Thematic relations

A forum for discussing linguistics or just languages in general.
Post Reply
G64
hieroglyphic
hieroglyphic
Posts: 42
Joined: 19 Jan 2016 23:36
Location: Italy

Thematic relations

Post by G64 » 05 Mar 2017 07:50

I've decided upon making a conlang that uses cases based on those, but I'm a layman in the topic, so I made this topic where hopefully you can help me by naming the thematic relation of a given noun in a given phrase. Of course this topic would be open to anyone who in interested in that: a sentence is posted, the noun of interest bolded, and others answer the question (or try to, offering their reasonments)
If someone is willing, this might as well make it to the Teach and Share section, with an entire guide about the topic

I beg perdon if a similar topic had already been posted (I was actually surprised to find close to nothing about thematic relations in such a forum)
Native: :ita: | Intermediate: :eng: | Basic: :lat: :esp:
Studied: :qya: (+all of the above)
Willing to study: :grc: :jpn: :heb: :rus:

(Linguistic noob, fear not to correct me)

User avatar
lsd
roman
roman
Posts: 910
Joined: 11 Mar 2011 21:11
Contact:

Re: Thematic relations

Post by lsd » 05 Mar 2017 08:27

My conlanging way is to simplify, as science search to unify the forces of the universe, I only admit one relation: concatenation...
Any speech is built as a concatenation of semantic primes...
But you can also follow a posteriori building even with ex nihilo roots... Natural languages are a source of inspiration...

User avatar
Adarain
greek
greek
Posts: 664
Joined: 03 Jul 2015 14:36
Location: Switzerland, usually

Re: Thematic relations

Post by Adarain » 05 Mar 2017 11:29

Have you looked at Ithkuil's terminology? Might be useful as a source of inspiration.
At kveldi skal dag lęyfa,
Konu es bręnnd es,
Mæki es ręyndr es,
Męy es gefin es,
Ís es yfir kømr,
Ǫl es drukkit es.

User avatar
Creyeditor
mongolian
mongolian
Posts: 4543
Joined: 14 Aug 2012 18:32

Re: Thematic relations

Post by Creyeditor » 05 Mar 2017 15:58

So, I always found the different thematic relations a bit vague and ill-defined. I don't know where exactly but I've seen somewhere (don't know where exactly) a schema based on two criteria, affectedness and control. Still somewhat controversial.

Affected and non-controlling: patient
Non-affected and controlling: agent
Affected and controlling: experiencer
Non-affected and non-controlling: theme
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

G64
hieroglyphic
hieroglyphic
Posts: 42
Joined: 19 Jan 2016 23:36
Location: Italy

Re: Thematic relations

Post by G64 » 07 Mar 2017 06:30

That is an interesting rule of thumb, the one I came up with was a from/through/to distinction, together with a subdivision of the roles in the physical, emotional, movement, change of state classes:
Agent, Force/Instrument/Patient (physical)
Stimulus/Content/Experiencer (emotional)
Source/Path/Goal (movement)
Agent, Force(again?)/Theme/Recipient (change of state

Plus two in's:
Location (in+movement)
Theme (in+change of state)
This second theme is used in phrases that involve "to be there" and the like verbs


But there's a bit of a chiasmus between yours and mine (and yours reflects the fact that, for example, experiencers are usually expressed as subjects in natlangs)
Native: :ita: | Intermediate: :eng: | Basic: :lat: :esp:
Studied: :qya: (+all of the above)
Willing to study: :grc: :jpn: :heb: :rus:

(Linguistic noob, fear not to correct me)

User avatar
Creyeditor
mongolian
mongolian
Posts: 4543
Joined: 14 Aug 2012 18:32

Re: Thematic relations

Post by Creyeditor » 07 Mar 2017 11:08

I think you really need at least the physical vs. emotional distinction.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

G64
hieroglyphic
hieroglyphic
Posts: 42
Joined: 19 Jan 2016 23:36
Location: Italy

Re: Thematic relations

Post by G64 » 07 Mar 2017 14:19

Creyeditor wrote:I think you really need at least the physical vs. emotional distinction.
Indeed I do, the scheme you saw just used a criterion to order the thematic relations, I'm not merging together any of them.
This way I have a handy way to represent them writing (I'm doing this when I take notes at school, just to practice recognizing them)
acute/macron/grave(/dot), written in the middle of the line of type are for from/through/to(/in).
Then I write over the previous circles a dot/circle/tilde/cross to represent physical/emotional/movement/change of state
Native: :ita: | Intermediate: :eng: | Basic: :lat: :esp:
Studied: :qya: (+all of the above)
Willing to study: :grc: :jpn: :heb: :rus:

(Linguistic noob, fear not to correct me)

G64
hieroglyphic
hieroglyphic
Posts: 42
Joined: 19 Jan 2016 23:36
Location: Italy

Re: Thematic relations

Post by G64 » 18 Mar 2017 06:38

Sorry for the double post :roll:

So, I was being taught impersonal verbs that express feelings in Latin, and it was something like:
ACC of the Experiencer, GEN of the Content (eg. I regret my action; me (ACC) actionis (GEN) piget)
Which should (?) be fine

But:
How would such a sentence look like with a Stimulus? "He made me regret my action"?
There's not a clear cut between Stimulus and Content in my head.

Another doubt:
Is also the de+ABL (I don't even know how English grammar calls it lol) a Content? (eg I spoke about justice; de iustitia locutus sum) Or does it stand on its own?
Native: :ita: | Intermediate: :eng: | Basic: :lat: :esp:
Studied: :qya: (+all of the above)
Willing to study: :grc: :jpn: :heb: :rus:

(Linguistic noob, fear not to correct me)

Curlyjimsam
sinic
sinic
Posts: 253
Joined: 01 Sep 2010 14:31
Location: UK

Re: Thematic relations

Post by Curlyjimsam » 31 Mar 2017 20:05

The problem with thematic relations is that a lot of people have had ideas about how they ought to be described, but there isn't a great deal of general agreement. Even amongst the most generally used terms (like "Agent", "Patient" or "Theme", "Experiencer" ...) there's still plenty of room for debate about what exactly they cover, whether we need all of them, whether we need others as well etc.

A better starting point might be to look at actual languages where cases are "thematically related" in some way and consider the ways in which they divide verbs up. "Split-S" or "active" languages are the most obvious type of language of this sort. I'm not sure there are any good free resources on these, but the article Active/agentive case marking and its motivations by Marianne Mithun is a good introduction to the basis of the case split in a range of split-S languages, if you can get hold of it. The book The Typology of Semantic Alignment is also very good, and goes into lots and lots of detail on lots and lots of languages.
Twitter: @jsbaker750
Website: seven-fifty.net

Post Reply