3.1.1 Spanish

The diachronic evolution of DOM in Spanish has received significant attention (Company, Reference Company2002; Laca, Reference Laca and Pensado Ruiz1995, Reference Laca and Company2006; Von Heusinger and Kaiser, Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005) and there is significant research as well on how it is currently expressed in different varieties (Brugè and Brugger, Reference Brugè and Brugger1996; Fábregas, Reference Fábregas2013; Leonetti, Reference Leonetti2004, Reference Leonetti2008; López, Reference López2012; Pensado, Reference Pensado2005; Rodríguez-Mondoñedo, Reference Rodríguez-Mondoñedo2007; Torrego, Reference Torrego1998; Von Heusinger and Kaiser, Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005; Weissenrieder, Reference Weissenrieder1985, Reference Weissenrieder1990). Descriptively, human and specific (definite) direct objects are obligatorily marked with the preposition a. In (7), which is also the preposition that marks goals in ditransitive constructions (dative case, e.g., Roberto le dio un libro a Patricia “Roberto gave Patricia a ring”), the preposition that appears with psych predicates (dative experiencers, e.g., A Juan le gusta el helado “Juan likes ice cream”), and the directional preposition used with locatives (e.g., Fueron a la tienda “They went to the store”). Human non-specific objects, as in (8), are optionally marked (depending on context). Inanimate objects – specific in (9) and non-specific in (10) – are not typically marked. The most succinct generalization in Spanish is that if the object is animate and specific it must be marked. Other objects are typically unmarked.

There is substantial variability in the use of DOM with inanimate objects (Fábregas, Reference Fábregas2013; López, Reference López2012). Some verbs like llamar “call” describir “describe” and seguir “follow” require DOM with inanimate objects (e.g., llamar a una ambulancia “call an ambulance,” seguir a un auto “follow a car”); inanimate proper nouns can be optionally marked by DOM, visitar a Brasil “visit Brasil” (Laca, Reference Laca and Company2006); the use of DOM with animals (non-human animate) is variable and depends on the size and or domesticity of the animal (matar un mosquito “kill a mosquito,” pasear al/ a un perro “walk the/a dog”); DOM can be used to disambiguate subject and object when both are inanimate (El submarino hundió al barco “The submarine sank the ship”); and when inanimate objects are topicalized in clitic left dislocations, they tend to favor DOM (A la película la terminé de ver ayer “The movie I finished watching yesterday”) in some varieties of Spanish (Montrul, Reference Montrul2013a).

For Torrego (Reference Torrego1998), definiteness, specificity, aspect, topicality, agentivity, and affectedness determine when objects are marked (see also Leonetti, Reference Leonetti2004 and Zubizarreta, Reference Zubizarreta1998). However, Von Heusinger and Kaiser’s (Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005) diachronic analysis suggests that as far as properties of the object are concerned, animacy and specificity are the most relevant parameters for DOM in Spanish and definiteness less relevant. All other factors (agentivity, affectedness, aspect, etc.) are contributing but not deterministic factors. López (Reference López2012) argues that within the object properties, animacy is the overriding feature that controls DOM in Spanish, at least with respect to the structure of nouns. The empirical evidence for his claim comes from the behavior of (non-specific) quantifiers like alguien “somebody,” nadie “nobody,” ninguno “no,” muchos “many,” pocos “few,” which require DOM when referring to animate objects (Vi a alguien “I saw somebody”/No vi a nadie “I saw somebody/I did not see anybody” vs. Vi algo “I saw something”/No vi nada “I did not see anything”).

Because our study is concerned with language change in the DOM system of Spanish in a language contact situation, an important question is whether the changes in DOM expression present in heritage language grammars are similar to or different from the changes already in progress in different Spanish varieties. I discuss next how DOM emerged along the Animacy and Definiteness scales, and the current variation attested in different monolingual or homeland varieties, before outlining the synchronic analyses of DOM assumed in this study.

Synchronic studies of regional varieties of Spanish (Argentina, Cuba, Mexico, Spain, US Spanish, Andean Spanish) from both sociolinguistic and experimental psycholinguistic perspectives have uncovered important variation with uses of DOM by monolingual (Arechabaleta Regulez, Reference Arechabaleta Regulez2019; Arechabaleta Regulez and Montrul, Reference Arechabaleta Regulez and Montrul2021; Bautista-Maldonado and Montrul, Reference Bautista-Maldonado and Montrul2019; Montrul, Reference Montrul2013a; Tippets, Reference Tippets2010) and bilingual speakers (Arechabaleta Regulez, Reference Arechabaleta Regulez2019, Reference Arechabaleta Regulez, Mardale and Montrul2020; Hur, Reference Hur, Mardale and Montrul2020; Montrul, Reference Montrul2004, Reference Montrul2014; Montrul and Bowles, Reference Montrul and Bowles2009; Montrul and Sánchez-Walker, Reference Montrul and Sánchez-Walker2013; Montrul et al., Reference Montrul, Bhatt and Girju2015), and it is important to understand how the current variation relates to diachronic DOM expansion. DOM did not exist in Latin: Latin direct objects were marked with accusative case but with no preposition. Among the Romance languages, only Spanish and Romanian developed full-fledged DOM systems, and the current tendency in most monolingual varieties of Spanish has been the gradual expansion of DOM along the Animacy and Definiteness scales, to inanimate definite objects. However, in bilingual varieties in contact with non-DOM languages we typically see the opposite of expansion, i.e., retraction through omission of DOM in core cases (animate, specific objects), as we confirm in this study. An exception to this pattern is a variety of Italian that exhibits DOM (Sicilian) in contact with Québec French that appears to retain DOM (D’Alessandro, Reference D’Alessandro2021).

Aissen (Reference Aissen2003) analyzed instances of DOM in the Medieval text Poema del Mio Cid. She found that a-marking with direct objects was less widely distributed then than it is in present day Spanish. In Medieval Spanish, common noun phrases referring to humans, both definite and indefinite, were not always marked, as shown in (11), whereas today they must be marked, as in (12).

Other diachronic studies (Company, Reference Company2001, Reference Company2002; Laca, Reference Laca and Company2006) have shown that a-marking is also advancing with inanimate objects in some Spanish varieties (see Table 3.3), such as Mexican Spanish, as in (13) and (14) (examples from Company, Reference Company2001: p. 147).

The evolution of DOM with inanimate objects was further confirmed by Von Heusinger and Kaiser’s (Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005) diachronic and synchronic investigation of Spanish DOM, using corpora from the Poema de Mio Cid and its contemporary translations, and synchronic analyses of different varieties of Spanish (Argentina, Uruguay, Peru, and Mexico) collected from written and oral corpora (informal interviews, short stories, and email messages to a newspaper). The purpose of their investigation was to trace whether the evolution of DOM in Old Spanish and in different Latin American varieties followed the definiteness scale and the animacy scale. Tables 3.4 and 3.5 show the distribution of DOM along the two scales (columns for definiteness and rows for animacy). Grey scale shows expansion.

Comparing Tables 3.4 and 3.5, we see that DOM extended from marking animate pronouns and proper names to marking animate and specific noun phrases. This transition was mediated by topicality because at first only topicalized specific noun phrases were marked, while non-topicalized ones were not. The expansion later included indefinite NPs with the features [± specific]. So, in the transition from Old to Modern Spanish, topicality and specificity were transitional triggers affecting definite and indefinite NPs, respectively.

The evolution of DOM in Latin American Spanish in Table 3.6 shows expansion along the Animacy scale, since inanimate definite objects are specific and can receive DOM marking.

Several semantic and syntactic factors facilitated the marking of specific inanimate objects. These include the lexical nature of the verb (verbs that seem to be followed by the directional preposition “a,” like seguir a or perseguir a “follow”), secondary predication (Consideran a las facturas fraudulentas “They consider the bills fraudulent”), the preverbal or topicalized position of the objects (A esta película ya la vi “This movie I already saw”), and clitic doubling (Lo vi a Juan “ I saw Juan”). Von Heusinger and Kaiser (Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005) also suggest that there is synchronic variation in the parameters that regulate DOM in Latin American Spanish. For example, while definiteness and specificity trigger DOM in the Spanish spoken in Argentina and Uruguay, animacy and specificity are the main triggers in other varieties.

Several studies have investigated the factors that may underlie variable uses of DOM in different varieties, such as peninsular Spanish and Argentinian Spanish (Dumitrescu, Reference Dumitrescu1997), and between Mexican, Argentinian, and Peninsular Spanish (Arechabaleta Regulez, Reference Arechabaleta Regulez2019; Arechabaleta Regulez and Montrul, Reference Arechabaleta Regulez and Montrul2021; Balasch, Reference Balasch2011; Tippets, Reference Tippets2010), to enable understanding of the evolutionary path of DOM. There are also studies of DOM variation within varieties, such as in the Spanish of Cuba (Alfaraz, Reference Alfaraz2011), of Argentina (Montrul, Reference Montrul, Rodríguez Louro and Colantoni2013b), of Mexico (Arechabaleta Regulez, Reference Arechabaleta Regulez2019; Arechabaleta Regulez and Montrul, Reference Arechabaleta Regulez and Montrul2021; Bautista-Maldonado and Montrul, Reference Bautista-Maldonado and Montrul2019), and of the United States (Arechabaleta Regulez, Reference Arechabaleta Regulez, Mardale and Montrul2020; Montrul, Reference Montrul2004, Reference Montrul2014; Montrul and Bowles, Reference Montrul and Bowles2009; Montrul and Sánchez-Walker, Reference Montrul and Sánchez-Walker2013; Montrul et al., Reference Montrul, Bhatt and Girju2015). Topicalization turns out to be a very significant driving factor in this evolution, as mentioned earlier. In Mexico and Argentina, definite inanimate objects tend to appear with DOM in production (e.g., choqué al auto “I crashed the car,” cocecharon al maíz “They harvested corn”), especially when these are topicalized (A esta plaza la patrocinan Aerolineas Argentinas y Ud. “This park is sponsored by Aerolineas Argentinas and you”) (Company, Reference Company2001; Dumitrescu, Reference Dumitrescu1997; Von Heusinger and Kaiser, Reference Von Heusinger and Kaiser2011). The acceptance of topicalized inanimate objects with DOM was confirmed by Montrul’s (Reference Montrul, Rodríguez Louro and Colantoni2013b) experimental study with 26 native speakers in Argentina. The native speakers completed a written grammatical judgment task testing, in simple clauses, animate and inanimate objects with and without DOM, with and without accusative clitic doubling,and with topicalizations. Montrul investigated whether topicalizations and accusative clitic doubling, which are common in Argentine Spanish, contributed to the extension of DOM to inanimate objects. The results showed categorical acceptance of DOM with animate objects and acceptablity to some degree with inanimate specific and non-specific objects, especially when these were clitic doubled and topicalized (A la casa la compró Mariela “DOM the house it Mariela bought”). Montrul’s experimental findings confirmed Dumitrescu’s (Reference Dumitrescu1997) observation of DOM in speech and street signs as well as the evolutionary paths described by Company (Reference Company2001) and Von Heusinger and Kaiser (Reference Von Heusinger, Kaiser, Von Heusinger, Kaiser and Stark2005).

Bautista-Maldonado and Montrul (Reference Bautista-Maldonado and Montrul2019) investigated DOM in native speakers of Mexican Spanish using an elicited production task and a written acceptability judgment task. They found that DOM with animate, specific direct objects is produced consistently, and is categorically accepted as grammatical. Except for non-specific NPs, all other NP types along the definiteness scale showed categorical DOM marking (Pronouns, Names, specific NPs, and quantifiers). They also confirmed that there is a slight tendency toward DOM marking with specific inanimate objects, although the expansion of DOM to inanimate objects with topicalizations was not examined in this study. Non-specific inanimate objects were preferred without DOM in the acceptability judgment task. An unexpected finding of this study, which goes against what has been reported in the literature, is that native speakers of Mexican Spanish did not produce or accept clitic-doubled personal pronouns with DOM, such as Lo vi a él “I saw him,” preferring vi a él, without the clitic, or lo vi, without the strong pronoun, instead). (Arechabaleta Regulez, Reference Arechabaleta Regulez2019; Arechabaleta Regulez and Montrul, Reference Arechabaleta Regulez and Montrul2021) tested young native speakers of Mexican Spanish with two oral production tasks, an acceptability judgment task, and a sentence processing task with eye tracking during reading to corroborate the expansion of DOM to inanimate objects in Mexican Spanish. Although the native speakers did not produce inanimate objects with DOM in the two production tasks, they rated them more acceptable (*Cocecharon al maíz “They harvested corn”) than unmarked animate objects (*Vieron Juan “They saw Juan”) in the acceptability judgment task. In the eye tracking processing task, the Mexican speakers showed sensitivity (measured by longer latencies) to DOM omission in sentences with animate objects (*El actor liberó el compañero con su llave “The actor freed his partner with is key”) but less sensitivity (shorter latencies) to the ungrammaticality of sentences with marked inanimate objects (*El joven movió al sofá a la calle para dormir “The young man moved the sofa to the street to sleep”). Since DOM innovations were detected in online processing more than in grammaticality judgments and oral production, Arechabaleta Regulez‘s (Reference Arechabaleta Regulez2019) and Arechabaleta Regulez and Montrul’s (Reference Arechabaleta Regulez and Montrul2021) results support the hypothesis that language variation and change may start with comprehension before it reaches production (Lundquist et al., Reference Lundquist, Rodina, Sekerina and Westergaard2016).

As we will cover in more depth in the rest of this book, retraction (omission) of DOM has also been reported in other varieties, particularly in Spanish in contact with English in the United States (Montrul, Reference Montrul2014; Montrul and Bowles, Reference Montrul and Bowles2009) and in Spanish in contact with Quechua in Perú (Sánchez, Reference Sánchez2003), both are bilingual varieties where Spanish is in contact with a non-DOM language. Yet, omission of DOM with animate and specific objects has also been attested in monolingual homeland varieties, as in Cuban (Alfaraz, Reference Alfaraz2011) and Dominican Spanish (Lunn, Reference Lunn, Lee, Geeslin and Clemens2002). For example, Alfaraz (Reference Alfaraz2011) reports significant omission of DOM with animate specific direct objects in spontaneous production, as (15) and (16) illustrate. The data came from younger and older speakers, and the younger speakers omitted DOM with animate objects almost 80% of the time compared to the older speakers, who omitted about 40% of the time. If these trends are replicated in other studies, they would indicate diachronic change in the monolingual variety.

In sum, leaving aside bilingual varieties for the moment, it seems that recent studies of monolingual varieties of Spanish have uncovered important variation in the expression of DOM with animate and inanimate objects.

3.1.2 Romanian

DOM in Romanian is marked with the preposition pe, which is homophonous with the locative preposition pe, as in (17), and the preposition appearing with some indirect objects, as in (18).

The two main parameters that regulate DOM in Romanian are animacy and referential stability (Ciovârnache and Avram, Reference Ciovârnache and Avram2013; Farkas and Von Heusinger, Reference Farkas and von Heusinger2003; Mardale, Reference Mardale2008, Reference Mardale2010a). Referential stability is related to how the specificity value of a given DP can change depending on discourse properties (Farkas and Von Heusinger, Reference Farkas and von Heusinger2003). Farkas and Von Heusinger (Reference Farkas and von Heusinger2003) proposed the referentiality stability scale, as in (19).

Following the referentiality scale in (19), pe-marking is obligatory with personal pronouns and proper names, as in (20), including those referring to personified animals (Farkas and Von Heusinger, Reference Farkas and von Heusinger2003; Române, Reference Române2005; Von Heusinger and Chiriacescu, Reference Von Heusinger, Chiriacescu, Espinal, Leonetti and McNally2009; Von Heusinger and Gáspár, Reference Von Heusinger and Gáspár2008). Names of cities (i.e. Chicago) are not pe-marked. Pe-marking is optional with referentially stable objects, as in modified definite animate DPs in (21) and indefinite animate DPs, in (22).

Pe-marking is generally ungrammatical with inanimate objects, both definite specific and indefinite DPs, as in (23) and (24).

In Romanian, animacy is less deterministic as a trigger of DOM than it is in Spanish because some specific inanimate objects can be DOM-marked. DOM (pe-marking) is also sensitive to definiteness, specificity, and topicalization (Cornilescu, Reference Cornilescu2000; Mardale, Reference Mardale2010b; Tigău, Reference Tigău2010). Strong pronominal direct objects are obligatorily marked with pe irrespective of animacy, as in (25b). Inanimate objects are usually not acceptable with DOM (Irimia, Reference Irimia2018); however, specific inanimate objects can also be differentially marked with pe, as is (26), and particularly when dislocated, as in (27a). Some speakers find that using DOM with inanimate objects in a post verbal position creates a semantic upgrading effect (27b) (Ticio and Avram, Reference Ticio and Avram2015: p. 387).

Summarizing, animate specific direct objects are marked with the preposition pe in Romanian, especially if they are expressed with names or pronouns. Definite animate objects (definite DPs) are optionally pe-marked because, semantically, they are less referentially stable, and inanimate specific objects can be pe-marked if dislocated. In addition, there are other syntactic constraints that determine the optionality of pe-marking with definite DPs. Romanian has post-nominal definite articles that are an enclitic suffix (e.g. băiat-ul ‘boy-the’). Definite DPs behave differently depending on the presence or absence of the definite enclitic article, and this is the result of a syntactic constraint independent of DOM that applies to all prepositions in Romanian (except for cu ‘with’). Therefore, constructions with a preposition + noun with a definite article suffix are ungrammatical, as the sentence in (28) shows. If the noun is suffixed with a definite article (băiat-ul ‘boy-the’), it cannot be pe-marked unless it is further modified with an adnominal expression, as in băiatul înalt (boy-the tall) (Dobrovie-Sorin, Reference Dobrovie-Sorin1994; Mardale, Reference Mardale2010b; Von Heusinger and Chiriacescu, Reference Von Heusinger and Chiriacescu2013). Some Romanian speakers use pe and drop the definite article leaving a bare noun, as in (29), while others drop pe and use the definite article, as in (30) (Von Heusinger and Chiriacescu, Reference Von Heusinger, Chiriacescu, Espinal, Leonetti and McNally2009; Von Heusinger and Onea Gáspar, Reference Von Heusinger and Gáspár2008).

When DOM is present with definite DPs, the nominal expression is referentially stable because DOM provides referential persistence in the discourse (Chiriacescu and Von Heusinger, Reference Chiriacescu and von Heusinger2010); when DOM is omitted, the expression has a role reading (e.g., the director, the king, the president), where the referent is not stable (Hill, Reference Hill2013). Some familiar functional expressions such as the mother, the teacher, the priest, or the boss are exceptions to the syntactic constraint on prepositions and may appear with pe (at least in spoken Romanian) (Von Heusinger and Gáspár, Reference Von Heusinger and Gáspár2008). Example (31) is acceptable under the reading that the boss is granted contextual uniqueness to its referent (role reading), though such expressions are rare:

Another distinctive syntactic characteristic of Romanian is the presence of accusative clitic doubling (CD), by which the object DP is doubled by an accusative clitic. Whenever a direct object is doubled by an accusative clitic, DOM (pe-marking) is required, as in (32) and (33). Clitic doubled objects without pe-marking are typically ungrammatical (Farkas and Von Heusinger, Reference Farkas and von Heusinger2003), but there is also some variation in native speakers of Romanian, as we discuss below.

The distribution of CD with postverbal objects also follows the definiteness and referentiality scales: while pronouns and proper names always occur with CD, there is a strong preference for CD with pe-marked definite direct objects as in (34), and some preference for CD with pe-marked animate indefinite direct objects, as in (35) (Von Heusinger and Gáspár, Reference Von Heusinger and Gáspár2008). Inanimate objects are ungrammatical with CD because CD requires DOM, andnon-specific inanimate objects cannot be pe-marked, as in (36).

There is disagreement regarding the use of CD in some contexts, particularly for definite modified objects, and this is likely due to register (higher registers prefer CD). In general, Gramatica Limbii Române (Reference Române2005), Von Heusinger and Chiriacescu (Reference Von Heusinger, Chiriacescu, Espinal, Leonetti and McNally2009), Von Heusinger and Gáspár (Reference Von Heusinger and Gáspár2008) all indicate that CD is obligatory with proper names and personal pronouns, regardless of whether the pronouns refer to animate or inanimate objects, and thus sentences as in (36) are ungrammatical:

With modified animate DPs, CD and DOM are optional (Farkas and Von Heusinger, Reference Farkas and von Heusinger2003). Gramatica Limbii Române (Reference Române2005) notes that CD is obligatory with modified animate DPs with strong quantifiers, while definite modified DPs are reported to be marginal for some speakers (Von Heusinger and Chiriacescu, Reference Von Heusinger, Chiriacescu, Espinal, Leonetti and McNally2009). For the purposes of the present study, I assume that CD and DOM are both required for proper names and personal pronouns, and optional with definite DPs, despite the variation that exists in the language. DOM and CD are ungrammatical with inanimate objects, especially if these are not specific.

David (Reference David2015) proposes that the clitic-doubled pe construction is prototypically employed in clauses involving direct objects with human referents but can be extended to clauses containing non-human referents, as the constructional polysemy branches out. CD with inanimate objects and pe was common in Old Romanian (Hill and Mardale, Reference Hill and Mardale2017). Unlike other Balkan languages, Romanian does not have a clitic doubling construction independent of DOM (all clitic-doubled constructions must have pe-marking), although this was possible in Old Romanian. Some pe-marked direct objects without CD are heard in present-day spoken Romanian spoken, but these are dispreferred in the higher registers. It appears that in the historical development of Romanian, CD and DOM were different constructions that developed independently (Cornilescu, Reference Cornilescu2000; Von Heusinger and Onea Gáspár, Reference Von Heusinger and Gáspár2008, but cf., Hill and Mardale, Reference Hill and Mardale2017). Because both CD and DOM mark discourse prominence, the two constructions became entrenched in present day Romanian (David, Reference David2015). Diachronically, object topicalization (fronting) and clitic left dislocation (CLLD) facilitated the merger of CD and DOM with animate objects into what today can be considered a merged construction. In this study, we examine the extent to which different speakers of Romanian, including heritage speakers, immigrants, and Romanian speakers in Romania, exhibit this close relationship between DOM and CD in their grammars.

3.1.3 Hindi

Hindi/Urdu (henceforth Hindi) is a SOV split ergative language that also exhibits DOM marked with the post-position –ko, as shown in (36). Hindi does not have definite articles, unlike Spanish and Romanian. Hence, when –ko appears with direct objects, it signals human animacy and specificity. –ko is obligatory with pronouns and human personal names; it is optional with animals (Butt, Reference Butt1993; Junghare, Reference Junghare1983; Magier, Reference Magier1987; Masica, Reference Masica1982; Mohanan, Reference Mohanan1993, Reference Mohanan1994a, Reference Mohanan, Butt, Holloway King and Ramchand1994b; Singh, Reference Singh1994).

Human, specific direct objects must be overtly marked with –ko, as in (37), and are ungrammatical without –ko marking. Non-human animate, specific direct objects may be optionally marked with –ko, as in (38).

Ko-marking is obligatory with specific indefinites. In general, non-specific human objects are marginally acceptable without -ko marking, as in (39).

Animate non-human non-specific objects are unmarked, as in (40a), and they are unacceptable with overt –ko marking (40b).

Inanimate, specific direct objects can also be optionally marked with –ko, as in (41):

Overt marking is optional with inanimate objects, but only when these are specific. Thus, both (42a,b) are possible, and while the object in (42a) can be interpreted as either indefinite or definite, the ko-marked one in (42b) must refer to an already mentioned banana:

Inanimate, non-specific objects, as in (43), are ungrammatical with –ko-marking. However, specificity has to be evaluated in context.

Object marking in Hindi has quite a different function for inanimate objects than it does for human objects: in the case of inanimate objects only, -ko marks definiteness. The claim that -ko is possible with indefinite animate objects, but not with indefinite inanimate objects, is supported by Singh’s (Reference Singh1994: p. 227) and Mohanan’s (Reference Mohanan1994a: p. 79) observations that the indefinite marker ek may occur with ko-marked objects if they refer to animate objects, but not if they refer to inanimate objects. However, Hindi sources in general vary considerably on the status of overt object marking on inanimate objects. Butt (Reference Butt1993) claims that ko-marking is possible for specific indefinite inanimate objects, whereas Singh (Reference Singh1994) mentions that some speakers disallow -ko with inanimate objects altogether. It is unclear whether this variation is due to dialect variation or to other linguistic factors, since context, word order, and aspect (in addition to overt case) play roles in fixing the specificity of inanimate direct objects in Hindi (Butt, Reference Butt1993; Mahajan, Reference Mahajan1990; Mohanan, Reference Mohanan1993, Reference Mohanan1994a; Singh, Reference Singh1994). To summarize, following Mohanan (Reference Mohanan1994a, Reference Mohanan, Butt, Holloway King and Ramchandb), Dayal (Reference Dayal2011), Kulikov et al. (Reference Kulikov, Malchukov and de Swart2006), and López (Reference López2012), the following generalizations for –ko marking as DOM are assumed in this study:

1. 1) pronouns, names and definite human DPs in object position are always marked with -ko

2. 2) -ko is obligatory with indefinite specific animate objects

3. 3) -ko is optional with inanimate objects depending on context and specificity

Thus, specificity more than definiteness and animacy is the main trigger for DOM in Hindi.