disambiguate a sentence. – ewawe Feb 17 '16 at 21:10
  • People using other languages surely do not need the signal for new word is now beginning, even though they don't speak without spaces either? Or is there some subtle way that this information is conveyed (without the use of an omnipresent 'n')? – Bobson Dugnutt Feb 17 '16 at 21:17
  • I'm not saying it's necessary information, but it would convey some information in principle. Ways natural languages actually convey word boundaries are varied and generally not absolute: linguistics.stackexchange.com/a/9907/5581 – ewawe Feb 17 '16 at 21:20
  • Closing as too broad sorry: what exactly counts as an example of this isn't defined clearly enough. – curiousdannii Feb 18 '16 at 7:31
  • 2
    According to your own link (the wikipedia entry on the Ninde language) this claim is completely wrong. The n- prefix seems to be a fossilised article before inanimate nouns only. – fdb Feb 18 '16 at 11:57
4

I think the answer has to be no, not from the perspective of observed facts of language, but from problems with the idea of "carrying information". Here's an example from (American) English. There is a rule of pronunciation that /t/ (for /k/ for some people) is pre-glottalized at the end of the syllable. So you could say that this rule "carries no information". Except: it informs you, when you encounter a glottalized t, that it is at the end of the syllable. It's not the same kind of information that adding the suffix -ed carries, in English verbs, but it is information. The concept of "rule" implies "information", since rules apply in some context, and the fact of applying a rule informs you that the item is in that context and not some other context. If there is absolutely no context, then you don't have a rule.

  • Perhaps "rule" was too broad, as I meant that it was applied in every context (as with the case of every word beginning with an 'n'). In this light, would the answer still be no? – Bobson Dugnutt Feb 17 '16 at 22:02
  • 2
    But the rule inserting n doesn't apply in every context, it just applies just at the beginning of the word -- not the middle or end. So, still no, since nothing "happens everywhere", it happens in a specific context, and thus the fact of the rule applying can be seen as information that you are in that context. – user6726 Feb 18 '16 at 0:13
2

What about the word final stress in modern French? It carries no information except for marking word boundaries.

  • 1
    Actually, French has a final stress only on the last syllable of a phrase. But this all depends on how you define "word". – fdb Feb 18 '16 at 14:09
  • Stress in a lot of languages helps to mark word boundaries. But many languages also have phenomena like words without stress, cliticization, and phrase stress rules that make it complicated to perfectly divide all words in a sentence. – ewawe Feb 18 '16 at 18:18

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.