3

Some languages combine pro-drop with null or zero morphemes – inflectional morphemes, more particularly. Turkish is an example of this. To illustrate, the verb istemek = to want is inflected as follows (in present continuous):

1Sg: istiyor-um
2Sg: istiyor-sun
3Sg: istiyor__
1Pl: istiyor-uz
2Pl: istiyor-sunuz
3Pl: istiyor-lar

Note that in the 3rd person singular, there is no morpheme – or at least not one that’s phonetically realised. So, in that sense, we have a null morpheme. (-yor- is the morpheme for continuous, but there’s no morpheme for person.)
Next, Turkish is a pro-drop language. To say e.g. I want to eat an apple, one does not have to use the first-person pronoun ben, as in Ben bir elma yemek istiyorum, but can instead say: Bir elma yemek istiyorum. Similarly, in the third person one need not use the pronoun o. To say He wants to eat an apple, one need not say O bir elma yemek istiyor, but can say:

(T) Bir elma yemek istiyor.

Now, (T) is the kind of sentence I am interested in, as it contains neither a personal pronoun nor an inflectional suffix. (More carefully, neither is realised at the surface.)

My question simply is whether anyone can supply another language that exhibits this – ideally a language where this happens in the first person singular. It’s fine if the language is merely null-subject and not fully pro-drop (i.e. allowing pronoun dropping only for subjects). Of course, I don’t want a language like Japanese or Korean, which lacks inflection in the first place.

1

2 Answers 2

2

Technically, English (and Turkish) imperatives are pro-drop and feature the null morpheme.

But what is arguably really going on here is that the Turkish present-progressive endings are like a compound with copula, and Turkish has null copula for 3rd-person singular. (The actual Turkish 3rd-person present tense is marked, eg etmek / eder.)

(This goes back to the etymology of -iyor. If I recall correctly, it is a relatively recent innovation, and had some literal meaning, which is why it works like a noun or adjective, not a suffix - no vowel harmony. And of course English present progressive also uses the copula.)

In that case, as we're counting a compound tense with the copula, then any pro-drop language with null copula and compound tenses would qualify. For example, Russian (past tense).

9
  • This is very interesting: thank you! Although, is Russian past tense a good example? From what I understand, pronouns are harder to drop there – exactly because the verb is inflected only for number and gender. Also, you say that Turkish 3rd-person present tense is marked. You mean because the inf is etmek, but the 3rd-person present tense is eder and not just et? And it’s istemek, but ister and not just iste? Does that automatically mean it’s marked for 3rd person, though? Couldn’t it be argued that it’s just marked as finite, or as finite and present tense?
    – MarkOxford
    Apr 3, 2019 at 20:04
  • In Russian one can definitely say «Пошел?». «Понравилось.» «Договорились.». Is it sometimes ambiguous? Yes. Apr 3, 2019 at 20:34
  • Yes, if you want to argue there is a null after -er in eder you can, but then many more languages have some conjugation that fits your criteria. Eg South Slavic present tense. Or many Armenian tenses. Apr 3, 2019 at 20:43
  • 1
    Actually, I think I just found one: Romanian. If the verb stem ends in a consonant, the first-person inflectional morpheme is null. Take alerga = to run. The stem is alerg- and the inflection goes: alerg, alergi, aleargă, alergăm, alergați, aleargă. Therefore, I run is just Alerg. I think that’s exactly the kind of example I was looking for.
    – MarkOxford
    Apr 4, 2019 at 9:14
  • 1
    Yes, Romanian, and also Armenian in some tenses. Like եկա, all other forms are like 1st-person plus something else. However... the whole premise of null morpheme is questionable, because there are stem changes (in Turkish too, eg et- becomes ed-). So we should distinguish between surface form and lemma. Like in English, 1st-person appears to take the lemma plus null morpheme, but then there is am. Apr 4, 2019 at 13:44
1

I've seen literature that advances the thesis that languages without "pro-drop" tend to have verb that inflect to distinguish some, but not all person/number categories (I think the typical examples used to illustrate this claim are English, French and German), while languages with "pro-drop" tend to either have inflected verbs that (mostly) distinguish between all the person/number categories, or verbs that do not inflect. This generalization is mentioned in the Wikipedia article on "Pro-drop languages".

This would imply that in a language with "pro-drop", a null affix would most likely only mark one person/number category, not several.

Based on what I've read, marking the third person with a null affix on the verb is not uncommon, although I don't have any numbers to give you. I haven't heard of this being a common strategy for marking the first person singular, although it probably exists in some pro-drop language.

Navajo is an example of a language with prefixes to mark subjects and objects where a third-person subject (singular or plural) is marked with a "null" affix.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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