Loading ...
Sorry, an error occurred while loading the content.
 

Marking both patient and agent on verbs only if both are animate

Expand Messages
  • BPJ
    This might be a totally weird idea, but here goes: A while ago Christophe mentioned that in Spoken French objects are marked on the verb if and only if they
    Message 1 of 11 , Oct 29, 2013
      This might be a totally weird idea, but here goes:

      A while ago Christophe mentioned that in Spoken French objects are marked
      on the verb if and only if they are definite, and that that is a fairly
      common pattern in natlangs. I thought that I might adopt that feature in
      Sohlob but naturally I soon came up with a weird twist! In Sohlob animacy
      is much more deeply entrenched in the grammar than definiteness, so what if
      it is animate objects that are marked? Then again the important categories
      in Sohlob are animacy *and* agency/patiency. Animates are inherently more
      agent-worthy and less patient-worthy than inanimates, and moreover the
      following holds:

      * Animate agent NPs have no marking.
      * Inanimate agent NPs always have marking.
      * Animate patient NPs have marking if the agent is animate.

      The only factor which makes me wonder to what extent marking on verbs
      should follow the same pattern is the fact that in the ancestor language
      the marker on animate patients and the marker on inanimate agents was the
      same _-ya_ clitic: it marked that either did not have its canonical
      participant role, but at the same time only one NP in each clause could be
      marked with that clitic and inanimate agents counted as more abnormal than
      animate patients, hence more marking-worthy, or else animate patients of
      inanimate agents were grammatically subjects -- the inanimate agent was
      marked both with the instrumental case and the _-ya_ inversion particle.
      Not that the verb had any passive marking or anything! In Sohlob the
      instrumental ending _-r_ and the inversion particle _-ya_ are merged as the
      ergative ending _-Vl_ and the _-ya_ on animate patients lives on as the
      accusative ending _-Vy_.

      Now my thought is that maybe *all* Animate participants and *no* Inanimate
      participants are marked on the verb? How naturalistic is that? Note that
      real-historically Sohlob has happily dropped NPs when recoverable from
      context without any participant marking on verbs.

      /bpj
    • Jörg Rhiemeier
      Hallo conlangers! ... That makes sense. ... Yes. For animates, the agent role is more typical; for inanimates, the patient role. Hence the common kind of
      Message 2 of 11 , Oct 29, 2013
        Hallo conlangers!

        On Tuesday 29 October 2013 17:39:23 BPJ wrote:

        > This might be a totally weird idea, but here goes:
        >
        > A while ago Christophe mentioned that in Spoken French objects are marked
        > on the verb if and only if they are definite, and that that is a fairly
        > common pattern in natlangs. I thought that I might adopt that feature in
        > Sohlob but naturally I soon came up with a weird twist! In Sohlob animacy
        > is much more deeply entrenched in the grammar than definiteness, so what if
        > it is animate objects that are marked?

        That makes sense.

        > Then again the important categories
        > in Sohlob are animacy *and* agency/patiency. Animates are inherently more
        > agent-worthy and less patient-worthy than inanimates,

        Yes. For animates, the agent role is more typical; for
        inanimates, the patient role. Hence the common kind of
        animacy-based split ergativity.

        > and moreover the
        > following holds:
        >
        > * Animate agent NPs have no marking.
        > * Inanimate agent NPs always have marking.
        > * Animate patient NPs have marking if the agent is animate.

        In many languages, inanimate NPs receive less marking on verbs
        than animate ones. In some older IE languages, neuter plural
        subjects go with singular verbs. In my conlang Old Albic,
        inanimate non-singular patients go with verbs marked for a
        singular patient.

        Also, in many languages, animate patient NPs receive a case
        marking that inanimate ones do not receive, such as the Spanish
        _a_, in origin a dative marker. PIE was like this, too: while
        non-neuter nouns have distinct nominatives and accusatives,
        neuter nouns have one form for both (which is endingless in
        the athematic declension). Again, my conlang Old Albic marks
        animate patients with a suffix -m, while inanimate patients are
        unmarked.

        Old Albic does not allow true inanimate agents; they are coded
        as instrumentals instead and the verb has a zero agent. Some
        languages, such as the Algonquian family, use an inverse maker
        on the verb if the patient outranks the agent on the animacy
        hierarchy.

        > The only factor which makes me wonder to what extent marking on verbs
        > should follow the same pattern is the fact that in the ancestor language
        > the marker on animate patients and the marker on inanimate agents was the
        > same _-ya_ clitic: it marked that either did not have its canonical
        > participant role, but at the same time only one NP in each clause could be
        > marked with that clitic and inanimate agents counted as more abnormal than
        > animate patients, hence more marking-worthy,

        Sure. Inanimate agents are much more "out of the ordinary" than
        animate patients. Inanimate objects, after all, usually do not
        act out of themselves; but animate beings are as likely to be
        affected by an action of something else as inanimate objects.

        > or else animate patients of
        > inanimate agents were grammatically subjects -- the inanimate agent was
        > marked both with the instrumental case

        As in Old Albic, and I think also in quite a few natlangs.

        > and the _-ya_ inversion particle.
        > Not that the verb had any passive marking or anything! In Sohlob the
        > instrumental ending _-r_ and the inversion particle _-ya_ are merged as the
        > ergative ending _-Vl_ and the _-ya_ on animate patients lives on as the
        > accusative ending _-Vy_.

        This is an interesting and plausible development.

        > Now my thought is that maybe *all* Animate participants and *no* Inanimate
        > participants are marked on the verb? How naturalistic is that?

        I think this is quite naturalistic. At least, it makes sense to
        me, and I guess that there are natlangs that do just that.

        --
        ... brought to you by the Weeping Elf
        http://www.joerg-rhiemeier.de/Conlang/index.html
        "Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1
      • BPJ
        ... marked ... animacy ... what if ... more ... Of course it s just such a system which developed in Sohlob, and arguably was in place already in Kijeb
        Message 3 of 11 , Oct 30, 2013
          Den 29 okt 2013 21:09 skrev "Jörg Rhiemeier" <joerg_rhiemeier@...>:
          >
          > Hallo conlangers!
          >
          > On Tuesday 29 October 2013 17:39:23 BPJ wrote:
          >
          > > This might be a totally weird idea, but here goes:
          > >
          > > A while ago Christophe mentioned that in Spoken French objects are
          marked
          > > on the verb if and only if they are definite, and that that is a fairly
          > > common pattern in natlangs. I thought that I might adopt that feature in
          > > Sohlob but naturally I soon came up with a weird twist! In Sohlob
          animacy
          > > is much more deeply entrenched in the grammar than definiteness, so
          what if
          > > it is animate objects that are marked?
          >
          > That makes sense.
          >
          > > Then again the important categories
          > > in Sohlob are animacy *and* agency/patiency. Animates are inherently
          more
          > > agent-worthy and less patient-worthy than inanimates,
          >
          > Yes. For animates, the agent role is more typical; for
          > inanimates, the patient role. Hence the common kind of
          > animacy-based split ergativity.

          Of course it's just such a system which developed in Sohlob, and arguably
          was in place already in Kijeb (although Kjb. Is strictly not the ancestor
          of Shl.) except that Kijeb didn't have an ergative ending distinct from the
          instrumental.

          >
          > > and moreover the
          > > following holds:
          > >
          > > * Animate agent NPs have no marking.
          > > * Inanimate agent NPs always have marking.
          > > * Animate patient NPs have marking if the agent is animate.
          >
          > In many languages, inanimate NPs receive less marking on verbs
          > than animate ones.

          Exactly.

          > In some older IE languages, neuter plural
          > subjects go with singular verbs.

          Is there any other than Greek where this is consistently the case? I don't
          remember.
          IIRC it has been claimed that neuter plurals were originally collective in
          meaning. I find that hard to believe though!

          > In my conlang Old Albic,
          > inanimate non-singular patients go with verbs marked for a
          > singular patient.

          In Sohlob number is marked only on determiners so that the simplest form is
          the indefinite plural. I think however that the indefinite singular marker
          -- historically a cliticized numeral ' one' -- can be omitted with
          inanimates but not with animates.
          Case marking goes on the whole NP and is the same for singular and plurals.
          Such number marking as exists was originally reduplication or suppletion of
          pronominal roots.

          >
          > Also, in many languages, animate patient NPs receive a case
          > marking that inanimate ones do not receive, such as the Spanish
          > _a_, in origin a dative marker.

          Sohlob uses the allative instead of the dative with inanimates. They can't
          really be agents so they can't be recipients, only receptacles!

          > PIE was like this, too: while
          > non-neuter nouns have distinct nominatives and accusatives,
          > neuter nouns have one form for both (which is endingless in
          > the athematic declension). Again, my conlang Old Albic marks
          > animate patients with a suffix -m, while inanimate patients are
          > unmarked.

          As already said Sohlob marks animate patients of animate agents as
          accusative and inanimate agents as ergative. When the agent is animate and
          the patient is inanimate both are unmarked.

          >
          > Old Albic does not allow true inanimate agents; they are coded
          > as instrumentals instead and the verb has a zero agent.

          Kijeb is like that too.

          > Some
          > languages, such as the Algonquian family, use an inverse maker
          > on the verb if the patient outranks the agent on the animacy
          > hierarchy.

          As said Kijeb marked inversion on the NPs. I don't know any natlang which
          does that however.

          >
          > > The only factor which makes me wonder to what extent marking on verbs
          > > should follow the same pattern is the fact that in the ancestor language
          > > the marker on animate patients and the marker on inanimate agents was
          the
          > > same _-ya_ clitic: it marked that either did not have its canonical
          > > participant role, but at the same time only one NP in each clause could
          be
          > > marked with that clitic and inanimate agents counted as more abnormal
          than
          > > animate patients, hence more marking-worthy,
          >
          > Sure. Inanimate agents are much more "out of the ordinary" than
          > animate patients. Inanimate objects, after all, usually do not
          > act out of themselves; but animate beings are as likely to be
          > affected by an action of something else as inanimate objects.
          >
          > > or else animate patients of
          > > inanimate agents were grammatically subjects -- the inanimate agent was
          > > marked both with the instrumental case
          >
          > As in Old Albic, and I think also in quite a few natlangs.
          >
          > > and the _-ya_ inversion particle.
          > > Not that the verb had any passive marking or anything! In Sohlob the
          > > instrumental ending _-r_ and the inversion particle _-ya_ are merged as
          the
          > > ergative ending _-Vl_ and the _-ya_ on animate patients lives on as the
          > > accusative ending _-Vy_.
          >
          > This is an interesting and plausible development.

          I actually retrofitted that from the Sohlob endings which go back to my
          very first conlang almost forty years ago where the nominative was -l and
          the accusative was i/j! I had no idea then that [l] might arise from [rj]
          and -r was the genitive, the only other case being a dative. The lang had a
          nullar number though and the sound /K/ which I had discovered on my own. I
          didn't know it was a lateral and wrote it 《çh》!

          >
          > > Now my thought is that maybe *all* Animate participants and *no*
          Inanimate
          > > participants are marked on the verb? How naturalistic is that?
          >
          > I think this is quite naturalistic. At least, it makes sense to
          > me, and I guess that there are natlangs that do just that.

          Actually I discovered that the sound changes add up to a strange pattern:
          verbs get marked for whether one or more than one animate entities are
          participating, regardless of their role so the verb gets plural marking if
          both participants are animate singular! I like that though!

          /bpj

          >
          > --
          > ... brought to you by the Weeping Elf
          > http://www.joerg-rhiemeier.de/Conlang/index.html
          > "Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1
        • Jörg Rhiemeier
          Hallo conlangers! ... Such a system may also have been in place, BTW, in Early PIE. The neuter nominative-accusative may have just been an absolutive, and
          Message 4 of 11 , Oct 31, 2013
            Hallo conlangers!

            On Thursday 31 October 2013 01:31:28 BPJ wrote:

            > Den 29 okt 2013 21:09 skrev "Jörg Rhiemeier" <joerg_rhiemeier@...>:
            > > Hallo conlangers!
            > [...]
            > > Yes. For animates, the agent role is more typical; for
            > > inanimates, the patient role. Hence the common kind of
            > > animacy-based split ergativity.
            >
            > Of course it's just such a system which developed in Sohlob, and arguably
            > was in place already in Kijeb (although Kjb. Is strictly not the ancestor
            > of Shl.) except that Kijeb didn't have an ergative ending distinct from the
            > instrumental.

            Such a system may also have been in place, BTW, in Early PIE.
            The neuter "nominative-accusative" may have just been an
            absolutive, and the instrumental would have done the duty of
            the ergative with inanimate nouns.

            > [...]
            > > In many languages, inanimate NPs receive less marking on verbs
            > > than animate ones.
            >
            > Exactly.
            >
            > > In some older IE languages, neuter plural
            > > subjects go with singular verbs.
            >
            > Is there any other than Greek where this is consistently the case? I don't
            > remember.

            I don't remember either. Greek definitely does, as you noted.
            I *think* Hittite does so too, but I can't find a reference now.

            > IIRC it has been claimed that neuter plurals were originally collective in
            > meaning. I find that hard to believe though!
            >
            > > In my conlang Old Albic,
            > > inanimate non-singular patients go with verbs marked for a
            > > singular patient.
            >
            > In Sohlob number is marked only on determiners so that the simplest form is
            > the indefinite plural. I think however that the indefinite singular marker
            > -- historically a cliticized numeral ' one' --

            Similar to Basque.

            > can be omitted with
            > inanimates but not with animates.
            >
            > Case marking goes on the whole NP and is the same for singular and plurals.
            > Such number marking as exists was originally reduplication or suppletion of
            > pronominal roots.
            >
            > > Also, in many languages, animate patient NPs receive a case
            > > marking that inanimate ones do not receive, such as the Spanish
            > > _a_, in origin a dative marker.
            >
            > Sohlob uses the allative instead of the dative with inanimates. They can't
            > really be agents so they can't be recipients, only receptacles!

            As does Old Albic!

            > > PIE was like this, too: while
            > > non-neuter nouns have distinct nominatives and accusatives,
            > > neuter nouns have one form for both (which is endingless in
            > > the athematic declension). Again, my conlang Old Albic marks
            > > animate patients with a suffix -m, while inanimate patients are
            > > unmarked.
            >
            > As already said Sohlob marks animate patients of animate agents as
            > accusative and inanimate agents as ergative. When the agent is animate and
            > the patient is inanimate both are unmarked.

            This makes perfect sense. For one of my Hesperic languages, I
            plan such a thing without noun classes but with an animacy
            hierarchy. If the subject outranks the object, both are unmarked.
            If their ranks are equal, the subject is unmarked and the object
            marked with the accusative (=dative) case. If the object outranks
            the subject, the subject is marked with the ergative (=genitive)
            case, and the object with the accusative (=dative) case.

            > > Old Albic does not allow true inanimate agents; they are coded
            > > as instrumentals instead and the verb has a zero agent.
            >
            > Kijeb is like that too.

            And some scholars suspect this for Early PIE, too.

            > > Some
            > > languages, such as the Algonquian family, use an inverse maker
            > > on the verb if the patient outranks the agent on the animacy
            > > hierarchy.
            >
            > As said Kijeb marked inversion on the NPs. I don't know any natlang which
            > does that however.

            I don't know, either. Inversion is usually marked on the verb.

            > [...]
            > > This is an interesting and plausible development.
            >
            > I actually retrofitted that from the Sohlob endings which go back to my
            > very first conlang almost forty years ago where the nominative was -l and
            > the accusative was i/j! I had no idea then that [l] might arise from [rj]
            > and -r was the genitive, the only other case being a dative. The lang had a
            > nullar number though and the sound /K/ which I had discovered on my own. I
            > didn't know it was a lateral and wrote it 《çh》!

            Old Albic contains very little that goes back to my first attempts
            at conlanging. But my first conlang had a gender system that
            worked in a similar way as that of Old Albic, though it was yet
            more similar to that of Novial (which I did not know back then)
            in having a single neuter gender for both animates of unspecified
            sex and inanimates.

            One of the things I discovered without knowing natlang precedents
            was Suffixaufnahme, which also eventually found its way into Old
            Albic.

            > [...]
            > > I think this is quite naturalistic. At least, it makes sense to
            > > me, and I guess that there are natlangs that do just that.
            >
            > Actually I discovered that the sound changes add up to a strange pattern:
            > verbs get marked for whether one or more than one animate entities are
            > participating, regardless of their role so the verb gets plural marking if
            > both participants are animate singular! I like that though!

            This is interesting.

            --
            ... brought to you by the Weeping Elf
            http://www.joerg-rhiemeier.de/Conlang/index.html
            "Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1
          • Christophe Grandsire-Koevoets
            ... In my conlang Moten, I use a similar trick but based on volitionality rather than animacy: with transitive verbs, the nominative case is used only if the
            Message 5 of 11 , Nov 1, 2013
              On 29 October 2013 21:09, Jörg Rhiemeier <joerg_rhiemeier@...> wrote:

              >
              > > or else animate patients of
              > > inanimate agents were grammatically subjects -- the inanimate agent was
              > > marked both with the instrumental case
              >
              > As in Old Albic, and I think also in quite a few natlangs.
              >
              >
              In my conlang Moten, I use a similar trick but based on volitionality
              rather than animacy: with transitive verbs, the nominative case is used
              only if the subject is volitional. Non-volitional subjects are marked with
              the instrumental (for instance, the difference between "see" and "look at"
              in Moten is the form of the subject only. The verb itself is _ipe|laj_ for
              both meanings). Since inanimates can never be volitional, they only ever
              appear as subjects of transitive verbs in the instrumental. The exception
              seems to be weather phenomena: when used as subjects of transitive verbs,
              they always appear in the nominative case, never in the instrumental. This
              makes them even more volitional than human beings, which *can* appear in
              the instrumental when they are experiencers!

              Note that this doesn't extend to intransitive verbs: those take subjects in
              the nominative case regardless of volitionality or animacy. So it seems
              it's the presence of an object that triggers the volitional rule for the
              nominative case.


              >
              > > Now my thought is that maybe *all* Animate participants and *no*
              > Inanimate
              > > participants are marked on the verb? How naturalistic is that?
              >
              > I think this is quite naturalistic. At least, it makes sense to
              > me, and I guess that there are natlangs that do just that.
              >
              >
              Agreed. I remember reading about a natlang that marked first and second
              person subject on the verb, but only marked third person subjects if they
              were animate (maybe even restricted to human beings). Inanimate third
              person subjects were not marked on the verb. I can't remember which
              language it was though...

              On 31 October 2013 01:31, BPJ <bpj@...> wrote:

              > Exactly.
              >
              > > In some older IE languages, neuter plural
              > > subjects go with singular verbs.
              >
              > Is there any other than Greek where this is consistently the case? I don't
              > remember.
              > IIRC it has been claimed that neuter plurals were originally collective in
              > meaning. I find that hard to believe though!
              >
              >
              Why? There's a lot of evidence that at least the neuter plurals in -a, as
              found in Latin and Greek for instance, have the same origin as the feminine
              singulars in -a, i.e the suffix *-h2, which originally formed derived
              abstract nouns and definitely had also a secondary collective meaning (see
              http://attach.matita.net/silvialuraghi/file/Origin%20of%20the%20feminine%20gender.pdf).
              It's not that weird: in languages with animate/inanimate gender
              distinctions, it's not unknown for the inanimate gender to lack number
              distinctions that are present with animates (see Japanese, for instance,
              which can use the -tachi associative plural suffix on animate nouns, but
              not on inanimate ones). Basically, inanimate nouns are always mass nouns.
              If the language also has a productive way of creating collective nouns,
              it's not that surprising to see it evolve as a true plural marking for
              inanimate nouns that lack one originally, but could allow it due to
              semantics. It's just a case of a derivational affix turning into an
              inflectional one, a development that is well known and common.

              But frankly, to me collectives turning into actual plurals makes a lot of
              sense to me!


              >
              > >
              > > Also, in many languages, animate patient NPs receive a case
              > > marking that inanimate ones do not receive, such as the Spanish
              > > _a_, in origin a dative marker.
              >
              > Sohlob uses the allative instead of the dative with inanimates. They can't
              > really be agents so they can't be recipients, only receptacles!
              >
              >
              Moten has something similar with the verbs _ja|zi|n_ and _joplej_ (which
              both mark "transfer". Actual translation depends on the participants). When
              used with an animate noun in the benefactive (which also works as dative),
              they take on the meaning "to give to". With an inanimate noun in the
              accusative case (which can take a spatial allative sense, made more clear
              by the optional prefix _mo-_), they take on the meaning "to put in/on".
              Naturally, inanimate nouns cannot be put in the benefactive, unless they
              are personified. Moten also has an originative form, opposite of the
              benefactive (giving the meaning "to take from" to the verbs mentioned
              above), and distinct from the delative (in Moten that's simply the genitive
              case, with the optional prefix _mo-_).
              --
              Christophe Grandsire-Koevoets.

              http://christophoronomicon.blogspot.com/
              http://www.christophoronomicon.nl/
            • BPJ
              Christophe: =========== Since inanimates can never be volitional, they only ever appear as subjects of transitive verbs in the instrumental. The exception
              Message 6 of 11 , Nov 1, 2013
                Christophe:
                ===========
                Since inanimates can never be volitional, they only ever
                appear as subjects of transitive verbs in the instrumental. The exception
                seems to be weather phenomena: when used as subjects of transitive verbs,
                they always appear in the nominative case, never in the instrumental. This
                makes them even more volitional than human beings, which *can* appear in
                the instrumental when they are experiencers!
                ===========
                Wheather phenomena are animate in Sohlob too, and so are an assorted bunch
                of other phenomena which a modern Terran Westerner would consider
                inanimate, includingbut but not limited to:

                - heavenly bodies (some of which are deities)
                - fire and volcanoes
                - water and bodies of water
                - plants
                - houses
                - musical instruments
                - books

                /bpj

                On Nov 1, 2013 10:40 AM, "Christophe Grandsire-rKoevoets" <
                tsela.cg@...> wrote:
                >
                > On 29 October 2013 21:09, Jörg Rhiemeier <joerg_rhiemeier@...> wrote:
                >
                > >
                > > > or else animate patients of
                > > > inanimate agents were grammatically subjects -- the inanimate agent
                was
                > > > marked both with the instrumental case
                > >
                > > As in Old Albic, and I think also in quite a few natlangs.
                > >
                > >
                > In my conlang Moten, I use a similar trick but based on volitionality
                > rather than animacy: with transitive verbs, the nominative case is used
                > only if the subject is volitional. Non-volitional subjects are marked with
                > the instrumental (for instance, the difference between "see" and "look at"
                > in Moten is the form of the subject only. The verb itself is _ipe|laj_ for
                > both meanings). Since inanimates can never be volitional, they only ever
                > appear as subjects of transitive verbs in the instrumental. The exception
                > seems to be weather phenomena: when used as subjects of transitive verbs,
                > they always appear in the nominative case, never in the instrumental. This
                > makes them even more volitional than human beings, which *can* appear in
                > the instrumental when they are experiencers!
                >
                > Note that this doesn't extend to intransitive verbs: those take subjects
                in
                > the nominative case regardless of volitionality or animacy. So it seems
                > it's the presence of an object that triggers the volitional rule for the
                > nominative case.
                >
                >
                > >
                > > > Now my thought is that maybe *all* Animate participants and *no*
                > > Inanimate
                > > > participants are marked on the verb? How naturalistic is that?
                > >
                > > I think this is quite naturalistic. At least, it makes sense to
                > > me, and I guess that there are natlangs that do just that.
                > >
                > >
                > Agreed. I remember reading about a natlang that marked first and second
                > person subject on the verb, but only marked third person subjects if they
                > were animate (maybe even restricted to human beings). Inanimate third
                > person subjects were not marked on the verb. I can't remember which
                > language it was though...
                >
                > On 31 October 2013 01:31, BPJ <bpj@...> wrote:
                >
                > > Exactly.
                > >
                > > > In some older IE languages, neuter plural
                > > > subjects go with singular verbs.
                > >
                > > Is there any other than Greek where this is consistently the case? I
                don't
                > > remember.
                > > IIRC it has been claimed that neuter plurals were originally collective
                in
                > > meaning. I find that hard to believe though!
                > >
                > >
                > Why? There's a lot of evidence that at least the neuter plurals in -a, as
                > found in Latin and Greek for instance, have the same origin as the
                feminine
                > singulars in -a, i.e the suffix *-h2, which originally formed derived
                > abstract nouns and definitely had also a secondary collective meaning (see
                >
                http://attach.matita.net/silvialuraghi/file/Origin%20of%20the%20feminine%20gender.pdf
                ).
                > It's not that weird: in languages with animate/inanimate gender
                > distinctions, it's not unknown for the inanimate gender to lack number
                > distinctions that are present with animates (see Japanese, for instance,
                > which can use the -tachi associative plural suffix on animate nouns, but
                > not on inanimate ones). Basically, inanimate nouns are always mass nouns.
                > If the language also has a productive way of creating collective nouns,
                > it's not that surprising to see it evolve as a true plural marking for
                > inanimate nouns that lack one originally, but could allow it due to
                > semantics. It's just a case of a derivational affix turning into an
                > inflectional one, a development that is well known and common.
                >
                > But frankly, to me collectives turning into actual plurals makes a lot of
                > sense to me!
                >
                >
                > >
                > > >
                > > > Also, in many languages, animate patient NPs receive a case
                > > > marking that inanimate ones do not receive, such as the Spanish
                > > > _a_, in origin a dative marker.
                > >
                > > Sohlob uses the allative instead of the dative with inanimates. They
                can't
                > > really be agents so they can't be recipients, only receptacles!
                > >
                > >
                > Moten has something similar with the verbs _ja|zi|n_ and _joplej_ (which
                > both mark "transfer". Actual translation depends on the participants).
                When
                > used with an animate noun in the benefactive (which also works as dative),
                > they take on the meaning "to give to". With an inanimate noun in the
                > accusative case (which can take a spatial allative sense, made more clear
                > by the optional prefix _mo-_), they take on the meaning "to put in/on".
                > Naturally, inanimate nouns cannot be put in the benefactive, unless they
                > are personified. Moten also has an originative form, opposite of the
                > benefactive (giving the meaning "to take from" to the verbs mentioned
                > above), and distinct from the delative (in Moten that's simply the
                genitive
                > case, with the optional prefix _mo-_).
                > --
                > Christophe Grandsire-Koevoets.
                >
                > http://christophoronomicon.blogspot.com/
                > http://www.christophoronomicon.nl/
              • Christophe Grandsire-Koevoets
                ... Interesting. But the weird thing about weather phenomena in Moten is that despite being always volitional, they are still treated as inanimates! That s to
                Message 7 of 11 , Nov 1, 2013
                  On 1 November 2013 13:00, BPJ <bpj@...> wrote:

                  > Christophe:
                  > ===========
                  > Since inanimates can never be volitional, they only ever
                  > appear as subjects of transitive verbs in the instrumental. The exception
                  > seems to be weather phenomena: when used as subjects of transitive verbs,
                  > they always appear in the nominative case, never in the instrumental. This
                  > makes them even more volitional than human beings, which *can* appear in
                  > the instrumental when they are experiencers!
                  > ===========
                  > Wheather phenomena are animate in Sohlob too, and so are an assorted bunch
                  > of other phenomena which a modern Terran Westerner would consider
                  > inanimate, includingbut but not limited to:
                  >
                  > - heavenly bodies (some of which are deities)
                  > - fire and volcanoes
                  > - water and bodies of water
                  > - plants
                  > - houses
                  > - musical instruments
                  > - books
                  >
                  >
                  Interesting. But the weird thing about weather phenomena in Moten is that
                  despite being always volitional, they are still treated as inanimates!
                  That's to say, in the few areas where Moten makes a definite distinction
                  between animate and inanimate nouns (mostly pairs of verbs, which often
                  have actually a slightly different distinction from just animacy), weather
                  phenomena are still considered inanimate!

                  That said, animacy is not an inflectional category in Moten. It's purely
                  semantic, and actually most of the areas where the distinction is made,
                  it's not quite an animate vs. inanimate one. Usually, the distinction is
                  between "human beings and animals bigger/stronger than human beings" vs.
                  "animals smaller than human beings and everything else". True animacy
                  distinction seems only to be done... well, with the volitional nominative
                  thing, with the exception of weather phenomena. And it's a sliding scale
                  rather than a black-and-white thing: small animals, while usually
                  considered animates, are "felt" less animate the smaller they become, and
                  thus less likely to exhibit volitionality. I don't think a Moten speaker
                  would ever treat bacteria as volitional Truly, Moten has some weird things
                  going on...

                  Maybe weather phenomena are actually a third pole to the "human beings and
                  animals bigger/stronger than human beings" vs. "animals smaller than human
                  beings and everything else" distinction. One piece of evidence for that is
                  that when asserting existence of something, "human beings and animals
                  bigger/stronger than human beings" use the verb _ispej_, while "animals
                  smaller than human beings and everything else" use _jaki_. Both can be
                  translated as "to exist", and are often used to mean "there is". As for
                  weather phenomena, they actually don't use either. Instead, they use the
                  verb _ivdaj_, which I usually translate as "to happen", but is restricted
                  to weather phenomena only.

                  But as I wrote above, animacy is not an inflectional category in Moten,
                  just a semantic one with some consequences in the choice of some verbs
                  and/or constructions. It features lots of shades of grey.
                  --
                  Christophe Grandsire-Koevoets.

                  http://christophoronomicon.blogspot.com/
                  http://www.christophoronomicon.nl/
                • Adam Walker
                  ... You are my hero!! Adam
                  Message 8 of 11 , Nov 1, 2013
                    On Fri, Nov 1, 2013 at 7:00 AM, BPJ <bpj@...> wrote:

                    >
                    > Wheather phenomena are animate in Sohlob too, and so are an assorted bunch
                    > of other phenomena which a modern Terran Westerner would consider
                    > inanimate, includingbut but not limited to:
                    >


                    > - books
                    >
                    >

                    You are my hero!!

                    Adam
                  • Padraic Brown
                    ... Seconded! Certainly in the World, many books are in fact animate beings! Semantically speaking, all those things on BPJ s list  would be considered
                    Message 9 of 11 , Nov 1, 2013
                      Adam Walker <carraxan@...> wrote:


                      >> Wheather phenomena are animate in Sohlob too, and so are an assorted bunch
                      >> of other phenomena which a modern Terran Westerner would consider
                      >> inanimate, includingbut but not limited to:
                      >
                      >> - books
                      >
                      > You are my hero!!

                      Seconded! Certainly in the World, many books are in fact animate beings! Semantically
                      speaking, all those things on BPJ's list  would be considered animate, or at least
                      potentiate, in many World languages on account of their capacity for "flowing" -- waters
                      flow, and thus rivers are animate; volcanoes explode (the lava simply flowing up into
                      the atmosphere) and are thus animate; thunder clouds and lightning flow through the air
                      and rain flows down it. Knowledge flows from the book and into the eyes and minds of
                      the reader. And sometimes lodges there, growing into a sort of parasitic life form all its
                      own, breeding, biding its time until its host is ready to transmit that knowledge on to
                      another host......

                      Padraic

                      > Adam
                    • Jörg Rhiemeier
                      Hallo conlangers! ... Old Albic has an elaborate system of degrees of volition, which I shall lay out in brief below. An agentive subject indicates full
                      Message 10 of 11 , Nov 1, 2013
                        Hallo conlangers!

                        On Friday 01 November 2013 10:40:25 "Christophe Grandsire-Koevoets" wrote:

                        > On 29 October 2013 21:09, Jörg Rhiemeier <joerg_rhiemeier@...> wrote:
                        > > > or else animate patients of
                        > > >
                        > > > inanimate agents were grammatically subjects -- the inanimate agent was
                        > > > marked both with the instrumental case
                        > >
                        > > As in Old Albic, and I think also in quite a few natlangs.
                        >
                        > In my conlang Moten, I use a similar trick but based on volitionality
                        > rather than animacy: with transitive verbs, the nominative case is used
                        > only if the subject is volitional. Non-volitional subjects are marked with
                        > the instrumental (for instance, the difference between "see" and "look at"
                        > in Moten is the form of the subject only. The verb itself is _ipe|laj_ for
                        > both meanings). Since inanimates can never be volitional, they only ever
                        > appear as subjects of transitive verbs in the instrumental.

                        Old Albic has an elaborate system of degrees of volition, which
                        I shall lay out in brief below.

                        An agentive subject indicates full volition:

                        (1a) Ibretasa Senantho am bul.
                        AOR-break-3SG:P-3SG:A Senantho-AGT the bowl
                        'Senantho broke the bowl (on purpose).'

                        This is the unmarked member of the paradigm.

                        A dative subject indicates that the subject acted accidentally:

                        (1b) Ibreatasa Senanthon am bul.
                        AOR-break-3SG:P-3SG:A Senantho-DAT the bowl
                        'Senantho broke the bowl (accidentally).'

                        This may indicate a situation where the bowl slipped out of
                        Senantho's hands and dropped to the floor, or something like
                        that.

                        Thirdly, the instrumental case indicates external cause:

                        (1c) Ibreta Senanthømi am bul.
                        AOR-break-3SG:P Senantho-INS the bowl
                        'Senantho broke the bowl (under external cause).'

                        An example would be a situation where Senantho is knocked over
                        by someone else and therein knocks over and breaks the bowl.
                        Note that the verb form is different: there is no longer any
                        agent in the clause.

                        Inanimate nouns may only take the last of the three degrees of
                        volition listed here:

                        (1d) Ibreta cheri am bul.
                        AOR-break-3SG:P stone-INS the bowl
                        'A stone broke the bowl.'

                        There must have been an external force that flung the stone
                        against the bull, hence only instrumental case is possible here.

                        With verbs of perception, the meaning of the degrees of volition
                        is that the agentive case marks deliberate observation, and the
                        dative case cursory perception:

                        (2a) Iselasa Senantho am chvanam.
                        AOR-see-3SG:P-3SG:A Senantho-AGT the dog-OBJ
                        'Senantho looked at the dog.'

                        (2b) Iselasa Senanthon am chvanam.
                        AOR-see-3SG:P-3SG:A Senantho-DAT the dog-OBJ
                        'Senantho saw the dog.'

                        *Patients* are something different still. Those are in the
                        objective case. Old Albic is a fluid-S language, wherein
                        some verbs (such as verbs of motion) may take either an agent
                        or patient subject:

                        (3a) Anacvamsa Mørdindo hadar.
                        AOR-arrive-3SG:A Mørdindo-AGT today
                        'Mørdindo arrived today.'

                        Mørdindo travelled actively.

                        (3b) Anacvama gratath Mørdindol hadar.
                        AOR-arrive-3SG:P letter-OBJ Mørdindo-PRT today
                        'Mørdindo's letter arrived today.'

                        The letter was carried by a messenger.

                        It may be noted that plural inanimate patients go with singular
                        marking on the verb:

                        (3c) Anacvama gratethim Mørdindølim hadar.
                        AOR-arrive-3SG:P letter-PL-OBJ Mørdindo-PRT-PL-OBJ today
                        'Mørdindo's letters arrived today.'

                        > The exception
                        > seems to be weather phenomena: when used as subjects of transitive verbs,
                        > they always appear in the nominative case, never in the instrumental. This
                        > makes them even more volitional than human beings, which *can* appear in
                        > the instrumental when they are experiencers!

                        Weather phenomena and some other things are animate in Old Albic,
                        too, and as one cannot ascribe degrees of volition to them, but
                        the agentive case is the unmarked member of the paradigm discussed
                        above, those are usually in the agentive case when they function
                        as agents.

                        > Note that this doesn't extend to intransitive verbs: those take subjects in
                        > the nominative case regardless of volitionality or animacy. So it seems
                        > it's the presence of an object that triggers the volitional rule for the
                        > nominative case.

                        Intransitive verbs, as long as they are action verbs, behave
                        the same way as transitive verbs in Old Albic: all three degrees
                        of volition occur. If the verb is fluid, there are thus *four*
                        different forms possible.

                        > > > Now my thought is that maybe *all* Animate participants and *no*
                        > >
                        > > Inanimate
                        > >
                        > > > participants are marked on the verb? How naturalistic is that?
                        > >
                        > > I think this is quite naturalistic. At least, it makes sense to
                        > > me, and I guess that there are natlangs that do just that.
                        >
                        > Agreed. I remember reading about a natlang that marked first and second
                        > person subject on the verb, but only marked third person subjects if they
                        > were animate (maybe even restricted to human beings). Inanimate third
                        > person subjects were not marked on the verb. I can't remember which
                        > language it was though...

                        I can't either, but I think that kind of marking isn't rare.

                        > On 31 October 2013 01:31, BPJ <bpj@...> wrote:
                        > > Exactly.
                        > >
                        > > > In some older IE languages, neuter plural
                        > > > subjects go with singular verbs.
                        > >
                        > > Is there any other than Greek where this is consistently the case? I
                        > > don't remember.
                        > > IIRC it has been claimed that neuter plurals were originally collective
                        > > in meaning. I find that hard to believe though!
                        >
                        > Why? There's a lot of evidence that at least the neuter plurals in -a, as
                        > found in Latin and Greek for instance,

                        Indeed, in all ancient IE languages including Hittite.

                        > have the same origin as the feminine
                        > singulars in -a, i.e the suffix *-h2, which originally formed derived
                        > abstract nouns and definitely had also a secondary collective meaning (see
                        > http://attach.matita.net/silvialuraghi/file/Origin%20of%20the%20feminine%20
                        > gender.pdf).

                        Yes. I am not sold by the idea that the feminine marker *-h2
                        is the same as the abstract/collective marker *-h2 rather than
                        just homophonous; but the abstract *-h2 and the collective/plural
                        *-h2 are connected.

                        > It's not that weird: in languages with animate/inanimate
                        > gender
                        > distinctions, it's not unknown for the inanimate gender to lack number
                        > distinctions that are present with animates (see Japanese, for instance,
                        > which can use the -tachi associative plural suffix on animate nouns, but
                        > not on inanimate ones). Basically, inanimate nouns are always mass nouns.

                        There is a sense of individuality to a perceptive living being
                        that inanimate objects just do not possess. It makes a difference
                        whether you have one person or many; it makes less of a difference
                        whether you have one stone or many.

                        > If the language also has a productive way of creating collective nouns,
                        > it's not that surprising to see it evolve as a true plural marking for
                        > inanimate nouns that lack one originally, but could allow it due to
                        > semantics. It's just a case of a derivational affix turning into an
                        > inflectional one, a development that is well known and common.

                        Sure.

                        > But frankly, to me collectives turning into actual plurals makes a lot of
                        > sense to me!

                        To me, too!

                        --
                        ... brought to you by the Weeping Elf
                        http://www.joerg-rhiemeier.de/Conlang/index.html
                        "Bêsel asa Éam, a Éam atha cvanthal a cvanth atha Éamal." - SiM 1:1
                      • BPJ
                        All talking things are naturally animate �to the Sohlçan, no matter what skill was used to make them talk.
                        Message 11 of 11 , Nov 3, 2013
                          All talking things are naturally animate �to the Sohlçan, no matter what
                          skill was used to make them talk.
                          On Nov 1, 2013 5:21 PM, "Adam Walker" <carraxan@...> wrote:

                          > On Fri, Nov 1, 2013 at 7:00 AM, BPJ <bpj@...> wrote:
                          >
                          > >
                          > > Wheather phenomena are animate in Sohlob too, and so are an assorted
                          > bunch
                          > > of other phenomena which a modern Terran Westerner would consider
                          > > inanimate, includingbut but not limited to:
                          > >
                          >
                          >
                          > > - books
                          > >
                          > >
                          >
                          > You are my hero!!
                          >
                          > Adam
                          >
                        Your message has been successfully submitted and would be delivered to recipients shortly.