Russian Linguistics

, Volume 13, Issue 3, pp 185–206 | Cite as

On preposition repetition: A study in the history of syntactic government in Old Russian

  • Emily Klenin


The results described above diverge somewhat from those obtained by Worth (1982), and suggest an explanation for the purely formal hierarchies established by Zaliznjak (1986).

The divergence from Worth (1982) lies primarily in our different interpretation of the “basic syntactic kernel” (Worth 1982, 503) of PrepRep, which, according to Worth, consists of the preposition itself, a noun plus its case marker (“governed by the preposition”), “plus one or more (adjectival or nominal) modifiers, any or all of which optionally replicate the preposition of the kernel structure”. This view of the syntactic kernel derives naturally from the Muscovite data, in which (N - Adj) PrepRep occurs freely, albeit optionally, as in the birchbarks. It is more problematic for a text such as the Suzdal Continuation, in which (N - Adj) PrepRep is clearly exceptional. In contrast with Worth (1982), it will be proposed here that the modifier is not included as such in any syntactic ‘kernel’ of PrepRep, and that the original structural description of PrepRep did not refer to nouns at all, but only to noun phrases. It will, specifically, be suggested that PrepRep was sensitive to NP boundaries within the domain of the PP, with all PP-internal NP boundaries potentially triggering a reiteration of P. This description is neutral as to whether the iteration of the preposition is merely a ‘spelling out’ of prepositional government within a single PP, or whether, instead, it represents an actual replication of the PP structure itself. (Thus, no solution is offered to the question of whether, for example,с Вла uмuром с Росmuславuчем consists, after the operation of PrepRep, of a single PP with an internal redundant spelling out of the governing P, or whether, instead, it actually consists of two PPs.) In particular, it will be proposed that Old Russian PrepRep in its basic form provided an unusual instance in which the morphosyntactic coding provided by P is, contrary to Worth (1982), demonstrably not entirely redundant on case-marking, but instead marks the external syntactic status of the NP (with respect to PP control), whereas case-marking codes the internal syntactic structure of the NP (with respect to agreement between modifier and head). Prepositional government, coded onto NPs externally in the form of the preposition, triggers the acquisition of NP-head (i.e., N) case-marking, which in turn is distributed to modifiers in the form of ‘internal’ case-marking across the NP; that the case-marking acquired by modifiers is essentially different from that acquired by NPs is shown by the limitations on the occurrence of PrepRep in modifier-head constructions. Old Russian PrepRep provides clear evidence that the government of NPs by Ps is essentially different from NP-internal case-marking, even though the two types of case-marking necessarily interact.

The earliest of the Muscovite and Novgorod examples of PrepRep date from the latter 13th century, and the majority are from the 14th–15th centuries; they are, generally speaking, original documents, not copies. The Laurentian Manuscript, although copied in the latter 14th century, is based on earlier texts, the last one from the beginning of the same century; it is known to be an extremely conservative manuscript, and probably reflects rather accurately the language of its sources (Klenin 1983b). Thus, its data potentially reflect an earlier stage of development than that offered by the other two corpuses. This impression is supported by other data, albeit fragmentary: whereas 12th-century Novgorod texts contain NP appositive PrepRep (example [7] above), ‘(adjectival) modifier + head’ PrepRep is attested only from the latter 13th century (examples [3] and [4]). In the First Novgorod Chronicle (Synod Copy), Istrina (1923, 146–47) cites numerous examples of NP appositive PrepRep throughout the text, but the only example she cites of modifier-head PrepRep is from 1231. Moreover, the Muscovite data, which extends about 60 years later than the material from the birchbarks, is also slightly more permissive, showing even occasional occurrences of modifier-initial modifier-head PrepRep. Overall, then, there is at least some reason to suspect that, in the earliest Old Russian, modifier-head PrepRep was a relatively minor subtype, perhaps restricted to Slavonic or syntactically marked environments, but that it expanded substantially in the Muscovite period. If, however, adjectival modifiers are less well represented than patronymics, titles, and the like, then it is not convincing to define the syntactic kernel of PrepRep as a modifier-head relationship. Thus, if Worth is accurate in his analysis of the Muscovite situation, then that situation must be essentially quite different from the one represented in the Suzdal Continuation, where the syntactic ‘kernel’ is represented by prepositions governing NPs in appositive relations to each other; from the little documentary evidence available, it seems plausible that the Muscovite situation represents a relatively late flowering of one type of PrepRep, at a period when the original phenomenon was waning.

What, then, was the original phenomenon? As discussed above, PrepRep in the Suzdal Continuation is largely a phenomenon of appositive NPs; even in the modifier-head type of PrepRep, the head is typically itself an NP, that is, a proper name. Thus, the typical PrepRep structure in the Suzdal Continuation can be depicted more or less as follows: or perhaps: Moreover, modifier-head PrepRep was apparently favored by NP nesting; that is, even in modifier-head PrepRep, the appearance of the repeated preposition was especially likely to be triggered by a phrase boundary within the prepositional object. Thus, PrepRep did not normally mark elements smaller than the NP, but rather marked NPs themselves as internal to PP. In this sense, and contrary to Worth's analysis, PrepRep is not redundant on case-marking, since case-marking, precisely, applies within NPs, which can acquire their global marking by virtue either of prepositional government or of their position with respect to the verb phrase. According to Worth (1982, 505) “the initial preposition can be the governer only of the noun to follow”, but, in the Suzdal Continuation, there seems to be no reason to assume that PrepRep refers to nouns at all; instead, it must refer to noun phrases. Rather than being redundant on case-marking, PrepRep in the Suzdal Continuation is the analogue within PP's and external to NPs to the sort of coding represented within NPs by case-marking.

In broader terms, in the context of the gradual shift within medieval Slavic away from prepositionless oblique case forms and toward the increased used of prepositions, it is likely that PrepRep represents a temporary stage in the establishment of the PP as a structural unit. As is well known, control of case by prepositions is, in Modern Russian as in English and other languages, complicated by PP-internal NP constructions, for example genitive possessives. In a PP such as к маmерu моеŭ nо∂ругu, the NP моеŭ nо∂ругu is obviously contained within the PP, but it derives its case-marking not from the preposition but from its status as a possessive dependent on the head N маmерu. In Modern Russian, the ‘traffic rules’, as it were, of the transmission of case-marking phenomena are well established; in Old Russian, when the use of PPs was still expanding (and when PPs may well have been less fully established in underlying structure than they were on the surface), it is quite likely that the relative opacity or transparency of NP boundaries was less fully automatized. Every PP-internal NP boundary potentially could be articulated by means of PrepRep, and this served to articulate the structure of the PP as a whole.

As noted above, it is unclear whether PrepRep represents merely a ‘spelling out’ of P control internally, within the PP, or whether it instead represents a replication of the whole PP structure itself. Although most earlier interpretations, including Worth (1982), seem to point to the former interpretation, the latter one is not without support. It can be argued that appositive NPs, when governed by prepositions, were actually construed as appositive PPs, with the second preposition omissible only if the two prepositional phrases with identical prepositions succeeded each other without intervening material. Some evidence supporting such an interpretation can be found in an example such as the following one:

(28) г Кыеву Свяmославу ко Всеволо∂uчю (s.a. 1190), where the omitted preposition should be the first element in a phrase к Свяmославу ко Всеволо∂uчю, but has been omitted, apparent under conditions of identity with the preposition of the immediately preceding prepositional phrase. Similarly, when 3 names are conjoined, the preposition can occur before the first and third conjuncts, for example

(29) къ Роману u Косmянmuну u къ Сmефану (s.a. 945). The Laurentian Manuscript contains some half dozen examples of prepositions being omitted before the intermediate NPs in a chain of conjuncts, but then re-appearing before the last conjunct. (The reverse does not occur.)

At this stage in the history of PrepRep, certain unstable constructions, in particular patronymic constructions, were still clearly felt as appositive NPs. As the appositive status of these NPs (and, indeed, their independent NP status) deteriorated, this sort of PrepRep can have been re-interpreted as NP-internal. At this point, even though prepositional government in Old Russian was sufficiently well established that PrepRep had lost its original function, it could develop a new function, as a marker of elements within the NP. In this new role, however, PrepRep really was redundant with case-marking for all modifiers except possessive genitives, and eventually it developed the marginal and largely stylistic value that it now enjoys in Russian dialects and occasionally in literary texts.

Although there is no way of providing anything approaching complete documentation for such a scenario, it fits well with the known facts, and it provides some explanation, on the level of superficial coding strategies, of why PrepRep is so much more likely, in Zaliznjak's corpus, if the phrase-initial element is a proper noun: if the P belongs with the immediately following NP, then a P − NP sequence is a ‘complete’ PP, and what follows the proper noun must be construed as starting a new NP, and hence, in terms of Old Russian, a new PP, although the P itself could be omitted. By extension, the word order constraint on PrepRep, namely its virtual absence from (Adj − N) constructions, as compared with its optional appearance in (Adj − N) strings, can be explained by assuming that, although a NP may well have no modifiers, only much less often does it occur without a head. Hence, a P − N sequence can be interpreted as P − NP and thus as a ‘complete’ PP, whereas a P − Adj sequence is normally read as ‘incomplete’, and therefore will not normally trigger PrepRep. A similar ‘coding strategy’ interpretation of the function of PrepRep was suggested in Worth (1982). Unexplained in Worth's interpretation, however, is the relatively greater use of PrepRep observed by Zaliznjak in (N + N) strings, and its near universality in strings with initial names or personal pronouns; these additional facts correlate with the significance of the NP in PrepRep, in that the more plausible the initial word is as head, the more likely is PrepRep to occur: NP (including proper nouns and personal pronouns) outranks N, and N outranks Adj; N in the context of (N − N) is more ‘NP-like’ than is N in the (N − Adj) context.10

Thus, the data of the Laurentian Manuscript add considerably to our store of information on Old Russian PrepRep. These data suggest that a distinction should be made between modifier-head PrepRep and appositive NP PrepRep. The former is highly restricted in the Suzdal Continuation, and is apparently particularly associated with Slavonic contexts. It is unclear whether this association is primarily based on syntactic criteria, in particular the nesting of NPs within PPs that is more characteristic of Slavonic than of more colloquial sources; structurally, modifier-head PrepRep is favored by the presence of an internal NP boundary within the object of the preposition, and, in this respect, the favored mode of modifier-head PrepRep is that which is structurally most similar to appositive NP PrepRep, which by definition applies to prepositional objects with internal NP boundaries. In all instances, syntactic complexity seems to favor PrepRep in the Suzdal Continuation. It has been suggested that the appositive NP type of PrepRep was historically primary, but was eventually overtaken by modifier-head PrepRep, in which the repeated preposition recurs NP-internally. This history is suggested as both plausible, in terms of available documentation, and consistent with what is known of the rise of analytic PP constructions is Slavic.


Noun Phrase Personal Pronoun Prepositional Phrase Proper Noun Common Noun 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Secondary Sources (Cyrillic)Google Scholar
  2. Борковский, Б. И.: 1949,Сuнmаксuс ∂ревнерусскuх грамоm (Просmое nре∂ложенuе), Львов: Львовский гос. ун-т.Google Scholar
  3. Борковский, В. И.: 1963, ‘Лингвистические данные новгородских грамот на бересте (из раскопок 1956–1957 гг.)’, Арциховский, А. Б. и Б. И.Борковский:Новθоро∂скuе θрамомы на бересmе. Из рацкоnок 1956–1957 θо∂а, Москба: Нзд-во АН СССР, pp. 167–322.Google Scholar
  4. Борковский, Б. И.: 1965, ‘Синтаксис’, Борковский, Б. И. С. Кузнецов:Исmорuческая θраммаmuка русскоθо языка, изд. 2-ое, Москва: Наука, pp. 338–547.Google Scholar
  5. Зализняк, А. А.: 1986, ‘Новгородские берестяные грамоты с лингвистичецкой точки зрения’, Янии, В. Л. и А. А. Зализняк:новθоро∂скuе θрамоmы на бересmе. Из раскоnок 1977–1983 θо∂ов. Комменmарuu u словоуказаmeль к бересmяным θрамомам (uз раскоnок 1951–1983 θо∂ов.), Москва: Наука, pp. 89–219.Google Scholar
  6. Истрина, Э. С.: 1923, ‘Цинтаксические явления Синодального списка 1-ой Новгородской летописи’,ИОРЯС 24 (1919),2, pp. 1–172;26 (1921), pp. 207–239.Google Scholar
  7. Карский, Э. Ф.: 1929/1962, ‘Наблюения в области синтаксиса Лабрентьевского списка летописи’, (1st published inИзвесmuя nо РЯС II, 1, 1–75.) Reprinted inТру∂ы nо белорусскомы u ∂руθuм славянскuм языкам, Москва: АН СССР, pp. 58–112.Google Scholar
  8. Лимонов, Ю. А.: 1967,Леmоnuсанuе Бла∂uмuро-Суз∂альско ŭ Русu, Ленинград: Наука.Google Scholar
  9. Лихачев, Д. С.: 1947,Русскuе леmоnuсu u uх кульmурно-uсmорuческоe значенuе, Москва — Ленинград: Изд-во АН СССР.Google Scholar
  10. Secondary Sources (non-Cyrillic)Google Scholar
  11. Culicover, Peter W., Thomas Wasow, Adrian Akmajian (eds.): 1977,Formal Syntax, New York: Academic Press.Google Scholar
  12. Fillmore, Charles: 1968, ‘The Case for Case’, Bach, Emmon and Robert T. Harms, eds.:Universals in Linguistic Theory, New York: Holt, pp. 1–88.Google Scholar
  13. Jackendoff, Ray: 1973, ‘The Base Rules for Prepositional Phrases’, Anderson, S.. and P. Kiparsky (eds.):A Festschrift for Morris Halle, New York: Holt, pp. 345–56.Google Scholar
  14. Jackendoff, Ray: 1977a, ‘Constraints on Phrase Structure Rules’, in: Culicover, Wasow, and Akmajian 1977, pp. 249–83.Google Scholar
  15. Jackendoff, Ray: 1977b,X′ Syntax: A Study of Phrase Structure, Cambridge, MA: MIT Press.Google Scholar
  16. Klenin, Emily: 1983a,Animacy in Russian. A New Interpretation, Columbus, Ohio: Slavica.Google Scholar
  17. Klenin, Emily: 1983b, ‘The Genitive-Accusative as a Slavonicism in the Laurentian Manuscript’, Flier, M. S., ed.:American Contributions to the Ninth International Congress of Slavists, Kiev, 1983. I. Linguistics, Columbus, Ohio: Slavica, pp. 161–70.Google Scholar
  18. McCawley, James D.: 1988,The Syntactic Phenomena of English, Chicago: University of Chicago.Google Scholar
  19. Selkirk, Elisabeth: 1977, ‘Some Remarks on Noun Phrase Structure’, in: Culicover, Wasow, and Akmajian 1977, pp. 285–316.Google Scholar
  20. van Riemsdijk, H. C.: 1978,A Case Study in Syntactic Markedness. The Binding Nature of Prepositional Phrases, Lisse: De Ridder.Google Scholar
  21. Worth, Dean S.: 1982, ‘Preposition Repetition in Old Russian’,International Journal of Slavic Linguistics and Poetics 2526, pp. 495–507 (Essays in Honor of Edward Stankiewicz).Google Scholar
  22. Worth, Dean S.: 1989, ‘Prepositional Repetition in the russian Funeral Lament’, Falchikov, M., C. Pike and R. Russell (eds.):Words and Images. Essays in honour of Professor (Emeritus) Dennis Ward, pp. 191–95.Google Scholar
  23. Primary Source MaterialsGoogle Scholar
  24. Лавр = Laurentian Manuscript of 1377. Photocopy made available through the courtesy of the Harvard Ukrainian Research Institute.Google Scholar
  25. ДДГ =Духовные u ∂оговорные грамоmы велuкuх u у∂ельнух князеŭ XIV–XVI вв., Москва — Ленинград: Нзд-во АН СССр, 1950.Google Scholar
  26. ГВНП =Грамоmу велuкого Новгоро∂а u Пскова, С. Н. Валк, ред., Москва — Ленинград: Лзд-⋎о АН СССР, 1949.Google Scholar
  27. СГ =Смоленскuе грамоmы XIII–XIV веков, Р. Н. Аванесов, рел. (подготовнли к печати Т. А. Сумникова и В. В. Лопатин), Москва: Изд-во АН СССР, 1963.Google Scholar
  28. Котощихин, Григорий:О Россuu в чарсованuе Алексел Мuхаŭловuча. Text and Commentary. A. E. Pennington, ed. Oxford: Clarendon Press, 1980.Google Scholar

Copyright information

© Kluwer Academic Publishers 1989

Authors and Affiliations

  • Emily Klenin
    • 1
  1. 1.University of CaliforniaLos Angeles

Personalised recommendations