Comitative and terminative in Votic and Lower Luga Ingrian.
1. IntroductionThere is no definite opinion on the number of morphological cases in the Votic language, as different sources present different case lists. Among other cases, Tsvetkov (2008 : 12) and Ariste (1968 : 33-35) mention the comitative (with the -ka -ka marker), and terminative (with the -ssa marker). Ahlqvist (1856 : 72-73) and Agranat (2007 : 40, 111) qualify the corresponding forms as postpositional constructions. (1) There are either little or no arguments for the chosen interpretation of the comitative and terminative in these grammars.
In his description of the Ingrian language, Laanest (HaaHecT 1978 : 224-226) lists the Lower Luga comitative [TEXT NOT REPRODUCIBLE IN ASCII] and terminative (s)sa among other morphological cases. He mentions the postpositional origin of both markers, but gives no further comments about their current status.
There are few articles discussing the status of the Votic comitative and terminative, and even less studies of the same question in Lower Luga Ingrian. All of the studies base on the materials collected half a century ago or earlier.
This paper aims at the following:
1. To clarify the status of the Votic comitative and terminative in the contemporary language using recently collected field data.
2. To define the status of the comitative and terminative in contemporary Lower Luga Ingrian.
3. To compare the comitative and terminative in these two languages. Such comparison is especially valuable as from our point of view the Lower Luga Ingrian is a specific convergent variety that developed as a language of interethnic communication between the original Votic population and the Ingrian population that settled in the same area (Rozhanskiy, Markus 2014). (2)
2. Background
In the paper about the Votic comitative, Ariste (1984 : 3) expresses his view that Ahlqvist (1856) did not consider the comitative as a case only because the corresponding marker has no harmonic variants. On the other hand, Ariste provides only one argument to support his opinion that the comitative is a case--occasional examples where the comitative marker is attached both to the noun and the dependent adjective. As Ariste also mentions, such examples occur in Votic only rarely.
Kettunen (1956 : 5) discusses the terminative in Finnic languages and concludes that "Der Terminative kommt als vollkommen produktiver Kasus nur in Estnischen". However, he only considers the -ni marker (or its variants in related languages), but mentions nothing about the -ssa marker. It can be noted that the status of the terminative with the -ni marker is also not evident. Kettunen lists the opinions of different researchers, but gives no arguments to support one of the opinions.
The most detailed discussion of the comitative and terminative is provided in (Oinas 1961). Oinas conducts a comparative analysis of postpositional cases in Finnic languages. He differentiates between postpositional case suffixes and postpositions that have not yet reached the status of case suffixes. Concerning Votic Oinas (1961 : 179) concludes that both the comitative and terminative are postpositional case suffixes. The same conclusion is made for Ingrian, although this is somewhat contradictory to the analysis on page 158, where the Soikkola Ingrian terminative is classified as a postposition. The status of the Lower Luga Ingrian terminative is not mentioned explicitly.
The criteria used by Oinas for distinguishing between a case and a postposition are (1) the structural identity of the forms with the older stock of suffixes; (2) the close juncture or internal open juncture between the added form and the head-word; (3) the occurrence of sound features characteristic of non-initial syllables; (4) the adaptation to the exigencies of vowel harmony; (5) the agreement of the attribute. However, Oinas mentions that "criterion 1 is less important, due to the fact that BF [Baltic-Finnic] languages have no fixed case suffix structure" and "criterion 5 is an indication of case suffixal status, but its absence does not negate the possibility of such status". Thus, 'criteria (2), (3) and (4) can be considered the most important in determining case suffixal status" (Oinas 1961 : 179).
It is easily noticeable that the main criteria used by Oinas are phonetic or morphological. On the other hand, Grunthal (2003 : 30) gives the priority to syntactic criteria. He refers to the opinion expressed in Plank 1992 : 19: "The distinction between case affixes and adpositions is mainly syntactic by nature in so far as case affixes are morphologically bound to words whose syntactic relations they encode, while adpositions express grammatical relations within a syntactic construction rather than with respect to an individual constituent". In Grunthal 2003 : 27 both the Votic comitative and terminative are qualified as case suffixes. Ingrian is not analyzed in this study.
A typological research (ApxunoB 2001 : 50-51) notes that the Finno-Ugric comitative is not a case in a full sense. Obviously, the reason for its ambiguous status is the postpositional origin, see e.g. HaaHecT 1975 : 58. (3) The Votic and Ingrian terminative also originates from a postposition, and thus poses a similar challenge for a researcher. The grammaticalisation of a postposition into a case marker can take centuries (cf. the history of the Estonian morphological comitative in Habicht 2000 : 43-48 and the analysis in Stolz, Stroh, Urdze 2006 : 364-366). Also, changes in the language can be for a long period of time ignored by grammars (cf. a citation from Oinas 1961 : 34: "For centuries Est[onian] descriptive grammars have not recognized the ka-comitative as a case").
3. Data and methods
The analysis presented in this paper is based both on texts collections and specially elaborated questionnaires that were recorded during field trips to Ingria in 2001-2013. In the questionnaires, the speakers were asked to translate sentences from Russian into their native language, or to evaluate if the sentences from the questionnaire sound correct. The Votic examples in this paper come from the Luuditsa variety.
The questionnaires were designed to check different criteria, which define the position of the comitative and terminative on a conventional scale between prototypical cases and postpositions. Unlike all previous research on the subject, we used criteria referring to all language levels: phonetics, morphology, and syntax.
It should be pointed out that a set of potential criteria for differentiating between case markers and postpositions is much longer than the one presented in this paper. However, a whole number of syntactic criteria cannot be tested on the contemporary language data. Both languages are on the verge of extinction, all the speakers are on average 80 years old, and Russian has been for years their main language of communication.
Consequently, the speakers do not use complicated syntactic constructions, and cannot easily evaluate if such artificially constructed sentences from the questionnaires are grammatically and semantically correct. In particular, the speakers were not able to produce or evaluate sentences with an emphatic particle -tsi inserted between the stem and the comitative or terminative marker; therefore we cannot use this criterion in the analysis.
4. Votic
4.1. Vowel harmony
Votic has vowel harmony: front and back vowels usually do not appear in the same form, and most suffixes have back-vocalic and front-vocalic variants.4 5 6 7 Case markers that contain a full vowel or a lateral approximant have two harmonic variants.
Back-vocalic Front-vocalic pehko-a bush-PART tutto-a daughter-PART bratko-AAd brother-ADALL [TEXT NOT REPRODUCIBLE IN ASCII] hare-ADALL
Postpositions are independent words and their phonetic structure is not influenced by the preceding noun.
kato [TEXT NOT REPRODUCIBLE IN ASCII] 'on the roof' ko takkan (5) 'behind the house' [TEXT NOT REPRODUCIBLE IN ASCII] 'on the hand' tsula takkan 'behind the village'
Both the comitative and terminative markers contain a full vowel (a or a), but they do not have front-vocalic variants (* -ka (6) or * -ssa), and they do not change if added to a front-vocalic stem:
Comitative: pojga-ka (7) 'with the son', tuto-ka 'with the daughter', meheka 'with the husband',
Terminative: kotto-ssa 'up to the house', mettsa-ssa 'up to the forest', jarve-ssa 'up to the lake'.
According to this criterion, both the comitative and terminative are similar to postpositions.
4.2. Prosodic independence
In Votic, the first syllable of a word usually carries the main stress. This is true also for postpositions, and they are qualified as separate words by the native speakers: na-vva takkan 'at the table'. On the other hand, case markers cannot be stressed, and they are not considered as separate words by the speakers: [TEXT NOT REPRODUCIBLE IN ASCII]-ssi (table-TRANS) 'into a table'.
The comitative marker is not stressed and constitutes one prosodic unit with the preceding stem. The speakers qualify comitative forms as one word, e.g. [TEXT NOT REPRODUCIBLE IN ASCII] 'with a table'.
The terminative marker ssa is prosodically independent because it carries its own stress. The speakers consider the terminative marker as a separate word: [TEXT NOT REPRODUCIBLE IN ASCII] 'up to the table'.
According to this criterion, the comitative is similar to case markers, while the terminative is similar to postpositions.
4.3. Phonetic structure
The following two criteria consider the phonetic structure of the markers. As mentioned above, there is a variation in the transcription of the comitative and terminative markers by different authors (cf. Table 3 in the Appendix). In most cases, this can be explained by the dialectal variation and/or the language change over a long time period. However, in some cases there is no obvious explanation. For instance, Agranat worked with the same Votic speakers as we did, but she transcribed the terminative marker with a short vowel (ssa) (ArpaHaT 2007 : 40). (8) Instrumental measurements showed that in our data the duration of the vowel in the terminative marker ranges from 200 to 300 ms. This length corresponds to the duration of long vowels in initial syllables, and this is an additional reason to consider the terminative marker in contemporary Votic as a separate word. The duration of all other vowel types is usually not more than 120-150 ms. The vowel in the comitative marker is on average 110 ms long.
This paper is based on our own Votic data only, thus we transcribe the comitative with a short vowel, and the terminative with a geminate and a long vowel.
a. An initial geminate
An independent word (including postpositions) cannot start with a geminate in Votic. On the other hand, case markers and other morphemes often have an initial geminate: [TEXT NOT REPRODUCIBLE IN ASCII] ELAT, [TEXT NOT REPRODUCIBLE IN ASCII]/ [TEXT NOT REPRODUCIBLE IN ASCII] ADALL, -ssi trans, etc.
The terminative marker ssa starts with a geminate, so according to this criteria it should be considered as a case suffix but not as a postposition. (9)
Elena Markus, Fedor Rozhanskiy
This criterion is not relevant for the comitative, as both case markers and postpositions can start with a single consonant.
b. A short final vowel
Case markers in contemporary Votic contain only short vowels, e.g. -ssi TRANS, -a/-a PART.
Postpositions can only end in a short vowel if they consist of more than one syllable, e.g. takka '(to) behind', uli 'across'. If a postposition is monosyllabic and ends in a vowel, this can only be a long vowel, e.g. mu 'by, along'.
The comitative marker is monosyllabic and has a short final vowel, so from that point of view its structure is typical for case markers but not for postpositions. It is important to note that the vowel in this marker was previously long (cf. Table 3 in the Appendix). The shortening of the vowel can be considered an argument for treating the comitative as a case marker in the contemporary language. On the other hand, the terminative preserved the long vowel, so it looks rather as a postposition.
4.4. Alternations on the border between the stem and the marker
In words of certain paradigmatic types, the stem-final a/a changes into e/e before case markers ([TEXT NOT REPRODUCIBLE IN ASCII], [TEXT NOT REPRODUCIBLE IN ASCII] 2011 : 62--63), cf.:
ejna hay.GEN pajka patch.GEN nurka corner.GEN norika bride.GEN ejne-ssa hay-ELAT [TEXT NOT REPRODUCIBLE IN ASCII] patch-ADALL nurke-z corner-INESS norike-sse bride-ELAT
Postpositions do not affect the final vowel of the preceding noun:
ejna pale hayGEN on 'on the hay' nurka takkan corner.GEN behind 'behind the corner' pajka pale patch.GEN on 'on a patch' norika tuvve bride.GEN towards 'towards the bride'
The comitative marker normally does not affect the preceding stem/ form; however, in our data there are several examples where the final vowel is changed before the comitative:
ejna hayGEN ejna-ka hay-coM
pajka patch.GEN pajka-ka patch-coM
but nurka corner.GEN nurke-ka corner-coM
norika bride.GEN norike-ka bride-coM
The change of the stem-final vowel is stable before the case markers, but not before the comitative (it depends both on the lexeme and the speaker).
The terminative marker never affects the preceding stem/form:
ejna hay.GEN/ILL ejna-ssa hay-TERM pajkka patch.ILL pajkka-ssa patch-TERM nurkka corner.ILL nurkka-ssa corner-TERM norikka bride.ILL norikka-ssa bride-TERM
Thus, according to this criteria the terminative behaves as a postposition, while the comitative occasionally functions as a case marker.
4.5. Combining with the stem vs. case form
In Votic, case markers always combine with either a weak grade or a strong grade stem depending on the paradigmatic class: ko-z (house-INESS), kotto-a (house-PART). Case markers are never attached to another case form, i.e. to the form that already contains a case marker.
Postpositions govern the genitive, the partitive or the illative forms (genitive is the most frequent): ko pale (house.GEN on) 'on the house', kottoa muta (house.PART along) 'along the house', kotto paj (house.iLL towards) 'towards the house'.
It should be pointed out that constructions "noun + postposition" and "stem + case marker" often look similar. Most postpositions govern the genitive form of a noun and case markers are attached to the stem, which often coincides with the genitive form (there is no special marker of the genitive in Votic), cf. ko-ssa (house-ELAT) 'from the house' and ko takkan (house.GEN behind) 'behind the house'. The same applies to the illative, as in Luuditsa Votic the -se/-se marker is not common (10) and therefore the illative forms are often identical to one of the stems, cf. kotto (house.ILL) 'into the house' and kotto paj (house.ILL towards) 'towards the house'. However, if a postposition governs the partitive form, there is no ambiguity, as the postposition is added after the case marker, cf. kotto (house.ILL) 'into the house' and kotto-a mutd (house-PART along) 'along the house'.
The terminative combines mostly with the genitive (11) or the short illative forms, (12) and in such examples we cannot define its status as a case marker or a postposition.
sohho-ssa swamp[ILL]-term ~ so-ssa swamp [GEN]-term 'up to the swamp'
kotto-ssa house[ILL]-TERM ~ ko-ssa house[GEN]-TERM 'up to the house'
However, we also have occasional examples of the terminative attached to the allative form, and such forms are unambiguously similar to postpositional phrases:
jegepera-le-ssa Jogopera-ALL-TERM 'up to the Jogopera village'
This criterion does not help to determine the status of the comitative marker, because it combines with the stem that is identical with the genitive. (13)
4.6. The order of markers in indefinite pronouns
In Votic, the morpheme order is fixed and does not allow any rearrangements. However, there is one exception in indefinite pronouns: in several case forms, the indefinite suffix -le can be placed both after and before the case markers:
miga-z-le what-INESS-INDEF ~ miga-le-z what-INDEF-INESS 'in something' tsene-ss-le who-ELAT-iNDEF ~ tsene-le-ssa who-INDEF-ELAT 'about someone'
It is not possible to place a postposition before the -le marker:
miga-le pal what-iNDEF on 'on something' (but not * miga pal-le)
From this point of view, the comitative marker behaves as a typical case marker: it occurs both before and after the indefinite suffix:
miga-ka-le what-coM-iNDEF ~ miga-le-ka what-INDEF-coM 'with something'.
On the contrary, the position of the terminative marker cannot vary in indefinite pronouns; it can only be placed after the indefinite suffix:
miga-le-ssa what-iNDEF-TERM 'until something' (but not *miga-ssa-le)
In this respect, the terminative behaves as a postposition.
4.7. Marking of adjectives in NP
In Votic there is case and number agreement of adjectives with nouns:
vana vetta pene-ssa pangi-sse
pour.IMP water.PART small-ELAT bucket-ELAT
'Pour water from the small bucket'
In a postpositional phrase, the postposition cannot be replicated after each constituent. The case and number of the adjective agrees with those of the noun (genitive, partitive or illative--depending on the postposition).
vene me-B sur-ta nusti-a jekki-a mu
boat go.PRS-3sG big-PART beautiful-PART river-PART along
'A boat goes along a large beautiful river'
In comitative constructions, there is usually only one comitative marker for the whole phrase:
tama ela-B nusti nore tuto-ka
3sg live.PRS-3sG beautiful.GEN young.GEN girl-coM
'He lives with a beautiful young girl'
However, adjectives can attach the comitative marker in emphatic constructions:14
tama ela-B nusti-ka nore-ka tuto-ka
3sg live.PRS-3sG beautiful-coM young-coM girl-coM
'He lives with a beautiful young girl'
In terminative constructions, only one terminative marker for the whole phrase is allowed. Dependant adjectives are either in the genitive or the illative form:
tama tul-i uvve/ute kajvo-ssa
3sg come-IMPF.3SG new.GEN/new.ILL well-TERM
'He came up to a new well'
Double marking is prohibited:
* tama tul-i uvve-ssa/ute-ssa kajvo-ssa 3sg come-iMPF.3sG new-TERM well-TERM
Thus, according to this criterion the terminative is a typical postposition, while the comitative can behave in both ways.
4.8. Conjoined NP
In a conjoined phrase, each noun is obligatory marked with the same case affix:
anna marje-D pet'a-nna i masa-nna
give.iMP berry-PLNOM Peter-ADALL and Mary-ADALL
'Give the berries to Peter and Mary'
A drop of the case marker is impossible:
*anna marje-D peta i masa-nna
give.iMP berry-PLNOM Peter and Mary-ADALL
In case of conjoined postpositional phrases, two constructions are possible:
a) one postposition for both PPs:
ire-D ele-ta kapi i ahjo takkan
mouse-PLNOM live-PRS.3PL cupboard.GEN and stove.GEN behind
'The mice live behind the cupboard and the stove'
b) a postposition for each of the conjoined PPs:
ire-D ele-ta kapi takkan i ahjo takkan
mouse-PLNOM live-PRS.3PL cupboard.GEN behind and stove.GEN behind
'The mice live behind the cupboard and behind the stove'
Similarly to case affixes, the comitative marker is obligatory added to each noun in a conjoined NP:
tama su-B lihha sona-ka ja pertsa-ka
3sg eat.PRS-3sG meat.PART salt-coM and pepper-coM
'He eats meat with salt and pepper.'
It is not possible to have just one comitative marker in this phrase:
*tama su-B lihha sona ja pertsa-ka
3sg eat.PRS-3sG meat.PART salt and pepper-coM
The same strategy is typical for constructions with the terminative:
sia tul-i-D jekke-ssd vaj merre-ssa
2sg come-iMPF-2sG river-TERM or sea-TERM
'Did you come up to the river or up to the sea?'
However, in our dataset we have occasional examples where only the last noun is marked:
?sia tul-i-D j[section]kke vaj merre-ssa
2sg come-iMPF-2sG river.iLL or sea-TERM
'Did you come up to the river or to the sea?'
Thus, in conjoined phrases the comitative always behaves as a case marker, while the terminative can occasionally behave as a postposition.
4.9. Summary
The characteristics of the comitative and terminative as compared to typical case markers and postpositions are summarized in Table 1. Every criterion for comparison is formulated in such way that it has a positive value for the case markers and a negative value for the postpositions. The following symbols are used in the Table:
+ the feature is present
- the feature is absent
+/- the feature is present, but there are certain limitations
[empty set] the feature is not relevant
5. Lower Luga Ingrian
Similarly to Votic, the Lower Luga dialect of Ingrian has special markers both for the terminative (ssd) and the comitative (-ka/-ka). (15) In the latter respect, Lower Luga Ingrian is different from Soikkola Ingrian, which does not have a separate marker for the comitative, and expresses the corresponding meaning either with the adessive or with a postposition (PoxaHCKHH 2010 : 82, 84).
Oinas (1961 : 158) suggests that the terminative marker was borrowed into Lower Luga Ingrian from Votic. In our data, most features of the -ssd marker are the same as in Votic, except for two (see 5.2. and 5.4). The comitative marker has more differences from the corresponding Votic marker. In Table 2 we summarize the features of the Lower Luga Ingrian comitative and terminative, but we discuss in detail only those characteristics that are not the same for Votic and Lower Luga Ingrian markers: the vowel harmony, the possibility to be attached to a case form, marking of adjectives in NP, and marking of nouns in a conjoined NP.
5.1. Vowel harmony
Unlike in Votic, the Ingrian comitative marker has two harmonic variants:
pojan-ka 'with the son', siun-ka 'with you', but tuton-ka 'with the daughter', kenen-ka 'with somebody'
According to this criterion, the Lower Luga Ingrian comitative is similar to case markers, but not to postpositions.
The terminative marker has only one back-vocalic variant (-ssa) similarly to Votic. It is therefore similar to postpositions.
5.2. Combining with the stem vs. case form
There are two possible strategies to construct a comitative form in Lower Luga Ingrian (both strategies are found in our field data, and both were also mentioned by Laanest (.HaaHecT 1966 : 106)). The variation depends partially on a sub-dialect, and partially on a concrete speaker.
The first strategy (labeled as Com (type 1) in Table 2) is to add the -ka/ -ka marker to the genitive form of a noun, e.g. saha-n-ka (saw-GEN-coM) 'with a saw', siu-n-ka (2SG-GEN-COM) 'with you'.
Another option (labeled as Com (type 2) in Table 2) is to attach the -ka/ -ka marker directly to the stem, e.g. saha-ka (saw-cOM) 'with a saw', siu-ka (2SG-COM) 'with you'.
In Ingrian, case markers are added to the stem, and synchronically there can only be one case marker in a form. On the other hand, postpositions combine with different case forms, most frequently with the genitive one. Hence, the comitative construction of the second type can be considered a case form, while the first type is a typical postpositional phrase. (16)
The Lower Luga Ingrian terminative marker is always attached to the illative form (not to the genitive or allative):
soho-ssa swamp [ILL]-term puhu-ssa tree[ILL]-TERM lauko-paiva-ssa saturday[ILL]-TERM
In those nouns that mark the illative with -sse, the latter is often dropped before the terminative, and in this case the terminative marker is added after the stem. However, examples with the terminative attached after the -sse also occur:
venne-ssa boat[ILL]-TERM ~ venne-sse-ssa boat-ILL-TERM 'up to the boat' katko-ssa cradle[iLL]-TERM ~ katko-sse-ssa cradle-iLL-TERM 'up to the (child's) cradle'
In this respect, the Lower Luga Ingrian terminative marker behaves as a postposition.
5.3. Marking of adjectives in NP
In Lower Luga Ingrian, there can be only one comitative marker for the whole NP:
han ella lusti-n nore-n tuto-n-ka
3sg live.PRS.3sG beautiful-GEN young-GEN girl-GEN-coM
'He lives with a beautiful young girl.'
Unlike in Votic, it is not possible to attach the comitative marker to each of the constituents even in emphatic constructions. The Lower Luga Ingrian speakers qualified such examples as unacceptable:
*han ella lusti-n-ka nore-n-ka tuto-n-ka
3sg live.PRS.3sG beautiful-GEN-coM young-GEN-coM girl-GEN-coM
Hence, from the point of view of this feature, the Lower Luga Ingrian comitative is similar to postpositions, but not to case markers.
The terminative marker behaves similarly to Votic: there is only one marker for the whole NP. The adjectives are marked with the illative:
ha man-i ute kaivo-ssa
3sg go-IMPF.3sG new.ILL well-TERM
'He went up to the new well.'
5.4. Conjoined NP
Unlike in Votic, not only the comitative but also the terminative obligatory marks all nouns in a conjoined NP:
ja-d sia lauko-paiva-ssa vai puhha-ssa
stay.PRS-2sG 2sg Saturday-TERM or Sunday-TERM
'Will you stay till Saturday or Sunday?'
In this respect, both the comitative and the terminative are similar to case markers.
5.5. Summary
The features of the Lower Luga Ingrian comitative and terminative are summarized in Table 2. Features that are different from those in Votic and were discussed in sections 5.1.-5.4. are highlighted in bold.
6. Conclusions
Both the Votic comitative and terminative have certain features of case suffixes, as well as those of postpositions. However, the comitative is closer to case suffixes, while the terminative mainly has postpositional characteristics. The traditional qualification of the two Votic markers as belonging to the same category (either cases or postpositions) can probably be explained by their similar historic development, but not by their synchronic status. (17)
Similar situation is observed in the Lower Luga dialect of Ingrian. The terminative marker is close to typical postpositions. The comitative marker has a number of differences from Votic, but the position of the two comitative types on a conventional scale between cases and postpositions is either the same or very close to that in Votic. However, the coexistence of two strategies for building comitative constructions clearly indicates that the comitative marker is still in the transition stage in Lower Luga Ingrian.
The position of the Votic and Lower Luga Ingrian comitative and terminative on a conventional scale between cases and postpositions is schematically plotted on Figure 1. (18)
The comparison of the Votic and Ingrian comitative demonstrates that features of adpositional cases do not necessarily develop in the same direction. It is known that the comitative with the -ka marker is a recent inno vation in Ingrian (Oinas (1961 : 49-50) and Laanest (HaaHecT 1966 : 106) consider it as a borrowing from Votic or Finnish). However, it is noticeable that in Ingrian the comitative marker developed harmonic variants, and thus became more similar to case markers. On the other hand, it does not mark the adjectives in NP, and from this respect it behaves as a postposition.
Intensive language contacts also influence the grammaticalization processes, and the general picture becomes even more blurred. The two types of Ingrian comitative probably represent two different outcomes of contact influence. Type 1 could have been borrowed as a postposition that could be added to genitive forms (19) (in this way the forms of the sahanka type appeared). Type 2 might have been adopted from Votic as ready comitative forms (like sahaka) and later the -ka was generalised as a productive marker.
Appendix
In Table 3 we list different variants of the spelling and opinions on the status of the Votic comitative and terminative as found in published sources. If this status is not stated explicitly by the author, we use the spelling of the form as a criterion (solid spelling corresponds to a case suffix, separate spelling corresponds to a postposition). Mustonen (1883) often separated the markers with a hyphen, and thus it is not clear whether he considered them as suffixes or postpositions. In the table, the sources are listed in the alphabetical and not chronological order, because the date of publication does not always correspond to the time period when the data was collected.
Addresses
Elena Markus
University of Tartu and Institute of Linguistics of the Russian Academy of Sciences E-mail: [email protected]
Fedor Rozhanskiy
University of Tartu and Institute of Linguistics of the Russian Academy of Sciences E-mail: [email protected]
Abbreviations
1-1st person, 2-2nd person, 3-3rd person, ADALL--adessive-allative, ALL--allative, COM--comitative, GEN--genitive, ELAT--elative, ILL--illative, INDEF--indefinite pronominal suffix, INESS--inessive, IMP--imperative, IMPF--imperfect, PART--partitive, PL--plural, PRS--present, PLNOM--nominative plural, POSTP--postposition, SG--singular, SUP--supine, TERM--terminative, TRANS--translative.
Setala 1964--E. N. Setalan vatjalaismuistiinpanot. Painokuntoon toimittaneet ja julkaisseet Lauri Posti ja Seppo Suhonen.--Memoria Saecularis E. N. Setala 27. II 1964, Helsinki 1964 (MSFOu 135), I--VIII, 1-93.
https://dx.doi.Org/ 10.3176/lu.2014.4.01
REFERENCES
Ahlqvist, A. 1856, Wotisk grammatik jemte sprakprof och ordforteckning, Helsingfors (Acta Societatis Scientiarum Fennicae V: I).
Alava, V. 1908, Vatjalaisia haatapoja, haalauluja ja -itkuja, Helsinki (Suomi IV: 7).
Alvre, P. 1990, Deklinatsiooni uhisjooni eesti-vadja-isuri-soome keelealal.--Itamerensuomalaiset kielikontaktit. Itamerensuomalainen symposium 7. kansainvalisessa fenno-ugristikongressissa Debrecenissa 27. 8.-1. 9. 1990, Helsinki (Kotimaisten kielten tutkimuskeskuksen julkaisuja 61), 9-16.
Atlas Linguarum Fennicarum. ALFE 1, Helsinki 2004 (SKST 800. Kotimaisten kielten tutkimuskeskuksen julkaisuja 118).
Ambus, A. 1962, Dialoogivormilised noidussonad vadjalastel.--ESA 8, 224-228.
--1964, Vadja fraseoloogiat.--ESA 10, 253-258.
--1970, Vadja tanu sapano vadjalaste endi kirjeldustes.--Laanemeresoom laste rahvakultuurist, Tallinn, 94-107.
Ariste, P. 1968, A Grammar of the Votic Language, Bloomington--The Hague (UAS 68).
--1984, Der Komitativ im Wotischen.--Linguistica et Philologica. Gedenk schrift fur Bjorn Collinder (1894--1983), Wien (Philologica Germanica 6), 1-10.
Grunthal, R. 2003, Finnic Adpositions and Cases in Change, Helsinki (MSFOu 244).
Habicht, K. 2000, Grammaticalization of Adpositions in Old Literary Estonian. --Estonian: Typological Studies IV, Tartu (Tartu Ulikooli eesti keele oppetooli toimetised 14), 19--58.
Heinsoo, H., Kuusk, M. 2004, No ku tuleb unohtamize kerd, ni taita ema tseli vai jaB.--FU 26, 17-35.
Kettunen, L. 1956, Die Herkunft des Terminativs, Genitivs, Instruktivs und Komitativs, Helsinki (AASF B 98, 2).
Kettunen, L., Posti, L. 1932, Naytteita vatjan kielesta, Helsinki (MSFOu 63).
Kokko, O. 2007, Inkerinsuomen pirstaleisuus. Eraiden sijojen kehitys murteen yksilollistymisen kuvastajana, Joensuu (Joensuun yliopiston humanistisia julkaisuja 48).
Kulikov, L. 2009, Evolutions of Case Systems.--The Oxford Handbook of Case, New York (Oxford Handbooks in Linguistics), 439-457.
Lahisukukielet. Finnic languages: Karelian, Ludic, Vepsian, Ingrian, Votic, Livonian, Helsinki 1983 (Suomi 119: 3).
Markus, E., Rozhanskiy, F. 2012, Votic or Ingrian. New Evidence on the Kukkuzi Variety.--Finnisch-Ugrische Mitteilungen 35, 77-95.
Mustonen, O. A. F. 1883, Muistoonpanoja Vatjan kielesta.--Virittaja, kielija kansatieteellisia lehtia, Porvoo (Kotikielen Seura I), 144-188.
Magiste, J. 1959, Woten erzahlen. Wotische Sprachproben, Helsinki (MSFOu 118).
Oinas, F. J. 1961, The Development of Some Postpositional Cases in BaltoFinnic Languages, Helsinki (MSFOu 123).
Plank, F. 1992, From Cases to Adpositions.--Aspects of English Diachronic Linguistics, Fasano (Biblioteca della ricerca. Cultura straniera 48), 17-61.
Stolz, T., Stroh, C., Urdze, A. 2006, On Comitatives and Related Categories. A Typological Study with Special Focus on the Languages of Europe, Berlin--New York (Empirical Approaches to Language Typology 33).
Rozhanskiy, F., Markus, E. 2014, Lower Luga Ingrian as a Convergent Language.--FINKA Symposium. On the Border of Language and Dialect. University of Eastern Finland, 4--6 June 2014, Joensuu, 36-37.
Suhonen, S. 1985, Wotisch oder Ingrisch?--Dialectologia Uralica. Materialien des ersten Internationalen Symposions zur Dialektologie der uralischen Sprachen 4.--7. September 1984 in Hamburg, Wiesbaden, 139-148.
Tsvetkov, D. 2008 [1922], Vadja keele grammatika, Tallinn.
[TEXT NOT REPRODUCIBLE IN ASCII]. 2007, [TEXT NOT REPRODUCIBLE IN ASCII], [TEXT NOT REPRODUCIBLE IN ASCII] (Mitteilungen der Societas Uralo-Altaica 26. Einheitliche Beschreibung der Dialekte uralischer Sprachen. Begrundet von Klara E. Majtinskaja und Wolfgang Veenker).
[TEXT NOT REPRODUCIBLE IN ASCII] A. B. 2001, [TEXT NOT REPRODUCIBLE IN ASCII] ([TEXT NOT REPRODUCIBLE IN ASCII]).
[TEXT NOT REPRODUCIBLE IN ASCII]. 2011, Tpagu^uoHHoe u HoBoe b [TEXT NOT REPRODUCIBLE IN ASCII] ([TEXT NOT REPRODUCIBLE IN ASCII]), [TEXT NOT REPRODUCIBLE IN ASCII].
[TEXT NOT REPRODUCIBLE IN ASCII] A. 1966, [TEXT NOT REPRODUCIBLE IN ASCII]. [TEXT NOT REPRODUCIBLE IN ASCII].
-- 1975, [TEXT NOT REPRODUCIBLE IN ASCII], 5-121.
-- 1978, [TEXT NOT REPRODUCIBLE IN ASCII].
[TEXT NOT REPRODUCIBLE IN ASCII]. 1930, [TEXT NOT REPRODUCIBLE IN ASCII] ([TEXT NOT REPRODUCIBLE IN ASCII]), 201-305.
[TEXT NOT REPRODUCIBLE IN ASCII]. 2011, [TEXT NOT REPRODUCIBLE IN ASCII]. [TEXT NOT REPRODUCIBLE IN ASCII]. Tom II. [TEXT NOT REPRODUCIBLE IN ASCII], [TEXT NOT REPRODUCIBLE IN ASCII].
[TEXT NOT REPRODUCIBLE IN ASCII] M. 3. 2005, [TEXT NOT REPRODUCIBLE IN ASCII] ([TEXT NOT REPRODUCIBLE IN ASCII]). [TEXT NOT REPRODUCIBLE IN ASCII], [TEXT NOT REPRODUCIBLE IN ASCII].
[TEXT NOT REPRODUCIBLE IN ASCII]. 2010, [TEXT NOT REPRODUCIBLE IN ASCII]. 1, 74-93.
* Supported by the Estonian Research Council grant IUT2-37 and the Russian Foundation for Humanities projects 12-04-00168a and 11-04-00153a.
(1) See Table 3 in the Appendix about the variants of the markers and the status of the Votic comitative and terminative as treated by different researchers.
(2) The Lower Luga Ingrian was also influenced by Estonian and the Ingrian dialect of Finnish (MycnnMOB 2005).
(3) The case markers that originated from postpositions are quite common typologically, cf. Kulikov 2009.
(4) In fact, this rule is more complicated, because i is a neutral vowel, which can appear both in front- and back-vocalic words, and e acts as a front vowel in suffixes and root-finally, but as a neutral vowel elsewhere. Additionally, some back-vocalic derivational suffixes can combine with front-vocalic roots and change the stem into back-vocalic, see [TEXT NOT REPRODUCIBLE IN ASCII] 2011 : 30-34 for more details. The marker of the 2P1 imperative -ka/-ga has no harmonic variants (note that one of its allomorphs is homonymous with the comitative marker).
(5) The geminated kk in takkan is the result of a recent Ingrian influence on Luuditsa Votic. Both this form and the more archaic takan are attested in our corpus.
(6) Oinas (1961 : 39) noted that there are front-vocalic allomorphs of the comitative marker (ka, ka) in the Kukkuzi and Jogopera Votic varieties. The Kukkuzi variety is a mixed Votic-Ingrian language (see Suhonen 1985; Markus, Rozhanskiy 2012 for more details), so the presence of the front-vocalic variants is not surprising. The front-vocalic comitative marker in Kukkuzi is mentioned also by Ariste (1968 : 33) and in Atlas Linguarum Fennicarum 2004. However, there are no examples of the front-vocalic comitative in the Jogopera variety neither in our field data nor in Ariste 1968 and Tsvetkov 2008. Probably such forms were recorded from the speakers whose language was influenced by Ingrian language (this explanation is offered also by Oinas).
(7) Here and below we separate the comitative and terminative markers from the head words with a hyphen. This way of transcription is conventional for this paper, and it does not indicate the status of the markers in any way.
(8) Consequently, in ArpaHaT 2007 the terminative is homonymous with the elative marker (both are transcribed as -ssa), while in our transcription the former one has a long vowel (-ssa), and the latter has a reduced or apocopated vowel (-ssa or -ss).
(9) Oinas (1961 : 151) mentions also the variant with an initial single consonant, but he notes that this variant is rare.
(10) The short illative form without the -se/-se marker is the basic one in Luuditsa Votic, but rare examples of the full illative (with the -se/-se marker) also occur.
(11) The combination of the terminative marker with the genitive form is probably a recent innovation. Ariste (1968 : 34-35) claimed that the terminative marker "is added to the illative or the allative" (the same is noted in Alvre 1990 : 13). In the contemporary Votic, the combination of the terminative with the allative form is very rare, while combinations with the genitive form are quite common.
(12) Ahlqvist (1856 : 117) has examples of the terminative marker attached to the full illative forms.
(13) Except that the stem final vowel can alternate before the comitative marker. This feature was discussed in 4.4., so we do not use it as a criterion here.
(14) This feature of the Votic comitative was already noted by Kettunen (1956 : 10) and Ariste (1968 : 34).
(15) Oinas (1961 : 49-50) mentions also variants of the comitative marker with a long vowel -kd, -ka, and a variant with the initial geminate -kka, but none of those are found in our data.
(16) In fact, such forms can be interpreted in two ways: either as forms where the comitative marker -ka/-ka is added to the genitive form or as forms with the comitative marker -nka/ -nka. Kulikov (2009 : 445) notes that it is sometimes nearly impossible to draw with accuracy the distinction between adpositional cases and multilayer case marking. Compare also the Karelian language, where the genitive marker is considered to be a part of the comitative suffix -nke that originated from the postposition kera (KoBaneBa, PognoHOBa 2011 : 85). However, it should be noted that there are no other examples in Ingrian that can be analyzed as containing two case markers in the same form.
(17) Oinas (1961 : 40, 152) noted that both the comitative and terminative were at first considered as postpositions, and later as case markers.
(18) The figure is only schematic, first of all because all the criteria that determine the status of the two markers are treated equally here, while in fact their value is not the same. On the other hand, it does not seem possible to evaluate exactly the weight of each criterion.
19 Kokko (2007 : 190) noted that the -ka comitative must have been borrowed into Lower Luga Ingrian and Finnish varieties as a postposition; otherwise it is not possible to explain the presence?? of the genitive ending -n before -ka.
Table 1
Votic comitative and terminative as compared to case markers and
postpositions
Criteria CASE COM TERM POSTP
1 Vowel harmony variants + - - -
2 Prosodic dependence + + - -
3a Possibility to start with + [empty set] + -
a geminate
3b A short final vowel (for + + - -
monosyllabic markers)
4 Alternations on the stem- + +/- - -
suffix border
5 Impossible to attach to + [empty set] - -
a case form
6 Free order of suffixes in + + - -
indefinite pronouns
7 Obligatory marking of + +/- - -
adjectives in NP
8 Obligatory marking of every + + +/- -
noun in a conjoined NP
Table 2
Lower Luga Ingrian comitative and terminative
as compared to case markers and postpositions
CASE COM COM
(type 1) (type 2)
1 Vowel harmony variants + + +
2 Prosodic dependence + + +
3a Possibility to start + [empty set] [empty set]
with a geminate
3b A short final vowel (for + + +
monosyllabic markers)
4 Alternations on the + [empty set] +/-
stem-suffix border
5 Impossible to attach to + - +
a case form
6 Free order of suffixes in + + +
indefinite pronouns
7 Obligatory marking of + - -
adjectives in NP
8 Obligatory marking of every + + +
noun in a conjoined NP
TERM POSTP
1 - -
2 - -
3a + -
3b - -
4 - -
5 - -
6 - -
7 - -
8 + -
Table 3
The status of the Votic comitative and terminative in published
sources
Comitative
Source Variants of Status
the spelling
Ahlqvist 1856 ka Postposition
Alava 1908 kaa Postposition
Ambus 1964 -ka Case
Ambus 1970 -ka Case
Ariste 1968 -ka, -ka, -ka Case
Kettunen, Posti 1932 -ka, -ka, -ga Case
Heinsoo, Kuusk 2004 -ka Case
Lahisukukielet 1983 -ka, -ka Case
Magiste 1959 -ka, -ka Case
Mustonen 1883 -kaa, -ka ? Postposition
Setala 1964 ka Postposition
Tsvetkov 2008 (1922) -ka Case
ArpaHaT 2007 ka Postposition
fleHcy 1930 ka, k Postposition
Ahlqvist 1856 sa Postposition
Alava 1908 saa Postposition
Ambus 1962 -ssd Case
Ambus 1970 -ssd, -sd Case
Ariste 1968 -ssd Case
Kettunen, Posti 1932 -ssd Case
Magiste 1959 -ssd Case
Mustonen 1883 -saa ? Postposition
Tsvetkov 2008 (1922) -ssaa Case
ArpaHaT 2007 ssa Postposition
[TEXT NOT REPRODUCIBLE sd Postposition
IN ASCII] 1930
Comitative
Source Examples with
page numbers
Ahlqvist 1856 p. 114: naise' ka
p. 116: wasara' ka
Alava 1908 p. 10: tamaa kaa,
Jumalaa kaa
Ambus 1964 p. 254: kaglaka
p. 257: tseleka
Ambus 1970 p. 98: ner/kaka
p. 98: migaka
Ariste 1968 p. 33: jangaka, izaka
Kettunen, Posti 1932 p. 18: tsellaka
p. 145: eppipoikika, sappega
Heinsoo, Kuusk 2004 p. 20: meheka, lautoika
Lahisukukielet 1983 p. 56: viljaka, viljaka, rahaka
p. 58: karjaka
Magiste 1959 p. 146: opezeka
p. 173: verkkoika
Mustonen 1883 p. 151: tama kaa
p. 153: vee-ka, enne-kaa,
Jumala-kaa
Setala 1964 p. 4: kalakorma ka
p. 9: lampaje ka, apezl ka
Tsvetkov 2008 (1922) p. 33: taloka
p. 37: rati ka
ArpaHaT 2007 p. 111: suur paa ka, kana ka
fleHcy 1930 p. 222: Uva ka, tsive k
Ahlqvist 1856 p. 108: Jumalale sa
p. 117: wasumase sa, kotose sa
Alava 1908 p. 23: uomnikkoos saa
p. 26: paivavalgetikkoos saa
Ambus 1962 p. 224: kenmissd
Ambus 1970 p. 97: kaukdnesd
p. 99: kervissd, nobdssd
Ariste 1968 p. 34: neundssd, mahdssd
Kettunen, Posti 1932 p. 146: surmassd
Magiste 1959 p. 177: peaviaaessd
p. 184 -185: paikennessa
Mustonen 1883 p. 146: vuhes-saa
Tsvetkov 2008 (1922) p. 33: talossaa p. 37: rattissaa
ArpaHaT 2007 p. 110: cullaa ssa
[TEXT NOT REPRODUCIBLE p. 238: restorind sa, suetsa sa
IN ASCII] 1930
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Markus, Elena; Rozhanskiy, Fedor |
|---|---|
| Publication: | Linguistica Uralica |
| Date: | Dec 1, 2014 |
| Words: | 6619 |
| Previous Article: | Evidentiality in dialects of Khanty. |
| Next Article: | Types of consonant alternation in the inflectional system of Soikkola Ingrian. |