Skip to main content

Investigations on mixed agreement: polite plurals, hybrid nouns and coordinate structures


This article investigates complex agreement patterns with the polite plural pronoun vi and so-called ‘hybrid nouns’ in Serbian. I show how many curious agreement phenomena are to a great extent determined by the inability of an agreement target to simultaneously agree with an exclusively semantic and an exclusively formal φ-feature of an agreement controller. Consequently, in some cases (e.g., the polite plural pronoun vi) masculine emerges as the default gender value, as a result of an independently motivated mechanism. I argue that an analysis based on these two factors allows for wider empirical coverage than the analysis developed in Wechsler and Hahm (2011) and Wechsler (2011) based on an Agreement Marking Principle. I also discuss the so-called ‘different pronoun hypothesis’, which Wechsler and Hahm (2011) propose to explain different types of agreement triggered by the polite plural pronoun. In light of some new facts, however, I argue that the ‘different adjective hypothesis’ in fact might be on the right track. Along the way, I also develop an analysis of gender agreement with coordinated phrases consisting of singular number conjuncts and suggest that gender in Serbian should be represented in terms of binary features [±masculine] and [±feminine].

This is a preview of subscription content, access via your institution.


  1. 1.

    Glosses: 1 = 1st person, 2 = second person, 3 = 3rd person, ACC = accusative, AUX = auxiliary verb, DAT = dative, F = feminine, GEN = genitive, INS = instrumental, LOC = locative, M = masculine, N = neuter, NOM = nominative, NUM = number, PL = plural, REFL = reflexive particle, SG = singular.

  2. 2.

    Note that among the latter, only third person singular and third person nominative plural pronouns actually distinguish for gender; see Sect. 4.1.

  3. 3.

    Wechsler and Zlatić (2000, 2003) recognize pragmatic agreement as well, which I ignore here, since it is not relevant for the purposes of this paper.

  4. 4.

    Wechsler and Hahm (2011) argue that, given the Agreement Marking Principle, the lexical representation for deca ‘children’ proposed in Wechsler and Zlatić (2000, 2003) could be slightly simplified by removing the plural NUMBER feature from the INDEX set. In particular, any Index target agreeing with this noun would appear in plural anyway as a consequence of the Agreement Marking Principle, since deca ‘children’ is notationally plural.

  5. 5.

    See also Arsenijević (2014) for discussion of similar facts.

  6. 6.

    I will focus on the noun braća in this article as a representative of this type of hybrid nouns, because the noun deca ‘children’ introduces an unnecessary complication. In particular, as already mentioned, deca has neuter plural semantic features, given its meaning, and feminine singular formal features, since it declines as a feminine singular noun. However, in nominative, neuter plural nouns and feminine singular nouns in Serbian happen to have a homophonous ending -a, which in principle makes it unclear what kind of agreement is involved with agreement targets ending in -a and agreeing with the nominative form of deca (this is also discussed in detail in Wechsler and Zlatić 2003, Sect. 3.3). There is no such complication with the noun braća ‘brothers’, however, since its semantic features are masculine plural, and masculine plural nouns and feminine singular nouns in Serbian have different endings in nominative (-i and -a, respectively). Thus, it can be assumed with reasonable certainty that agreement targets ending in -a and agreeing with the nominative form of braća agree for formal (feminine singular) features.

  7. 7.

    Note that the form of the participle, which in (26) is feminine singular, does not affect prenominal adjectival agreement. That is, even in examples like (i), where the verb žive ‘live’ does not show any gender agreement, the prenominal adjective still must be strictly formal (i.e., feminine singular). A separate study which I conducted with 12 different informants (7 females/5 males) also confirms this:

    1. (i)
  8. 8.

    Note, however, that plural non-nominative adjectives and third person pronouns do not distinguish for gender. I return to this point in the next section.

  9. 9.

    There are other types of interesting nouns in Serbian that could be tested in this way, such as masculine nouns of profession which may be used of women (e.g., pesnik ‘poet’). There is, however, strong tendency for many speakers of modern Serbian to use feminine nouns in such cases (e.g., pesnikinja ‘female poet’), which can be an important intervening factor. For a discussion of some nouns of profession in Serbian see Wechsler and Zlatić (2003:Chap. 4) and for a general discussion of the contrast between male and female nouns of profession see Bobaljik and Zocca (2010).

  10. 10.

    The feminine (formal) agreement is preferred for predicative adjectives and participles as well, although to a bit lesser degree. For predicative adjectives 36 speakers prefer the feminine form vs. 4 speakers who chose the masculine pattern (1 speaker finds them equally acceptable), while for participles 32 speakers favor the feminine form as opposed to 9 speakers who chose the masculine form (1 speaker again find the two patterns equally acceptable). Notice that since in plural gender is distinguished only in nominative (as discussed in detail in the next section), plural secondary-predicate accusative or instrumental adjectives cannot be used in this respect.

  11. 11.

    One exception to this seems to be the noun govno ‘shit, crap’, which ends in -o in nominative (and should therefore be classified as Declension IM), but allows both extended and non-extended stems (e.g., DAT: govnet-u or govn-u).

  12. 12.

    It is not clear whether these two instrumental endings are always completely interchangeable—there is variation among speakers and the syntactic context seems to matter (see Stanojčić and Popović 1992:82 for some discussion).

  13. 13.

    Haspelmath (2006) argues that the formal concept of markedness as used here and in references given above is inadequate and should be replaced by frequency asymmetries in usage. However, it has been shown that this proposal faces a number of problems, which I cannot go into here in detail—I refer the reader to Bobaljik and Zocca 2010, Nevins (2011) and references therein for relevant discussion. It is useful to note, though, that frequency of use is more likely a symptom, rather than a cause, of grammatical asymmetries, as observed in Nevins (2011:420, footnote 2). As Nevins notes, it has been known since Greenberg (1966:45) that frequency counts for the category of person are highly unreliable, owing to genre-dependence of the texts chosen for counting (see also Bobaljik 2008:Sect. 3.1 on the (lack of) correlation between attested person categories and the functional load). At the same time, Bobaljik and Zocca (2010:154–156) provide clear evidence, which is based on actual corpus studies of gender markedness, that Haspelmath’s proposed correlation between frequency and markedness effects is “strikingly not supported” (Bobaljik and Zocca 2010:156).

  14. 14.

    The assumption is that different morphological features carry different levels of cognitive significance and therefore exist in some type of hierarchical relation; the Person > Number > Gender hierarchy is, for instance, a common example (e.g., Greenberg 1966; Harley and Ritter 2002 etc.). It has also been proposed that there are subhierarchies within features; e.g., first and second person are more highly ranked than third person (Silverstein 1985). For evidence from production and processing in support of feature hierarchies see Carminati (2005) and Malovrh and Lee (2010).

  15. 15.

    Note that in Bulgarian and Macedonian pronouns have only two cases in addition to nominative: (full and short forms of) accusative and dative, while adjectives have no case. However, only singular forms of both pronouns and adjectives make gender distinctions.

  16. 16.

    I come back to this point in Sect. 4.2.1; see Nevins (2011) for a detailed discussion of dual and gender in different dialects of Slovene.

  17. 17.

    Wechsler and Zlatić (2003) discuss the noun devojčurak ‘small girl’, which declines as Declension IM and denotes a female individual, but this noun crucially cannot trigger feminine agreement (*lepa FEM devojčurak)—the masculine agreement is obligatory lep MASC devojčurak. Note also that unlike nouns like vojvoda, devojčurak is clearly morphologically complex: it is based on the root devojk- ‘girl’ and the diminutive suffix -urak, which contributes the Declension IM specification.

  18. 18.

    This is just one possible implementation of the masculine-as-default logic; for a somewhat different formal characterization of similar ideas, see Bobaljik and Zocca (2010).

  19. 19.

    For this to happen, the noun would, in addition to the DCL III (feminine) diacritic, have to be specified for the DCL IM diacritic, which would trigger masculine agreement. I argue there is no DCL IM diacritic, but even if there were, nouns can be specified only for one DCL diacritic (i.e., declension), not two or three. DCL diacritics determine nominal case endings, and if it were possible for an inanimate noun to be specified simultaneously for two different DCL declensions, we would expect it to have a mixed type of case endings—e.g., some masculine and some feminine. This never happens, however; i.e., if a noun like knjiga ‘book’ (Declension III) ends in -u in accusative singular, its other case endings always come from the same set and are completely predictable (see Table 5).

  20. 20.

    The impoverishment rules in question would probably have to involve a more elaborate, binary-value, case feature system, such as those developed in Calabrese (2006) and Halle and Vaux (1997), for instance. A detailed presentation of such an approach is, however, beyond the scope of this article.

  21. 21.

    An extreme case of this is undeclined nouns like Miki ‘Miki’ or Meri ‘Marry’, discussed in Wechsler and Zlatić (2001), which are similar to Declension IV nouns. These nouns (generally female loan names) have just one form, but they trigger regular feminine agreement on adjectives, just like standard Declension III nouns (see Table 5):

    1. (i)

    On the current analysis, these nouns are specified for a full set of features at the point when agreement happens, so that features like [FEM], [SG], and [ACC] are copied on the agreement target. However, in the case of this particular set of nouns, all features in their inflectional suffix are deleted prior to VI, which results in the complete feature neutralization in their form; i.e., these nouns will have a single morpho-phonological form (there are also undeclined adjectives like braon ‘brown’ or fer ‘fair’ and similar assumptions can be made for them as well). Wechsler and Zlatić (2001) also observe that undeclined nouns are incompatible with oblique cases like dative or instrumental, unless they are modified by a regularly inflected adjective (see (ii) below). Note that this kind of incompatibility does not arise when the oblique case is assigned by a preposition.

    1. (ii)

    This shows that oblique cases like instrumental and dative (assigned by verbs) are subject to a condition whereby they must be morpho-phonologically realized at PF within the NP they are assigned to (see Despić 2013). Such a condition would in spirit be very similar to Wechsler and Zlatić’s (2001:550) Case Realization Constraint and would create the same effect. For more on the difference between oblique cases assigned by verbs and prepositions see Despić (2013), Bošković (2006), Franks (1995) etc.

  22. 22.

    Note that Croatian behaves differently from Serbian in this respect. That is, in contrast to the majority of Serbian speakers, who reject the masculine agreement in plural, the majority of Croatian speakers actually seem to prefer it to the feminine pattern. This indicates that the markedness constraint in (43a) does not apply in Croatian, which shouldn’t be surprising given the discussion of the contrast between Serbian and Russian from above. That is, whether or not a markedness constraint and a related impoverishment rule will apply in a language may depend on a number of different factors, which I don’t have much to say about (recall that even in Serbian there are some speakers (3 out of 42 in this study), who choose the masculine pattern in plural). However, this analysis makes a clear prediction about the direction in which markedness may accumulate: there shouldn’t be any speakers (Croatian or Serbian), who strictly follow the declension (feminine) agreement in singular (e.g., lepa vojvoda) and choose strictly masculine (semantically based) agreement in plural (e.g., lepi vojvode)—to the best of my knowledge this prediction is borne out.

  23. 23.

    Sometimes a particular number value on the suffix may trigger morpho-phonological processes on the root/stem of certain nouns, but that is a completely separate issue; e.g., the root/stem of the masculine noun rob ‘slave’ is extended by -ov in plural (i.e., rob-ov-i), while that does not happen with masculine nouns like konj ‘horse’ (i.e., konj-i) or zub ‘tooth’ (i.e., zub-i).

  24. 24.

    For example, if exclusively formal features are properties of vocabulary items, which are inserted post-syntactically, while exclusively semantic features are present in syntax, then agreement with these two types of features would happen in separate components of the grammar, which would (at least in part) derive (53). But, (53) is clearly compatible with any system that assumes a deep grammatical division between these two types of features (Index/Concord in HPSG, or interpretable/uninterpretable in Minimalism). I leave these questions for future research.

  25. 25.

    Recall from the discussion in Sect. 2.2.1 (Table 1) that the feminine singular pattern is somewhat preferred in examples like (46)–(47). At this point I can only speculate that a possible cause of this might be analogy with prenominal agreement, which is exclusively feminine singular.

  26. 26.

    It could be the case that similarly to Declension IV nouns (see Sect. 3.3.2), at the point when agreement happens, 1st and 2nd person pronouns are specified for all features, including gender, but that gender is deleted prior to VI in the context of 1st and 2nd person features. Gender would then be neutralized in the pronoun’s form, but it would be visible in the form of the agreeing adjective (unless further impoverishment happens, as with plural non-nominative adjectives).

  27. 27.

    3rd person pronouns encode gender distinction, which can be easily accounted for by assuming that 3rd person is unmarked and therefore does not trigger accumulation of markedness. Also, since pronouns do not have declensions, gender that appears with pronouns is always semantic; i.e., there are no mismatches of the sort we see with the vojvoda-type nouns:

    1. (i)
      a. On je došao. b. Ona je došla.
        He AUX.3 arrived.M.SG   She AUX.3 arrived.F.SG
        ‘He arrived.’   ‘She arrived.’
      c. *On je došla. d. *Ona je došao.
        He AUX.3 arrived.F.SG   She AUX.3 arrived.M.SG
  28. 28.

    This would of course have to be independently supported; for instance, one could expect predicate phrases in French to be different.

  29. 29.

    Wechsler and Hahm (2011) also note that (75a) (their (53b)) is possible for some speakers. Native speakers I consulted find examples like (75a) much better than (74). Many of them allow both (75a) and (75b), but some of them prefer (75b), possibly due to analogy with the contrast in the primary predicate agreement pattern (e.g., (58)).

  30. 30.

    Coordination of plural conjuncts in Serbian (and other Slavic languages) has been examined in detail, especially in the context of first and/or last conjunct agreement (Arsenijević and Mitić 2015; Bošković 2009; Marušič et al. 2007 and references therein). For agreement with CoordPs consisting of singular number conjuncts see Wechsler and Zlatić (2003:Chap. 8) and references therein.

  31. 31.

    Binary feature systems have been proposed for other domains as well. For instance, Nevins (2011) argues that the number system in languages with singular, plural and dual are based on features [±singular], [±augmented], where dual is be represented with the combination [–singular, –augmented] (for a similar type of analysis of languages with singular, plural and paucal see Despić 2013). In the domain of person, binary feature systems based on [±speaker], [±hearer] (Bobaljik 2008 and references therein) or [±participant], [±author] (Nevins 2007; Harbour 2006) have been proposed.

  32. 32.

    As discussed in Franks (1995) and Wechsler and Zlatić (2003:Chaps. 7 and 8), Serbian QPs like pet glumica ‘five actresses(gen)’ also trigger neuter singular agreement. Furthermore, CoordP consisting only of QPs of this type also triggers neuter singular agreement in Standard Serbian. I interpret this to mean that CoordP based only on QP-conjuncts is not automatically assigned plural number by an independent mechanism, in contrast to regular NP-coordination. In other words, CoordP based exclusively on QP-conjuncts is a QP itself and triggers QP-like agreement (while CoordP based on NP-conjuncts has NP-like properties). The reason why this is neuter singular, I suggest, is that QPs are simply not specified for features like [GEN] and [NUM] that characterize regular NPs, which should not be surprising—QPs do not have declensions and cannot be pluralized (or de-pluralized). Thus, they are expected to trigger the same type of agreement as infinitives in (83), for instance. Alternatively, one could argue that they are underlyingly specified for features neuter singular (e.g., Franks 1995:115 but see Wechsler and Zlatić 2003:163 for some criticism of this view). I leave further exploration of these possibilities for future work, since morpho-syntactic properties of Serbian and generally Slavic QPs are notoriously complex and are clearly outside of the scope of this paper (for discussion of QPs and paucal number in Serbian, see Despić (2013); for the relationship between QPs and case see Bošković (2006) and the abovementioned references).

  33. 33.

    This point becomes particularly clear in the domain of binary-value person features. Thus, [+speaker] is not a simple identification with the speaker, i.e., inclusive [+speaker, +hearer] does not mean ‘someone who is the speaker and the hearer at the same time’ (see Bobaljik 2008 and Watanabe 2013).

  34. 34.

    Recall from Sect. 3.3.1 that the apparent neuter vs. non-neuter distinction arises in Serbian accusative plural, but this is qualitatively different from the two-way gender neutralization in Polish, Slovak and Slovenian, which happens in nominative non-singular, and covers both adjectives and pronouns. Serbian nominative has the standard three-way gender distinction in both singular and plural, and the neuter vs. non-neuter contrast in accusative plural is an effect of two independent principles: (i) the general gender neutralization in plural non-nominative, which, among other things, removes gender distinction between masculine and feminine in accusative plural, and (ii) the general nominative = accusative syncretism in neuter, which makes neuter plural accusative syncretic with neuter plural nominative, by removing the [ACC] feature before the gender impoverishment can apply (see again (39)). But as the discussion of Serbian pronouns from Sect. 3.3.1 shows, gender impoverishment is complete even in accusative plural, exactly when there is no nominative = accusative syncretism in neuter singular; i.e., the nominative singular neuter pronoun ono is non-syncretic with accusative singular pronoun (nje)ga, and therefore gender is completely neutralized in accusative plural (i.e., nj(ih) for all three genders).

  35. 35.

    For instance, replacing [NEUT] with [–masculine, –feminine] would not undermine the common assumption that among gender values neuter is most marked; in fact, Nevins (2011) convincingly shows that dual, which is also represented with a combination of two minus values ([–singular, –augmented]), is more marked than singular and plural.

  36. 36.

    Similar can be said about double-mismatch hybrids braća ‘brothers’ and deca ‘children’, which also trigger the default masculine plural agreement when they are coordinated with feminine singular nouns. That is, even though they have [FEM] [SG] formal features, they do not trigger feminine plural agreement when they are coordinated with regular feminine singular nouns:

    1. (i)

    The reason is quite simple: like tata in (89b–c), these nouns are marked for two different gender values (semantic and formal) and, therefore, whenever they are coordinated they automatically introduce conflicting gender information within CoordP. As a result, default masculine is triggered. This effect is most obvious when braća and deca are not the last conjunct, because of the “last conjunct agreement” effect. That is, in certain context the participle can agree exclusively with the last conjunct of a CoordP (instead of the whole CoordP) if the last conjunct is plural (this has been the subject of extensive research in recent years; see Arsenijević and Mitić 2015; Bošković 2009, Marušič et al. 2007 and references therein). Since braća and deca are hybrids that do have a [PL] specification (unlike tata or vojvoda), to control for this effect, it is more appropriate not to put them in the last conjunct.

  37. 37.

    Given the general goal of this paper, I have to leave spelling out technical details of the formal relationship between [PL]SEM and [PL]FORM for some other occasion. What is important for our purposes is that [PL]FORM is always morpho-phonologically identical to [PL]SEM (there’s no special [PL]FORM inflectional suffix different from [PL]SEM) and that [PL]FORM always entails [PL]SEM; that is, whenever the [PL]FORM interpretation is possible, the [PL]SEM interpretation is available as well (as in (91a)). The opposite is not true: there are cases like (93) where just [PL]SEM is available. Also, there are no case where just [PL]FORM is possible; i.e., the opposite of (93). This clearly indicates that [PL]SEM is the norm and that [PL]FORM is an extra option available in certain contexts (e.g., (91a)). The statement in (95) simply says that this extra option is not available in the context of accumulated markedness created by the presence of non-nominative case, which is consistent with the fact that other features, like gender, are neutralized in the same context.

  38. 38.

    Note that the general nominative = accusative syncretism in neuter does not have any bearing on the facts involving the honorific pronoun, whose use is pragmatically restricted to male and female individuals.

  39. 39.

    Note that facts similar to those in (99)–(100) have also been reported independently by Arsenijević (2014); I believe that the analysis developed here can be extended to them as well.

  40. 40.

    For such speakers the patterns in Tables 17 and 18 seem to co-exist. Speakers who would accept only the pattern in Table 18 and reject the one in Table 17 would be predicted to find (99a) possible only with the honorific interpretation; I haven’t been able to find such speakers (which should not be surprising given that the pattern in Table 18 is somewhat old-fashioned).

  41. 41.

    Note that the statement in (95), in particular the part that [PL]FORM is unavailable in the context of non-nominative cases, would not necessarily have to hold for other Slavic languages, given that they vary with respect to markedness thresholds (as discussed in Sect. 3). But if it did hold, one possible prediction would be that Slavic languages that neutralize gender in nominative plural, would prefer singular (mixed) agreement with the polite plural pronoun, since the nominative form of plural adjective/participle would also be limited to [PL]SEM, given markedness considerations. Studies done by Corbett (1983, 2000, 2006), as discussed in Wechsler (2011:1001–1002), suggest that this might be true: Bulgarian, Macedonian, Ukrainian, Belorussian, and Russian (long form adjectives) all favor the mixed agreement pattern. Russian short form adjectives, on the other hand, prefer plural agreement but they are special in many ways. Their syntactic distribution is limited and they preserve only the nominal endings of the nominative case (Timberlake 1993:845). Thus, they only have 4 forms: 3 singular forms for each gender and one plural form. But given this, the plural form might be analyzed as an elsewhere form (along the lines of Serbian instrumental; Table 17) without any feature specification, including plural, which would make it compatible with the honorific pronoun. This is, of course, just a speculation at this point and I leave a more careful investigation of these issues for future work.


  1. Arregi, K., & Nevins, A. (2012). Morphotactics: Basque auxiliaries and the structure of Spellout. Dordrecht: Springer.

    Book  Google Scholar 

  2. Arsenijević, B. (2014). The agreement of adjectives with the honorific pronoun in Serbo-Croatian. Philologia Mediana, VI/6, 39–50.

    Google Scholar 

  3. Arsenijević, B., & Mitić, I. (2015). On the (in)dependence of gender with respect to number in agreement with coordinated subjects: An experimental study. Journal of Slavic Linguistics, 24(1), 41–69.

    Article  Google Scholar 

  4. Bailyn, J. (2001). The case of slavic predicates. In A. Strigin et al. (Ed.), ZAS occasional papers in linguistics (pp. 1–26). Berlin: Zentrum fur allgemeine Sprachwissenschaft.

    Google Scholar 

  5. Baker, M. C. (2008). The syntax of agreement and concord. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  6. Bobaljik, J. D. (2008). Missing persons: A case study in morphological universals. The Linguistic Review, 25, 203–230.

    Article  Google Scholar 

  7. Bobaljik, J. D. (to appear). Distributed morphology, In Encyclopaedia.

  8. Bobaljik, J. D., & Zocca, C. L. (2010). Gender markedness: The anatomy of a counter-example. Morphology, 21, 141–166.

    Article  Google Scholar 

  9. Bonet, E. (1991). Morphology after syntax: Pronominal clitics in Romance. Doctoral dissertation, MIT, Cambridge, MA.

  10. Bonet, E. (1995). Feature structure of Romance clitics. Natural Language and Linguistic Theory, 13, 607–647.

    Article  Google Scholar 

  11. Bošković, Ž. (2006). Case checking vs. case assignment and the case of adverbial NPs. Linguistic Inquiry, 37, 522–533.

    Article  Google Scholar 

  12. Bošković, Ž. (2009). Unifying first and last conjunct agreement. Natural Language and Linguistic Theory, 27(3), 455–496.

    Article  Google Scholar 

  13. Bowers, J. (1993). The syntax of predication. Linguistic Inquiry, 24, 591–656.

    Google Scholar 

  14. Bowers, J. (2001). Predication. In M. Baltin & C. Collins (Eds.), The handbook of contemporary syntactic theory (pp. 299–333). Cambridge: Blackwell.

    Chapter  Google Scholar 

  15. Calabrese, A. (2005). Markedness and economy in a derivational model of phonology. Berlin: de Gruyter.

    Book  Google Scholar 

  16. Calabrese, A. (2006). On absolute and contextual syncretism. Remarks on the structure of paradigms and on how to derive it. In A. Bachrach & A. I. Nevins (Eds.), Inflectional identity (pp. 156–205). London: Oxford University Press.

    Google Scholar 

  17. Calabrese, A. (2008). On markedness and underspecification in morphology. Ms., University of Connecticut.

  18. Calabrese, A. (2011). Investigations on markedeness, syncretism and zero exponence in morphology. Morphology, 21(2), 283–325.

    Article  Google Scholar 

  19. Carminati, M. N. (2005). Processing reflexes of the feature hierarchy (person > number > gender) and implications for linguistic theory. Lingua, 115, 259–285.

    Article  Google Scholar 

  20. Comrie, B. (1975). Polite plurals and predicate agreement. Language, 51(2), 406–418.

    Article  Google Scholar 

  21. Corbett, G. (1983). Hierarchies, targets and controllers: Agreement patterns in Slavic. London: Croom Helm.

    Google Scholar 

  22. Corbett, G. (1991). Gender. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  23. Corbett, G. (2000). Number. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  24. Corbett, G. (2006). Agreement. Cambridge: Cambridge University Press.

    Google Scholar 

  25. Croft, W. (1990). Typology and universals. Cambridge: Cambridge University Press.

    Google Scholar 

  26. Despić, M. (2010). Markedness and marked features in Serbian. In W. Browne, A. Cooper, A. Fisher, E. Kesici, N. Predolac, & D. Zec (Eds.), Proceedings of the formal approaches to Slavic linguistics 18 (FASL 18): The second Cornell meeting (pp. 1–20). Ann Arbor: University of Michigan, Michigan Slavic Publications.

    Google Scholar 

  27. Despić, M. (2013). A note on paucal, agreement and case. In A. Podobryaev (Ed.), Proceedings of the formal approaches to Slavic linguistics 20 (FASL 20): The second MIT meeting (pp. 57–72). Ann Arbor: University of Michigan, Michigan Slavic Publications.

    Google Scholar 

  28. Despić, M. (2016). Coordinating gender: What can coordinate structure agreement tell us about gender? Studies in Polish Linguistics, 11(1), 1–25.

    Google Scholar 

  29. Embick, D. (2000). Features, syntax, and categories in the Latin perfect. Linguistic Inquiry, 31, 185–230.

    Article  Google Scholar 

  30. Embick, D., & Noyer, R. (2007). Distributed morphology and the syntax-morphology interface. In G. Ramchand & C. Reis (Eds.), The Oxford handbook of linguistic interfaces (pp. 289–324). Oxford: Oxford University Press.

    Google Scholar 

  31. Franks, S. (1995). Parameters of Slavic morphosyntax. New York: Oxford University Press.

    Google Scholar 

  32. Friedman, V. A. (1993). Macedonian. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 249–305). London: Routledge.

    Google Scholar 

  33. Greenberg, J. (1966). Language universals, with special reference to feature hierarchies. The Hague: Mouton.

    Google Scholar 

  34. Halle, M. (1997). Distributed morphology: Impoverishment and fission. In B. Bruening, Y. Kang, & M. McGinnis (Eds.), MIT working papers in linguistics (pp. 425–449). Cambridge: MIT.

    Google Scholar 

  35. Halle, M., & Marantz, A. (1993). Distributed morphology and the pieces of inflection. In K. Hale & S. J. Keyser (Eds.), The view from building 20 (pp. 111–176). Cambridge: MIT Press.

    Google Scholar 

  36. Halle, M., & Vaux, B. (1997). Theoretical aspects of Indo-European nominal morphology. In J. Jasanoff, C. Melchert, & L. Oliver (Eds.), Mír curad: Studies in honor of Calvert Watkins (pp. 223–240). Innsbruck: Innsbrucker Beiträge zur Sprachwissenschaft.

    Google Scholar 

  37. Haspelmath, M. (2006). Against markedness (and what to replace it with). Journal of Linguistics, 42, 25–70.

    Article  Google Scholar 

  38. Harbour, D. (2006). Person hierarchies and geometry without hierarchies or geometries. Queen Mary’s Occasional Papers Advancing Linguistics 6. Available at

  39. Harley, H., & Ritter, E. (2002). Person and number in pronouns: A feature-geometric analysis. Language, 78, 482–526.

    Article  Google Scholar 

  40. Heim, I. (2008). Features on bound pronouns. In D. Harbour, D. Adger, & S. Béjar (Eds.), Phi theory (pp. 35–56). Oxford: Oxford University Press.

    Google Scholar 

  41. Herrity, P. (2000). Slovene: A comprehensive grammar. London: Routledge, Taylor and Francis.

    Google Scholar 

  42. Ingram, D. (1978). Typology and universals of personal pronouns. In J. H. Greenberg (Ed.), Universals of human language, Vol. III. Word structure (pp. 213–248). Stanford: Stanford University Press.

    Google Scholar 

  43. Jakobson, R. (1941). Kindersprache, Aphasie und allgemeine Lautgesetze. Uppsala: Almqvist & Wiksell.

    Google Scholar 

  44. Jakobson, R. (1984). Structure of the Russian verb. In L. R. Waugh & M. Halle (Eds.), Russian and Slavic grammar—studies 1931–1981. Berlin: de Gruyter (Translation of “Zur Struktur des Gender markedness russischen Verbums” in Charisteria Gvilelmo Mathesio qvinqvagenario a discipulis et Circuli Lingvistici Pragensis sodalibus oblata, Prague, 1932.).

    Chapter  Google Scholar 

  45. Kiparsky, P. (1973). “Elsewhere” in phonology. In S. R. Anderson & P. Kiparsky (Eds.), A Festschrift for Morris Halle (pp. 93–106). New York: Holt, Reinhart and Winston.

    Google Scholar 

  46. Malovrh, P. A., & Lee, J. F. (2010). Connections between processing, production and placement: Acquiring object pronouns in Spanish as a second language. In B. VanPatten & J. Jegerski (Eds.), Second language processing and parsing (pp. 231–256). Amsterdam: Benjamins.

    Chapter  Google Scholar 

  47. Marušič, F., Nevins, A., & Saksida, A. (2007). Last conjunct agreement in Slovenian. FASL, 16, 210–227.

    Google Scholar 

  48. Mayo, P. (1993). Belorussian. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 887–946). London: Routledge.

    Google Scholar 

  49. Nevins, A. (2007). The representation of third person and its consequences for person-case effects. Natural Language and Linguistic Theory, 25, 273–313.

    Article  Google Scholar 

  50. Nevins, A. (2011). Marked targets versus marked triggers and impoverishment of the dual. Linguistic Inquiry, 42(3), 413–444.

    Article  Google Scholar 

  51. Noyer, R. (1997). Features, positions and affixes in autonomous morphological structure. New York: Garland Publishing. Revised version of 1992 MIT doctoral dissertation.

    Google Scholar 

  52. Noyer, R. (1998). Impoverishment theory and morphosyntactic markedness. In S. G. Lapointe, D. K. Brentari, & P. M. Farrell (Eds.), Morphology and Its Relation to Syntax (pp. 264–285). Stanford: CSLI.

    Google Scholar 

  53. Pollard, C., & Sag, I. (1994). Head-driven phrase structure grammar. Stanford/Chicago: CSLI/University of Chicago Press.

    Google Scholar 

  54. Priestly, T. M. S. (1993). Slovene. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 388–454). London: Routledge.

    Google Scholar 

  55. Sadowska, I. (2012). Polish: A comprehensive grammar. London: Routledge/Taylor and Francis.

    Google Scholar 

  56. Scatton, E. A. (1993). Bulgarian. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 188–247). London: Routledge.

    Google Scholar 

  57. Silverstein, M. (1985). Hierarchy of features and ergativity. In P. Muysken & H. van Riemsdijk (Eds.), Features and projections (pp. 163–233). Dordrecht: Foris.

    Google Scholar 

  58. Short, D. (1993a). Czech. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 455–532). London: Routledge.

    Google Scholar 

  59. Short, D. (1993b). Slovak. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 533–592). London: Routledge.

    Google Scholar 

  60. Shevelov, G. Y. (1993). Ukrainian. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 957–998). London: Routledge.

    Google Scholar 

  61. Stanojčić, Ž., & Popović, L. (1992). The grammar of the Serbian language [gramatika srpskoga jezika]. Beograd: Zavod za udžbenike i nastavna sredstva.

    Google Scholar 

  62. Stevanović, M. (1962). The grammar of Serbo-Croatian for senior grades of high school [Gramatika srpskohrvatskog jezika za više razrede gimnazije]. Cetinje: Obod.

    Google Scholar 

  63. Swan, O. E., & Gálová-Lorinc, S. (1990). Beginning Slovak. Columbus: Slavica Publishers, Inc.

    Google Scholar 

  64. Timberlake, A. (1993). Russian. In B. Comrie & G. Corbett (Eds.), The Slavonic languages (pp. 827–886). London: Routledge.

    Google Scholar 

  65. Watanabe, A. (2013). Person-number interaction: impoverishment and natural classes. Linguistic Inquiry, 44(3), 469–492.

    Article  Google Scholar 

  66. Wechsler, S. (2011). Mixed agreement, the person feature, and the index/concord distinction. Natural Language and Linguistic Theory, 29, 999–1031.

    Article  Google Scholar 

  67. Wechsler, S., & Hahm, H.-J. (2011). Polite plurals and adjective agreement. Morphology, 21, 247–281.

    Article  Google Scholar 

  68. Wechsler, S., & Zlatić, L. (2000). A theory of agreement and its application to Serbo-Croatian. Language, 76, 799–832.

    Article  Google Scholar 

  69. Wechsler, S., & Zlatić, L. (2001). Case realization and identity. Lingua, 111, 539–560.

    Article  Google Scholar 

  70. Wechsler, S., & Zlatić, L. (2003). The many faces of agreement. Stanford: CSLI.

    Google Scholar 

  71. Zwicky, A. M. (1978). On markedness in morphology. Die Sprache, 24, 129–143.

    Google Scholar 

Download references


For discussion of some of the ideas presented here, I wish to thank Andrea Calabrese, Jonathan Bobaljik, the audiences at the FASL 18 Conference (Cornell, May 2009), the Utrecht Syntax Interface Lecture Series (October 2016), the Göttingen Linguistics Colloquium (October 2016) and the Cornell Linguistics Colloquium (February 2017), as well as the participants of my seminars at the Cornell. For assistance with data collection I am very thankful to Alen Bešić, Biljana Čubrović, Ivanka Jelić and Ivana Mitić. Finally, I want to thank two anonymous reviewers and the Editors for their careful and helpful suggestions.

Author information



Corresponding author

Correspondence to Miloje Despić.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Despić, M. Investigations on mixed agreement: polite plurals, hybrid nouns and coordinate structures. Morphology 27, 253–310 (2017).

Download citation


  • Mixed agreement
  • Markedness and marked features
  • Hybrid nouns
  • Honorific pronouns
  • Index/Concord