The Default Case in Standard Arabic

The default Case is a common phenomenon in Universal Grammar (UG). There are some languages which require that all Noun Phrases have Case. For these languages default Case meets something that has become known as the Case Filter (Rouveret and Vergnaud 1980). This is to say, if a particular Noun Phrase is not assigned a Case in association with some specification in some other part of the grammar, then default Case assignment principle can apply. Typical cross-linguistic default Cases are Nominative or Genitive, though the value of the default Case can vary from one language to another. While the default Case in English is accusative, it is nominative in most languages. The default mechanism which assigns this value is only invoked when the structural mechanism is not applicable. This paper argues, by citing multiple cross-linguistic examples, that assumption of a default Case in a language accounts for a better understanding of its syntactic and morphological structure. Based on Schütze‟s (2001) proposal for English, it develops a theory to account for the default Case in Standard Arabic (SA). It argues that nominal expressions in SA do not receive nominative Case by assignment of other syntactic means. As such, its mechanism does not interact with the Case Filter, which is assumed to be a syntactic constraint. This paper shows that diverse phenomena in the distribution of nominative nominal expressions in SA can be treated using default Case. Previous studies have ample evidence that such phenomena from other languages have proved that instances for default Case are common, and furthermore, that there are opportunities within the Case framework to reduce the cross-linguistic differences in Case patterns in the event of choosing a default Case.


Introduction
This paper defends a particular notion of default Case as part of Universal Grammar and explores its impact on the mechanism of morphological Case and its association with abstract Case. The major argument is based on a kind of default Case which contains the form of NPs in the PF component that is somehow independent and permitted at the stage of syntactic level, and provides a reasonable explanation of several facts about the way that morphological Case is assigned. In this regard, this study proposes the approach of Emonds (1985, 229) which says that "If possible, I would like to avoid recourse to a "fall-back" morphological Case, at least for German, and conserve the notion that whenever Case marking fails, then the case filter applies". This proposal aims to be a standard one for default, and examines a few similarities between some set of constituents in respect of their behavior in general, where features of these constituents of those sets do not share with each other the fact that they are not part of any natural classes within their category. In this example, the constituents are nominal because their Case is not specified at the syntax level, and is thus sent to the procedure of the morphological Spell-Out without any sign of the specific Case mark which they could be assigned to. In this regard, default Case has no significant role to play in licensing these NPs because licensing abstract Case is one of the requirements of syntax. Rather, when we realize that licensing and morphological Case is both independent mechanisms, the shortcoming in Emonds" proposal above should not appear: default Case in this proposal cannot protect a nominal expression from violating the Case Filter.

Theoretical Views on Case
The early 1990s marked a shift from GB (Government and Binding Theory) to Minimalism as the stated approach of choice for syntacticians. Commenting on the way most Minimalist works, Chomsky emphasizes that Minimalist is a program, not a theoretical approach. Contrary to formal frameworks to this idea, Chomsky thus focuses on affording accurate features of grammar at abstraction and generalization levels.
For Minimalism, the fundamental questions are to examine whether the Case theory deals with differences between nominative and accusative Case assignment, and whether it is possible to have a uniform theory for assigning these Cases. One approach (Chomsky, 1991) assumes to assimilate accusative Case assignment to the same type of structural configuration as nominative, namely Spec(ifier)-Head configuration. More precisely, generative theories of early 1990s contributed a basic grammar of the Spec-Head configuration in the theory of agreement, assuming that full agreement takes place only in Spec-Head configurations (see Koopman, 2006 for a recent defense) (Note 1). The other approach claims that all Case assignments are characterized by c-command (a conventional abbreviation of constituent-command) and locality (in Chomsky, 2000, the relationship is named Agree) (Note 2). The theory tends to consider Agree perspective since Koopman & Sportiche (1991) assume that subjects move from a lower position so Spec,IP (in the most recent work, Spec,TP).
The model of grammar in Minimalist syntax proposes that abstract Case features construct a system of uninterpretable formal features, located at the centre of linguistic coding of what International Journal of Linguistics ISSN 1948-5425 2020 Chomsky (2004, P.7) defines as "duality of semantics", one part is thematic structure and the other is information structure. Case features are determined via an agreement operation in the course of derivation, while under the most recent version, they are subject to a Probe-Goal system, a feature-checking mechanism. DPs and pronouns enter the syntax with an unvalued Case-feature, and serve as a Goal whose Case feature is valued in the course of the derivation via agreement relations with a c-commanding Probe which has some set of interpretable features. The following features state this operation (cited in Radford & Ramos, 2001, P.2): 1. An unvalued case-feature on a (noun or pronoun expression serving as a) Goal is valued as specified below (and deleted) via agreement in person and number with a c-commanding Probe (= higher head) which carries a specific set of interpretable features; the Goal"s Case Feature is valued as: In general, there are at least two proposals in literature. The first is proposed by Schütze (1997) and Chomsky (2001), which suggests that Case is licensed to DPs as a reflexing of the value of phi-features on the head of Case checking. The other speculation is proposed, by many including Pesetsky & Torrego (2004), which assumes DPs analogue of tense in a verbal system. Case may thus be understood as licensed on DPs as a result of Valuing a [uT] feature on D by T (u = uninterpretable).
Some nominal expressions carry uninterpretable feature to be checked, and others carry no Case feature at all. Each structure generated by syntactic operation of grammar is subsequently sent to Spell-out at the PF (Phonetic Form) level. Assuming unvalued feature in derivation is required to be valued in the course of syntax, the said derivation will crash. This is because the PF level is not able to spell out uninterpretable features (Chomsky, 2006, P.13). So, when DPs are sent to PF, the uninterpretable Case feature has already been checked, while those DPs with no Case feature can survive at PF component, given that they do not have any uninterpretable Case feature that are needed to be checked.
The morphology inserts Case affixes in accordance with features on DPs; hence, for SA -un is the Nom Case feature, -an is the ACC one. As a result, the Nom (-un) suffix is added to the nominative DP (Note 3).
However, this paper assumes the Minimalist approach of Chomsky"s work (1995) and the Distributed Morphology (Halle & Marantz, 1994). The fundamental points of this combination are the appearance of a post-syntactic Spell-Out feature and the postulate that all Vocabulary Insertions are part of that component. The notion of default Case is advanced and defined by Marantz (1991): "The default Case forms of a language are those that are used to spell out nominal expressions and are not associated with any Case feature assigned or otherwise determined by syntactic mechanisms." ISSN 1948-5425 2020 However, this study does not tend to completely follow what was proposed in Marantz (1991), it will rather assume, for the purpose of concreteness and familiarities of syntactic distribution of Case assignment, that these forms are more often shown in the syntax and not in the post-syntax proper, as proposed in Marantz (1991) (Note 4).

Literature Review
The default Case is a common phenomenon in Universal Grammar (UG). There is sufficient evidence that supports the existence of default Case in various languages. Within the generative grammar, for example, Bayer (1984) assumes that nominative Case is the default Case in Icelandic languages. He proposes that DPs in Icelandic languages are assigned NOM Case by a default mechanism when there is no Case assigner available in the structure to assign Case to an overt DP. Bayer (1984, P. 245) suggests a similar proposal for German. Radford (2016) and Schütze (2001) propose that accusative is the default Case in English. Radford (2016) argues that the assumption of a default accusative Case in English can account for the accusative Case of the pronoun me in the following dialogue: b. The best athlete, her/*she, should win.
Schütze asserts that these DPs are not located in argument positions, and there is no Case assigner available in the structure which suggests that they get their ACC Case via a default rule. Another kind of argument in support of his claim comes from ellipsis. Schütze shows that English pronouns in elliptical constructions that have no visible verb appear in ACC form, even the fact that their sematic feature is associated with subjects. The structures in (4b) and (4c) ISSN 1948-5425 2020 www.macrothink.org/ijl b. Me/*I neither.

International Journal of Linguistics
The third evidence comes from coordination construction. Schütze illustrates that the following examples show that English pronouns bear ACC Case even if they serve as the second conjunct of the conjoined subject of the clauses. Consider the following example: 5. Did your parents or him/*he pick up Mary?
One could suggest that conjunction construction in English is a case assigner. However, Schütze emphasizes that there is no significant evidence for this suggestion. Furthermore, Halmari & Regetz (2011) argue against this position. They assert that conjunctions combine two or more different expressions into one syntactic unit, but they are outside Case-assignment and agreement relations. The fourth evidence, cited in Schütze (2001), is that modified pronouns need to be ACC, even if they occur in subject positions: 6a. The real me/*I is finally emerging.
b. Dear me/*I c. Lucky us/*we! He asserts that the only reasonable account of such phenomenon is to invoke the default Case mechanism.
Other languages provide evidence for the existence of default Case. As mentioned above, the default Case in German is NOM. This can be found in left-dislocation environment (Grohmann, 2000).
this-NOM guest when has him-ACC the maitre d" greeted "This guest, when did the maitre d" greet him?
The analysis of Grohmann treats this structure as including base generation in which the first element is Hanging Topic Left Dislocation (HTLD), and not Contrastive Left Dislocation (CLD), which includes movement and prohibits Case marking mismatch. The major difference between them involves that in contrast to the latter type, the former type necessitates a prosodic break after the first element. In HTLD, left-dislocated topics with NOM Case are associated with non-NOM positions such as prepositional objects or objects. According to Grohmann, it is more plausible to say that there is no Case assigner for the left-dislocated topic in (6). In addition to German, default NOM Case is also found in left-dislocation DPs in Icelandic (Sigurðsson, 2004), and Greek (Anagnostopoulou, 1997), among other languages. As for SA, various researchers such as Fassi Fehri (1993, P. 45), Ouhalla (1994), Mohammad (2000) and Aoun et al. (2010) assume that NOM is the default Case. Fassi Fehri (1993, P. 45) observes that "subjects in SVO sentences receive default nominative only in the absence of external governors, otherwise, they receive specific structural cases from the latter." However, this paper will cite three kinds of arguments supporting the claim that NOM is the default Case in SA, discuss the assignment of default International Journal of Linguistics ISSN 1948-5425 2020 and provide a new analysis of the mechanism of default in SA based on Distributed Morphology of Schütze (2001).

Default Case in Arabic
This section argues that nominative (NOM) is the default Case in SA. It examines three kinds of environments in SA where NOM nominal expressions can occur (Note 5).

Left Dislocated Topics
Left dislocated topics refer to a DP which occurs in a sentence-initial position, and is associated with a pronoun in a lower position. Left Dislocation does not obey movement constraints, and thus it is not a result of movement, more likely, it is base-generated (Note 6).
In SA, SVO is often considered as an instance of left-dislocation. The topic phrase is associated with a pro (null pro subject, called "little pro") whose features are recoverable from suffixes on the verb. 7. ?al-mariiD-u zur-tu-hu the-patient-NOM visit-1S-him "The patient, I visited him." The example in (3) shows that the Case assigned to the left-dislocated DP, ?al-mariiD-u (the patient), is independent of the one assigned to its resumptive pronoun. While the left-dislocated DP appears in the nominative Case, the resumptive pronoun appears in the accusative Case -hu. This suggests that there is no connection between the left-dislocated DPs in SA and their corresponding arguments through Case concord. A question arises here: how do left-dislocated DPs get their nominative Case?
In order to answer this question, it is essential to first consider the process of Case mechanisms. The morphological Case marking can be seen as a result of the sequence of a number of mechanisms. First, Case can be assigned by a head (e.g. verbs or Ts). Second, it can be the result of matching a relative nominal expression semantically, and more precisely, of a left-dislocated DP matching the Case of its associated constituent by an inherent lexical mechanism (Ura, 2001) (Note 7). Third, Case can be assigned to a head; D within its projection. This is evident by adjectives and nouns within a DP in Latin (a proposal was typically dubbed concordial Case by Blake, 1994). Strictly speaking, under such an assumption, nominative Case feature cannot co-occur with DPs, since left-dislocated DPs are not within their thematic domain.
Fourth, Case assignment can be applied by default, a mechanism which arises when other mechanisms can be ruled out. More specifically, while other three mechanisms are introduced in the course of syntactic derivation, the default mechanism is purely a morphological component.
This study strongly argues for the fourth mechanism recommending nominative Case being the default Case. The reason for this argument to apply the default mechanism is that no other International Journal of Linguistics ISSN 1948-5425 2020 mechanisms can be applied here. In other words, default Case mechanism is applied only in environments where others are ruled out. Given that, it is now required to move to the next question.
The compelling reason for not reducing the nominative Case of left-dislocated DPs in SA to the inherent mechanism is due to the fact that the nominative Case feature that applies to left-dislocated DP, ?al-mariiD-u "the patient", is different from the one associated with the thematic position. It is associated with the resumptive pronoun, hu "him", to which the theta-role of Theme is assigned the accusative Case by a lexical verb. This suggests that the nominative Case those left-dislocated DPs in SA occurs is not taken as an inherent Case. The left-dislocated DP (as in 7) cannot receive structural nominative Case, as there is no Case assigner with assigning ability in the structure. In other words, it is obvious that there is no overt nominative Case assigner with the assigning ability to assign the nominative Case to the left-dislocated topic, and thus no structural Case can be assigned.
To sum up, the theoretical and descriptive discussion argues that nominative Case that assigned to the left-dislocated topics in SA is actually a default Case which is typically assigned in this language in the absence of other available Case assigners.

Ellipsis
Nouns and pronouns in elliptical context that have no overt verb in SA are nominative, even though they are associated with a syntactic object or prepositional object. The examples in (8b) illustrate possible answers to the question in (8a). The nominative nouns/pronouns have the grammatical function of objects. Note that the standing third singular masculine in SA is huwa "he" in the nominative form, while it is ?iyya:yhu "him" in the accusative or genitive form.
8a. man qaabal-ta l-yawm-a who met.3SM-you.1SM the-today-ACC "Who did you meet today?" b. huwa / *?iyyahu he.NOM / *him.GEN The elliptical structures arguably contain surface deletion of vP. Moreover, v is not just a silent head; it should be rather absent, or its features are far away from those of a finite clause. One could suggest a special T that assigns NOM to its object in (8b), but it is not possible to verify its existence. It is assumed that the null hypothesis could be that Case mechanism of v is absent here and NOM is default. This analysis could be extended to the elliptical clauses of (9a) in which the nominative pronoun (9b) has the grammatical function of a prepositional object.

Verbless Sentences
Verbless sentences in SA consist of a topic, and a comment (predicate). First, it is important to see how it is a topic and not a subject and then to see how verbless sentences are finite clauses and not small clauses. Finally, this will lead to the conclusion that both topic and comment in such constructions are assigned their nominative Case via the default mechanism; they have no other structural assigner. The following reference represents a typical example of this construction: 10. ?aT-taalib-u shaaTir-un the-student-NOM smart-NOM "The student is smart." Considering topics vs. subjects in SA, it has been assumed that SA has both preverbal and postverbal subjects, and that they are different with respect to agreement features. While the former is triggering person, gender, and number agreement, the latter is triggering only person and gender agreement. The following pair of example illustrates this fact, respectively: 11a. ?aT-tullaab-u jaa?uu/*jaa?a  ISSN 1948-5425 2020 question that arises here is why examples like (11a) have full agreement. One possible answer is that a gap is expected to share its features with the topic and to trigger agreement features in a similar way. This would argue that the null subject is not a gap, but it is rather a resumptive pro. The following examples suggest that resumptive pro subject triggers full agreement features: 13a. laqad ?arsala ?aR-risaalat-a PCL sent.3SM the-letter-ACC "He sent the letter." b. laqad ?arsaluu ?aR-risaalat-a PCL sent.3PM the-letter-ACC "They sent the letter." Assuming that both sentences have a resumptive pro subject, verbs have full agreement with a resumptive pro subject. This supports the argument that a gap is not possible in postverbal subject position, but a pro is indeed fine.
Furthermore, Albalushi (2010) provides another piece of evidence. He shows that the fact that the initial NP is resumed by a resumptive pronoun within a coordinate structure island suggests that the preverbal NP is not a subject in Spec TP, but it is rather a topic in a peripheral position.
14. ?al-walad-u huwa wa ?akh.uu-hu marDaa the-boy-NOM he and brother.NOM-his sick "The boy, he and his brother are sick." Here, the boy in the first conjunct is resumed by a resumptive pronoun in the second conjunct. If this true, then the NP in (10) has the syntactic properties of what is often associated with topics.
Considering the second premise that verbless sentences are finite clauses and not small clauses, the first piece of empirical evidence comes from the contrast between the following examples.

International Journal of Linguistics
Secondly, verbless sentences which have a present-tense interpretation can contain temporal adverbs (Eisele, 1988). If temporal adverbs must be anchored to a syntactic projected tense node, the above example (15b) provides an additional evidence for the tense presence in verbless sentences in SA. Thirdly, Benmamoun (2000) argues that verbless sentences in SA are not tenseless, since they can co-occur with the complementizer ?inna which selects finite clauses. Note that non-finite sentences are selected by "?an". Now, although verbless sentences in SA are more likely tensed clauses and sign the finiteness properties, the nominative Case appearing on the preverbal DPs is acutely the default Case that is typically assigned to topics where no overt copula verb occurs. The claim that topics in verbless constructions get default NOM Case is supported by the fact that topics realize ACC Case where overt copula verb occurs. Consider the following: 17. kaana ?aT-taalib-u shaaTir-an was the-student-NOM smart-ACC "The student was smart."

Analyses and Conclusion
That nominative Case is a default Case in SA has been proposed and examined by a few authors such as Ouhalla (1994), Mohammad (2000), Soltan (2007), and Aoun et al (2010). Additionally, Emonds (1985) and Hudson (1995) claim that languages have a morphological Case system mark NOM on subjects. This study has attempted to show that the distribution of nominative NPs is influenced by some criteria that could be set independently. It has examined that these criteria are connectively very narrow and cannot be associated normally with other criteria. The following criteria that are related to the range of differences have been examined in this paper: 3. Is the nominative Case assigned to topics in verbless constructions structural?
Following Schütze (1997), this study has assumed that these criteria must be determined in the post-syntactic level. This suggests that the NOM Case positions in SA might be caseless positions in syntax, which infers that NPs can receive structural mechanism without getting Case, and that some NPs need no structural mechanism. In other words, it is more plausible to say that the Case Filter may not be determined as a morphologically condition that applies to all overt NPs. Under Chomsky"s (1995) work, NPs are supposed to inter the derivation with an uninterpretable Case feature (see section 2), however, some NPs optionally might have morphological Case features at that level too. Just the last features can make the differences between NOM, ACC, GEN, and so on; just these would have a default; and languages behave differently in respect of their select of default.
In SA, in contrast to English pronouns (see Emonds, 1985, P. 297), the morphological Case depends on grammatical or abstract Case, thus, the distributions of Case marking found in SA are often generated via mechanisms for morphological Case. It is obvious that nominative NPs in SA are not realizing structural NOM Case, obtained at PF, as they are being out of the scope of any Case assigner. Therefore, this study has proposed that in the distribution of NOM Case, the abstract Case cannot reproduce the pattern of Case distribution, and that the default Case mechanism is eventually the last resort. This is consistent with Jones (1988) who also claims that the notion of default Case has the potential of turning the Case Filter into a vacuous notion. As long as one restricts default Case to morphological realization, then a noun phrase that does not bear an abstract Case feature in syntax gets a default morphological Case at PF. A question that arises here is: how is the default Case affix inserted? Following Schütze (2001), it is further assumed that such affix insertion can be a result of Elsewhere Insertion and Feature Filling, respectively. That is, the Elsewhere Insertion are the default Case forms, otherwise, it is possible to have an explicit feature-filling system that takes place prior to vocabulary insertion. This system may need a formalization such as "If you have a NP with no Case feature, assign the NOM Case on it". In terms of Elsewhere Insertion, one could be dealing with default forms; in this sense default Case is the descriptive mechanism of referring to that group of forms that are not independent in the syntax. On the other hand, Feature Filling is one of the functions of Spell-Out that provides a Case feature to NPs that have no one. Schütze asserts that this is parallel to other functions attributed to morphology such as Distributed Morphology. In this sense, affixes are added to stems, not because it is a syntactic requirement, but rather because this is one of the arbitrary features about how a particular language happens to spell-out.