Differential marking of direct objects in Monastirli džudezmu: a case study in Judeo-Spanish morphosyntax
-
Alice Corr
Abstract
This article presents a case study on the differential marking of direct objects (DOM) in Balkan Judeo-Spanish, an endangered language (sub)group that still lacks detailed, systematic documentation of its (comparative) morphosyntax. Countering recent claims that the phenomenon is frequently absent in Judeo-Spanish, I demonstrate the robust presence of so-called a-marking as a highly systematic, multi-dimensional DOM strategy in early 20th century fieldwork recordings from the understudied dialect of Monastir (present-day Bitola, North Macedonia). I show that, in our corpus, a-marking in transitive (S)V(S)O(S) structures is primarily regulated by (grammatical) animacy/person and (syntactic) definiteness, such that indefinite DOs are excluded and specificity plays no role. Theoretically, the empirical distribution of a-marking broadly conforms with, yet – crucially – cannot be fully subsumed under, scale-based hierarchies that model DOM in terms of referential prominence. Rather, our findings support the conclusion that a-marking of the DO occurs if and only if the argument is syntactically specified for animacy/person and definiteness. From the typological perspective, the (non-)coincidence of Monastirli a-marking with a co-referential accusative clitic exhibits a hitherto undescribed distribution distinct from other patterns documented for (Balkan) Romance. The present article thus not only advances description of, and inquiry into, the cross-linguistic landscape of DOM, but contributes more broadly to redressing the conspicuous absence of empirical and theoretical investigation into Judeo-Spanish dialect syntax.
1 Introduction
This article examines the morphosyntactic marking of direct object (DO) arguments in a west Balkan dialect of džudezmu (‘Ladino/Judeo-Spanish’),[1] based on a hand-annotated corpus of early 20th century oral texts. In the Romance group, differential object marking (DOM), a cross-linguistic grammatical strategy for differentiating subsets of DOs, has classically been understood to be realized through an adnominal ‘flagging’ (Haspelmath 2005, 2019) mechanism, viz. the so-called prepositional accusative (Diez 1882; Meyer-Lübke 1899; Spitzer 1928; Rohlfs 1971), that morphosyntactically distinguishes highly-referential, animate DOs to the general exclusion of inanimate arguments. Since the prepositional status of this prenominal element has been contested (Torrego 1998; Ormazabal/Romero 2013; Hill/Mardale 2021, i.a.), in this article I will henceforth use the term adnominal DOM or a-marking (where a < Latin ad ‘to(wards), at’), instead of prepositional accusative. This animacy-based marking mechanism can also optionally co-occur with a co-referential clitic, a pattern sometimes referred to as differential object indexing (DOI, following Iemmolo 2010), although the distribution of DOI in Romance is much more restricted, typically occurring only in a subset of contexts in which adnominal DOM is available.
In the Ibero-Romance group, the robust system of DOM attested in Castilian/Spanish (Central Ibero-Romance)[2] is considered to be a distinguishing grammatical characteristic of these varieties, especially vis-à-vis Western and Eastern IberoRomance varieties (e.g. Catalan, Portuguese) in which the availability of the phenomenon is more restricted. Despite 500 years of little to no contact following the expulsion of Sephardim from Spain in 1492, a popular and scholarly impression persists that Judeo-Spanish and Spanish/Castilian morphosyntax remain broadly commensurable. Yet, contrary to a productive and enduring academic interest in the properties of Spanish DOM and its internal (both dialectal and diachronic) and external (e.g. cross-linguistic) variation, the availability and distribution of the phenomenon in Judeo-Spanish has gone almost entirely unremarked in the literature.
The overarching aim of this study is to redress the absence of empirical investigation into both the availability of DOM and its structural characterization in Judeo-Spanish (cf. §1.1). Although adnominal DOM was well established in Old Spanish, and on the increase, in the centuries prior to the 1492 expulsion (Laca 2006; Romero Heredero 2021), Bunis (2015, 403; 2021 a, 402) has claimed that a-marking is “frequently absent” in Judeo-Spanish as early as the beginning of the 17th century. Following up on these claims, recent fieldwork conducted by the present author presents a different empirical picture in which DOM remains widespread amongst today’s Ladino speakers,[3] albeit with a non-trivial degree of heterogeneity in the data patterns observed.
To clarify the empirical scenario for the Judeo-Spanish group, in this article I confine our attention to the structural profile of adnominal DOM in a single dialect of the language in its (late) modern period (ca. 1914‒present, per Bunis’s 2021 b, §3.3 periodization). The linguistic material of the present article draws from a set of oral texts collected in 1927 by the American philologist Max A. Luria during fieldwork undertaken with the Sephardic community of Monastir (present-day Bitola), located in the southwestern region of what is today known as North Macedonia, close to its border with present-day Greece. Monastir (or Manastır) was a town and province in the former Ottoman Empire in which Iberian Jews settled from the 1500 s onwards (Benbassa/Rodrigue 2000, 9; Vidaković-Petrov 2011, 550; Mihajlovski 2021, 146–150) following the Jewish population’s expulsion from Spain and Portugal in the 15th‒17th centuries. Amongst its speakers, the Judeo-Spanish dialect of Monastir was referred to as džudezmu (cf. fn1), a designation I will use here interchangeably with Judeo-Spanish/Ladino when referring to the linguistic branch of that name, adopting the adjective Monastirli to indicate the local variety. Once a thriving regional centre of Ottoman Sephardim, Jewish life – and, in turn, the linguistic vitality of Monastirli dialect – was progressively and rapidly curtailed during the Holocaust, coming to a permanent end following the March 1943 deportation of Monastir’s Jewish population to Treblinka death camp. As such, this study involves the examination of the last (and only) monographic work on the dialect based on in situ fieldwork with džudezmu-speaking Sephardim from Monastir.[4]
1.1 Aims and claims
As part of a broader endeavour to advance investigation into the (comparative) grammar of Judeo-Spanish, an endangered language (branch) which still lacks a detailed, systematic description of its (comparative) morphosyntax (Corr 2020), the present article reports the results of a case study into the (differential) adnominal coding of DO arguments in a newly-compiled, manually-annotated corpus derived from early 20th century fieldwork materials from Monastir recorded in Luria (1930).
Empirically, the goal of the present article is to establish the distribution of Monastirli a-marking and the nominal parameters that regulate its occurrence. Theoretically, its purpose is to evaluate in the first instance a) whether the strategy is operative as an adnominal DOM strategy on a par with well-studied systems elsewhere in Romance and cross-linguistically, and, if so, b) how this system can be modelled in accordance with current approaches to (adnominal) DOM. Beyond these main objectives, I further augment our study by identifying the (non-)coincidence of Monastirli a-marking with a co-referential accusative clitic.
In this article, I make the following key claims. First, this study confirms the robust presence of a-marking as a highly systematic, multidimensional DOM strategy in the spoken language of Monastirli džudezmu. Specifically, I show that a-marking in transitive (S)V(S)O(S) structures is licensed as a function of the animacy/person and definiteness of the DO, such that indefinite DOs are systematically unmarked and specificity plays no role. Second, I show that these nominal factors are operative as DOM triggers in terms of their grammatical, rather than semantic, properties. Despite the broad conformity of our findings with scale-based hierarchies that capture DOM in terms of referential prominence (cf. §2.1), the main theoretical claim is that the licensing of DOM in Monastirli is regulated grammatically, and warrants modelling in those terms. Specifically, I propose that a-marking of the DO occurs iff the argument is syntactically specified for animacy and definiteness (respectively associated with person and D in formal frameworks, although I do not pursue a formal account here). Finally, this study makes a typological contribution by the attendant finding – to be developed in future research – that the (non-)coincidence of a-marking and a co-referential clitic in Monastirli exhibits an empirical distribution that is predicted, but hitherto undescribed, diverging from other documented patterns in the (Balkan) Romance group.
1.2 Structure of the article
The present article is structured as follows. Section 2 surveys the mechanisms and regulating factors of DOM in (Judeo-)Spanish and cross-linguistically, focusing on nominal triggers that condition a-marking in (Judeo-)Spanish. Section 3 details the methodological procedures and data sources used in the study. Section 4 sets out a characterization of the structural properties and distribution of DOM in our corpus in terms of the nominal-internal syntactic-semantic features of the DO. Section 5 offers general remarks and discusses the wider implications of our findings in relation to nominal parameters, referential scales and the theoretical modelling of Monastirli DOM. Section 6 concludes.
2 Background
2.1 Differential argument marking
Extensive theoretical and typological investigation has established the robust presence of differential argument (i.e. not necessarily object) marking strategies across the world’s languages (Bossong 1982, 1985; Sinnemäki 2014; Seržant/Witzlack-Makarevich 2018; Börstell 2019, i. a.). The phenomenon is nonetheless subject to considerable variation both in terms of the morphosyntactic mechanisms deployed, and the triggers of, DOM/DAM (for ‘differential argument marking’). Haspelmath (2005, 2013, 2019) identifies two major strategies for overt argument marking cross-linguistically: flagging on the (nominal) argument itself via case-markers and adpositions, and indexing at the verbal or clausal level. The majority of existing studies have considered DOM/DAM in terms of the former, and to the exclusion of the latter (Haspelmath 2021, 131), although whether or not (differential) argument indexing can or should be subsumed under the banner of DOM/DAM has not been resolved in the literature.[5] To avoid introducing further complexity into the investigation, I will treat them separately in the present article.
In terms of triggers, it has been well established that the licensing of DOM/DAM can be sensitive not only to the ‘local’ (Laca 2006) semantico-syntactic properties of the nominal expression (e.g. animacy, definiteness; nominal type and category), but also ‘global’ syntactic properties beyond the NP/DP itself.[6] These include (in Spanish and elsewhere) factors relating to the verbal event, e.g. verb class, transitivity, affectedness, telicity/aspect (Bello 1847; Pottier 1968; Hopper/Thompson 1980; Torrego 1998; Næss 2004, i. a.); the complexity of the predicate/clause type (e.g. clause union contexts, López 2012); as well as information-structural/discourse-configurational factors (e.g. topicalization) at a higher structural level (Laca 1995; Escandell Vidal 2009; Dalrymple/Nikolaeva 2011). Descriptive – and theoretical – characterizations of DOM/DAM are further complicated by dialectal, stylistic and idiolectal differences regarding when the mechanism can, or must, be deployed.
The gradient nature of the expression of DOM/DAM has profitably been captured by theoretical models invoking scales (e.g. Aissen 2003; Silverstein 1976; cf. Witzlack-Makarevich/Seržant 2018 for a detailed overview and references). These model the empirical correlation between the referential prominence (Haspelmath 2021) of the (here) DO and (the likelihood of) its differential coding, whereby the higher an argument places on a given scale, the greater the likelihood of DOM. The notion of referential prominence offers an umbrella term which encompasses the various semantico-syntactic properties identified as possible regulating factors in DOM, including animacy, person, definiteness, specificity and topic/discourse accessibility, as summarized in (1a–c):
(1) Types of scales for DOM (from Irimia/Pineda 2022, 343)
a. Animacy/person
1/2 > 3 > proper name > human > animate > inanimate
b. Specificity/definiteness
pronoun > name > definite > specific indefinite > non-specific
c. Topic accessibility scale
active > accessible > unused > brand-new anchored > brand-new unanchored
These scales, while in many ways a descriptive tool, have a certain degree of predictive power, to the extent that they represent implicational hierarchies. (On the explanatory (non)significance of these models, cf., e.g. Bickel/Witzlack-Makarevich 2008; Klein/de Swart 2011, 4.) Theoretically, functional motivations in terms of prominence and/or disambiguation predominate in the scholarship on DOM, although an alternative treatment is favoured in the formal literature, in which explanation for the phenomenon is sought in terms of its formal structural properties (for an overview of competing models of DOM under functional, typological and formal frameworks, cf. Bárány/Kalin 2020).
2.2 Adnominal DOM in general Spanish
In Spanish, as with a number of other Romance varieties which license DOM,[7] differential marking of the DO is realized through a reflex of the Latin preposition ad ‘to(wards), at’, a prenominal (case-)marker[8] which is homophonous in many Romance varieties with the dative preposition. In general Spanish, the precise conditions responsible for regulating DOM continue to be debated in the literature. As a rule, the realization of a-DOM correlates with nominal-internal features of the DO; namely, (i) animacy, conventionally characterized as an intrinsic property of nominals (e.g. Klein/de Swart 2011, although cf. §5.2.1); and (ii) properties associated with a high degree of referentiality or individuation (Timberlake 1975, 124), as modelled via the prominence scales for animacy/person (1a) and specificity/definiteness (1b) (cf. also Aissen 2003, 459; Romero Heredero 2022, 2). For our purposes, definiteness will be treated as a syntactic property of nominals, linked on formal theoretical treatments with the syntactic feature D (cf. §5.2.2), although from an interpretative point-of-view it is prototypically associated with uniqueness (i.e. it denotes an entity which is presupposed to exist, where that entity is unique) and object-reference.[9] Specificity is conceptualized here primarily as a discourse-pragmatic property or effect having to do with the recoverability of a referent (cf. Enç 1991, 24), or, more broadly, referential anchoring (von Heusinger 2019, 98–102). Following Leonetti (2004, 77), a specific reading is obtained “when the choice of referent is presented as heavily constrained” whilst a non-specific reading occurs “when the choice is presented as being relatively free”.[10] Thus, whilst definiteness and specificity are interpretatively close (cf. e.g. Enç 1991), they are taken here as distinct, mutually independent properties.
Adnominal DOM in Spanish is understood to have developed out of an earlier topic-marking strategy (Pensado 1995), which, by the time of the 1492 expulsion, had developed into a two-dimensional DOM system combining person/animacy with (optional)[11] definiteness and (incipient)[12] specificity. Modern-day varieties of Castilian/Spanish have extended further along the (person/animacy) and definiteness/specificity scale, albeit to varying degrees. The general observation for Spanish/Castilian is that definite and/or specific, human DOs are invariably a-marked (2a–d), whereas as the individuation of an animate DO decreases, the requirement for DOM declines in step (2d–f).
(2) a. La profesora te conoce *(a) ti.
the teacher you=knows dom you
‘The teacher knows you.’
b. La profesora conoce *(a) {Eliezer/ estos (alumnos)/ los alumnos}.
the teacher knows dom Eliezer these students the students
‘The teacher knows {Eliezer/these students/the students}.’
c. La profesora conoce *(a) todos (los estudiantes).
the teacher knows dom all the students
‘The teacher knows all the students.’
d. La profesora busca (a) {varios/unos/dos} alumnos.
the teacher knows dom several/some/two students
‘The teacher is looking for {a number of/some/two} [±specific] students.’
e. La profesora busca (a) estudiantes con buen expediente académico.
the teacher knows dom student with good record academic
‘The teacher is looking for [non-specific] students with a good academic record.’
f. La profesora busca (*a) estudiante con buen expediente académico.
the teacher knows dom student with good record academic
‘The teacher is looking for a [non-specific] student with a good academic record.’
The referentiality/individuation condition also manifests itself in quantified expressions, insofar as the ‘weak/strong’ distinction (Milsark 1974; McNally 2020) appears to be at work in the licensing of Spanish a-marking with animate DOs, as illustrated by obligatory a-marking of nominals introduced by ‘strong’ universal quantifiers (2c), versus its optionality with ‘weak’ existential quantifiers (2d). Notably, the interpretation of unmarked DOs is restricted to a non-specific, or ‘weak’ reading, but the reverse does hold; that is, a-marking, although it correlates with specificity, does not guarantee a specific, or ‘strong’, reading.
Traditional grammars formulate the animacy trigger in Spanish as a (loose) humanness restriction that allows for DOM with nonhuman referents on a case-by-case basis (e.g. personification of inanimate entities), involving a complex cluster of subjective criteria in the case of non-human animates (e.g. proximity-to-humanness, domestication, speaker affect for the referent and/or its class). Note, however, that this characterization overlooks various (other) semantico-referential and structural factors that interact with animacy in the licensing of DOM in general Spanish.[13] Amongst nominal triggers, for example, animacy properties seemingly override (in)definiteness/(non-)specificity with indefinite Q(uantifier)-pronouns (although the pronominal status of the DO is also a factor, cf. López 2012, 13–14, 63–64):
(3) a. No he visto *(a) nadie en todo el día.
not have.1sg seen dom nobody in all the day
‘I’ve not seen anyone all day.’
b. Quiero conocer *(a) alguien que me trate bien.
want.1sg know.infdom someone that me=treats.sbjv well
‘I want to meet someone who treats me well.’
By contrast, wider structural factors outside the direct object NP/DP can overrule the animacy generalization, as illustrated for general Spanish by the suppression of a-marking on animate DOs in possessive tener constructions (4a–b), and the licensing of a-marking with the same verb (4c) and with inanimates (4d) in secondary predication environments:[14]
(4) a. Tengo (*a) una mujer maravillosa.
have.1sgdom a woman marvellous
b. Tengo (a) la mujer más maravillosa del mundo.
have.1sgdom the woman most marvellous of.the world
c. Tengo *(a) la mujer más maravillosa aquí conmigo.
have.1sgdom the woman most marvellous here with.me
d. Veo al avión aterrizar.
see.1sgdom.the aeroplane land.inf
Adnominal DOM can also co-occur, under varying conditions, with a co-referential clitic (i.e. DOI, as introduced in §1). The licensing conditions for the indexing clitic are different from a-marking, and are typically understood to involve a dependency whereby the co-referential clitic is realized iff a-marking is already present on the object argument itself.[15] For example, clitic indexing of the DO is required alongside a-marking with personal pronouns (5a), whereas with proper names its availability varies dialectally (5b), such that it is considered ungrammatical in general Spanish, but permitted in many contemporary American Spanish varieties:
(5) a. (*Lo) vi a él.
acc.m = saw.1sg dom him
‘I saw him.’
b. (%Lo) vi a Mario.
acc.m = saw.1sg dom Mario
‘I saw Mario.’
The strategy developed in Spanish/Castilian from the 16th century onwards, thereby postdating the emergence of a-marking by several centuries (Melis 2018, 106, and references therein), and, crucially for our purposes, the watershed date of 1492, i.e. it would not have been a feature of the pre-expulsion dialects spoken by Iberian Jews.
2.3 Judeo-Spanish DOM
In Judeo-Spanish, adnominal DOM is morphologically realized, like Spanish, by a reflex of Latin ad. In most cases, it takes the form of the prenominal marker a, but also (unlike Spanish, though like Catalan) occurs as an before vowels (cf. also Quintana Rodríguez 2006, 130ss.). In Monastirli, it may also undergo vowel raising (from [a] to [e], and occasionally [i]), a phonological process characteristic of the variety in general (cf. e.g. Romero 2011, 91). Unlike the extensive investigation of DOM in (non-Jewish) Spanish dialects, the presence and behaviour of DOM in Judeo-Spanish has garnered little attention in the scholarly literature, and is rarely afforded more than passing mention (e.g. Fischer, et al. 2018, 90ss.; Gabriel/Grünke 2022, 262). This conspicuous lacuna is plausibly explained by the prevailing popular and scholarly impression that the grammar(s) of Judeo-Spanish and (Old) Castilian/Spanish have remained, to all intents and purposes, one and the same (e.g. Crews 1935, 28; Agard 1950, 204; Benaim 2011, 71; Cárdenas 2004, 1; Çiprut 2020, 11), with the result that Judeo-Spanish syntax has not been deemed a priority for further academic study. This investigatory inattention to Judeo-Spanish sentence-level grammar (and its (non-)coincidence with Castilian/Spanish) may also explain the discrepancy, identified in Section 1, between Bunis’s (2015, 403; 2021 a, 402) claims that a-marking is “frequently absent” in the language from the 1600 s onwards, versus the finding that it remains widespread in my 21st century fieldwork data.
On the occasions that the phenomenon is described in the literature on the language, Judeo-Spanish a-marking is invariably portrayed – in academic (e.g. Bunis 2015, 403; 2021 a, 402; Joseph 2019, 269), descriptive (e.g. Marín Ramos 2018, 27) and instructional (e.g. Koen-Sarano 2002, 28; Markova 2008, 62; Varol 2008, 69) materials alike – as a preposition that marks ‘personal’ or ‘human’ DOs. Whilst the characterization of DOM in these texts is presumably not intended to be comprehensive, it is striking that none accurately describes the empirical generalization for general Spanish (i.e. all omit the crucial restriction of Spanish adnominal DOM to referential/specific animate DOs), especially given that the extension of the Spanish pattern to Judeo-Spanish DOM is often either implicitly assumed (e.g. Bunis 2015, 403) or explicitly articulated (Joseph 2019, 269). The assumption of morphosyntactic commensurability between the two groups – both in general and in relation to DOM – seems premature at best, considering the (uncontroversial) empirical claim that Spanish syntax is anything but monolithic. Moreover, it puts the Judeo-Spanish literature at odds with the specialist literature on Ibero-Romance DOM, where it is shown that the branch exhibits extensive dialectal variation. It is also at odds with my ongoing investigation (cf. fn3) into the comparative landscape of Judeo-Spanish DOM, whose results present a heterogeneous picture for the branch (and externally, vis-à-vis Spanish/Castilian) across synchronic and diachronic dimensions.
3 Methods and materials
Our primary source in the present investigation is Luria (1930), one of the more substantial (single-dialect) works of early 20th century Judeo-Spanish dialectology studies, and only monographic work, on Monastirli dialect. His study collates n=33 oral texts (n=23 prose; n=4 dialogue; n=6 romances/ballads), plus a collection of riddles, and another of proverbs, obtained from n=13 interviewees fitting the traditional profile of the dialectology informant (e.g. born in Monastir, raised by Monastir-born parents). For the study reported here, I examined a subset of Luria’s (1930) interview data, producing a hand-annotated corpus comprising tokens drawn from the first 23 prose texts of the oral material (henceforth, ‘the Luria corpus’). The dialogues (all fictional, and composed by a single speaker) were also manually inspected for tokens of DOM, but yielded no results. The remaining primary material was excluded from the present analysis to control for stylistic/register variation.
In order to investigate the regulating factors of Monastirli DOM with respect to the semantic-syntactic properties of the DO, the Luria corpus was analyzed for occurrences of DOs in (S)V(S)O(S) structures, which were concurrently filtered according to the grammatical category/structural status of the nominal argument (full NP/DP; strong/tonic, i.e. non-clitic, pronouns; clitic-only tokens). Owing to the well-established effects of verbal/clausal factors such as valency and information-structural dislocations on the (non-)licensing of DOM, I controlled for these variables by constraining the main analysis to (S)V(S)O(S) structures, i.e. transitive structures, where the DO occurs to the right of V, a position I will refer to as postverbal, a descriptive term referring to its linear position.[16] I did not otherwise control for the verbal predicate, with the exception of existential habere and possessive tenere constructions, which preliminary analysis had also confirmed as obligatory [–DOM] contexts. No restrictions were imposed on the (non-)realization of the subject argument, nor its position.
4 Adnominal DOM in Monastirli
In what follows, I divide the presentation and discussion of the Monastirli data in terms of the [±definiteness] status of the DO argument, an organizational structure which corresponds, virtually without exception, to the presence versus absence of DOM in this variety. Unless otherwise stated, the clausal environment for all tokens discussed are (S)V(S)O(S) structures (cf. §3). Similarly, all DO tokens discussed are animate unless otherwise specified.
We start by considering highly-referential, definite(-like) expressions (which invariably trigger DOM in Monastirli), proceeding heuristically along the scalar ranking of referential prominence (Haspelmath 2021) in the following order: personal pronouns (§4.1); pronominal (§4.2.1) and non-pronominal demonstrative expressions (§4.2.2); common nouns with a definite article (Art+NP) or possessive determiner (Poss+NP) (§4.3); and strong quantifier expressions (§4.4). Absent from this presentation are proper nouns, since these were unattested in DO position in the Luria corpus, a matter I address in Section 5.3. Finally, Section 4.6 presents the findings regarding indefinite(-like) DOs.
Whilst the main purpose is to provide a qualitative structural characterization of Monastirli DOM, I include numerical information where these data are particularly instructive and have a bearing on the qualitative analysis; generally, my reporting of quantitative data is included in instances where tokens for a particular condition are few, and thus wider generalizations cannot be reliably drawn, or should be treated with caution.
4.1 Personal pronouns
Personal pronouns are grammatical items lacking in descriptive content which single out uniquely identifiable individuals from the wider (endophoric and/or exophoric) discourse environment. Moreover, first and second person forms of reference are canonically used to denote speech participants in the non-linguistic (i.e. necessarily exophoric) context, and are thus conceptualized as human/animate by default (cf. e.g. Harley/Ritter 2002). As such, personal pronouns rank highly in terms of their referential prominence.
In the Monastirli data, a-marking is obligatory with personal pronouns in (S)V(S)O(S) structures:[17]
(6) a. Purque mi cunesin toduz ami.
because me=know.3sg all dom me
‘Because everybody knows me.’
b. para ti cumer ati vini
in.order.to you=eat.infdom you came.1sg
‘I came in order to eat you’
c. la vaz âfirrar aneye sole
her=go.2sg catch.infdom her alone
‘You’re going to catch her alone’
d. Y él mus cumande amuzotrus.
and he us= rule.3sgdom us
‘And he rules us.’
In the Luria corpus, tokens of strong pronominal DOs are overwhelming singular in number; second- and third-person plural tokens are unattested. Interpretatively, all tokens for this condition denote animate referents, such that inanimate referents are unattested with (strong) pronominal DOs in the Luria corpus. Furthermore, all tokens of postverbal personal pronouns (as exemplified above) are obligatorily indexed by a co-referential clitic.
4.2 Demonstrative DPs
Demonstrative expressions are another context highly propitious for DOM marking. Despite not involving person reference like the pronominal tokens in §4.1, demonstrative DPs are a highly referential class of nominal expression that fix reference via non-descriptive content to a single demonstratum within certain contextual parameters anchored to a deictic centre. In Monastirli džudezmu, demonstrative expressions in the nominal group can function as pronouns (e.g. aqueye ‘that.fsg [one]’) or as XPs formed by a prenominal determiner and lexical NP (e.g. aqueye mujer ‘that woman’).[18] Since the pronominal/non-pronominal distinction is known to be relevant to the licensing of DAM/DOM cross-linguistically (Comrie 1989, 195, cit. Andriani 2023, 28), I will deal with demonstrative pronouns (§4.2.1) and full lexical DPs (§4.2.2) in turn.
4.2.1 Demonstrative pronouns
In our corpus, all instances of demonstrative DO pronouns in (S)V(S)O(S) clauses involve human referents, and all are flagged by the prenominal marker in this environment:
(7) a. ¿Qué lu queris matar aésti?
why him= want.2sg kill.infdom this.msg
‘Why do you want to kill this man?’
b. Yamó an aquel qui li vindi̯ó il mi̯oyu.
called.3sgdom that.msg who to.him=sold.3sg the advice
‘He called the man (lit. ‘that one’) who sold him the piece of advice.’
The restriction of a-marking to human referents in these data suggest that pronominal forms are avoided for, or are otherwise incompatible with, inanimate entities for the purpose of demonstrative reference in Monastirli, analogous to our findings regarding the animacy criterion observed with strong personal pronouns in object position (§4.1). Unlike with personal pronouns, however, clitic indexing with pronominal demonstratives appears to be optional, albeit possibly preferred, inasmuch n=3 tokens (of a total n=4 animate DOs) involve a co-referential clitic.
4.2.2 Dem+NP
The relevance of the animacy criterion to the licensing of adnominal DOM in Monastirli is further underscored once non-pronominal DO expressions are considered. As exemplified in (8a–b), animate (human) lexical DOs introduced by a demonstrative determiner are obligatorily flagged by DOM in transitive structures:
(8) a. La mató anestemujer.
her.acc= killed.3sgdom this.f woman
‘He killed this woman.’
b. Antiz di matar anestiombri [...]
before of kill.infdom this.m man
‘Before killing this man [...]’
With inanimate demonstrative Dem+NPs, by contrast, adnominal DOM in the same environment is systematically absent:
(9) a. In no yivandu aqueyusmaderus [...]
in not bringing those.mpl logs
‘By not bringing those pieces of wood [...]’
b. Tumandu aqueyes parás [...] y dalez a lus provis
take.ptcp those.fpl coins and give.imp=them to the poor.pl
‘Take those coins and give them to the poor!’
Like their pronominal counterparts, no tokens of non-human animate Dem+NP expressions occur in the Luria corpus, such that the (ir)relevancy of the humanness criterion to the licensing of a-marking of demonstrative DOs in Monastirli cannot be ascertained. Notwithstanding this gap in our data, the foregoing empirical results nonetheless robustly confirm that animacy is the key regulating factor in the licensing of a-marking with (postverbal) demonstrative DOs in Monastirli transitive structures across pronominal and non-pronominal items alike. Finally, as with the pronominal DOs of §4.2.1, we can observe that the co-occurrence of the indexing clitic with human/animate Dem+NP expressions is optional.
4.3 Definite and possessive lexical DPs (Art/Poss+NP)
Moving down the animacy/definiteness scales, we now consider argumental expressions composed of a prenominal definite determiner (here, the definite article or possessive determiner) and a lexical N. Given their commensurability, both in general and in our findings in relation to DOM, I treat definite (Art/Poss+NP) common nouns together for the purposes of exposition.
In the Romance group, the licensing of DOM with common nouns headed by definite determiners is subject to considerable variation (cf. fn7). Yet the picture for Monastirli appears to be broadly similar to the general (Modern) Spanish pattern. (Recall from §2.2 that, in Modern Spanish, the licensing of DOM with lexical NPs introduced by a definite or possessive determiner is primarily contingent on the humanness/animacy of the DO.) Thus, in Monastirli, DOs involving human referents in transitive (S)V(S)O(S) clauses are obligatorily flagged by DOM (10a–b), in contrast to the (apparent) optionality of DOM with nonhuman animate DOs (10c–d), and its generalized absence with inanimate DOs (10e):
(10) a. Il turcu ’stá comu un locu bušcandu alǧidi̯ó
the Turk is like a crazy searching dom.the Jew
‘The Turk is searching like crazy for the Jew’
b. Queru ver a tu mujer
want.1sg see.infdom your wife
‘I want to see your wife’
c. no lu matarun alli̯ón
not it.m.acc=killed.3pldom.the lion
‘they didn’t kill the lion’
d. Dispu̯és tumó il curdiricu para si yivar
then took the lamb.dim in.order.to refl=take.inf
‘Then she seized the lamb in order to take it away with her’
e. [v]amuz a tumar las parás y impu̯és muz vinimuz in caze.
go.1pl to take.inf the money and then refl=come.1pl in home
‘We are going to take the money and then return home.’
At first blush, the Monastirli data coheres with the corresponding pattern in general Spanish, in which a-marking of definite Art/Poss+NPs is obligatory for human referents but available (per traditional grammars) with non-human animates according to the degree of anthropomorphism subjectively ascribed to the DO in question. The empirical situation, however, is subject to further fine-grained variation – both within Monastirli and with comparative respect to Spanish – than the foregoing generalizations suggest, as I discuss below in relation to (i) the parameter of animacy (§§4.3.1–4.3.2) and (ii) the availability of a co-referential clitic (§4.3.3).
4.3.1 Adnominal DOM with human referents
Firstly, whilst the obligatoriness of DOM with human referents in Monastirli is an extremely robust pattern, it is not exceptionless. In our corpus, n=2 tokens of (non-coordinated) animate DOs involving definite lexical NPs with a prenominal determiner appear in unmarked form:
(11) a. Arricujendu las criatures para liz mandar a la scolie
pick.up.ptcp the children in.order.to them=send.inf to the school
‘Pick up the children to send them to school’
b. L’afalaġó [...] qui si tomi il mu̯ertu
her=flattered that refl= take.sbjv.3sg the dead
‘She charmed her into taking the dead child’
Whilst broader generalizations cannot be inferred from a sample of this size, it is striking that the two exceptions involve non-adult referents, and, in the case of (11b), a non-living referent, since both the adult/non-adult and living/non-living distinctions are relevant to (linguistic) animacy, and are distinctions that can be encoded grammatically in languages cross-linguistically (including, notably, suppression of a-marking in Modern Spanish in a small minority of highly-restricted contexts, cf. Ormazabal/Romero 2013, 224).
With regard to the question of adult/non-adult referents, it is of particular interest that the only other definite token of the lexeme criature(s) ‘child(ren)’ in DO position is also unmarked by DOM in the Luria database:[19]
(12) Déšimi y yiré a ver a mi mujer y mis si̯eti cri̯atures
leave.imp=me and go.fut.1sg to see.infdom my wife and my seven children
‘Leave me be and I will go and see my wife and my seven children’
It is unclear what we can infer from this token, however. On the one hand, the token involves two conjoined DOs, where the non-adult referent (i.e. criatures ‘children’) is the second conjunct, and we know that, in general Spanish, DOM-marking is symmetric for co-ordinated DOs (Saab/Zdrojewski 2021, 864). On the other hand, it is unknown whether the symmetric requirement of Spanish/Castilian extends to (Monastirli) Judeo-Spanish. Consideration of quantitative data for this condition does not provide further insight, insofar as there are only n=3 other tokens of conjoined definite human DOs in our corpus, n=2 of which involve the overt flagging of DOM on both constituents (13a–b), whereas n=1 patterns like (12) in a-marking only the first conjunct (13c):
(13) a. Il padri las tumó a laz fijes y a la mujer y caminó
the father acc.fpl= took.3sgdom the daughters and dom the wife and walkedinadilantriforward
‘The father took his daughters and wife and set off ahead’
b. Esti patu mi queri cumer a mí y alaguerfanique
this duck me=wants eat.infdom me and dom the orphan.dim
‘This duck wants to eat me and the orphan’
c. Mi trusitis un patu qui me queri cumer amí y la guerfanique
me=brought.2sg a duck that me=wants eat.infdom me and the orphan.dim
‘You’ve brought me a duck that wants to eat me and the orphan’
Interpretatively, it is nonetheless conspicuous that the a-marked DOs involved in (13a–c), viz. daughters (fijes) and (female) orphan (guerfanique), are lexically close to criatures (‘children’), and often pick out (as is the case here) non-adult referents, although only the unmarked criatures denotes a non-adult referent by definition.
Note that the licensing of DOM in such examples is not a matter of kinship, since DO tokens of kinterms introduced by possessive (14a) and definite (14b) determiners in the corpus share the same pattern as common nouns:
(14) a. lu mató amiirmanu
acc.msg= killed.3sgdom my brother
‘he killed my brother’
b. Il provi impisó a cafrar alrey, alpadri,
the poor began.3sg to curse.inf dom.the king dom. the father
a la mujer, al papú, a lamaná
dom the wife dom.the grandfather domthe grandmother
‘The impoverished man began to curse the king, his [own] father, his wife, his grandfather, his grandmother’
(Bare kinship terms, which typically pattern with proper names, are unattested in DO position in the Luria corpus.)
4.3.2 DOM with non-human animate referents
With respect to the licensing of a-marking with non-human animate common nouns, we initially observed that its usage with definite (Art/Poss+NP) expressions appears to be optional in Monastirli in light of examples such as (10c–d). Closer inspection of the source material, however, suggests that it is doubtful whether there is, in fact, any optionality for this condition. This is because the only tokens where non-human animate DOs are regularly flagged by DOM pertain to animal stories where the creatures involved are treated as sentient and attributed fully human-like qualities (15a). By contrast, tokens of non-human animate DOs in human-centered narratives (where animals are not subject to anthropomorphism) remain systematically unmarked (15b):
(15) a. Él d’aínde no lu vidu alli̯ón
he still not acc.msg=saw.3sgdom.the lion
‘He still didn’t see the lion’
b. Bušcó ilḥamór
searched.3sg the donkey
‘He searched for the donkey’
With these observations in mind, it seems appropriate to characterize the animacy restriction on Monastirli a-marking as a humanness requirement, with the proviso that we cannot determine, based on the available data, whether this characterization applies only to cases of non-human referents whose animacy is lexically specified (i.e. by the descriptive content of lexical N), as with (15a–b), or whether it also extends to referents whose (non-human) animacy status is contextually derived from the wider discourse, such as those in (16a–b):
(16) a. Reyis ez il qui cumande atodus
kings is he who rules dom all.mpl
‘Kings are the one[s] who rule over all [creatures]’
b. Y no ay máz grandi di muzotrus qui mus cumande a muzotrus?
and not there.is more big of us that us= rule.3sg dom us
‘And there’s no one greater amongst us [animals] to rule over us?’
4.3.3 Postverbal DOs with co-referential clitic
A further property characterizing Monastirli DOM is the role of the co-referential accusative clitic in the marking of DOs in the Luria corpus. The first empirical observation is that the indexing clitic is possible with Art/Poss+NPs in Monastirli irrespective of animacy, as illustrated for human (17a–b) and non-human (17c) animate referents, as well as inanimate DOs (17d):
(17) a. Aġore yo voy ir y lu tupar alazei̯teru
now I go.1sg go.inf and him.acc=find.infdom.the oil.seller
‘Now I’m going to go and find the oil-seller’
b. La yamó ala vi̯eje il dispót
her.acc=called dom the old.woman the bishop
‘the bishop called the old woman’
c. La vidu la ġayine il irmanu
it.fsg.acc=saw the hen the brother
‘The brother saw the hen’
d. Tumó la pi̯edre y lu tapó ilpodzu
took the stone and it.msg.acc= covered the well
‘He picked up the stone and covered up the well’
Informal inspection of the dataset confirms that the relative incidence of the indexing clitic is much higher for animate DOs than inanimate DOs, with which it is overall infrequent (cf. also fn20). What’s more, the occurrence of the co-referential clitic is not reliant on the presence of adnominal DOM, as verified by its optional but frequent recurrence in tokens of unmarked non-human animate DOs (17c), as well its occasional appearance with inanimate DOs (17d), which do not admit a-marking in Monastirli.
4.4 Strong quantifier expressions
Since Milsark (1974), quantifier expressions can be subdivided according to a ‘weak/strong’ distinction, which, as already observed in §2.2, bears on the licensing of flagging DOM in general Spanish. Namely, ‘strong’ quantifiers – associated with reference to (sets of) individuals whose existence is, e.g. presupposed or retrievable from the discourse – obligatorily trigger a-marking on animate DOs (e.g. Sp. veo *(a) cada mujer ‘see.1sg dom each woman’) in general Spanish, a pattern in line with a wider tendency observed elsewhere in Romance.
In the Luria corpus, strongly-quantified animate DOs are attested (only) with the universal quantifier todo(s) ‘all’, occurring both in pronominal form (with an animacy restriction) and as the head of a full lexical DP:
(18) a. Dispu̯és qui luz yivarun imprezuz intodus [...]
after that them.acc= took imprisoned.mpldom all.mpl
‘After they took them all prisoner [...]’
b. Il rey aǧuntó un día a toduz luz grandis dil riinadu.
the king assembled one day dom all.mpl the.mpl great.mpl of.the kingdom
‘One day, the king assembled all the vizirs of the kingdom.’
In all such cases, the presence of the differential marker is exceptionless, and contrasts with the obligatory absence of adnominal DOM with inanimate DOs:
(19) a. Ya ti lo vo a mircar todul’ azeiti
already you.dat= it.acc.msg= go.1sg to buy.inf all.msg the oil
‘I’m going to buy all your oil off you’
b. Yo cun este cavese pidrí todelamunede
I with this head lost.1sg all the money
‘I, with this head [of mine], lost all the money’
Finally, we observe that an indexing clitic is possible with (pronominal/non-pronominal) strong quantifier expressions in Monastirli, and, like the lexical DPs in §4.3.3, is not subject to an animacy restriction.
4.5 Interim summary: Monastirli DOM with highly referential DOs
In the preceding subsections, I have examined the distribution of adnominal DOM in transitive clauses with a postverbal DO, describing its occurrence in relation to the syntactic-semantic properties of the a-marked DO. We have seen that the overt flagging of ‘highly-referential’ nominals such as personal pronouns, demonstratives, definite DPs, and strong quantifier expressions in Monastirli is strongly correlated with the humanness/animacy properties of the DO. Thus whereas human referents invariably require DOM (with a highly limited set of exceptions, cf. §4.3.1), adnominal DOM is systemically ruled out with inanimate entities, and is restricted with non-human animate (lexical) DOs, which generally incur a-marking only in highly anthropomorphized contexts (§4.3.2). Although our principal focus has been to establish the empirical distribution of adnominal DOM, I have also made note of its (non)coincidence with DOI. We find that the co-occurrence of an indexing clitic is obligatory only with (non-dislocated) personal pronouns, but is otherwise optional in all other environments surveyed. Crucially, moreover, I have shown that clitic indexing occurs with non-human animate DOs and, occasionally, inanimate DOs, even in the absence of a-marking.[20]
4.6 Indefinite(-like) expressions
Having established the occurrence of a-marking with ‘highly referential’ nominals in the Luria corpus, this section now turns to consider Monastirli DOM in relation to indefinite(-like) expressions, dealing with non-pronominal (§4.6.1) and pronominal (§4.6.2) indefinites in turn. Indefinite nominal expressions are a heterogeneous grouping of nominal items with distinct semantic and syntactic properties which, following Leonetti’s (2012) taxonomy for Modern Castilian/Spanish, includes lexical NPs introduced by an indefinite determiner, as well as bare nouns (with indefinite-like readings) and indefinite pronouns. The label ‘indefinite determiner’ is deployed here as a pre-theoretical term for items also known as ‘weak’ determiners (Milsark 1974; cf. also §2.2), including indefinite articles, cardinal numerals and indefinite quantifiers (cf. also Leonetti 2012, 292–6; Brasoveanu/Farkas 2016, 257–67).
4.6.1 Non-pronominal indefinite expressions
The headline finding from our empirical investigation of indefinite expressions in the Luria corpus is that DOM is systematically ruled out in non-pronominal indefinite expressions in Monastirli, irrespective of whether the lexical NP is introduced by an indefinite article (as with (20a–c)) or another indefinite determiner with more specialized readings (e.g. algún ‘some’, otru ‘(an)other’, (20d–e)), all of whose counterparts in Modern Spanish are compatible with adnominal DOM.
(20) a. Un díe la mujer vidu unazei̯teru.
one day the woman saw.3sg an oil.seller
‘One day, the wife saw an oil-seller.’
b. Si tumó otre ves une mujer la cuale ere firmoze, buene y uneste
refl= took.3sg again a woman the which was beautiful, good and honest
‘He once again took a wife who was beautiful, good and honest’
c. ¿Cu̯andu yo vo tiner il mazál di cumermi unben adám?
when I go.1sg have.inf the luck of eat.inf=me a human being
‘When am I going to be lucky enough to eat a human being?’
d. ¿Nunque muzotrus pudemus matar algúnben adám?
never we can kill.inf some human being
‘We can never kill a human being?’
e. Al díe mircó otrupatu.
on.the day bought.3sg other duck
‘The next day he bought another (live) duck.’
As the specific (20a–b, e) and nonspecific (20c–d) readings of the unmarked DO in the examples above show, specificity has no bearing on the licensing of adnominal DOM in Monastirli with indefinite expressions.
Moreover, the incompatibility of Monastirli DOM with indefinites even extends to structural environments which in Modern Spanish can override semantics factors regulating DOM, such as with secondary predicates (21a–b):[21]
(21) a. Vidu imfrenti umombri pinsandu
saw opposite a man think.ptcp
‘He saw a man opposite thinking’
b. Si dišó di pasar unabufne
refl= let.3sg of pass.inf a buffalo
‘They let a buffalo pass by’
Crucially, these structures do trigger a-marking of definite DOs, both human (22a) and nonhuman (22b), in Monastirli, as well as with personal pronouns (22c):
(22) a. Il haver dil maridu [...] la veyi alamujer diznude.
the friend of.the husband acc.3sg=sees dom the wife naked
‘The husbandi’s friend saw hisi wife in the nude.’
b. Vierun alcavayu vinir
saw.3pldom.the horse come.inf
‘They saw the horse arrive’
c. Qui mi dišéš vuzotruz ami favlar.
that me=let.sbjv.2pl you.pldom me speak.inf
‘Let me speak.’
That is, the availability of DOM with definite DOs in (22a–c) underscores that secondary predicate contexts are (potential) environments in DOM for Monastirli.
Finally, the absence of a-marking with indefinite expressions also extends to determiner-less lexical NPs, e.g. bare nominals, which are systematically unmarked by DOM irrespective of plurality in Monastirli (cf. Fábregas 2013, 20 for general Spanish DOM and its behaviour with bare indefinites):
(23) a. Ningune ves no mirquimus cavayu
no time not bought.1pl horse
‘We have never bought a horse’
b. In meyu caminu iscuntrarun ladruním.
in middle way found.3pl robbers
‘En route, they came across robbers.’
4.6.2 Indefinite(-like) pronouns
Our final parameter for consideration within the referentiality scale is the behaviour of Monastirli DOM with respect to animate indefinite(-like) Q-pronouns. Our investigation of the Luria corpus, however, only yields n=2 such tokens in (S)V(S)O(S) structures, both of which pattern differently with respect to a-marking.
The first finding is that, contrary to the expected behaviour for a Romance variety, the negative animate pronoun (ningunu(s) ‘nobody’) does not require a-marking in postverbal position (24a) in Monastirli:
(24) a. No veyi ningunus.
not saw.3sg nobody
‘He did not see anyone.’
b. yir al cazal sin mircar nade
go.inf to.the village without buy.inf nothing
‘go [back] to the village without having bought [lit. buying] anything’
Thus in (24a–b), both negative pro-forms appears unmarked with respect to DOM, despite the animacy of the referent in (24a), which is especially striking considering that this is an (obligatory) environment for DOM even in Ibero-Romance varieties that are otherwise considered ‘non-DOM’ languages (e.g. European Portuguese Não vi *(a) ninguém ‘not saw.1sg dom nobody’).
By contrast, we do find DOM with the animate relative/wh-pronoun ‘who’ in n=1 token (25a) from the Luria corpus, yielding a contrast with its non-animate counterpart (25b):
(25) a. Mete aġore aquen queris tú.
put.imp now dom who want.2sg you
‘Now put who you want [to work].’
b. Sinti̯endu il rey lucu̯e il ǧidi̯ó dizíe
hearing the king the what the Jew said
‘The king, on hearing what the Jew said’
In Castilian/Spanish, (25a) constitutes a compulsory environment for DOM, even with non-specific DOs (thus patterning with the quantified expressions of §4.4). The presence of the differential marker in (25a), which likely has a non-specific reading (‘whomever you want’),[22] suggests that the same characterization may extend to Monastirli also. However, contextual ambiguity remains such that the alternative reading for this token (i.e. ‘the [specific] person you want’) cannot be ruled out. Nevertheless, the general findings of the present study in relation to the irrelevance of specificity contrasts to Monastirli DOM cast doubt on the possibility that specificity should exceptionally play a role in the differential marking of quen ‘who’ in (25a).
5 Discussion
In what follows, Section 5.1 opens our discussion with a brief synthesis of the study and some general reflections on the comparative implications of our findings with respect to Castilian/Spanish and in the typological context of the Romance and Balkan groups. Section 5.2 is dedicated to a closer consideration of the key nominal parameters relevant to Monastirli DOM. Discussing animacy and definiteness in turn, I identify for both parameters an asymmetry between interpretative and grammatical factors, and I argue that only the latter operates as DOM-triggering properties in Monastirli džudezmu. The case for a syntactic treatment of these data is further developed in Section 5.3’s discussion, in which I consider at length the suitability of the scale-based approach as a theoretical framework for modelling the findings of our case study. Despite the latter’s broad descriptive compatibility with our empirical scenario, I ultimately conclude in favour of the benefits of a formally-oriented, configurational account for Monastirli DOM.
5.1 General remarks
In the preceding sections, I have set out a structural characterization of adnominal DOM in Monastirli with respect to the semantico-syntactic properties of differentially-marked postverbal DOs. Table 1 summarizes the empirical distribution of adnominal DOM in (S)V(S)O(S) clauses in terms of the nominal category/type and descriptive animacy status of the DO in Table 1 (the annotation n/a is used for non-attested conditions).
Availability of adnominal DOM with postverbal DOs in the Luria corpus
| Nominal category/type | [+animate] | [–animate] | |
| [+human] | [–human] | ||
| 1st / 2nd personal pronouns (strong) | Obligatory | Obligatory | n/a |
| 3rd personal pronouns (strong) | Obligatory | Obligatory | n/a |
| Proper names (bare / Artdef + NP) | n/a | n/a | n/a |
| Demonstrative pronouns (Dem + Ø) | Obligatory | n/a | n/a |
| Demonstrative Dem + NP | Obligatory | Optional | Absent |
| Artdef + lexical NP | Obligatory* | Optional | Absent |
| Poss + lexical NP | Obligatory | n/a | Absent |
| Universal quantifier (todos + Ø) | Obligatory | n/a | n/a |
| Universal QP (todos + NP) | Obligatory | n/a | Absent |
| Relative/wh-pronoun | Attested | n/a | Absent |
| Negative pronoun | Absent | n/a | n/a |
| Artindef + lexical NP [+specific] | Absent | ||
| Artindef + lexical NP [–specific] | Absent | ||
| Bare lexical NP | Absent | ||
From our analysis, it emerges that animacy (§5.2.1) and definiteness (§5.2.2) are defining nominal properties of Monastirli DOM that go hand-in-hand in the licensing of a-marking in this variety. In this regard, Monastirli a-marking behaves in line with the typological type observed in Castilian/Spanish, and associated with the Romance group more broadly, in which a-marking is harnessed as an animacy-based strategy (Hill/Mardale 2021) for the differential marking of highly-referential DOs. This pattern corresponds to a two-dimensional DOM system combining the implicational hierarchies of animacy/person (1a) and specificity/definiteness (1b) within the scale-based framework (although cf. §5.3).
With respect to the (comparative) grammatical landscape of Judeo-Spanish, the attestation of widespread a-marking as the mechanism of adnominal DOM in our 20th century corpus is consistent with my recent fieldwork findings (cf. §1, §2.3) indicating that a DOM system is maintained across the language branch to the present day. By contrast, the suggestion that (Monastirli) Judeo-Spanish should be differentiated from general Castilian/Spanish owing to absence of a-marking in the former finds no empirical support. However, our investigation has shed light on various areas of divergence with respect to general Spanish. Notably, the licensing of adnominal DOM in Monastirli diverges from general Spanish in its insensitivity to, i.a., specificity, secondary predication environments; and (paceFriedman 2008, 38) in the distribution of a co-referential clitic, whose occurrence we have documented with a wider range of nominal expressions vis-à-vis those permitted in general Spanish, and in the absence of both a-marking and animacy. Moreover, although our focus has been on nominal factors (cf. §5.2 below), I have not found any evidence in the present case study to suggest that properties relating to the verbal/event domain had an effect on the (non-)expression of Monastirli a-marking,[23] unlike what has been shown for Spanish/Castilian (e.g. Torrego 1998; García García 2018; Romero Heredero 2022).
From the typological perspective, the (non-)coincidence of Monastirli a-marking with a co-referential accusative clitic exhibits a hitherto undescribed distribution distinct from other patterns documented for Romance. Whilst Monastirli admits a-marking without clitic indexing, and vice versa, elsewhere in the Romance group clitic doubling (cf. fn15) is only permitted iff a-marking (the majority pattern), or is otherwise the only differential mechanism available (as seen in, e.g. Aromanian, Megleno-Romanian, which exhibit DOI characteristic of the Balkan group). Indeed, the decoupling of a-marking from clitic indexing in Monastirli yields an empirical pattern redolent of Balkan object reduplication (Tomić 2006; Friedman 2008, i.a.), displaying parallels in particular with the definiteness-based system found in Macedonian. Whilst the correspondence is not total,[24] the empirical similarities are nonetheless indicative of areal influence on Monastirli džudezmu (and, thus, Judeo-Spanish morphosyntax more broadly). These comparative considerations provide clear avenues of investigation for future research (as I have begun to develop in research currently in progress).
5.2 Nominal triggers of adnominal DOM
5.2.1 Animacy
In line with the orthodox (though not uncontested) characterization of Romance a-marking as an animacy-based strategy for the realization of DOM (cf. fn13), our analysis confirms the centrality of the animacy status of the DO in regulating the appearance of adnominal DOM in Monastirli. Without exception, inanimate DOs are never a-marked in our dataset, a strong empirical claim circumscribed only by the caveat that certain contexts in which inanimate DOs are a-marked in Spanish were absent from our corpus (i.e. without the relevant data, we cannot rule these out as possible a-marking environments for Monastirli). Dataset limitations notwithstanding, the analysis of the Luria dataset strongly points to animacy as a necessary condition for the occurrence of adnominal DOM. Note, however, that it is not a sufficient one: rather, adnominal DOM is strongly correlated with humanness, insofar as a-marking is systematic only for (definite) DOs with a human referent. By contrast, the licensing of adnominal DOM with non-human animates occurs as a function of the anthropomorphism attributed to the referent in question, an alternation which reveals itself in the Luria corpus in the regularity of a-marking with non-human animate DOs in animal tales, versus the opposite tendency in texts with human-centred narratives. In other words, although I have described the Monastirli data thus far in terms of a tripartite taxonomy (human/non-human animate/inanimate), our findings suggest that animacy is operative as a binary category in regulating the (non-)expression of (adnominal) DOM in this variety. Significantly, neither descriptive/lexical information nor extra-linguistic knowledge of a referent’s animacy is sufficient for a DO to be assigned a positive value for animacy.
From a theoretical perspective, the insensitivity of a-marking to biocultural/conceptual animacy challenges approaches to DOM that treat the category as a (semantically) ‘inherent’, unvarying property of nominals (e.g. Klein/de Swart 2011). However, the Monastirli data clearly demonstrate a difference between ‘intrinsic’ (e.g. lexico-semantic, real-world) animacy, and animacy as a DOM-triggering property. Crucially, theoretical treatments which conflate the two would not be able to model the empirical patterns of Monastirli, since ‘intrinsic/inherent’ animacy is not a sufficient predictor of adnominal DOM (where all other conditions, e.g. definiteness, are met) in the dialect. These facts, I contend, strongly indicate that the licensing of adnominal DOM in Monastirli involves grammaticalized animacy. This conclusion is also supported by the data from functional items with grammatically-encoded animacy, such as strong 1st/2nd personal pronouns (cf. §4.1), since these always mandate (adnominal) DOM in Monastirli. In other words, whereas lexically-supplied animacy or contextual knowledge may favour – but cannot, crucially, guarantee – a referent’s categorization as [+animate], (morpho)syntacticallysupplied animacy seemingly triggers an a-marking requirement.
Key corroboration that adnominal DOM’s sensitivity to animacy in Monastirli is best captured as a grammatical property (rather than in terms of semantic or lexical intrinsicness, vel sim.) comes from the rare tokens in our corpus of unmarked DOs involving definite human referents. These cases (previously discussed in §4.3.1) correspond to a small number of definite lexical expressions whose humanness/animacy is supplied via the descriptive content of N (e.g. criatures ‘children’) and which, apparently exceptionally, undergo a process of ‘de-animation’ (Ormazabal/Romero 2013, 224) that renders them non-animate/non-sentient for the purposes of animacy-triggered a-marking. Importantly, the items subject to ‘demotion’ in Monastirli form a lexically-coherent grouping not only in the context of our dataset, but correspond to a restricted range of animate NPs exhibiting analogous grammatical behaviours with respect to incorporation and other deanimation processes cross-linguistically (cf. Ormazabal/Romero 2013, 224 and references therein).
The significance of these ‘de-animated’ items – i.e. lexically human definite DOs that are unmarked for DOM – is twofold. On one level, these tokens further underscore that ‘semantic’ animacy (as with non-human animates) does not necessarily translate into a positive value for grammatical [+animacy] in Monastirli. On a second level, the coincidence of these apparent exceptions with an independently identified, cross-linguistic grouping indicates that their atypical behaviour with respect to a-marking in Monastirli cannot be dismissed as mere language-specific idiosyncrasy. Moreover, that this cross-linguistic grouping is unified on the basis of shared grammatical behaviours is an especially compelling, albeit subtle, piece of evidence that further underscores the likelihood that the absence of a-marking in their Monastirli counterparts has grammatical origin.
On our approach, then, a-marking as a function of animacy is not the result of a gradient assessment of the referent’s animacy status (in Monastirli), but is determined by a syntactic dyad that assigns DO arguments as [±animate] (and thus visible/invisible for the purposes of a-marking in accordance with the positive/negative value for that property).[25] Although my purpose is not to pursue a (formal) theoretical implementation here, configurational treatments of animacy-based DOM in terms of syntactic Person-related features (e.g. Cornilescu 2000; Rodriguez-Mondoñedo 2007; Irimia 2020 a, 2020 b i.a.) offer a plausible candidate for the relevant syntactic correlate of animacy that is consistent with the grammatical approach to Monastirli a-marking proposed here.
5.2.2 Definiteness
We have observed that definiteness is a definitional property of (adnominal) DOM in Monastirli džudezmu. In our study, the definiteness of tokens has primarily been assessed heuristically on the basis of the presence of formal marking conventionally associated with this property (e.g. definite determiners and other relevant morphoyntactic indicators), as opposed to the semantic content of the NP/DP. Nonetheless, our examination of the Monastirli data shows that, whilst a-marking of DO tokens positively correlates with (morpho-)syntactic indicators of definiteness, we find no equivalent correspondence with semantic notions of definiteness (fn9; cf. §5.3 for discussion). I interpret this finding as key evidence that Monastirli DOM’s definiteness requirement concerns a syntactic (rather than semantic) property.
On this basis, I propose that (adnominal) DOM is licensed in Monastirli if and only if the DO argument is syntactically specified for definiteness. In formal syntactic theory, this corresponds to the structural projection of the functional category D (conventionally lexicalized by a determiner),[26] whose presence in the extended nominal architecture is responsible for turning NPs into individual-denoting arguments, following Longobardi (1994, 2005) and related work. Nominal constituents projecting D-level structural architecture include pronouns, demonstratives, and lexical expressions headed by the definite article or possessive determiner (cf., e.g. Postal 1969; Longobardi 1994; Abney 1987): in other words, the class of ‘highly referential’ nominals that systematically incur a-marking in DO position in the Luria corpus.
Definiteness is a thus necessary condition for a-marking of DOs in Monastirli – indefinite expressions are systematically unmarked for DOM – but, as with animacy, not a sufficient one, since adnominal DOM is only available with DOs that fulfil the animacy conditions outlined in §5.2.1. The persistence of the definiteness restriction seemingly extends to at least some indefinite(-like) animate pronouns in Monastirli, at least when these occur postverbally in transitive structures: recall that the negative pro-form ningunu(s) ‘nobody’ remained unmarked for DOM in (24a), contrasting with the obligatory a-marking of its counterpart in Spanish/Castilian (where, it is typically assumed, the animacy of the pronoun suffices to trigger compulsory a-marking, cf. e.g. Rodríguez-Mondoñedo 2007; López 2012). The conclusion that definiteness ‘trumps’ animacy in this environment, however, can only be tentative, pending further investigation with a more sizeable sample of [+animate] pronominal quantifiers than the n=2 tokens of postverbal DOs corresponding to this condition retrieved from the Luria dataset.
5.3 Modelling Monastirli DOM: (beyond) referential scales
A further outcome of our study is the finding that the Monastirli data display compatibility with the implicational hierarchies of animacy/person (1a) and specificity/definiteness (1b), in line with scale-based frameworks that model DOM in terms of referential prominence (cf. §2.1). On this approach, the ‘cut-off’ for a-marking in Monastirli lies between animate/inanimate for the first scale (26a), and between definite and (specific) indefinite for the second (26b):[27]
(26) a. Animacy/person
1/2 > 3 > proper name > human > animate > *inanimate
b. Specificity/definiteness
pronoun > name > definite > *specific indefinite > *non-specific
With regard to the first scale, a noticeable gap in our dataset is the absence of proper nouns as DO arguments. Based on our findings, our expectation is that these would incur DOM (for human referents at least) in Monastirli, in line with the implicational scale for animacy/person in (1a)/(26a). Data retrieved from (broadly contemporaneous) sources beyond our corpus provide empirical support for this prediction:
(27) Yo no lo vide aYoseph (C19, Monastir)
I not him.acc= saw.1sgdom Yoseph
‘I did not see Yoseph’ (Hadar 2016, 148)
Given that toponyms could be a-marked in Old Spanish, and, to a certain extent, today (Fábregas 2013, 40ss.), an outstanding empirical question is whether this holds of Monastirli too (and, if so, how such a finding would fit with the foregoing characterization of adnominal DOM as a binary parameter constrained by descriptive humanness/grammatical animacy).
With regard to the second scale, a crucial result is that definiteness, not specificity, emerges as the key property relevant to Monastirli DOM. That is, whilst all (non-quantified) a-marked DOs in our dataset are indeed specific, the wholesale absence of adnominal DOM with indefinite DOs confirms that the specificity status of the DO argument is ultimately orthogonal to the possibility of its differential coding in Monastirli džudezmu. From a typological and theoretical perspective, the separation of definiteness and specificity in the Monastirli data is instructive. At the broadest level, their decoupling underscores the conceptual and empirical necessity of disentangling semantic and syntactic notions of referentiality (Klein/de Swart 2011, 4), or, indeed, definiteness, both in general, and in relation to DOM (e.g. Danon 2001).
Notably (as far as our study is concerned), because all a-marked DOs in our corpus are both morphosyntactically-specified for definiteness and referentially specific, syntactic definiteness cannot be separated from semantic definiteness (fn9) in a-marked tokens. However, the reverse does not hold with indefinite DOs, since these can be referentially specific (i.e. semantically ‘definite’) in the absence of syntactic definiteness. As indefinite DOs are systematically unmarked for DOM, we can infer that a-marking in Monastirli is insensitive to semantic criteria for definiteness, allowing us, in turn, to corroborate our proposal (cf. §5.2.2) that definiteness operates as a syntactic property in relation to Monastirli DOM.
The divorcing of definiteness/specificity in Monastirli is also revealing in terms of the contested status of specificity in relation to Spanish/Castilian DOM. Although the empirical facts between the two are different, the insensitivity of Monastirli DOM to specificity and/or absence of specificity effects in the dialect are consistent with theoretical hypotheses that reject a role for specificity as a factor in the licensing of DOM in Spanish/Castilian (cf. fn10) and elsewhere (e.g. Irimia 2020 a, §3).
5.3.1 Beyond scales
Despite the appeal of the scale-based framework for modelling the differential marking of DO arguments in Monastirli, our data also highlight the limitations of the approach. One major issue for scale-based models is how to account for the activation and interaction of the two hierarchies relevant to Monastirli a-marking. Notably, the occurrence of a-marking in Monastirli generally requires the conjunction of two scales for most, but – crucially – not all, cases. For example, the attestation of the prenominal marker on a non-specific animate wh-pronoun (24a) implicates the activation of the animacy/person scale, but not the definiteness/specificity one. That in some circumstances only one scale suffices, whereas other scenarios require both, is not readily accommodated on a scale-based treatment (cf., e.g. Irimia/Pineda 2022, 346, and elsewhere).
The problem is made only more prominent if we treat the co-occurrence of the accusative clitic with its associate as a form of differential marking (operative in parallel or in concert with the adnominal strategy). Here again, the indexing clitic mostly involves both scales, yet its availability in Monastirli with unmarked non-human animates and inanimates implicates the involvement of the definiteness/specificity scale without animacy/person (i.e. the reverse scenario to the previous one). The challenge for the scale-based framework is thus exacerbated because the approach must now account not only for the issue of asymmetric activation in the first place, but also for the empirical observation that two DOM mechanisms with broadly the same triggers (cf. fn20) show opposite behaviours with respect to single-scale licensing (here, animacy-only a-marking vs. definite-only DOI). The absence of a dependency between a-marking and clitic indexing in Monastirli further complicates the matter.
A more basic challenge that Monastirli presents for the scale-based approach is that the framework is a) founded on the assumption that DOM is fundamentally gradient in nature, and is b) designed to account for the phenomenon in those terms (e.g. Aissen 2003; von Heusinger/Kaiser 2007). However, the ascription of gradience does not cohere with the empirical facts for Monastirli a-marking (although it is arguably more applicable to clitic indexing as a differential strategy), insofar as the properties pertinent to adnominal DOM operate as obligatory triggers of the prenominal marker in the variety. More precisely, adnominal DOM is not merely made available but is systematically guaranteed in Monastirli by the combination of syntactically-specified animacy/person and definiteness as nominal-internal properties of the DO, the joint presence of which admits neither optionality nor exceptions with respect to the licensing of a-marking in the variety. (Indeed, apparent exceptions, as with a-marked anthropomorphized DOs or unmarked deanimated human referents, have ultimately been shown here to prove the rule.) In short, the chief advantage of the scale-based model, viz. its ability to handle gradience, is neutral in benefit to a theoretical treatment of a-marking in our dataset.
Conversely, a syntactic approach (along the lines set out in §5.2) is well suited to account for the kind of systematicity observable in the Monastirli data reported here. Pursuing a formal account of Monastirli DOM in terms of the complexity of the internal structure of differentially-marked DOs (Cornilescu 2000; López 2012; Irimia 2020 a, 2020 b; Hill/Mardale 2021; Irimia/Pineda 2022; Corr 2023, i.a.) will be a productive line of enquiry in future investigative endeavours. Examination of ‘global’ (clausal/verbal) factors (cf. §2.1) afforded less attention in the present article, particularly cases of information-structural displacement (a variable undiscussed here), should also be undertaken in order to determine their effect on the licensing of DOM in Monastirli džudezmu.
6 Conclusions
This article has contributed new empirical data from an understudied dialect (Monastirli) of an understudied language (Judeo-Spanish), enriching our understanding of microvariation in the morphosyntactic coding of core arguments in this language branch, and against the wider cross-linguistic backdrop concerning the mechanisms and triggers of DOM. First, the study has demonstrated the unequivocal presence of a robust, rule-governed DOM system in Monastirli Judeo-Spanish, the occurrence of which I have shown to be predictable as a function of the nominal properties of the DO argument. Restricting attention to adnominal marking of DOs in transitive (S)V(S)O(S) structures, I have proposed that Monastirli a-marking aligns typologically with the general trend in Romance, in which an (erstwhile) preposition – here, a reflex of Latin ad – is exploited as the mechanism for an animacy-based (Hill/Mardale 2021) DOM strategy that flags subsets of DO, regulated by a two-dimensional system combining (grammatical) animacy/person and syntactically-specified definiteness. Whilst the empirical distribution of a-marking in Monastirli can broadly be modelled in terms of referential prominence via implicational hierarchies known as scales (Aissen 2003, i.a.), the key finding that the licensing of DOM in our corpus is determined by grammatical, rather than semantic, properties, strongly suggests that a satisfactory account of Monastirli’s DOM system is best achieved by a syntactic treatment of the phenomenon. In favour of that conclusion, I have shown, amongst other findings, that (i) lexico-semantic and/or contextually-derived animacy/humanness correlates with, but is not a sufficient predictor of, a-marking in Monastirli (where all other conditions, e.g. definiteness, are met), and (ii) indefinite NP/DPs are systematically unmarked for DOM, irrespective of its specific/referential interpretation. Although a formal analysis is not a goal of the present article, I have situated our findings theoretically through the proposal that Monastirli a-marking is licensed iff the DO is syntactically specified for animacy and definiteness, corresponding in formal terms to the presence of [+person] and [+D] in the nominal structure. Finally, I have augmented the paper’s contribution by reporting attendant findings arising from our study, such as the (novel) distribution of clitic indexing in Monastirli vis-à-vis a-marking, and offered comparative observations with respect to Spanish/Castilian, Romance and the Balkan group, considerations I have begun to address in research currently in progress.
7 Abbreviations
* ungrammatical form or usage
% form or usage not universally accepted as grammatical
= cliticizes to
acc accusative
Art article
DAM differential argument marking
dat dative
def definite
Dem demonstrative
dim diminutive
DO direct object
DOM differential object marking, differential object marker
DP determiner phrase
f feminine
fut future
iff if and only if
imp imperative
indef indefinite
inf infinitive
m masculine
N(P) noun (phrase)
O object
pl plural
Poss possessive
ptcp participle
Q(P) quantifier (phrase)
refl reflexive
S subject
sbjv subjunctive
sg singular
V verb
XP unspecified phrasal category
Acknowledgments
This study was made possible thanks to research funding via the British Academy and Leverhulme Trust (grant number SRG1819\191358), with additional financial support from the University of Birmingham. I would like to thank the staff and students of the Department of Linguistics at New York University, who generously hosted me as a Visiting Scholar between February and June 2022. I am especially grateful to Gary Thoms, Teresa Colaizzo and colleagues at NYU’s Office of Global Services for facilitating the visit. I must also thank audiences at Spadisyn VI (University of Alcalá) and LSRL53 (INALCO, Paris), where earlier versions of this work were presented. I would also like to recognize the invaluable resources and expertise of New York Public Library’s Research Services, particularly the General Research Division and the Dorot Jewish Division, where I conducted primary archival research for this article. Finally, these acknowledgments would be incomplete without recognizing the memory of the Sephardic community of Monastir (c.1497–1943), and the members of this community whose linguistic expertise is represented in the interviews recorded by Max A. Luria: Moís Calderón, Mošé Carsola, Haïm Catán, Sara Coén, Yacób Coén, Mier Elías, Yošoa Grasianí, Naḥman Leví, Yahuda Negrín, Yosef Negrín, David Nisán, Moïšé Pardo and Sabataí Pesa.
8 References
Abbott, Barbara, Definiteness and indefiniteness, in: Horn, Lawrence R./Ward, Gregory (edd.), The Handbook of Pragmatics, Oxford, Wiley Blackwell, 2004, 122–149.10.1002/9780470756959.ch6Search in Google Scholar
Abney, Steven, The English Noun Phrase in its Sentential Aspect, PhD thesis, Massachusetts Institute of Technology, 1987.Search in Google Scholar
Agard, Frederick B., Present-Day Judaeo-Spanish in the United States, Hispania 33:3 (1950), 203–210.10.2307/334026Search in Google Scholar
Aissen, Judith, Differential Object Marking. Iconicity vs. economy, Natural Language and Linguistic Theory 21 (2003), 435–483. 10.1023/A:1024109008573Search in Google Scholar
Andriani, Luigi, Differential Object Marking in Barese. Old and new perspectives, in: Irimia, Monica-Alexandrina/Mardale, Alexandru (edd.), Differential Object Marking in Romance. Towards Microvariation, Amsterdam, John Benjamins, 2023, 24–55.10.1075/la.280.02andSearch in Google Scholar
Bárány, András/Kalin, Laura, Introduction, in: Bárány, András/Kalin, Laura (edd.), Case, agreement, and their interactions, Berlin/Boston, De Gruyter, 2020, 1–26.10.1515/9783110666137-001Search in Google Scholar
Bello, Andrés, Gramática de la lengua castellana destinada al uso de los americanos, Imprenta del Progreso, Santiago de Chile, 1847. Search in Google Scholar
Benaim, Annette, Sixteenth-century Judeo-Spanish testimonies, Leiden, Brill, 2011.10.1163/9789004210189Search in Google Scholar
Benbassa, Esther/Rodrigue, Aron, Sephardi Jewry. A history of the Judeo-Spanish community, 14th–20th centuries, Berkeley/Los Angeles/London, University of California Press, 2000.10.1525/9780520354456Search in Google Scholar
Bickel, Balthasar/Witzlak-Makarevich, Alena, Referential scales and case alignment. Reviewing the typological evidence, in: Malchukov Andrej L./Richards, Marc (edd.), Scales, Leipzig, Institut für Linguistik, 2008, 1–37. Search in Google Scholar
Börstell, Carl, Differential object marking in sign languages, Glossa: a journal of general linguistics 4(1):3 (2019), 1–18.10.5334/gjgl.780Search in Google Scholar
Bosque, Ignacio, Demonstratives, definiteness effects and the type-token distinction, Revista de Estudos Linguísticos da Universidade do Porto 1 (2021), 347–365.10.21747/16466195/lingespa16Search in Google Scholar
Bossong, Georg, Historische Sprachwissenschaft und empirische Universalienforschung, Romanistisches Jahrbuch 33 (1982), 17–51. 10.1515/9783110244908.17Search in Google Scholar
Bossong, Georg, Empirische Universalienforschung. Differentielle Objektmarkierung in den neuiranischen Sprachen, Tübingen, Gunter Narr, 1985. Search in Google Scholar
Brasoveanu, Adrian/Farkas, Donka F., Indefinites, in: Aloni, Maria/Dekker, Paul (edd.), The Cambridge Handbook of Formal Semantics, Cambridge, Cambridge University Press, 2016, 238–266.10.1017/CBO9781139236157.009Search in Google Scholar
Brugè, Laura/Brugger, Gerhard, On the accusative “a” in Spanish, Probus 8 (1996), 1–51. 10.1515/prbs.1996.8.1.1Search in Google Scholar
Bunis, David M., Judezmo (Ladino), in: Kahn, Lily/Rubin, Aaron D. (edd.), Handbook of Jewish Languages, Leiden, Brill, 2015, 365–450. 10.1163/9789004297357_014Search in Google Scholar
Bunis, David M., Judezmo/Ladino/Judeo-Spanish, in: Edzard, Lutz/Tirosh-Becker, Ofra (edd.), Jewish Languages: Text specimens, grammatical, lexical, and cultural sketches, Wiesbaden, Harrassowitz Verlag, 2021, 392–481 (=2021a).Search in Google Scholar
Bunis, David M., Judeo-Spanish (Judezmo, Ladino), Oxford Research Encyclopedias, Linguistics 2021 (=2021b).10.1093/acrefore/9780199384655.013.428Search in Google Scholar
Bužarovska, Eleni, The Contact Hypothesis Revised. DOM in the South Slavic Periphery, Journal of Language Contact 13:1 (2020), 57–95.10.1163/19552629-bja10003Search in Google Scholar
Cárdenas, John, Judeo-Spanish and the Living Museum Claim. A Synchronic View of a Diachronic Dilemma, California Linguistic Notes 29:1 (2004), 1–22.Search in Google Scholar
Caro Reina, Javier, Properhood of human unique nouns in Romance languages, Isogloss 8(5):9 (2022), 1–31.10.5565/rev/isogloss.226Search in Google Scholar
Çiprut, Henri, El žudéo-españól ez un amánte pára mi – Judeo-Spanish is my lover; performer: Henri, camera/interview: Ioana Aminian, transcription/translation: Ioana Aminian, Deyvi Papo, Imane Sghiouar, editor: Deyvi Papo, retrieved from www.oeaw.ac.at/VLACH, ID- number: ladi1251TRV0020a, 2020.Search in Google Scholar
Cornilescu, Alexandra, Notes on the prepositional accusative in Romanian, Bucharest Working Papers in Linguistics (2000), 91–106.Search in Google Scholar
Corr, Alice, Grammatical variation and language contact in Judeo-Spanish: typological and glottopolitical considerations, paper given at LCAHM School Research Seminars, University of Birmingham, 18 November, 2020.Search in Google Scholar
Corr, Alice, Judeo-Spanish differential object marking and microvariation in Balkan Romance, paper given at Linguistic Symposium on Romance Languages (LSRL 53), Paris (INALCO, Université Paris Cité, Université Sorbonne Nouvelle, and Université Paris Saclay), 29 June, 2023.Search in Google Scholar
Crews, Cynthia M., Recherches sur le judéo-espagnol dans les pays balkaniques, Paris, Droz, 1935.Search in Google Scholar
Dalrymple, Mary/Nikolaeva, Irina, Objects and information structure, Cambridge, Cambridge University Press, 2011.10.1017/CBO9780511993473Search in Google Scholar
Danon, Gabi, Syntactic definiteness in the grammar of Modern Hebrew, Linguistics 39:6 (2001), 1071–1116.10.1515/ling.2001.042Search in Google Scholar
De Hoop, Helen/Malchukov, Andrej, Case-marking strategies, Linguistic Inquiry 39 (2008), 565–587.10.1162/ling.2008.39.4.565Search in Google Scholar
Di Tullio, Ángela/Saab, Andrés/Zdrojewski, Pablo, Clitic Doubling in a Doubling World. The Case of Argentinean Spanish Reconsidered, in: Gallego, Ángel J. (ed.), The Syntactic Variation of Spanish Dialects, New York, Oxford University Press, 2019, 215–244.10.1093/oso/9780190634797.003.0008Search in Google Scholar
Diez, Friedrich, Grammatik der romanischen Sprachen, Bonn, E. Weber, 1882.Search in Google Scholar
Enç, Müvert, The semantics of specificity, Linguistic Inquiry 22:1 (1991), 1–25.Search in Google Scholar
Escandell Vidal, Victoria, Differential object marking and topicality. The case of Balearic Catalan, Studies in Language 33:4 (2009), 832–884.10.1075/sl.33.4.02escSearch in Google Scholar
Fábregas, Antonio, Differential Object Marking in Spanish. State of the art, Borealis 2:2 (2013), 1–80. 10.7557/1.2.2.2603Search in Google Scholar
Fischer, Susann/Gabriel, Christopher/Kireva, Elena, Towards a typological classification of Judeo-Spanish. Analyzing syntax and prosody of Sofia Judezmo, in: Braunmüller, Kurt/Höder, Steffen/Kühl, Karoline (edd.), Stability and Divergence in Language Contact. Factors and Mechanisms, Amsterdam/Philadelphia, John Benjamins, 2014, 77–108.10.1075/silv.16.05fisSearch in Google Scholar
Fischer, Susann/Navarro, Mario/Vega Vilanova, Jorge, Clitic Doubling and Language Contact, in: Parodi, Teresa (ed.), Proceedings of the VIII Nereus International Workshop (Referential properties of the Romance DP in the context of multilingualism), Arbeitspapier 129, Fachbereich Sprachwissenschaft, Universität Konstanz, 2018, 87–102.Search in Google Scholar
Friedman Victor A., Balkan object reduplication in areal and dialectological perspective, in: Kallulli, Dalina/Tasmowski, Liliane (edd.), Clitic doubling in the Balkan languages, Amsterdam, John Benjamins, 2008, 35–63.10.1075/la.130.05friSearch in Google Scholar
Gabriel, Christopher/Grünke, Jonah, Unmarked use of marked syntactic structures. Possessives and fronting of non-subject XPs in Bulgarian Judeo-Spanish, in: Garassino, Davide/Jacob, Daniel (edd.), When data challenges theory: unexpected and paradoxical evidence in information structure, Amsterdam/Philadelphia, John Benjamins, 2022, 239–270.10.1075/la.273.08gabSearch in Google Scholar
García García, Marco, Nominal and verbal parameters in the diachrony of Differential Object Marking in Spanish, in: Seržant, Ilja A./Witzlack-Makarevich, Alena (edd.), Diachrony of Differential Argument Marking, Berlin, Language Science Press, 2018, 209–242. Search in Google Scholar
Ghomeshi, Jila/Paul, Ileana/Wiltschko, Martina (edd.), Determiners. Universals and variation, Amsterdam/Philadelphia, John Benjamins, 2019.Search in Google Scholar
Hadar, Gila, La luz de la shjina i un espirito muevo ke nasio en el korason de Reyna Cohen, una mistikana judia en el empesijo del siglo 20, in: Díaz-Mas, Paloma/Martín Ortega, Elisa (edd.), Mujeres sefardíes lectoras y escritoras, siglos XIX–XXI, Madrid-Frankfurt, Iberoamericana-Vervuert, 2016, 179–195.10.31819/9783954878864-008Search in Google Scholar
Harley, Heidi/Ritter, Elizabeth, Person and number in pronouns. A feature-geometric analysis, Language 78:3 (2002), 482–526.10.1353/lan.2002.0158Search in Google Scholar
Harris, Tracy K., Death of a Language: The History of Judeo-Spanish, Newark, University of Delaware Press, 1994.Search in Google Scholar
Haspelmath, Martin, Argument marking in ditransitive alignment types, Linguistic Discovery 3:1 (2005), 1–21.10.1349/PS1.1537-0852.A.280Search in Google Scholar
Haspelmath, Martin, Argument indexing. A conceptual framework for the syntax of bound person forms, in: Bakker, Dik/Haspelmath, Martin (edd.), Languages across boundaries: Studies in memory of Anna Siewierska, Berlin/Boston, De Gruyter Mouton, 2013, 197–226.10.1515/9783110331127.197Search in Google Scholar
Haspelmath, Martin, Indexing and flagging, and head and dependent marking, Te Reo 62:1 (2019), 93–115.Search in Google Scholar
Haspelmath, Martin, Role-reference associations and the explanation of argument coding splits, Linguistics 59:1 (2021), 123–174.10.1515/ling-2020-0252Search in Google Scholar
Heim, Irene, 2019, Definiteness and indefiniteness, in: Portner, Paul/von Heusinger, Klaus/Maienborn, Claudia (edd.), Semantics – Noun Phrases and Verb Phrases, Berlin/Boston, De Gruyter Mouton, 2019, 33–69.10.1515/9783110589443-002Search in Google Scholar
Heusinger, Klaus von, Specificity, in: Portner, Paul/von Heusinger, Klaus/Maienborn, Claudia (edd.), Semantics – Noun Phrases and Verb Phrases, Berlin/Boston, De Gruyter Mouton, 2019, 70–111.10.1515/9783110589443-003Search in Google Scholar
Heusinger, Klaus von/Kaiser, Georg A., Differential object marking and the lexical semantics of verbs in Spanish, in: Kaiser, Georg A./Leonetti, Manuel (edd.), Proceedings of the Workshop “Definiteness, Specificity and Animacy in Ibero-Romance Languages”. Arbeitspapier 122, Fachbereich Sprachwissenschaft, Universität Konstanz, 2007, 85–110.Search in Google Scholar
Hill, Virginia/Mardale, Alexandru, The diachrony of differential object marking in Romanian, Oxford, Oxford University Press, 2021.10.1093/oso/9780192898791.001.0001Search in Google Scholar
Hopper, Paul J./Thompson, Sandra A., Transitivity in grammar and discourse, Language 56:2 (1980), 251–299. 10.1353/lan.1980.0017Search in Google Scholar
Iemmolo, Giorgio, Topicality and Differential Object Marking. Evidence from Romance and beyond, Studies in Language 34 (2010), 239–272. 10.1075/sl.34.2.01iemSearch in Google Scholar
Iemmolo, Giorgio/Klumpp, Gerson, Introduction to the special issue “Differential Object Marking. Theoretical and empirical issues”, Linguistics 52 (2014), 271–279.10.1515/ling-2013-0062Search in Google Scholar
Irimia, Monica Alexandrina, Variation in differential object marking. On some differences between Spanish and Romanian, Open Linguistics 6 (2020), 424–462, (=2020a).10.1515/opli-2020-0110Search in Google Scholar
Irimia, Monica Alexandrina, Types of structural objects. Some remarks on differential object marking in Romanian, in: Bárány, András/Kalin, Laura (edd.), Case, Agreement, and their Interactions: New Perspectives on Differential Argument Marking, Berlin/Boston, De Gruyter, 2020, 77–126, (=2020b).10.1515/9783110666137-003Search in Google Scholar
Irimia, Monica Alexandrina/Mardale, Alexandru (edd.), Differential Object Marking in Romance. Towards microvariation, Amsterdam, John Benjamins, 2023.10.1075/la.280Search in Google Scholar
Irimia, Monica Alexandrina/Pineda, Anna, Differential object marking in Catalan, Linguistic Variation 22:2 (2022), 325–385.10.1075/lv.20009.iriSearch in Google Scholar
Jaeggli, Osvald, Topics in Romance syntax, Dordrecht, Foris, 1981.10.1515/9783112420225Search in Google Scholar
Joseph, Brian D, Can there be language continuity in language contact?, in: Doron, Edit/Rappaport Hovav, Malka/Reshef, Yael/Taube, Moshe (edd.), Language Contact, Continuity and Change in the Genesis of Modern Hebrew, Amsterdam, John Benjamins, 2019, 257–286.10.1075/la.256.10josSearch in Google Scholar
Kabatek, Johannes/Obrist, Philipp/Wall, Albert, The third wave of studies on DOM in Romance, in: Kabatek, Johannes/Obrist, Philipp/Wall, Albert (edd.), Differential Object Marking in Romance. The third wave, Berlin/Boston, De Gruyter, 2021, 3–19.10.1515/9783110716207Search in Google Scholar
Klein, Udo/de Swart, Peter, Case and referential properties, Lingua 121:1 (2011), 3–19.10.1016/j.lingua.2010.07.005Search in Google Scholar
Koén-Sarano, Matilda, Kurso de djudeo-espanyol (ladino) para adelantados (trans. Gloria J. Ascher), Beer Sheva, Ben Gurion University of the Negev, 2002.Search in Google Scholar
Kolonomos, Žamila, Poslovice, Izreke i Priče Sefardskih Jevreja Makedonije / Proverbs, sayings and tales of the Sephardi Jews of Macedonia, Savez jevrejskih opština Jugoslavlje, 1979.Search in Google Scholar
Laca, Brenda, Sobre el uso del acusativo preposicional en español, in: Pensado, Carmen (ed.), El complemento directo preposicional, Madrid, Visor, 1995, 61–91.Search in Google Scholar
Laca, Brenda, El objeto directo. La marcación preposicional, in: Company Company, Concepción (ed.), Sintaxis histórica de la lengua española. Primera parte: La frase verbal, Mexico DF, UNAM/Fondo de Cultura Económica, 2006, 423–475.Search in Google Scholar
Ledgeway, Adam, Differential object marking in the dialects of Southern Italy. An overview, Caplletra 74 (Primavera, 2023), 181–282.10.7203/caplletra.74.26041Search in Google Scholar
Leonetti, Manuel, Indefiniteness and Specificity, in: Hualde, José Ignacio/Olarrea, Antxon/O’Rourke, Erin (edd.), The Handbook of Hispanic Linguistics, Blackwell, 2012, 285–305.10.1002/9781118228098.ch15Search in Google Scholar
Leonetti, Manuel, Specificity and Differential Object Marking in Spanish, Catalan Journal of Linguistics 3:1 (2004), 75–114.10.5565/rev/catjl.106Search in Google Scholar
Leonetti, Manuel, Specificity in clitic doubling and in Differential Object Marking, Probus 20:1 (2008), 33–66.10.1515/PROBUS.2008.002Search in Google Scholar
Longobardi, Giuseppe, Reference and proper names. A theory of N-movement in syntax and Logical Form, Linguistic Inquiry 25 (1994), 609–665.Search in Google Scholar
Longobardi, Giuseppe, Toward a unified grammar of reference, Zeitschrift für Sprachwissenschaft 24 (2005), 5–44.10.1515/zfsw.2005.24.1.5Search in Google Scholar
López, Luis, Indefinite objects. Scrambling, choice functions and differential marking, Cambridge, MA, MIT Press, 2012.10.7551/mitpress/9165.001.0001Search in Google Scholar
Lyons, Christopher, Definiteness, Cambridge, Cambridge University Press, 1999.Search in Google Scholar
Luria, Max A., A Study of the Monastir Dialect of Judeo-Spanish based on oral material collected in Monastir, Yugo-Slavia, PhD thesis, Columbia University, 1930.Search in Google Scholar
Malchukov, Andrej, Animacy and asymmetries in differential case marking, Lingua 118 (2008), 203–221.10.1016/j.lingua.2007.02.005Search in Google Scholar
Malchukov, Andrej/de Hoop, Helen, Tense, aspect, and mood based differential case marking, Lingua 121 (2011), 35–47.10.1016/j.lingua.2010.07.003Search in Google Scholar
Malchukov, Andrej/de Swart, Peter, Differential case marking and actancy variations, in: Malchukov, Andrej/Spencer, Andrew (edd.), The Oxford handbook of case, Oxford, Oxford University Press, 2009, 339–355.10.1093/oxfordhb/9780199206476.013.0023Search in Google Scholar
Marín Ramos, Ferrán, Grámatica básica de Djudeo-espanyol, MS Publishers, 2018.Search in Google Scholar
Markova, Alla, Beginner’s Ladino, New York, Hippocrene Books, 2008.Search in Google Scholar
McNally, Louise, Strong and weak nominals, in: Matthewson, Lisa/Meier, Cécile/Rullmann, Hotze/Ede, Thomas Zimmermann (edd.), Companion to Semantics, Wiley Blackwell, 2020, 1–20.10.1002/9781118788516.sem084Search in Google Scholar
Melis, Chantal, Spanish indexing DOM, topicality, and the case hierarchy, in: Seržant, Ilja A./Witzlack-Makarevich, Alena (edd.), Diachrony of differential argument marking, Berlin, Language Science Press, 2018, 97–128.Search in Google Scholar
Meyer-Lübke, Wilhelm, Grammatik der Romanischen Sprachen III. Romanische Syntax, Leipzig, Reisland, 1899.Search in Google Scholar
Mihajlovski, Robert, The Religious and Cultural Landscape of Ottoman Manastır, Leiden/Boston, Brill, 2021.10.1163/9789004465268Search in Google Scholar
Milsark, Gary Lee, Existential sentences in English, PhD dissertation, Massachusetts Institute of Technology, 1974.Search in Google Scholar
Næss, Åshild, What markedness marks. The markedness problem with direct objects, Lingua 114 (2004), 1186–1212.10.1016/j.lingua.2003.07.005Search in Google Scholar
Ormazabal, Javier/Juan Romero, Differential object marking, case and agreement, Borealis: An International Journal of Hispanic Linguistics 2:2 (2013), 221–239.10.7557/1.2.2.2808Search in Google Scholar
Pensado, Carmen, La creación del complemento directo preposicional y la flexión de los pronombres personales en las lenguas románicas, in: Pensado, Carmen (ed.), El complemento directo preposicional, Madrid, Visor Libros, 1995, 179–233. Search in Google Scholar
Postal, Paul, On so-called “pronouns” in English, in: Reibel, David A./Schane, Sanford A. (edd.), Modern studies in English, Englewood Cliffs, NJ, Prentice Hall, 1969, 201–224.Search in Google Scholar
Pottier, Bernard, L’emploi de la préposition “a” devant l’objet en espagnol, Bulletin de la Société de Linguistique 63 (1968), 83–95.Search in Google Scholar
Quintana Rodríguez, Aldina, Geografía lingüística del judeoespañol. Estudio sincrónico y diacrónico, Bern, Peter Lang, 2006.10.3726/978-3-0351-0275-8Search in Google Scholar
Rodríguez-Mondoñedo, Miguel, The syntax of objects. Agree and differential object marking, PhD thesis, University of Connecticut, 2007.Search in Google Scholar
Rohlfs, Gerhard, Autour de l’accusatif prépositionnel dans les langues romanes, Revue de Linguistique Romane 35 (1971), 312–334.Search in Google Scholar
Romero Heredero, Diego, Telicity and Differential Object Marking in the history of Spanish, in: Kabatek, Johannes/Obrist, Philipp/Wall, Albert (edd.), Differential Object Marking in Romance. The third wave, Berlin/Boston, De Gruyter, 2021, 315–338.10.1515/9783110716207-011Search in Google Scholar
Romero Heredero, Diego, Marcado diferencial de objeto y semántica verbal en español, Berlin/Boston, De Gruyter, 2022.10.1515/9783110751734Search in Google Scholar
Romero, Rey, Variation in Balkan Judeo-Spanish final clauses, Ianua. Revista Philologica Romanica 11 (2011), 89–99.Search in Google Scholar
Saab, Andrés/Zdrojewski, Pablo, On the Nonexistence of Asymmetric DOM in Spanish, Linguistic Inquiry 52:4 (2021), 852–866.10.1162/ling_a_00389Search in Google Scholar
Schikowski, Robert/Iemmolo, Giorgio, Commonalities and differences between differential object marking and indexing, University of Zurich, Department of Comparative Linguistics, 2015, 1–35.Search in Google Scholar
Silverstein, Michael, Hierarchy of features and ergativity, in: Dixon, Robert M.W. (ed.) Grammatical categories in Australian languages, Canberra, Australian Institute of Aboriginal Studies, 1976, 112–171.Search in Google Scholar
Sinnemäki, Kaius, A typological perspective on Differential Object Marking, Linguistics 52 (2014), 281–313.10.1515/ling-2013-0063Search in Google Scholar
Spitzer, Leo, Rum. “P(r)e”, Span. “a” vor persönlichem Akkusativobjekt, Zeitschrift für romanische Philologie 48 (1928), 423–432.10.1515/zrph.1928.48.1.77Search in Google Scholar
Suñer, Margarita, The role of agreement in clitic-doubled constructions, Natural Language and Linguistic Theory 6 (1988), 391–434.10.1007/BF00133904Search in Google Scholar
Timberlake, Alan, Hierarchies in the Genitive of Negation, The Slavic and East European Journal 19:2 (1975), 123–138.10.2307/306765Search in Google Scholar
Tomić, Olga Mišeska, Balkan Sprachbund morpho-syntactic features, Dordrecht, Springer, 2006.10.1007/1-4020-4488-7Search in Google Scholar
Torrego, Esther, The dependencies of objects, Cambridge, MA, MIT Press, 1998.10.7551/mitpress/2337.001.0001Search in Google Scholar
Varol-Bornes, Marie-Christine, Le judéo-espagnol vernaculaire d’Istanbul, Bern, Peter Lang, 2008.Search in Google Scholar
Vidaković-Petrov, Krinka, Corpus of Sephardic Folklore from Macedonia, in: Grandakovska, Sofija (ed.), The Jews of Macedonia and the Holocaust. History, Theory and Culture, An International Research Project, Institute in Social Sciences and Humanities Research “Euro-Balkan”, Skopje, 2011, 539–558.Search in Google Scholar
Witzlack-Makarevich, Alena/Seržant, Ilja A., Differential Argument Marking. Patterns of variation, in: Seržant, Ilja A./Witzlack-Makarevich, Alena (edd.), Diachrony of Differential Argument Marking, Berlin, Language Science Press, 2018, 1–40.Search in Google Scholar
Zdrojewski, Pablo, Caso /a/ caso: una teoría para la marcación diferencial de objetos del español, PhD thesis, University of Buenos Aires, 2023.Search in Google Scholar
© 2024 the author(s), published by Walter de Gruyter GmbH, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
Articles in the same Issue
- Frontmatter
- Frontmatter
- Aufsätze
- À la recherche du neutre protoroman. Réflexions à partir de l’expérience de rédaction et de révision de l’article */'ram‑u/ s.n. ‘branche ; branchage’ du DÉRom
- Senior domno Orbita confirmat. Evolución sociopragmática de los tratamientos don y señor en el iberorromance primitivo
- Aproximación estadística al reparto de <b> y <v> en manuscritos latinos de autores hispanos
- A ver (si): diacronía de un marcador del discurso en español
- Exclamativas encubiertas en español
- Más las oigo, mejor me parecen: comparativas proporcionales idiolectales
- Differential marking of direct objects in Monastirli džudezmu: a case study in Judeo-Spanish morphosyntax
- Penitencia cuaresmal como medicina del alma. Nueva edición de un sermón de Cuaresma de San Vicente Ferrer
- Amor y diálogo en la trama trovadoresca
- Origine dell’it. anche e dei suoi correlati romanzi
- Miszellen
- Die Wortgeschichte von fr. banqueroute, dt. Bankrott und it. bancarotta
- Délocutifs français d’origine bretonne
- Besprechungen
- Elena Liverani/Alessandro Parenti, Il dizionario spagnolo-italiano di Nicolao Landucci (1562), Biblioteca dell’Archivum Romanicum, Serie 2: Linguistica, vol. 64, Firenze, Olschki, 2022, VI + 352 p.
- Josephine Klingebeil, Codex oblitus. Curtius Rufus in der italienischen Übersetzung von Lodrisio Crivelli: Analyse zur Edition des Mscr.Dresd.Ob.47 (Beihefte zur Zeitschrift für romanische Philologie, 464), Berlin/Boston, De Gruyter, 2022, XVIII + 338 p.
- Il manoscritto Saibante-Hamilton 390, Edizione critica diretta da Maria Luisa Meneghetti, coordinamento editoriale di Roberto Tagliani, Roma, Salerno Editrice, 2019, CCXVI + 616 p.
Articles in the same Issue
- Frontmatter
- Frontmatter
- Aufsätze
- À la recherche du neutre protoroman. Réflexions à partir de l’expérience de rédaction et de révision de l’article */'ram‑u/ s.n. ‘branche ; branchage’ du DÉRom
- Senior domno Orbita confirmat. Evolución sociopragmática de los tratamientos don y señor en el iberorromance primitivo
- Aproximación estadística al reparto de <b> y <v> en manuscritos latinos de autores hispanos
- A ver (si): diacronía de un marcador del discurso en español
- Exclamativas encubiertas en español
- Más las oigo, mejor me parecen: comparativas proporcionales idiolectales
- Differential marking of direct objects in Monastirli džudezmu: a case study in Judeo-Spanish morphosyntax
- Penitencia cuaresmal como medicina del alma. Nueva edición de un sermón de Cuaresma de San Vicente Ferrer
- Amor y diálogo en la trama trovadoresca
- Origine dell’it. anche e dei suoi correlati romanzi
- Miszellen
- Die Wortgeschichte von fr. banqueroute, dt. Bankrott und it. bancarotta
- Délocutifs français d’origine bretonne
- Besprechungen
- Elena Liverani/Alessandro Parenti, Il dizionario spagnolo-italiano di Nicolao Landucci (1562), Biblioteca dell’Archivum Romanicum, Serie 2: Linguistica, vol. 64, Firenze, Olschki, 2022, VI + 352 p.
- Josephine Klingebeil, Codex oblitus. Curtius Rufus in der italienischen Übersetzung von Lodrisio Crivelli: Analyse zur Edition des Mscr.Dresd.Ob.47 (Beihefte zur Zeitschrift für romanische Philologie, 464), Berlin/Boston, De Gruyter, 2022, XVIII + 338 p.
- Il manoscritto Saibante-Hamilton 390, Edizione critica diretta da Maria Luisa Meneghetti, coordinamento editoriale di Roberto Tagliani, Roma, Salerno Editrice, 2019, CCXVI + 616 p.
