Ulcha language

based on Tsintsius 1982, Schmidt 1928, Lidija Sem 1986, Poppe 1958,  Sunik 1958, Sunik 1949, Awrorin 1951, Doerfer 1956, Miščenko 2014, Zhang 1998, An 1986, Wasilewič 1949, Gerasimowa 2002, Wseurkajinskyj perepys naselennja 2001, Wserossijskaja perepis'naselenija 2010

Starostin memorial readings XV, 22.10.20. Traditionally in Tungus-Manchu linguistic studies there are a few questions left unanswered: what is exact genetic clusterisation of the idioms; what are the most reliable isoglosses to clusterise the most of the know TM idioms; and how each intermediary proto-language could look like

The paper contains a description of the variety of Russian used by bilingual speakers of Southern Tungusic languages (some Nanai dialects and Ulch). The morphosyntactic contact-induced peculiarities of their speech are the focus of this paper, while cases of phonetic and lexical interference are discussed in less detail. The study is based on the data of a small oral corpus provided with a manual annotation of contact-induced features. One of the practical aims of the study is to test and improve the system of annotation used in the corpus.

The paper deals with marginal surprising uses of the impersonal construction attested in Nanai and Ulch (Southern Tungusic, the Russian Far East). In these languages, a dedicated R-impersonal construction with no overt subject develops secondary marginal non-impersonal uses with the first person pronoun in the subject position. The main question of the paper is how to explain such uses in a diachronic perspective. I propose the following hypothetical scenario. Initially, the construction under consideration is true impersonal with the main function of deleting or defocusing of the subject. Then, it develops additional modal and aspectual semantics. The modal or aspectual function becomes more foregrounded, than the valency-changing one, and the next step of this reinterpretation from a valency-changing device to a modal one is the possibility to have an overt subject.
IMPS > IMPS + (ASPECT/MODALITY) > ASPECT/MODALITY (+IMPS) > ASPECT/MODALITY (uses with 1st person pronouns).
The special status of the first person subject seems to be motivated by its strong predisposition to the modal and habitual contexts, covered by this construction (cf. 'it is necessary to do > I need to do'). In the paper, this hypothesis is verified on the text data of Ulch.

The paper deals with the morpheme kə̄n (~qān) in Ulch (Tungusic, Khabarovsk Krai). This morpheme has the limitative (restrictive) meaning (‘only’). The morphosyntactic features of kə̄n are quite unusual. In some uses it occupies the slot of derivational suffixes, in some others it behaves rather as a postfix or an enclitic. Similar cases of affix reordering in different languages were described by M. Haspelmath in terms of the “externalization of inflection”. M. Haspelmath proposes for them the diachronic path AINFL+B > B+AINFL motivated by the general semantically based “inflection-outside-derivation principle”. One of Haspelmath’s predictions implied by this principle is that the diachronic process in question is unidirectional, i.e. the reverse change is impossible. The data of Ulch, however, that “postfixal” uses of kə̄n are rather innovative, i. e. the ongoing change kə̄n+INFL > INFL+kə̄n (internalization of inflection?) is attested.

The paper deals with a construction with the verb ‘finish’ attested in Ulch (Tungusic, Khabarovsk Krai). The verb xod- ‘to finish’ can be used in Ulch a) with “simultaneous” converb (-mi~-məri), b) with “non-simultaneous” converb (-rə). The first construction has the regular phasal meaning ‘to finish V’; the second one, which will be discussed here, refers to the end of the event-sequence, not to the end of the event itself (‘to conclude by V’).
It is structurally and semantically similar to English to end up V-ing or Spanish acabar-construction (cf. Olbertz 1998). The hod-construction is not unique for Ulch: it is used e.g. in Dzhuen Nanai. In Ulch the construction reveals a high degree of grammaticalization and it can be considered as a full member of the grammatical system.
The semantics of the construction includes a clear nuance of counter-expectation. It signals that the event V a) is the last event in the sequence and b) it breaks somehow the expected event-sequence.
A more general question will be discussed on the data of Ulch. How do such markers and meanings covered by them fit into the pool of existing crosslinguistic notions (“comparative concepts”, Haspelmath 2010)? 1) They can be treated as an extension of the phasal semantic domain (‘to begin, to continue, to finish V’, cf. e.g. Engerer 2014), cf. the label “outer phase” proposed in (Plungian 1999). 2) They are close to “phasal polarity” items (such as already, still, yet), which are also described in terms of (counter)-expectation, cf. (van Baar 1997, Auwera 1998). 3) They form a semantic pair to markers referred as inconsequentials (Haiman 1988); frustratives (e.g. Sparing-Chávez 2003), avertives (Kuteva 1998): ‘V in vain’, ‘V (and then anti-V)’, ‘almost V (but however did not V)’, cf. the overview of the whole semantic cluster under the label “counter-to-fact TAM categories” in Kuteva et al. (2015). 4) They are in some respect close to adversative connectives (such as English but), cf. very similar cases in the study on adversatives (Malchukov 2004).
The data on the Ulch ‘finish’-construction involved in the study are of three types: elicited data, text data from our own field records collection, text data from published texts (Sunik 1985).

The paper deals with the restrictive (limitative) marker kə̄n 'only' in Ulch (Southern Tungusic).  This marker has untrivial positional features: it can attach before inflectional suffixes (as a derivational affix) or after them (as a postfix or enclitic). One might postulate here the process of affix reordering "infection + morpheme X > morpheme X + inflection" described in Haspelmath (1993) as "externalization of inflection". However, there is an evidence that external uses of kə̄n are innovative compared to internal ones, not vice versa, and this direction of diachronic development is very unexpected.