Posted by & filed under Podcast.

George and William have a discussion of those times when the subject isn’t in the case you might expect it to be in.

Links and Resources:

14 Responses to “Conlangery #96: Where did my Nominative go?”

  1. Billy James Brightraven

    Your remark about there being possibly existing a system wherein you use the same verb but changing the meaning depending on the marking on the “subject” reminded of something I once read on Wikipedia. I managed to track it down. “Tsova-Tush, a Caucasian language is an active language. According to Holisky (1987), there are 31 intransitive verbs where the argument is always marked as patientive and which refer to uncontrollable states (“be hungry”, “tremble”, etc.), and 78 intransitive verbs with an agentive argument (“walk”, “talk”, “think”). These form a split-S subset of the verbs. The rest of the verbs form a fluid-S system; for instance, a single verb root can be interpreted as “slip” when used with a patientive argument, and as “slide” with an agentive argument.”


  2. Christophe Grandsire-Koevoets

    Just listened to you podcast. Great as usual! It’s always nice to hear how morphology (cases), syntax (subject and agreement) and semantics (agents, patients, experiencers etc.) needn’t parallel each other in a single language, leading to very interesting (and fun!) phenomena like so called “quirky subjects”. More podcasts about such things would be more than welcome :).

    Just a question about the tests of subjectness you mentioned in the beginning: how do you do test for subjectness in a pro-drop language that lacks verbal agreement? My own Moten has “quirky subjects” (more on that below), but is just as pro-drop as Japanese is (so conjunction reduction doesn’t really work here, especially since it lacks clause-level conjunctions!) and doesn’t have any form of verbal agreement. Without these two, how can I tell that my “quirky subjects” are indeed subjects, other than because they usually translate as such in English?

    To be clear about what I mean, in my conlang Moten, intransitive verbs always take a subject in the nominative case. However, transitive verbs take a nominative subject only when the subject is volitional, i.e. purposefully accomplishes the action. When the subject is merely an experiencer, or is an agent but lacks volition, then the subject is in the instrumental form (I’m not calling it a case for reasons that are beyond this small comment. For all intents and purposes here you can consider it an instrumental case).

    Interestingly, Moten does exactly what you were talking about with verbs of perceptions: see/watch, hear/listen, etc. Those pairs are single verbs in Moten, and the form of the subject determines the meaning of the verb. This extends, however, to many more verbs than just perception. The verb idavi|zi, for instance, means “to be happy/content with”, with an instrumental subject. With a nominative subject, it means “to thank for (sthg)”. Another example is igi|noj, which means “to feel, to sense” with an instrumental subject, but “to understand” with a nominative one. And my personal favourite is izgeboj, which means “to work on (sthg)” with a nominative subject, but “to tire from (sthg)” when used with an instrumental one! 😉

    Finally, as Romance languages do with verbs like Spanish “gustar” and French “plaire”, Moten inverses the subject and object (compared to equivalent English verbs) with many verbs referring to mental activities. This is done not only with “to like” (Moten iteo|l translates indeed as “to please”), but also with “to think of” (in Moten jelojmaj translates as “to be pondered by”) and with “to remember” (jelojmastu|l, equivalent to “to be remembered by”), to take only two examples. If you add the different subjects (which can be used since all those verbs are transitive), you get a very interesting view of the way a psyche works :).

    • wm.annis

      I’m afraid I didn’t read everything about determining subjecthood with quirky subjects. I was willing to accept reports that previous scholars had satisfactorily determined that they really were subjects. Even if it lacks clause level conjunctions, there must be some construction in Moten that answers to sequencing in expressions such as “I went to the store and [ I ] bought a cabbage.” That might give a clue.

      Reflexive possession (“I didn’t see my own father”) might provide a clue, too, if you have some way to manage that with PRO-drop habits.

      • Christophe Grandsire-Koevoets

        Yes, there is a Moten construction that allows one to sequence expressions like your example. It consists of removing the auxiliary verb in the first clause (the closest Moten has to clause-level conjunction is indicating that an auxiliary verb works for two clauses). I cannot translate your example exactly, but close enough, so here it is:

        Subedon jagi, dlodan ja|zi|n etok.: “I went to the store and got a pear” (I don’t have a word for “cabbage” yet, nor a word for “to buy”).

        This would be the most fluent way of saying such a thing, and it means word for word “to the store go, a pear get/give/ was”. Notice the absence of a subject in both clauses. The use of jagi in the first clause already implies a first person subject (Moten verbs of motion are speaker-oriented, as in Japanese), and the use of ja|zi|n in that way in the second clause also implies a first person subject (this is actually a rather complex subject and I will focus one of my blog post in the future about how Moten translates “to give”, “to receive” and related verbs).

        You could add a subject in the second clause, especially if you wanted an instrumental subject to indicate that you didn’t ask for a pear: subedon jagi, koga dlodan ja|zi|n etok: “I went to the store and received a pear”, but I don’t know whether the absence of subject in the first clause could really be attributed to conjunction reduction. Just plain pro-drop nature would be enough to explain it, especially since the verb already implies a 1st person subject.

        Reflexive possession will work though. Moten has a reflexive pronoun vike that seems to refer back to the subject of the clause, no matter whether it is in the nominative or instrumental case. I’m not completely sure how it works though: as you surmised, the pro-drop nature of Moten means that vike doesn’t get much use.

      • Christophe Grandsire-Koevoets

        Interestingly, I didn’t know about κάμνω when I created izgeboj! Nice to hear that my thoughts about semantic extensions are naturalistic enough to show up in actual natlangs! It’s always great to get more ANADEWs! 😉

  3. Matthew

    French has an interesting verb – “falloir” – that can take a quirky subject. It means “to be needed, to be necessary”, and it always takes “il” as a dummy subject. To translate “I need a car”, you could say, “Il me faut une voiture”, or literally “It to me is necessary a car”.

  4. Chickenduck

    For most listeners, probably a more familiar example than Icelandic would be German, which does it the same way as Icelandic (though not as extensively):


    Mir ist langweilig = I’m bored (lit. to me is boring)

    Mir ist übel = I feel sick (lit. to me is bad/unpleasant)

    Dem Mann wurde es langsam zu blöd, und er ist weggegangen. = The man eventually got sick of it, and he left. (lit. To the man it slowly became too stupid, and he went away).


    And of course, the humorous mistake every anglophone student of German makes, saying “Ich bin heiß”, when correct is actually “MIR ist heiß”.

    (“Ich bin heiß” also exists, but has the meaning “I’m horny”, like in English “on heat”)

    • Philip Newton

      And such lovely constructions as “Hier wird nicht getanzt”, which looks like a passivised intransitive verb with an unexpressed subject — literally, something like “Here, is not danced” but meaning “One does not dance here”.

  5. Syntax

    Yup, Irish has something similar. I was nodding my head when William was talking about use of nouns rather than verbs.

    The boy loves the girl = Love is at the boy to the girl.
    He wants milk = Milk is from him.
    He is hungry = Hunger is on him.
    He is sad = Sadness is on him.


    • Chickenduck

      I love the Celtic conjugated prepositions and what you can do with them!!

      Like in Irish… “I speak Irish” is “Tá Gaeilge agam” which literally glosses as “There is Irish at me”.

      (I hope I spelled it right… I’ve only been learning Irish for a short time!)

  6. Chickenduck

    Oh and of course – “Tá a fhios agam”, which glosses as “there is – knowledge – at me”.

    Irish for “I don’t know.”

    • Chickenduck


      Sorry – obviously it’s Irish for “I know”. Not for DON’T know!! Where’s the edit button?!


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.