adnominal-person morphology
4.5. Adnominal syntax and secondary declension
This section describes the morphological phenomena involved in Erzya secondary de-clension. The discussion developed sets out to illustrate that, whereas various
modi- ers can become main items in an NP when the contextually predictable head noun is dropped, there are two basic strategies for marking MWN (modi ers without nouns): (i)
ZERO marking, and (ii) SOD (Speaker-Oriented Demonstrative) marking. Grammars of Erzya present both of these strategies to different degrees, but usually their treatment of
MWN is delivered in several separate sections with no connections drawn. The treatment of genitive-case personal pronouns has been associated with SOD strategies in Evsev'ev's grammar of Erzya (1963[1928/29]), whereas Agafonova (2000) presents personal pro-nouns and re exive/intensive pronouns with ZERO marking. Upon closer scrutiny it be-comes apparent that Agafonova's re exive/intensive pronoun charts include members from two different paradigms; the re exive/intensive paradigms are intermingled with genitive + SOD strategies. (Cf. GMYa 1962 I: 232; GMYa 1980: 267; Mosin & Bajushkin 1983: 116; Pall 1996: 18-19; Zaicz 2006) Thus after presenting background information on secondary declension, and demonstrating that both marking strategies are attested with locative modi ers, I provide an overview on the compatibility of various modi er types with secondary declension. And this I follow up with a morphological inspection of the genitive-form personal and re exive/intensive pronouns as rendered in MWN or secondary-declension forms.
4.5.1. Background
In my article ON QUANTIFICATIONINTHE ERZYA LANGUAGE (Rueter, forthcoming), I have noted that Erzya nominal-syntax structure entails symmetric marking of case. Case marking in turn requires the choice of one declension type from a selection of three, whereupon it is rendered with postposed orientation on the phrase- nal head, or in the absence of this constituent, on the nal constituent of the phrase. Thus, in addition to the simple noun phrase consisting of only a head noun, the Erzya NP can also be represented by numerous combinations of premodi ers with and without an NP head. First let us examine the simple head-noun NPs and NPs with single modi er, see (65).
(65) kudo+ś
house/home_N+NOM.DEF.SG
‘the/that/this house/home’
Head nouns can be preceded by single modi ers representing adjectives, quanti ers, spatial modi ers and determiners.
Adjective + Noun + Cx (66) pokš kudo+ś
big_A.ABS house/home_N+NOM.DEF.SG
‘the/that/this big house’
(67) jakśt́eŕe kudo+ś
red_A.ABS house/home_N+NOM.DEF.SG
‘the/that/this red house’
Quantifying modi er + Noun + Cx (68) źaro vina+ś
that-much_Q.ABS liquor_N+NOM.DEF.SG
‘the much liquor’
Spatial modi er + Noun + Cx (69) oš+so kudo+ś
town_N+INE house/home_N+NOM.DEF.SG
‘the/that/this house in town’
Determiner + Noun + Cx (70) iśt́amo kudo+ś
such_PRO-DEF.ABS house/home_N+NOM.DEF.SG
‘such a house’
(71) t́e kudo+ś
this_PRO-DEM-PROX.ABS house/home_N+NOM.DEF.SG
‘this house’
(72) ńe kudo+t́+ńe
this_PRO-DEM-PROX.ABS house/home_N+PL+DEF.PL.NOM
‘these houses’
Genitive modi er + Noun + Cx (73) a. vańa+ń kudo+ś
Vanya_PRP+GEN house/home_N+NOM.DEF.SG
‘Vanya's house’
b. čuvto+ń kudo+ś
wood_N+GEN house/home_N+NOM.DEF.SG
‘the/that/this wooden house’
c. sonze kudo+ś
he/she/it_PRON-PERS-3SG.GEN.POSS-3SG house/home_N+NOM.DEF.SG
‘the/that/this wooden house’
In examples (66–73) we can observe simple modi er structures in:
NP = N, A + N, Q + N, SPATIAL + N, DET + N and GEN-ATTR + N
Alternate ordering of head and modi er(s) will induce the addressee to perceive a com-plete sentence, although in context NP looking sequences can also be interpreted as com-plete sentences. Hence the upper-case letters in (76–77) indicate non-neutral, perhaps focus, predicate position.
(74) kudo+ś pokš
house/home_N+NOM.DEF.SG big_A.NOM.SG
‘the/that/this house [is] big.’
(75) kudo+ś oš+so house/home_N+NOM.DEF.SG town_N+INE
‘the/that/this house [is] in town.’
(76) POKŠ kudo+ś
big_A.NOM.SG house/home_N+NOM.DEF.SG
‘[now] this house is BIG’
(77) OŠ+SO kudo+ś
town_N+INE house/home_N+NOM.DEF.SG
‘[now] this house in IN TOWN’
Let us now observe a combination including both spatial and adjectival modi ers. Here it will be observed that the adjective directly precedes the NP head, and that the spatial modi er is also preposed in Erzya, preceding the adjective. The inessive marking in the locative modi er is indicative of an inessive NPošso ‘in town’ that premodi es the noun phrase pokš kudoś ‘the big house’.
(78) oš+so pokš kudo+ś
town_N+INE big_A.ABS house/home_N+NOM.DEF.SG
‘the/that/this big house in town’
Noun Phrase Constituent Ordering
The basic constituent ordering in the noun phrase can be outlined as determiner + quanti er + adjective + noun. This outline may appear oversimpli ed, but it seems to address a large portion of noun phrases in the Erzya corpora. Thus symmetry in NP
expansion strategies goes generally uncompromised, i.e. the modi ed NPs can be further modi ed with quanti ers, determiners and even NPs or adpositional phrases in modi er-case forms (e.g. GEN, INE, ELA, PROL, COMP, ABE, TRNSL, LOC). Adjectives can co-occur with quanti ers or determiners or both, see (79–81).
(79) a. kavto pokš kudo+so
two_NUM-CARD.ABS big_A.ABS house/home_N+INE
‘in two big houses’
b. kavto pokš kudo+t́+ńe+se
two_NUM-CARD.ABS big_A.ABS house/home_N+PL+DEF.PL+INE
‘in the/those/these two big houses’
(80) a. t́e pokš kudo+so+ńt́
this_PRON-DET.ABS big_A.ABS house/home_N+INE+DEF.SG
‘in this big house’
b. ńe pokš kudo+t́+ńe+se
these_PRON-DET-PL.ABS big_A.ABS house/home_N+PL+DEF.PL+INE
‘in these big houses’
*(81) a. t́e kavto pokš kudo+so+ńt́
this_PRON-DET-SG.ABS two_NUM-CARD.ABS big_A.ABS house/home_N+INE+DEF.SG
‘in the/that/this two big house’
b. ńe kavto pokš kudo+t́+ńe+se
these_PRON-DET-PL.ABS two_NUM-CARD.ABS big_A.ABS house/home_N+PL+DEF.PL+INE
‘in the/those/these two big houses’
In the examples above the grammatical category of number has an in uence on the dis-tribution of determiners, quanti ers and the remainder of the NP. A numeral (two and above) can co-occur with a head in an inde nite declension form, or it can appear with a head in a de nite declension form, i.e. the NPkavto kudoso ‘in two houses’ differs from the NPkavto kudot́ńese ‘in the two houses’ in matters of de niteness. Since de niteness is an entailment of demonstrative pronouns, it will be noted that quanti ers indicating numbers larger than one can only co-occur with the plural demonstrative pronoun ńe
‘these (anaphoric)’ and not its singular counterpart t́e ‘this’, see (81).
Another quali cation of constituent order addresses the genitive attributes, usual-ly indicating material and spatio-temporal source, purpose, and meronymy. Both adjec-tives and genitive attributes can be used separately as modi ers, but when they co-occur, the symmetric strategy assists greatly in disambiguation, see (82–85), i.e. the ordering kośke tumoń ‘dry oak’ sets off a premodifying genitive attribute NP to peŋgt́ ‘ re-wood’
while the reverse ordering tumoń kośke ‘of oak, dry’ indicates that the head has two modi ers – an adjective kośke ‘dry’ and a preceding genitive attribute tumoń ‘of oak’
NP. Let it suf ce that we observe the following examples, derived from (Kolyadyonkov 1940: 24, 52; Bartens 1999: 111).
(82) tumo+ń peŋg+t́
oak_N+GEN re-wood_N+ PL.NOM
‘oak(en) re-wood’
(83) kośke peŋg+t́
dry_A.ABS re-wood_N+ PL.NOM
‘dry re-wood’
(84) kośke tumo+ń peŋg+t́
dry_A.ABS oak_N+GEN re-wood_N+PL.NOM
‘ re-wood cut from dry oak’
(85) tumo+ń kośke peŋg+t́
oak_N+GEN dry_A.ABS re-wood_N+ PL.NOM
‘dry oak re-wood’
In a similar vein we can attest other NP modi ers and their placement before the head noun in examples from Mikhail Bryzhinski, see (86–87).
(86) kečaj son+ś+kak eź
Kechai_N-PRP.NOM.SG himself_PRON-PERS-3SG+REFL+CLT not_V-NEG-PRETI.PRED-3SG soda, ko+v eskeĺ+i
know_V.CONNEG where_PRON-INTER-SPATIAL+LAT stride_V+IND.PRES.PRED-3SG viŕ+ga+ńt́, seĚej diࡓ vete -kotó
forest_N+PROL+DEF.SG, tall_A.ABS and_CONJ fi ve_NUM-CARD.ABS -six_NUM-CARD.ABS seĎ+eĔ eþkelma+so Ğado ije+Ĕ
fathom_N+GEN thickness_N+INE hundred_NUM-CARD.ABS year_N+GEN þuvt+t+ne+Ĕ jutko+va
tree_N+PL+DET.PL+GEN among/between_POP+PROL
(Bryzhinski, M. Kirdazht manuscript) ‘Even Kechai himself didn’t know where he was walking through the forest, among the tall trees ve [or] six fathoms around and hundreds of years old.’
(87) kolmo či͔+t́ ti͔ńeŋk three_NUM-CARD.ABS day_N+PL.NOM you_PRON-PERS-2PL.DAT.POSS-2PL eź pečt́avo alaša+Ĕ pulo+Ĕ not_V-NEG-PRETI.PRED-3SG cross_V.CONNEG horse_N+GEN tail_N+GEN keĎe+se te Ďej+eĞ́ !
width_N+INE this_PRON-DEM.ABS river_N+NOM.DET.SG
(Bryzhinski, M. 1983: 90) ‘For three days you have not managed to cross this river that is the breadth of a horse's tail!’
Hence we can assume that in addition to the following ordering for NPs, there might also be room for double or triple embedding. Thus the simplex NP consists of possible determiners, quanti ers and adjectives, and a complex NP might consist of an NP
embedded in either an NP or adpositional phrase in one of the modi er cases. Evidence from Bryzhinski's texts indicates even more complexity, see below.
Simple NP = (DETERMINER) (QUANTIFIER) (ADJECTIVE) NOUN NP with single embedding
= NP [GEN | INE | ELA | PROL | COMP | ABE | TRNSL] + NP
= Adpositional phrase[INE | ELA | PROL | COMP | LOC] + NP NP with double embedding
vet́e%koto seĺ+eń ečkelma+so čuvto
= NP[NP[NP[Q N[GEN]] + N[INE]] + N]
alaša+ń pulo+ń keĺe+se t́e ĺej
= NP[NP[NP[NP[N [GEN]] + N[GEN]] + N[INE]] + DET N]
Symmetric case marking and head noun deletion
Case marking symmetry in Erzya, it must be stressed, is so persistent that the markers might also be viewed as enclitics. If, for example, the NP head is contextually predictable, it may also be deleted, whereupon the modi er closest to the NP- nal position becomes the new locus for case marking, see (88), and compare with (66–73), above. Other attestations of this phenomenon, known here as SECONDARY DECLENSION, can be found in (Evsev'ev 1963: 51, 101–103, 126, 129–132, 134–135, 162; Collinder 1969: 231;
Imaikina 1996: 27–32; Grebneva 2000: 107–108; Agafonova 2000: 139–141, 143–145;
Ermuškin 2004: 54; Keresztes 2005: 369–379; Zaicz 2006: 194–197 (who even mentions, without example, tertiary declension); Gil WALS feature/chapter 61 ADJECTIVESWITHOUT NOUNS.)
Adjective + Ø + Cx (88) a. pokš+oś
big_A.ABS+N.NOM.DEF.SG
(Bargova 1996: 68) ‘the/that/this big one’
b. jakśt́eŕe+ś
red_A.ABS+N.NOM.DEF.SG
(Lukyanov 1955: 9) ‘the/that/this red one’
Quantifying modi er + Ø + Cx c. źaro+ś
that-much_Q.ABS liquor_N+NOM.DEF.SG
(Abramov 1980: 18) ‘that much’
Spatial modi er + Ø + Cx d. oš+so+t́+ńe
town_N+INE+N.PL+ DEF.PL.NOM
(Abramov 1988: 359) ‘the/those/these ones in town’
Determiner + Ø + Cx e. iśt́amo+ś
such_PRO-DEF.ABS+N.NOM.DEF.SG
(Bryzhinski I. 1955: 74) ‘one such …’
*f. t́e+ś
this_PRO-DEM-PROX.ABS+N.NOM.DEF.SG
‘this one’
Genitive modi er + Ø + Cx
*g. vańa+ń+eś
Vanya_PRP+GEN+N.NOM.DEF.SG
‘Vanya's one’
*h. čuvto+ń+eś
wood_N+GEN+N.NOM.DEF.SG
‘the/that/this wooden one’
i. sonze+ś
he/she/it_PRON-PERS-3SG.GEN.POSS-3SG+N.NOM.DEF.SG
(Kirillov 1987: 74) ‘his/hers/its’
As can be observed in (88) predictable, de nite head-noun deletion is not an option attested for all NP types. While head deletion affords iśt́amo ‘such’ the role of de nite pronoun, an analogous solution is not available for the demonstrative pronoun t́e ‘this’.
The inde nite genitive modi ers, although unable to accommodate for this speci c va-riety of predictable head-noun deletion, have means to compensate, e.g. the genitive-form modi er fuses orthographically with the equivalent of the speaker-oriented (distal) demonstrative pronoun śe ‘that’ before undergoing declension, which is not always de-monstrative in type (cf. Ermuškin 2004: 57; Evsev'ev 1963: 126). (See also 89–92 and section 4.2.1.1. GENITIVE.)
Genitive modi er + Pron-dem-dist + Ø + Cx (89) t́išaj+eń+śe+ś
Tishai_PRP+GEN+PRON-DEM-DIST+N.NOM.DEF.SG
(Abramov 1989: 78) ‘the/that/this one of Tishai's’
(90) čuvto+ń+śe+ś
wood_N+GEN+PRON-DEM-DIST +N.NOM.DEF.SG
(Zhuravlov 1999: 119) ‘the/that/this [wooden one | one of wood]’
(91) ked́+eze ćora+ń+śe+d́e staka
hand_N+POSS-3SG>NOM.SG man_N+GEN+PRON-DEM-DIST+N.ABL heavy_A.NOM.SG
(Abramov 1987: 41) ‘her [Maryusha's] hand was heavier than that of a man's (Maryu-sha hit Vasya unexpectedly hard)’
(92) paŕak, ńe+t́ ĺed́śt́+ema+t́+ńe+d́e maybe_PRT, these_PRON-DEM-PROX+PL remember_V+N+PL+DEF.PL+ABL maŕav+i t́e+ń śokś+eń
feel_V+IND.PRES.PRED-3SG to_POP+POSS-1SG autumn_N+GEN pizeme+ś tundo+ń+śe+ks
rain_N+NOM.DEF.SG spring_N+GEN+PRON-DEM-DIST+N.TRNSL
(Chetvergov 1992: 91) ‘Maybe, it's these memories that make the autumn rain feel like spring [rain] to me’
Rueter (2003: 165–166) provides an extensive enumeration of semantic properties asso-ciated with the target of genitive marking including: material, place, time, purpose, indi-vidual-to-group and group-to-inferable-capacity. This collection of semantic properties attributed to the target of genitive marking can be augmented with that of the animate possessor, as noted by Evsev'ev (1963: 126), see (93).
(93) ki+ń šapka+ńt́ jomavt+i͔ŋk `
who_PRON-INTER+GEN cap_N+GEN.DEF.SG lose_V+IND.PRETI.PRED-2PL>3 ivan+oń+śe+ńt́ iĺi
Ivan_N-PRP+GEN+PRON-DEM-DISTAL+GEN.DEF.SG or_CONJ pet́a+ń+śe+ńt́?
Petya_N-PRP+GEN+PRON-DEM-DISTAL+GEN.DEF.SG
(Evsev'ev 1963: 126) ‘Whose cap did you lose: Ivan's or Petya's?’
With the addition of the possessor function, on the one hand, and the possibility of inde nite declension, on the other, we can establish the morphological inde nite genitive as a modi er phrase followed by a SODPRONOUN, which has a syntax-motivated parameter for overriding the demonstrative-declension requirement. Since the inde nite genitive modi er can be used with both referential and non-referential nouns, our next question is whether the genitive construction can be attested for personal pronouns, as well. In fact, Evsev'ev (1963: 162) describes possessive pronouns in the de nite declension with regular morphology that correlates directly to the SOD pronoun strategy attested in (93), compare table (4.65).
Table 4.65 Genitive-case personal pronouns with SOD secondary nominative forms or according to Evsev'ev the possessive pronouns in the de nite declension
P Semi
phonetic Orthographic
Morphologic Gloss
1 SG ɦɨɧɶɞɡɟɫɶ moń+śe+ś PRON-PERS-1SG.GEN+DEM-DIST+N.NOM.DEF.SG
PL ɦɢɧɟɤ-ɫɟɫɶ mińek+śe+ś PRON-PERS-1PL.GEN.POSS-1PL+DEM-DIST+N.NOM.DEF.SG
2 SG ɬɨɧɶɬɶ-ɰɟɫɶ tońt́+śe+ś PRON-PERS-2SG.GEN.POSS-2SG+DEM-DIST+N.NOM.DEF.SG PL ɬɵҥɤ-ɫɟɫɶ ti͔ŋk+śe+ś PRON-PERS-2PL.GEN.POSS-2PL+DEM-DIST+N.NOM.DEF.SG
3 SG ɫɨɧɡɷ-ɰɟɫɶ sonze+śe+ś PRON-PERS-2SG.GEN.POSS-2SG+DEM-DIST+N.NOM.DEF.SG PL ɫɵɧɫɬ-ɫɟɫɶ si͔nst+śe+ś PRON-PERS-2PL.GEN.POSS-2PL+DEM-DIST+N.NOM.DEF.SG
(Adapted from Evsev'ev 1963: 162) One peculiarity here, however, is that Evsev'ev does not provide a 3SG pronoun form corresponding to that of sonze+ś ‘his/hers/its’ as shown above in (88i).
Evsev'ev (1963: 101–103, 126, 129–132, 134–135, 162) deals with the phenom-enon of noun-head deletion in three separate instances. His rst mention of it addresses the variety in which words declined in the inde nite inessive, translative and compara-tive cases can be in ected a second time in the demonstrative declension, see Склонение определенных имен… ‘Declension of de nite nouns …’; this variety is observed in the inessive word form oš+so+ś town_N+INE+N.NOM.DEF.SG ‘the one in town’. The second mention introduces both the adjective-modi er ašo+ś white_A+N.NOM.DEF.SG ‘the white one’ and the inde nite-genitive modi er form čuvto+ń+śe+ś wood/tree_N+GEN+PRON
-DEM-DIST+N.NOM.DEF.SG ‘the wooden one, the one of wood’. In dealing with the latter Evsev'ev considers the segments +śe+ś to be a reduplicated ś element, a view held by some scholars even today. This interpretation might be countered. In regular declension of the ablative no linking vowel is present, whereas this derivation is regularly repre-sented in +śe+d́e+. The third mention of the phenomenon deals with genitive-form per-sonal pronouns ti͔ŋk+śe+ś you_PRON-PERS-2PL.GEN.POSS-2PL+PRON-DEM-DIST+N.NOM.DEF.SG
‘the/that/this one of yours’. Instead of repeating the idea of a reduplicated ś segment, Evsev'ev volunteers a vernacular Russian-language parallel in vaš%to where the Russian possessive pronoun vaš ‘your (2PL)’ is combined with the demonstrative particle to ‘that’
or et ‘this’ (cf. Lyons 1999: 48–49). This clari cation by Evsev'ev speaks in favor of the distal-demonstrative interpretation and can be supported with evidence in Erzya of other demonstratives used in post-genitive-modi er position, see t́e ‘this’ in (94) and śet́e
‘and/now this’ in (95). (The editors of MW have considered the 1SG form mońćit́ińt́ to be an analogy of the 2SGtońćit́ińt́. It is, but then it is not a genitive form of the re exive/
intensive 2SG pronoun with secondary genitive de nite singular declension, rather the genitive form of the 2SG personal pronoun followed by a speaker-oriented contextual demonstrative in śet́e, followed by the secondary genitive de nite singular declension.)
(94) ruz+oń koj%kona govor+t+ne+s+kak Russian_N+GEN some_PRON-INDEF.ABS dialect_N+PL+DEF.PL+ILL+CLT
sova+ś finno%ugra+ń t́e enter_V+IND.PRETI.PRED-3SG Finno-Ugrian_N+GEN this_PRON-DEM-PROX.ABS val+oś “mečka” forma+so –
word_N+NOM.DEF.SG “mechka_N.ABS” form_N+INE –
“ĺevks marto avaka ovto” smuśt́+se.
“off-spring_N.ABS with_POP female_N.ABS bear_N.ABS” meaning_N+INE
(Bryzhinski M 1991: 157) ‘This Finno-Ugrian word, in the form “mechka”, has even be-come part of the lexicon in some Russian dialects in the sense “sow bear with cub(s)”.’
(95) [mońćit́ińt́ ]
a moń+śe+t́e+ńt́
but_CONJ I_PRON-PERS-1SG.GEN+PRON-SOD+PRON-DEM-PROX+GEN.DEF.SG
viŕ ava+ńeń usk+ik
forest_N.ABS mother_N+DAT haul_V+IND.PRETI.PRED-2SG>3SG
(MW 1992: 1282) ‘but mine, you’ve taken to the Mother of the Woods.’
Interim conclusions
In Erzya there are two modi er-without-noun marking strategies, and in Evsev'ev's grammar these have not been joined in one section, nor have they in grammars since then. On the basis of what has been demonstrated above with regard to declension in
NPs where the head noun has been deleted in contextual circumstances allowing for predictable identi cation, we can draw the following interim conclusions on what modi ers are attested in SECONDARYDECLENSION and their types:
ZERO = Simple shift of declension locus to main item of NP: Adjectives: ašo skal+oś ‘the white cow’ => ašo+ś ‘the white one’
Quanti ers: źaro vina+ś ‘so much liqour’ => źaro+ś ‘so much’
Spatial modi ers: ošso lomań+t́+ńe ‘the people in the town’ => ošso+t́+ńe ‘the ones in town’
Determiners: iśt́amo boćka+ś ‘a barrel like that’ => iśt́amo+ś ‘one like that’
Genitive-form personal pronouns: sonze kŕepośt́+eś ‘his/her/its stronghold’ => sonze+ś
‘his/hers/its’
SODPRONOUN -śe = Speaker-oriented (distal) demonstrative pronoun following main item of NP and subsequent shift of declension locus to that pronoun:
Nouns in inde nite genitive: kšńi+ń kojme+ś ‘the shovel of iron’ => kšńi+ń+śe+ś ‘the iron one’
Genitive-form personal pronoun: moń kudo+ś ‘my house; that house of mine’ =>
moń+śe+ś ‘mine; that one of mine’