Switch Reference in Mongolian
Contact

2. Outline of Switch-Reference in Mongolian

The following outline is entirely based on the works mentioned. I am particularly endebted to Tserenpil-Kullmann and Bayarmaa, which are the source of a good proportion of the example sentences in this post. (Many but not all sentences are so marked.)

TL;DR:

Same-subject marking between main and subordinate clauses is obligatory through use of the reflexive suffix in:

  • ALL subordinate clauses ending in a verb in an oblique case, except where blocked by a very small number of main-clause verbs.
  • MOST subordinate clauses with subordinating converbs — although some coordinating converbs don't require the reflexive suffix to mark 'same subject'.
  • SOME subordinate clauses featuring so-called postpositions, depending on the postposition.

Differential subject marking assists in distinguishing the subject of the subordinate clause from that of the main clause.

Coordinating converbs do not take the reflexive suffix. 'Same subject' over strings of coordinate clauses is partly dependent on the converb used.

Failure to apply SR when required results in ungrammatical sentences.

The following is a characterisation of SR in Mongolian.

1. SR in Mongolian operates biclausally between a subordinate clause (the marked clause) and a main clause (the reference clause). It serves to identify the subject of the subordinate clause as being the same as or different from that of the main clause. The subordinate clause may be either a daughter clause (core argument of the sentence) or an adsubordinate clause (e.g., adverbial clause).

SR in the sense that Jacobsen defined it (obligatorily indicated by a morpheme) does not operate over cosubordinate constructions (clause chains), although a couple of converbs are related to SR.

2. SR in Mongolian involves two separate grammatical systems.

a) Possessive markers (suffixes / particles) are the most consistently observable markers of SR.

A reflexive-possessive suffix (hereinafter 'reflexive suffix') on a subordinate clause indicates that the clause retains the subject of the main clause, i.e., 'same-subject'. The subject of the subordinate clause is unexpressed.

If there is no reflexive suffix, the subject is different from that of the main clause. Non-reflexive possessive particles, which are in a complementary distribution with the reflexive suffix, explicitly indicate 'different subject' but are not always present.

Although the reflexive suffix is the most consistent and visible marker of SR, it is not always available to mark 'same subject', either because it is not applicable grammatically or because it is blocked by a small number of verbs in the main clause.

b) Differential subject marking in a subordinate clause, whereby the subject of the clause bears a case other than the Nominative (either the Genitive or Accusative), plays a role in signalling a switch of subjects. When the subject of a subordinate clause bears Accusative or Genitive case, it contrasts with the Nominative subject of the main clause, thus indicating 'different subject'.

Differential subject marking in subordinate clauses has two features (Guntsetseg 2016):

While Accusative marking can serve to highlight the difference between the subject of the subordinate clause and that of the main clause, it is not a definitive marker of 'different subject'. Differential subject marking must be interpreted within the grammatical context. For example, the subject of the subordinate clause may be unexpressed, the subordinate clause may contain an ordinary Accusative, or the main clause may have a fronted Accusative.

Leaving aside certain gaps and exceptions, the general operation of SR in Mongolian may be schematically represented as follows:

Same/
different subject
Subject of main clauseSubject of subordinate clauseReflexive/
possessive marking
SAME SUBJECTNominative(Omitted)Reflexive suffix
DIFFERENT SUBJECTNominativeDifferential subject markingNo marking OR
Possessive Particle

3. A requirement for 'same subject' or 'different subject' may be associated with certain converbs in both ad-subordinate and co-subordinate clauses, but this must be regarded as a specific grammatical property of the converbs in question.


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!