This article explores the low IP area of Najdi Arabic, with special reference to a construction involving a post-verbal object that is resumed on the verb (as in ‘The man saw-her the woman’). We provide evidence that the object is a topic in the low IP area in such cases. However, the position of the object in the low IP area is masked by the movement of the verb to T0 or, in some cases (if T0 is lexically filled) to the head of the low Focus Phrase. We also provide evidence that Najdi Arabic low topics exclusively express given, accessible information (i.e., familiar topics), which is found to be a prominent characteristic that distinguishes them from high topics (in the CP domain). High topics are shown to be ambiguous with respect to their specific discourse-informational value (aboutness/shifting vs. contrastiveness vs. familiarity). Additionally, we offer evidence that low topics in Najdi Arabic are recursive, dominated by a low Focus Phrase (FocP > TopP*). At a general level, the study demonstrates that topics in the low IP area are more restrictive in their informational content than previously believed.
The cartographic approach to syntactic structure has been increasingly used as a tool to investigate the nature of the structural and functional make-up of natural languages (Belletti 2004, 2005; Cinque 1999, 2002; Cinque and Rizzi 2008; Rizzi 1997, 2004 among many others). The cartography of syntactic structures is a framework that “seeks to determine the number, type, and layering of the functional heads in clause structure as well as in the internal structure of nominal expressions and other phrases” (Brugé et al. 2012: 3). A body of related research has shown that cartography is not only a theoretical enterprise that accounts for the presence of functional elements but also a syntactic approach that unearths the hierarchical projections and their structural properties that are bypassed by other alternatives, e.g., ones adopting a non-split CP view of sentence structure (see Haegeman 2012; Shlonsky 2015; Svenonius 2014). The validity of this approach also crystalizes in situations where the role of discourse and information structure in sentence processing and formation is manifested. This is mainly achieved because cartography is a restrictive theory that assumes “a one-to-one relation between position and interpretation” (Cinque 1999: 132).
This article makes use of one of the advancements of the cartographic approach, namely the low IP area framework (Belletti 2004, 2005) in order to explore a set of intriguing structures from Najdi Arabic (NA) and position them in the wider theoretical context. This exploration provides evidence from NA showing that the low IP area subsumes functional discourse projections whose hierarchical distribution and interpretive content are not completely identical to their counterparts in the CP domain. For instance, Topic Phrase in the low IP area of NA is shown not to dominate Focus Phrase, implying that the low IP area is not quite a mirror image of the syntactic structure at the left periphery, as assumed by some (see Belletti 2004). The low IP area is found to be a more restrictive domain where a particular type of information is projected and interpreted.
The following discussion is structured as follows. Section 2 contains a summary of the data relevant to the analysis. It is followed by an analysis of the data using the low IP area framework to account for the syntactic constructions under discussion. Section 4 provides evidence that low topics in NA express a specific informational value (i.e., familiar topical information), as opposed to high CP topics, which are shown to be ambiguous in their interpretation unless some syntactic element, such as certain topic particles, is used. Section 5 contains the conclusion.
2 Setting the scene
In this section, we provide a descriptive overview of the (un)marked word order in NA. Second, we discuss a particular construction which we term as “clitic-object construction”, where the postverbal object is resumed on the verb by a pronominal clitic that mirrors the ɸ-content of the object.
Najdi Arabic (NA) is an Arabic dialect spoken predominantly in the middle and northern regions of the Arabian Peninsula (Abboud 1964; Ingham 1994). NA has attracted the attention of many researchers, especially in the last two decades (Alshamari 2017a, 2017b; Alshammari 2018; AlShammiry 2007; Alsweel 1981; Ingham 2010; Jarrah and Zibin 2016; Lewis 2013). Word order patterns, pragmatic and discourse-interpretive properties of discourse particles, information structure, multiple wh-constructions, adverbial clauses, verbal system, and genitive constructions are among the main issues that have been explored in this regard (see Alshamari 2017a for an overview).
Almost all studies on NA syntax converge on the assumption that NA is an SVO language, where the preverbal subject is neutrally interpreted as the grammatical subject of the sentence, without expressing any informational content relating to discourse saliency/giveness or newness. For instance, AlShammiry (2007: 6) treats the preverbal subject in (1a) as a neutral subject (i.e., the subject is shown to express no discourse value, e.g., a topic or a focus), to which the property expressed in the predicate of the clause is potentially attributed. This results in the subject-predicate construction. The sentences in (1b)–(1d) include other instances of SVO clauses (all examples in this article are from NA; unless otherwise stated):
Alshamari (2017a: 19) mentions that an SVO sentence forms a felicitous answer to a what-happens question, i.e., a complete thought expressed by a proposition with a neutral-description reading (i.e., a thetic reading; cf. Kuno 1973). An answer to a what-happens question would thus be expressed through a form of the unmarked word order used in a given language, which is the SVO word order in NA. This essentially implies that the subject is discourse free in SVO clauses in NA. Additionally, Alshamari (2017a) mentions that the SVO word order in NA is predominantly used in embedded contexts, which are crosslinguistically often used as diagnostics to identify the unmarked word order of a given language (see, e.g., Bader and Häussler 2010; Jarrah 2019; Vikner 1994).
Other word order patterns are nonetheless still productive in NA, particularly the VSO word order, which is viewed as a common marked word order across NA subdialects (Abboud 1964; Ingham 1994). This contrasts with Standard Arabic (SA), where VSO is regarded as the predominant, unmarked word order (Aoun et al. 2010; Fassi Fehri 1993; Parkinson 1981; Soltan 2007). Alshamari (2017a, 2017b) and Alshammari (2018), among others, propose that VSO clauses in NA are syntactically derived through the movement of the lexical verb to a left-peripheral position, undergoing focalization, while the rest of the clause expresses the background of the proposition (i.e., the presupposition). The following sentence is an example of VSO order taken from Alshamari (2017a: 20):
An approximate interpretive translation of (2) would be ‘It was seeing that Firas was doing with the car in the yard.’
VSO word order in NA is thus used to focus on the eventuality described in the clause, which is derivationally achieved by focalizing the verb, i.e., the verb moves to the head of Focus Phrase in the left periphery (see Ouhalla 1994 for an analysis along these lines).
Other marked word orders are also attested in NA. For instance, the following sentences are examples of the marked order OSV:
Note that in OSV clauses, the verb may carry a resumptive clitic of the fronted object. When this clitic appears on the verb, it is indicative that the object serves as a topic (an element whose informational content is given/old, as it is recoverable from the preceding discourse). On the other hand, when this resumptive clitic is not used, the fronted object is interpreted as a contrastive focus. The resumptive clitics are widely analyzed as pronouns that are Ā-bound by their antecedents that are merged in their surface position (Demirdache 1991; Shlonsky 1992; Wahba 1984; see Choueiri 2002 for related discussion).
In NA, a resumptive clitic of the object also appears on the verb in OVS clauses, signaling topicality of the fronted object:
The presence of the resumptive clitic mirroring the object on the verb, and indicating topicality of the preverbal object, is also attested in Standard Arabic (SA), whose topics are distributionally identified through their sentence-initial position, and morphologically by nominative case (Albuhayri 2019; Ouhalla 1997; Soltan 2007; among others). Consider the following OVS sentences from SA.
The presence of the sentence-initial object marked with nominative case is interpreted as a topical element that is base-generated in the left periphery. Its thematic position is filled with a resumptive clitic that is incorporated into the verb during sentence derivation. Ouhalla (1997) refers to the constructions in (5) as instances of “left-dislocated phrases” (LD-phrases) (see also Soltan 2007). Aoun and Benmamoun (1998) refer to them as “clitic left-dislocation” (CLLD), which is a familiar construction in Romance languages, including Italian and Spanish (Cinque 1990). Whatever the terminology used for the constructions in (5) is, the common view regarding the position of the fronted object is that it is not an outcome of movement.
An important point to be mentioned here is that NA exhibits highly productive instances of a construction where the clitic still occurs on the verb while the object appears postverbally. This is clearly shown in the following examples (the clitic is boldfaced).
The examples in (6) share some distinctive morphological and interpretive properties. Firstly, the verb bears a clitic that, on the surface, doubles the ɸ-content of the postverbal object. Secondly, the object should be interpreted as a discourse-given element, à la Reinhart (1981). One piece of evidence that indicates an intimate link between the presence of this object doubling clitic on the verb and the givenness of the object comes from the incompatibility of this clitic with nonspecific, indefinite objects, which do not convey given information (see, among others, Chafe 1976; Rizzi 2004). This incompatibility between the clitic and the indefinite/nonspecific object is shown in the following examples.
Additionally, clitics of this type are incompatible with postverbal objects that are contrastively interpreted as implying corrective focus, one type of contrastive focus (see Jarrah 2020; Jarrah and Abusalem 2021; Ouhalla 1997; Steube 2001; Zimmermann 2008). Consider the following two examples, where the clitic is banned, as the corrective reading of the object is enforced.
The fact that the clitic on the verb is not compatible with correctively focused elements, which are definite in (8), strongly indicates that the occurrence of this clitic on the verb should not be attributed to the definiteness of the object. We propose that clitics of this type only appear on the verb when the postverbal object is a topic. Consider the following dialogue, which shows that the object (in the last utterance) is a topic.
Speaker D’s utterance includes an instance of an SVO clause where the verb is resumed by a non-fronted object, in a construction that we will henceforth call “clitic-object construction”. Being discourse-accessible, the object in Speaker’s D utterance is a topic rather than a focus, or even a neutral element. That is because the object in such cases expresses given information that is well-established (salient) in the preceding discourse.
The following dialogue is another example where the interpretive property of the topical object in clitic-object constructions becomes apparent.
The dialogue in (10) is about food, which has acquired a discourse status as “given” at this stage of the conversation. Speaker D’s utterance contains the DP ʔalfetˤa:jer ‘the pies’, which expresses discourse-given information as well, as it is a hyponym of the topic referent food. The constituent ʔalfetˤa:jer is resumed on the verb through the clitic (-hin).
The generalization made at this point is that the object in clitic-object constructions expresses discourse-given information (interpreted as a topic). The interpretation of the object in such cases is, ceteris paribus, similar to the examples where the object appears sentence-initially and is resumed on the verb (see the examples in (3)). An important question that has to be resolved here is why the verb is still resumed by a resumptive clitic that is associated with the postverbal object.
3 Low topics in NA
Given that the object in clitic-object constructions is interpreted as a topic, we propose that the object in these constructions is a topic that is located in the low IP area. In other words, the object is not situated in its thematic position but is base generated in the Spec(ifer position) of Topic Phrase that is projected in the low IP area. Belletti (2004, 2005 argues that natural languages subsume a discourse-related field that is situated above vP and below TP. This field includes recursive Topic Phrase and non-recursive Focus Phrase. The low IP area is schematically represented in the following tree diagram in Example 11.
As shown in the schematic representation in (11), Topic Phrases can be projected between TP and vP. In fact, there has been a lot of crosslinguistic research arguing in favor of the existence of this area in many languages (e.g., Italian, Chinese, English, Jordanian Arabic, etc.; cf. Duncan et al. 2018; Garzonio 2006; Jarrah and Abusalem 2021; Paul 2005; Poletto 2006). However, there is no research that attempts to test whether or not NA (as well as other varieties in the Arabian Peninsula) projects this area. The null hypothesis is that the low IP area is present in all languages, given it is provided by UG (see Belletti 2004, 2005). Jarrah and Abusalem (2021) argue extensively for the presence of this area in Jordanian Arabic (JA) grammar. Their main evidence comes from the movement of the verb and the discourse value of the subject in VSO clauses. Examining a two-million-word corpus of naturally occurring data from JA, supported by grammaticality judgments from 50 JA speakers, Jarrah and Abusalem (2021) find that the subject in VSO clauses in JA is mostly either a definite DP or a modified, indefinite DP (representing about 96% of all tokens of the subject in VS[O] clauses in their corpus). Consider the following VSO examples from JA that include definite subjects (12a), a specific, indefinite subject (12b), and a modified, indefinite subject (12c) (Jarrah and Abusalem 2021: 130).
Jarrah and Abusalem (2021) take the status of the subject in VSO clauses in JA as evidence that a certain informational value (i.e., a topic or a focus) is assigned to the post-verbal subject in VSO clauses. Coupled with other tests including the linear position of the subject with respect to the past tense copula ka:n and high vP and IP adverbials, Jarrah and Abusalem (2021) propose that the subject in VSO clauses is a topic or contrastive focus that is positioned in a discourse field between TP and vP.
We propose that NA, similar to JA and other languages with corresponding structures, projects a low IP area with a recursively-layered Topic Phrase as well as Focus Phrase. Due to the fact that the object in clitic-object constructions is interpreted as a topic, we claim that the object is first merged in this area as a low topic. This indicates that the object maintains a higher structural position than the lexical verb in sentence derivation, as the low IP area is located above vP (see 11). However, this high position of the low topic relative to the lexical verb is masked by the independently motivated movement of the verb to T0 in Arabic (see Benmamoun 2000; Ouhalla 1994). This is schematically shown in the following tree (the lexical verb in Arabic adjoins to v0 prior to its movement to T0; silent copies appear between < >).
Given that T0 is structurally higher than all elements in the low IP, the verb appears to the left of the topical object, generating the SVO word order in the surface. In other words, movement of the verb to T0 is what makes the object appear to the right of the verb on the surface.
Let us now turn out attention to instances where T0 is lexically supported by the past tense copula ka:n. In such cases, the verb may appear to the left of the topical object (14) as well as to the right of it (15).
We can see that in (14) the object still occurs to the right of the verb, although T0 is lexically supported by ka:n. For such cases, we propose that the verb in (14) moves to the head of Focus Phrase that is projected above Topic Phrase in the low IP area (below T0), which is filled with ka:n. Recall that the low IP area includes Topic Phrase, which is dominated by Focus Phrase (see 11). Evidence that the verb in (14) is located in the low Focus Phrase comes from the fact that the verb in such cases is felicitously interpreted as a contrastive focus; otherwise, the sentence is deemed unacceptable by NA native speakers. The verb must be associated with a unique intonational pattern; it takes a rising contour, which then falls towards the end of the sentence.
The following dialogue presents an example with T0 which is lexically supported by ka:n, while the verb appears to the left of the topical object. The verb is interpreted as a contrastively focused element.
In Speaker D’s utterance, the lexical verb ʔijsa:wi is not tensed (T0 is lexically supported by ka:n). Given the fact that contrastive focus in Arabic is marked by movement (Moutaouakil 1989; Ouhalla 1997), we cannot only assume that the verb adjoins to little v0 (because the verb adjoins to v0, irrespective of the informational value assigned to the verb). We propose that the lexical verb in Speaker D’s utterance moves to Focus Phrase that is situated in the low IP area. Because the object is situated in the low topic, a resumptive clitic on the verb is used.
The presence of contrastive focus in the low IP area is attested in a number of languages, including JA. Jarrah and Abusalem (2021: 137) mention that “[t]here is strong evidence that the low IP domain in JA includes not only (recursive) Topic Phrase, but also a projection hosting contrastive Focus Phrase”. Their main evidence comes from the fact that the subject in VSO clauses in JA can be contrasted with another element. For example, sentence (17a) expresses that it was the lady, not her daughter, who visited us, whereas (17b) states that it was Isaam, not Ali, who stole the car (the two examples have been taken from Jarrah and Abusalem 2021: 137–138).
Following Molnár (2002), Jarrah and Abusalem (2021) argue that the fact that the subject in VSO clauses can be contrasted with another element gives rise to a contrastive focus reading. The postverbal subject in the two examples in (17) negates the value assigned to a variable in the assertion structure of the previous discourse, which introduces an alternative value for such a variable (see Zubizarreta 1998).
By contrast, when the verb is not contrastively focused (as shown in 15, reproduced below as (18)), it appears to the right of the object, which we attribute to the lack of verb movement to Focus Phrase. The object is still a low topic that is resumed on the verb.
One point to underscore here is that when the verb expresses (non-contrastive) new information, it appears to the right of the topical object, as shown in the following pair.
The verb does not express contrastive or corrective information; hence it remains in situ (given that T0 is lexically supported by ka:n). As already mentioned in Moutaouakil (1989), among others, elements that express new information remain in situ in Arabic grammar. This is significant here because it provides evidence that the object in Speaker B’s answer is not located in its thematic position, as it appears to the left of the verb that expresses new information. Notice that the fact that the elements expressing new information (which are not contrastive) do not appear in the low IP area is at variance with Belletti’s (2004) work on Italian, where low foci express new information, while high foci express contrast.
A related observation that deserves mentioning at this point is the fact that the object in clitic-object constructions can be extraposed. An adverbial can intervene between the verb and the topical object, as shown in the following example.
The extraposition of the topical object is a compelling argument that the verb and the object are not underlyingly adjacent; they are not located under the same XP. Given that low Topic Phrase is recursive, the low IPA area is thus predicted to accommodate more than one topical element. We propose that PP bissa:ħah ‘in the yard’ is a low topic in (20), hence its position to the left of the topical object.
As a matter of fact, PP bissa:ħah ‘in the yard’ can be interpreted as a topic even if it appears to the right of the topical object, as shown in the following dialogue.
The PP bissa:ħah ‘in the yard’ is a topic expressing accessible information that is already well-established in the previous discourse. Assuming our analysis to clitic-object constructions in NA to be valid (i.e., the object is a low topic) as well as the recursivity of low topics in the low IP domain, the PP bissa:ħah ‘in the yard’ functions as a topic. This is expected under the proposal that low topics, like high topics, are recursive, hence the possibility of having stacked topics following each other in the sentence. Consider the following tree structure (irrelevant details are skipped).
The tree in (22) implies that the PP bissa:ħah ‘in the yard’ in Speaker A’s last sentence in (21) does not adjoin to VP. Rather, it is located in the low IP area, dominated by Topic Phrase, whose Spec is filled with the topical object. This line of analysis provides us with an account of why elements such as bissa:ħah ‘in the yard’ and the topical object both express topicality, while they are situated in a postverbal position. In our account, topicality is only expressed by elements which are located in a designated Topic Phrase; however, the movement of other sentential elements, including the verb, to a higher structural position gives rise to the surface output where such elements remain in situ.
Further evidence supporting our view regarding adverbials that are located between the topical object and the verb as topics comes from the fact that the relevant sentences are deemed unacceptable by native speakers if such adverbials express new information. This is shown in the following dialogue, where the PP adverbial expresses new information:
The fact that the PP min dʒaddah ‘from Jeddah’ in Speaker B’s answer appears to the right of the verb is expected, as the PP expresses new information that is not morphosyntactically marked in Arabic grammar. Locative PPs in Arabic typically appear to the right of the object on the surface (see Jarrah 2017b). Additionally, the fact that the PP min dʒaddah is not permitted to appear to the left to the verb is significant, as it provides evidence that adjuncts in NA grammar do not adjoin to the low IP area unless they are assigned a certain discourse value (topical or contrastive).
The assumption that the low IP area includes a recursive Topic Phrase is discussed in Belletti (2004) for Italian. For instance, Gianni and the PP al direttore ‘to the director’ in (23) are analyzed as topics located in the low IP area (Belletti 2004: 38):
According to Belletti, Gianni and the PP al direttore ‘to the director’ are pronounced with a downgrading intonation that is not normally found with true subjects (filling Spec,TP).
In a related vein, Jarrah and Abusalem (2021) maintain that JA provides evidence that low topics are recursive. For instance, the subject in the following VSO clause is followed by a locative adverbial bi-rbid, which is analyzed as a topical element because its surface position (between the subject and the object) is not the same as the discourse neutral position in the Arabic clause for locatives (Jarrah and Abusalem 2021: 136).
Jarrah and Abusalem (2021) mention that non-topical locative PPs in JA (and in Arabic in general; Jarrah 2017b) are normally located at the end of the clause, as shown in the following example (Jarrah and Abusalem 2021: 136).
This discussion reveals that adverbials that appear between the verb and the object are low topics whose presence in the sentence provides evidence for the recursivity of low topics in NA.
With this being the case, clitic-object constructions are similar to Clitic Right Dislocation (CLRD) discussed in Cecchetto (1999) and Frascarelli (2000, 2004. For instance, Cecchetto (1999) provides several pieces of empirical evidence that the topic position involved in right dislocation (including the right-dislocated object) is indeed low in the clause structure, i.e., the VP periphery.
Cecchetto (1999) argues that CLRD does not involve adjunction of the dislocated XP to the right of the same node that hosts a left dislocated XP (an assumption known as the “mirror hypothesis”). He also shows that a CLRD sentence cannot be the result of double topicalization movement (cf. Kayne 1994). Under the latter analysis, the right dislocated XP is left dislocated, giving rise to an ordinary clitic left dislocation (CLLD) sentence. Afterwards, the IP that includes the base-generation position of the dislocated XP is dislocated, i.e., moved to a higher CP position. Based on a cluster of several apparently unrelated differences between CLLD and CLRD, including anti-reconstruction effects and ECP effects, Cecchetto (1999) argues that the object in CLRD constructions is positioned in the VP periphery which Belletti (2004) later argues to be the same as the low IP area.
Before we discuss our assumption that low topics in NA express a unique value of given information (i.e., familiar topical information), we would like to discuss a related construction in Levantine Arabic that appears to be amenable to our analysis developed for the object in clitic-object constructions in NA. In Levantine Arabic, the accusative object can be matched by the clitic that appears on the verb and is prefixed with a dative preposition, as shown in the following example (Shlonsky 1997: 195) (for ease of reference, we term these constructions as “object clitic-doubling” (OCD) constructions.
According to Shlonsky (1997), the preposition la is used to assign case to the postverbal object, as the accusative case on the verb is assigned to the clitic. However, the true nature of the object in such constructions is investigated in passing. There is evidence showing that OCD constructions in Levantine Arabic call for a similar treatment of clitic-object constructions in NA. The object is a low topic in the two constructions. OCD constructions are optional. The object is not always marked with la. In fact, there are some cases where la is prohibited, namely when the object is an indefinite, nonspecific element, as shown in the following example.
This ban on the use of la in conjunction with indefinite, nonspecific objects is remarkable; it supports the topicality analysis of the object in OCD constructions. Additionally, the object in OCD constructions can be extraposed as long as the intervening PP is interpreted as a topic.
Following our analysis developed for clitic-object constrictions in NA, we propose that the object in OCD constructions is a low topic that is base-generated in the low IP area. la functions as a topic marker flagging the object when it functions as a topic in the low IP area.
It should also be mentioned that object marking of this type is attested in many languages in the world. Maltese is a prominent example in this regard. As mentioned in Döhla (2016), the object in Maltese is marked with an object marker especially when it is a proper noun or a stressed personal pronoun, as shown in the following examples.
The object is also marked when its reference has a high degree of topicality. The following sentence is a case in point.
As a first approximation, we can say that when the object is marked with lil, this can be indicative of the object being a low topic. However, this generalization is compounded by the fact that focused elements are also marked with lil, as shown in the following example:
The marking of unspecific but focal referents with lil can be interpreted as indicating that the low IP area in Maltese is filled with some element – a focus or a topic.
In the following section, we provide evidence that low topics in NA express familiar information.
4 Low topics are familiar topics
In this section, we argue that low topical objects and PPs in NA do not express contrastively given information. Frascarelli and Hinterhölzl (2007) propose a typology of topics in Italian and German. They argue that topics are not homogenous; rather they come with different types of discourse information. Frascarelli and Hinterhölzl (2007) identify three different categories of topics. Building on Frascarelli and Hinterhölzl (2007), Jiménez-Fernández and Miyagawa (2014: 284) offer the following definitions of these three types of topics (see also Alshamari 2017a).
In simpler terms, an aboutness topic is the discourse topic (i.e., what the whole discourse talks about). A contrastive topic is selected from a well-established set of alternatives that are all known to the interlocutors. A familiar topic is a discourse salient element that does not induce such a set of alternatives.
An example of a contrastive topic is given in the following sentence where the speaker picks out Balqees out of her other family members, who are all known to the speaker and the hearer.
The most important point here is that available evidence indicates that only familiar topics project in the low IP area. Firstly, if a contrastive topic were enforced to appear postverbally in clitic-object constructions, the grammaticality of the resulting sentence is degraded, as shown in the following example.
On the other hand, sentence (36) is acceptable under the interpretation that the speaker respects Balqees without inducing other alternatives (i.e., Balqees’s other sisters) which create oppositional pairs with respect to other topics.
Likewise, low topics in NA do not express aboutness/shifting information. Aboutness/shifting topics are normally marked with as for in English. Alshamari (2017b) argues that aboutness topics in NA are marked with the expression binisbah li which, as speakers indicated, cannot be used in conjunction with low topics in NA, as shown in the following sentence.
Alshamari (2017a: 73) also proposes that aboutness topics in NA (as is the case in other languages) may not be associated with the rest of the accompanying sentence by a thematic relation, as shown in the following sentence.
The DP llijaqah ‘the stamina’ is not assigned any theta role, e.g., a theme, patient, or experiencer of the verb. This property of aboutness topics not being associated with the rest of the sentence via a thematic relation can be accommodated under Bianchi and Frascarelli’s (2010) framework, according to which aboutness topics are not even part of the left periphery of the host sentence. Rather, they are part of Conjunction Phrase (πP), which is headed by the paratactic (π), which is the implementation of the speech act conjunction that links discourse segments with each other (see Jarrah et al. 2019 for the implementation of this idea in the analysis of some constructions in Jordanian Arabic). This implies that aboutness topics are not part of the syntactic structure of the sentence; hence their thematic relation to the rest of the sentence, if any, is accidental. On these grounds, we suggest that aboutness topics are not projected in the low IP, which is normally viewed as a mirror image of the left periphery (see Belletti 2004). As the left periphery does not include aboutness topics, the low IP does not include them, either. Additionally, it is hard to think of a low topic that is not related to the verb with a thematic relation.
Since low topics cannot be contrastive or aboutness, we can formulate the generalization that low topics in NA are all familiar, unlike high topics, which can be contrastive as well as familiar topics.
One important remark here is that the ambiguity of high topics no longer holds when designated topic particles are used. Alshamari (2017a) argues that NA subsumes a set of particles which he calls ‘topic particles’ whose main function is to mark the element that functions as a topic and, and to spell out the relevant topical feature. For instance, the particle tara is a contrastive topic particle, the particle tigil is a familiar topic particle, whereas mar is an aboutness/shifting topic particle (see Alshamari 2017a for motivation and analysis). Consider the following examples:
By contrast, ambiguity never arises in conjunction with low topics, as the latter invariably express familiar information. This property of low topics in NA is consistent with Jarrah and Abusalem’s (2021) remark that low topics in Jordanian Arabic require a high degree of contextual anaphoricity. This may explain why topic particles are not used in conjunction with low topics, as the latter have not been found to be ambiguous. If the use of topic particles is in fact triggered in order to resolve an ambiguity that arises in the context of high topics (which are ambiguous unless disambiguated by a proper particle), then this would be desirable, as it can be taken as evidence in favor of the Strong Minimalist Thesis (see Chomsky 2001, 2004), which views language as an optimal solution to legibility conditions.
One final question that could be raised in this article is why contrastive topics do not merge in the low IP in NA. There is indeed evidence indicating that the main reason behind the ban on the presence of contrastive topics in the low IP area in NA is that the IP area is truncated. Jarrah and Abusalem (2021) argue that low topics in JA can dominate a contrastive focus. They mention that “[s]entence […] is significant as it shows that the low Topic Phrase can also dominate the low Focus Phrase in JA” (p. 145). The sentence they refer to is mentioned in (40):
The object DP ʔilbe:t is a low topic that c-commands (hence appears to the left of) the contrastively focused subject ʃarikat ʔilxadda:ʃ ‘Khadash Company’. However, the NA sentence corresponding to (40) is ungrammatical for most speakers, under the contrastive-focus interpretation of the subject:
Once the topical object appears to the right of the contrastively focused subject and is resumed on the verb by a clitic, the grammaticality of the sentence improves significantly:
The contrast between (41) and (42) implies that low topical elements do not dominate a low contrastive focus. This is significant because it implies that the topic space above Focus Phrase is not present in the low IP area in NA. Our take on these data is that the low IP area in NA is truncated. These facts, coupled with the assumption made in Frascarelli and Hinterhölzl (2007) that contrastive topics (and aboutness topics) are projected above Focus Phrase in the CP domain, provide us with a reason why contrastive topics are not permissible in the low IP area in NA. The area where contrastive topics are predicted to project in is not present in the low IP area.
In this article, we have investigated clitic-object constructions in NA, which are shown to be subsumed under an interpretive-based approach. We have shown that what appears on the surface as an object in its base-generation position is in fact a low topic that is situated in the low IP area of the clause. This is consistent with Belletti’s (2004, 2005 model, which is shown to offer a theoretically satisfactory account of the intricacies of the clitic-object constructions in NA. The present study has shown that this model requires an empirical refinement as far as NA is concerned. It has been shown that low topics only express familiar information. Contrastive (and aboutness) topics are shown not to be allowed in this zone of the clause, in sharp contrast to the left periphery of the sentence. We regard this as evidence showing that low topics in natural languages are more restricted than previously believed.
We are extremely grateful to Volker Gast for his insightful editorial advice as well as to two anonymous reviewers for their insightful comments and remarks that considerably enhanced the overall quality of the study. All remaining errors and shortcomings are ours.
Abboud, Peter Fouad. 1964. The syntax of Najdi Arabic. Austin, TX: University of Texas dissertation.Search in Google Scholar
Albuhayri, Salem. 2019. Information structure in Standard Arabic verbal sentences. Milwaukee, WI: University of Wisconsin dissertation.Search in Google Scholar
Alshamari, Murdhy. 2017a. Topic particles in the North Hail dialect of Najdi Arabic. Newcastle: Newcastle University dissertation.Search in Google Scholar
Alshamari, Murdhy. 2017b. A feature-based analysis of the syntax of the clause-initial particle ʁedɪ in North Hail Arabic. Poznan Studies in Contemporary Linguistics 53(3). 305–343. .10.1515/psicl-2017-0012Search in Google Scholar
Alshammari, Ahmad. 2018. The syntax of temporal and conditional adverbial clauses in Najdi Arabic. Newcastle: Newcastle University dissertation.Search in Google Scholar
AlShammiry, Khalaf. 2007. The clause structure of Turaif Arabic. Lawrence, KS: University of Kansas dissertation.Search in Google Scholar
Alsweel, Abdulaziz Ibrahim. 1981. The verbal system of Najdi Arabic: A morphological and phonological study. Seattle, WA: University of Washington dissertation.Search in Google Scholar
Belletti, Adriana. 2004. Aspects of the low IP area. In Luigi Rizzi (ed.), The Structure of IP and CP: The cartography of syntactic structures, 16–51. Oxford: Oxford University Press.Search in Google Scholar
Benmamoun, Elabbas. 2000. The feature structure of functional categories: A comparative study of Arabic dialects. Oxford: Oxford University Press.Search in Google Scholar
Bianchi, Valentina & Mara Frascarelli. 2010. Is topic a root phenomenon? Iberia International Journal of Theoretical Linguistics 2. 43–88.Search in Google Scholar
Brugé, Laura, Anna Cardinaletti, Giuliana Giusti, Nicola Munaro & Cecilia Poletto (eds.). 2012. Functional heads. Oxford: Oxford University Press.10.1093/acprof:oso/9780199746736.001.0001Search in Google Scholar
Büring, Daniel. 1999. Topic. In Peter Bosch & Rob van der Sandt (eds.), Focus: Linguistic cognitive, and computational perspectives, 142–165. Cambridge: Cambridge University Press.Search in Google Scholar
Chafe, Wallace. 1976. Givenness, contrastiveness, definiteness, subjects, topics, and point of view. In Charles Li (ed.), Subject and topic, 25–55. New York, NY: Academic Press.Search in Google Scholar
Chafe, Wallace. 1987. Cognitive constraints on information flow. In Russel Tomlin (ed.), Coherence and grounding in discourse, 21–51. Amsterdam & Philadelphia: John Benjamins.10.1075/tsl.11.03chaSearch in Google Scholar
Chomsky, Noam. 2001. Derivation by phase. In Michael Kenstowicz (ed.), Ken Hale: A life in language, 1–52. Cambridge, MA: The MIT Press.Search in Google Scholar
Chomsky, Noam. 2004. Beyond explanatory adequacy. In Adriana Belletti (ed.), Structures and beyond: The cartography of syntactic structure, vol. 3, 104–131. New York, NY: Oxford University Press.Search in Google Scholar
Choueiri, Lina. 2002. Issues in the syntax of resumption: Restrictive relatives in Lebanese Arabic. Los Angeles, CA: University of Southern California dissertation.Search in Google Scholar
Cinque, Guglielmo. 1990. Types of Ā-dependencies. Cambridge, MA: MIT Press.Search in Google Scholar
Cinque, Guglielmo. 1999. Adverbs and functional heads. Oxford: Oxford University Press.Search in Google Scholar
Cinque, Guglielmo. 2002. Functional structure in DP and IP. Oxford: Oxford University Press.Search in Google Scholar
Demirdache, Hamida. 1991. Resumptive chains in restrictive relatives, appositives, and dislocation structures. Cambridge, MA: Massachusetts Institute of Technology dissertation.Search in Google Scholar
Döhla, Hans-Jörg. 2016. The origin of differential object marking in Maltese. In Gilbert Puech & Benjamin Saade (eds.), Shifts and patterns in Maltese, 149–174. Berlin & Boston: De Gruyter Mouton.10.1515/9783110496376-007Search in Google Scholar
Duncan, Philip T., Travis Major & Mfon Udoinyang. 2018. Searching high and low for focus in Ibibio. In Jason Kandybowicz, Travis Major, Harold Torrence & Philip T. Duncan (eds.), African linguistics on the Prairie: Selected papers from the 45th annual conference on African linguistics, 269–288. Berlin: Language Science Press.Search in Google Scholar
Fassi Fehri, Abdelkader. 1993. Issues in the structure of Arabic clauses and words (Studies in Language and Linguistic Theory 29). Dordrecht: Kluwer Academic Publishers.10.1007/978-94-017-1986-5Search in Google Scholar
Frascarelli, Mara. 2000. The syntax-phonology interface in focus and topic constructions in Italian (Studies in Language and Linguistic Theory 50). Dordrecht: Kluwer Academic Publishers.10.1007/978-94-015-9500-1Search in Google Scholar
Frascarelli, Mara. 2004. Dislocation, clitic resumption and minimality: A comparative analysis of left and right topic constructions in Italian. In Reineke Bok-Bennema, Bart Hollebrandse, Brigitte Kampers-Manhe & Petra Sleeman (eds.), Romance languages and linguistic theory 2002, 99–118. Amsterdam & Philadelphia: John Benjamins.10.1075/cilt.256.07fraSearch in Google Scholar
Frascarelli, Mara & Hinterhölzl Roland. 2007. Types of topics in German and Italian. In Susanne Winkler & Kerstin Schwabe (eds.), On information structure, meaning and form, 87–116. Amsterdam & Philadelphia: John Benjamins.10.1075/la.100.07fraSearch in Google Scholar
Garzonio, Jacopo. 2006. Focus in the IP: The particle ma in Florentine. Rivista di Grammatica Generativa 31. 19–38.Search in Google Scholar
Givón, Talmy. 1983. Topic continuity in discourse: An introduction. In Talmy Givón (ed.), Topic continuity in discourse: A quantitative cross-language study, 5–41. Amsterdam & Philadelphia: John Benjamins.10.1075/tsl.3.01givSearch in Google Scholar
Haegeman, Liliane. 2012. Adverbial clauses, main clause phenomena, and composition of the left periphery: The cartography of syntactic structures, vol. 8. Oxford: Oxford University Press.10.1093/acprof:oso/9780199858774.001.0001Search in Google Scholar
Ingham, Bruce. 2010. Information structure in the Najdi dialects. In Jonathan Owens & Alaa Elgibali (eds.), Information structure in spoken Arabic, 75–92. London: Routledge.Search in Google Scholar
Jarrah, Marwan. 2017a. A criterial freezing approach to subject extraction in Jordanian Arabic. Canadian Journal of Linguistics/Revue canadienne de linguistique 62. 411–448. https://doi.org/10.1017/cnj.2017.19.Search in Google Scholar
Jarrah, Marwan & Nimer Abusalem. 2021. In favour of the low IP area in the Arabic clause structure: Evidence from the VSO word order in Jordanian Arabic. Natural Language & Linguistic Theory 39(1). 123–156. https://doi.org/10.1007/s11049-020-09474-y.Search in Google Scholar
Jarrah, Marwan, Sharif Alghazo & Moh’d Al Salem. 2019. Discourse functions of the wh-word Su: in Jordanian Arabic. Lingue e linguaggio 18. 291–318.Search in Google Scholar
Jarrah, Marwan & Aseel Zibin. 2016. Syntactic investigation of nunation in Haili Arabic. SKY Journal of Linguistics 29. 39–62.Search in Google Scholar
Kayne, Richard 1994. The antisymmetry of syntax (Linguistic Inquiry Monograph 25). Cambridge, MA: MIT Press.Search in Google Scholar
Kuno, Susumu. 1973. The structure of the Japanese language (Current Studies in Linguistics 3). Cambridge, MA: MIT Press.Search in Google Scholar
Kuno, Susumu. 1976. Subject, theme, and the speaker’s empathy: A re-examination of relativization phenomena. In Charles N. Li (ed.), Subject and topic, 417–444. New York, NY: Academic Press.Search in Google Scholar
Lambrecht, Knut. 1994. Information structure and sentence form: Topic, focus and the mental representations of discourse referents. Cambridge: Cambridge University Press.10.1017/CBO9780511620607Search in Google Scholar
Lewis, Robert. 2013. Complementizer agreement in Najdi Arabic. Lawrence, KS: University of Kansas dissertation.Search in Google Scholar
Mohammad, Mohammad A. 2000. Word order, agreement and pronominalization in Standard and Palestinian Arabic (Current Issues in Linguistic Theory 181). Amsterdam & Philadelphia: John Benjamins.10.1075/cilt.181Search in Google Scholar
Molnár, Valéria. 2002. Contrast from a contrastive perspective. In Hilde Hallelgard, Stig Johansson, Bergljot Behrens & Cathrine Fabricius-Hansen (eds.), Information structure in a cross-linguistic perspective, 147–161. Amsterdam & New York: Rodopi.10.1163/9789004334250_010Search in Google Scholar
Ouhalla, Jamal. 1994. Verb movement and word order in Arabic. In David Lightfoot & Norbert Hornstein (eds.), Verb movement, 41–72. Cambridge: Cambridge University Press.10.1017/CBO9780511627705.004Search in Google Scholar
Ouhalla, Jamal. 1997. Remarks on focus in Standard Arabic. In Mushira Eid & Robert R. Ratcliffe, (eds.) Perspectives on Arabic linguistics X: Papers from the tenth Annual symposium on Arabic Linguistics, 9–45. Amsterdam & Philadelphia: John Benjamins.10.1075/cilt.153.04ouhSearch in Google Scholar
Parkinson, Dilworth. 1981. VSO to SVO in Modern Standard Arabic: A study in diglossia syntax. Al-’Arabiyya 14. 24–37.Search in Google Scholar
Pesetsky, David. 1987. Wh-in-situ: Movement and unselective binding. In Eric Reuland & Alice ter Meulen (eds.), The representation of (in)definiteness, 98–129. Cambridge, MA: MIT Press.Search in Google Scholar
Poletto, Cecilia. 2006. Old Italian scrambling: The low left periphery of the clause. In Patrick Brandt & Petra Schumacher (eds.), Form, structure, and grammar: A festschrift presented to Günther Grewendorf on the occasion of his 60th birthday, 209–229. Berlin: Akademie Verlag.10.1524/9783050085555.209Search in Google Scholar
Rizzi, Luigi. 1997. The fine structure of the left periphery. In Liliane Haegeman (ed.), Elements of grammar, 281–337. Dordrecht: Kluwer Academic Publishers.10.1007/978-94-011-5420-8_7Search in Google Scholar
Rizzi, Luigi. 2004. Locality and the left periphery. In Adriana Belletti (ed.), Structure and beyond, 223–251. Oxford: Oxford University Press.Search in Google Scholar
Shlonsky, Ur. 1992. Resumptive pronouns as a last resort. Linguistic Inquiry 23. 443–468.Search in Google Scholar
Shlonsky, Ur. 1997. Clause structure and word order in Hebrew and Arabic: An essay in comparative Semitic syntax. Oxford: Oxford University Press.Search in Google Scholar
Shlonsky, Ur (ed.). 2015. Beyond functional sequence: The cartography of syntactic structures, vol. 10. Oxford: Oxford University Press.10.1093/acprof:oso/9780190210588.001.0001Search in Google Scholar
Soltan, Usama. 2007. On formal feature licensing in minimalism: Aspects of Standard Arabic morphosyntax. College Park, MD: University of Maryland dissertation.Search in Google Scholar
Vikner, Sten. 1994. Finite verb movement in Scandinavian embedded clauses. In David Lightfoot & Norbert Hornstein (eds.), Verb movement, 117–147. Cambridge: Cambridge University Press.10.1017/CBO9780511627705.008Search in Google Scholar
Wahba, Wafaa. 1984. WH-constructions in Egyptian Arabic. Urbana-Champaign, IL: University of Illinois at Urbana-Champaign dissertation.Search in Google Scholar
Zubizarreta, Maria. 1998. Prosody, focus, and word order. Cambridge, MA: MIT Press.Search in Google Scholar
© 2022 Murdhy Alshamari and Marwan Jarrah, published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.