10. The Subject in Subordinate Clauses
10.1.1. Same subject
In cases where the subject of the verb in the subordinate clause is the same as that in the main clause, the subject is always omitted in the subordinate clause.
10.1.2. Different subject
Where the subject of a subordinate clause is different from that of the main clause, a phenomenon known as 'differential subject marking' comes into play. The following explanation, including examples, is based on Guntsetseg 2016, Differential Case Marking in Mongolian.
For a daughter clause ending in a verbal noun, that is, a clause that occupies a core role within the sentence, three cases are possible on the subject of the subordinate clause: Nominative, Accusative, or Genitive.
Tujaa 〖Dorž / Dorž.ACC / Dorž.GEN Germany.DIR go.PST.ACC〗 know.PST
Tuyaa knew that 〖Dorj went to Germany〗.
ᠲᠤᠶᠠᠭᠠ ᠳᠣᠷᠵᠢ / ᠳᠣᠷᠵᠢ ᠵᠢ ᠭᠧᠷᠮᠠᠨ
ᠤᠷᠤᠭᠤ ᠶᠠᠪᠤᠭᠰᠠᠨ ᠢ ᠮᠡᠳᠡᠭᠰᠡᠨ᠃᠃
Туяа Дорж / Доржийг / Доржийн Гэрман руу явсныг мэдсэн.
Guntsetseg considers the Genitive to be due to the functioning of the verb form (verbal noun) as a noun. In other environments, such as converbial clauses or a clauses ending in gež, only the Nominative and Accusative are possible. Nevertheless, in such clauses the Genitive does serve to indicate that the subject of the subordinate clause differs from that of the main clause.
Guntsetseg's analysis concentrates on subordinate clauses that allow only the Nominative or Accusative subject marking. The following is a subordinate clause ending in a postposition:
Tujaa 〖Dorž / Dorž.ACC come.PST.GEN after〗 go.PST
Tuyaa went 〖after Dorž came〗.
ᠲᠤᠶᠠᠭᠠ ᠳᠣᠷᠵᠢ / ᠳᠣᠷᠵᠢ ᠵᠢ ᠢᠷᠡᠭᠰᠡᠨ ᠦ
ᠳᠠᠷᠠᠭᠠ ᠶᠠᠪᠤᠭᠰᠠᠨ᠃
Tуяа Дорж / Доржыг ирсний дараа явсан.
This also applies to clauses ending in the quotative gež:
Tujaa 〖Dorž / Dorž.ACC Germany.DIR go.PST〗 that know.PST
Tuyaa knew 〖that Dorj went to Germany〗.
ᠲᠤᠶᠠᠭᠠ ᠳᠣᠷᠵᠢ / ᠳᠣᠷᠵᠢ ᠵᠢ ᠭᠧᠷᠮᠠᠨ
ᠤᠷᠤᠭᠤ ᠶᠠᠪᠤᠭᠰᠠᠨ ᠭᠡᠵᠦ ᠮᠡᠳᠡᠭᠰᠡᠨ᠃
Tуяа Дорж / Доржийг Герман руу явсан гэж мэдсэн.
similarly for ad-subordinative clauses (adverbial clauses) with subordinating converbs:
Tuyaa went 〖as soon as Dorž came〗.
ᠲᠤᠶᠠᠭᠠ ᠳᠣᠷᠵᠢ / ᠳᠣᠷᠵ ᠵᠢ
ᠢᠷᠡᠮᠡᠭᠴᠡ ᠶᠠᠪᠤᠭᠰᠠᠨ᠃
Tуяа Дорж / Доржыг ирмэгц явсан.
This also applies in the case of the subordinating converb -val/-bal:
〖If Tuyaa comes〗 I will not come.
ᠪᠢ ᠲᠤᠶᠠᠭᠠ / ᠲᠤᠶᠠᠭᠠᠠ ᠵᠢ
ᠢᠷᠡᠪᠡᠯ ᠢᠷᠡᠬᠦ ᠦᠭᠡᠢ᠃
Би Tуяа / Туяааг ирвэл ирэхгүй.
Guntsetseg identifies two factors triggering differential subject marking in subordinate clauses.
The first is adjacency of the subject of the subordinate clause to the subject of the main clause. According to this factor, if the subject of the subordinate clause is immediately adjacent to the subject of the main clause, there is a strong tendency for the subject of the subordinate clause to take the Accusative case, thus serving to differentiate it from the subject of the main clause which is in the Nominative case.
1sg 〖?2sg / 2sg.ACC Germany.DIR go.PST.ACC〗 know.PST
I knew 〖that you went to Germany〗.
ᠪᠢ ?ᠴᠢ / ᠴᠢᠮᠠᠶᠢ ᠭᠧᠷᠮᠠᠨ ᠤᠷᠤᠭᠤ
ᠶᠠᠪᠤᠭᠰᠠᠨ ᠢ ᠮᠡᠳᠡᠭᠰᠡᠨ᠃
Би чамайг Герман руу явсныг мэдсэн.
If the subject of the main clause is not adjacent to the subject of the subordinate clause, the tendency to use the Accusative case diminishes:
〖2sg / 2sg.ACC Germany.DIR go.PST.ACC〗 1sg know.PST
I knew 〖that you went to Germany〗.
ᠴᠢ / ᠴᠢᠮᠠᠶᠢ ᠭᠧᠷᠮᠠᠨ ᠤᠷᠤᠭᠤ
ᠶᠠᠪᠤᠭᠰᠠᠨ ᠢ ᠮᠡᠳᠡᠭᠰᠡᠨ᠃
Чи / чамайг Герман руу явсаныг би мэдсэн.
The second factor triggering differential subject marking is what Guntsetseg calls 'Referentiality'. The tendency for the subject of the subordinate clause to feature differential subject marking is strongest for personal pronouns. This tendency diminishes along a 'Referentiality scale', from personal names (still strong but somewhat weaker) to definite noun phrases (either Accusative or Nominative are possible) to indefinite noun phrases (Nominative is preferred) to 'very weak indefinite noun phrases' (Nominative only allowed).
Guntsetseg gives as an example of the latter in the following sentence:
〖who one person / who one person.*ACC house.DAT 3POSS thieving do.PST.ACC〗 Tujaa know.PST
Tuyaa knew 〖that someone had committed a theft at her home〗.
ᠬᠡᠨ ᠨᠢᠭᠡ ᠬᠦᠮᠦᠨ / ᠬᠦᠮᠦᠨ ᠢ ᠭᠡᠷ ᠲᠤ ᠨᠢ
ᠬᠤᠯᠠᠭᠠᠢ ᠬᠢᠭᠰᠡᠨ ᠢ ᠲᠤᠶᠠᠭᠠ ᠮᠡᠳᠡᠭᠰᠡᠨ᠃
Хэн нэг хүн / *хүнийг гэрт нь хулгай хийснийг Туяа мэдсэн.
The implications for SR are that differential subject marking in the subordinate clause plays a key role in differentiating it from the subject of the main clause ('different subject'), especially where no reflexive marking is available. The tendency to use differential subject marking is strongest when a) the subject of the subordinate clause is adjacent to that of the main clause and b) the subject of the subordinate clause is a personal pronoun and, to a somewhat lesser degree, a personal name.
It is possible for a subordinate clause to contain two instances of the Accusative case: one a normal Accusative, the other a result of direct subject marking. However, Guntsetseg also notes that the use of differential subject marking to highlight the 'different subject' relationship tends to have a suppressant effect on the ordinary Accusative.
Guntsetseg speculates that the requirement for differential subject marking on the subjects of subordinate clauses is becoming stronger, spreading from adjacent positions to non-adjacent positions and moving further along the 'Referentiality scale'. This suggests that it is assuming greater importance as a marker of SR.
10.2. Interpreting the Subject
Unlike its neighbour Chinese, a highly topic-prominent language in which the notion of 'subject' is problematic, Mongolian is a language in which identification of the grammatical subject is of paramount importance.
SR plays a role in this since the identification of 'same subject' involves a rigid definition of grammatical subject. For instance, in the following sentence, the grammatical subject of the main sentence is ödör 'day'; that of the subordinate clause is tüünij bij 'his body'. Hence the use of tusam rather than tusmaa.
〖day pass.FUT as〗 he/she.GEN body improve.SIMUL be.FUT
〖The more days go by〗 the better he is feeling.
ᠡᠳᠦᠷ ᠥᠩᠭᠡᠷᠡᠬᠦ ᠲᠤᠰᠤᠮ
ᠲᠡᠡᠭᠦᠨ ᠦ ᠪᠡᠶᠡ ᠳᠡᠭᠡᠷᠡᠳᠡᠵᠦ ᠪᠠᠢᠨᠠ᠃
Өдөр өнгөрөх тусам түүний бие дээрдэж байна.
In the written language, at least, parts of the body must be treated as different subjects from the owner, as in:
〖When (he) gets up quickly〗 his head spins.
ᠬᠤᠷᠳᠤᠨ ᠪᠣᠰᠬᠤᠤᠯᠠᠷ ᠲᠡᠭᠦᠨ ᠦ
ᠲᠣᠯᠤᠭᠠᠢ ᠡᠷᠭᠢᠳᠡᠭ᠃
Хурдан босохлоор түүний толгой эргэдэг.
In the spoken language, however, this restriction may be relaxed.
〖quickly get-up.WHEN.REFL〗 he/she.GEN head spin.HAB
〖When (he) gets up quickly〗 his head spins.
ᠬᠤᠷᠳᠤᠨ ᠪᠣᠰᠬᠤᠤᠯᠠᠷ ᠵᠢᠠᠨ ᠲᠡᠭᠦᠨ ᠦ
ᠲᠣᠯᠤᠭᠠᠢ ᠡᠷᠭᠢᠳᠡᠭ᠃
Хурдан босохлоороо түүний толгой эргэдэг.
Mongolian requires awareness at all times of the subject of the sentence, even if it is not explicitly mentioned. The following sentence is a simple illustration of this. The subject of the main clause is xösög 'mounts/animals'. The subject of the long subordinate clause is mal.čin ajl 'herding household', which is not explicitly mentioned in this sentence but appeared in the previous sentence. Xelengüüt lacks the reflexive suffix because subject of the subordinate clause ('herding household') and that of the main clause ('mounts') are different.
Within the subordinate clause, the subordinate clause nüüx-d-ee 'when move' carries the reflexive suffix because the subject of nüüx 'move', mal.čin ajl 'herding household', is the same as that of the entire subordinate clause.
〖(herding household)〖move.FUT.DAT.REFL〗 dog.GEN.REFL trough.ACC felt.DAT wrap.N last.GEN camel.DAT load.ANT "last.čijn camel.GEN load respect" say-SIMUL say.MOM〗 mount move.HAB EXCL.
〖(Herding households) 〖when moving〗 wrap their dog's trough in felt and load it on the last camel, and the moment (they) say: "The last camel's load is the honour"〗, the mounts set out.
(ᠮᠠᠯᠴᠢᠨ ᠠᠢᠯ) ᠨᠡᠭᠦᠬᠦ ᠳᠦ ᠪᠠᠨ
ᠨᠣᠬᠠᠢᠨ ᠦ ᠪᠡᠨ ᠢᠳᠡᠭᠦᠷᠢ ᠵᠢ
ᠢᠰᠡᠭᠡᠢ ᠳᠦ ᠪᠣᠭᠤᠨ ᠰᠡᠭᠦᠯᠴᠢ ᠵᠢᠨ
ᠲᠡᠮᠡᠭᠡᠨ ᠳᠦ ᠠᠴᠢᠶᠠᠯᠠᠭᠠᠳ᠄ "ᠰᠡᠭᠦᠯᠴᠢ ᠵᠢᠨ
ᠲᠡᠮᠡᠭᠡᠨ ᠦ ᠠᠴᠢᠶᠠ ᠬᠦᠨᠳᠦ ᠪᠡᠨ"
ᠭᠡᠵᠦ ᠬᠡᠯᠡᠩᠭᠦᠲᠡ ᠬᠥᠰᠬᠡ ᠬᠥᠳᠡᠯᠳᠡᠭ ᠠᠪᠠᠢ᠃
(Малчин айл) Нүүхдээ нохойныхоо идүүрийг эсгийд боон сүүлчийн тээмээнд ачаалаад: "Сүүлчийн тэмээний ачаа хүндээ" гэж хэлэнгүүт хөсөг хөдөлдөг авай.
The following sentence describes the death of a young, impulsive wild goat who butted his horn so that it grew inwards. At the very end of the sentence is a simile for the goat's sudden realisation of the result of his impulsiveness, comparing the departure of this fleeting thought to a bird startled by someone sitting down. The main clause of this segment is šuvuu ürgen nisex 'bird is startled and flies'. Because the subordinating converb suunguut 'the moment sits down' does not have a reflexive suffix here, the action of sitting cannot be the bird's and can only be that of someone else.
〖〖(horn) brain.GEN brown envelope pierce.FUT moment.DAT just〗〖 "1sg.GEN horn wrongly grow-PST" say.PST〗 sudden realisation wild-goat.GEN head.DAT one come-in.SIMUL come.ANT forever disappear.PST 3POSS〗 〖bird 〖(someone) sit.MOM〗 be-startled.n fly.FUT.GEN〗 same be.PST
〖The instant (the horn) pierced the envelope of the brain〗, 〖the sudden realisation entered the wild goat's brain 〖that "my horn grew wrong"〗, and disappeared forever〗, was 〖like a bird startled and flying away 〖the moment (someone) sits down〗.
ᠲᠠᠷᠢᠬᠢᠨ ᠦ ᠪᠣᠷᠤ ᠬᠠᠯᠢᠰᠤ ᠴᠣᠭᠤᠷᠠᠬᠤ
ᠬᠣᠷᠤᠮ ᠳᠤ ᠯᠠ "ᠮᠢᠨᠦ ᠡᠪᠡᠷ ᠪᠤᠷᠤᠭᠠ
ᠤᠷᠭᠤᠵᠡᠢ" ᠭᠡᠭᠰᠡᠨ ᠲᠠᠮ ᠲᠤᠮᠬᠠᠨ ᠲᠠᠮᠤ
ᠲᠤᠮᠠᠬᠠᠨ ᠤᠬᠠᠭᠠᠷᠠᠯ ᠶᠠᠩᠭᠢᠷ ᠦᠨ
ᠲᠣᠯᠤᠭᠠᠢ ᠳᠦ ᠨᠢᠭᠡ ᠣᠷᠤᠵᠤ ᠢᠷᠡᠭᠡᠳ ᠡᠭᠦᠷᠢᠳᠡ
ᠵᠠᠮᠬᠠᠷᠠᠭᠰᠠᠨ ᠨᠢ ᠰᠢᠪᠠᠭᠤ ᠰᠠᠭᠤᠩᠭᠤᠲᠠ
ᠦᠷᠭᠦᠨ ᠨᠢᠰᠬᠦ ᠵᠢᠨ ᠠᠳᠢᠯ ᠪᠥᠯᠦᠭᠡ᠃
Тархины бор хальс цоорох хоромд л "Миний эвэр буруу ургажээ" гэсэн там тумхан ухаарал янгирын толгойд нэг орж ирээд үүрд замхарсан нь шувуу суунгуут үргэн нисэхийн адил бөлгөө.
2. Outline of Switch-Reference in Mongolian
3. Possessive Forms in Mongolian; their role in SR
- 4.1. Verb Forms that take Case Endings
4.2. Daughter clauses
4.3. Other Predicate Forms
4.4. Gedeg (Complementiser)
5. SR in Ad-subordinate (Adverbial) Clauses
6. Ad-subordinate Clauses with Postpositions
- 6.1. The reflexive attaches directly to the postposition
6.2 The reflexive suffix attaches to the verb form preceding the postposition
7. Verbs that Block the Reflexive Suffix
10. The Subject in Subordinate Clauses
- 10.1.1. Same subject
10.1.2. Different subject (differential subject marking)
10.2. Interpreting the Subject
Sorry, I'm now using Disqus for comments. If you'd prefer not to use Disqus, please send me an email and I'll list your comments separately. Thanks!