The inclusion of discourse markers (DMs), which are commonly depicted as syntactically non-integrated elements, within the domain of grammaticalization has been a controversial topic in the last two decades. The emergence of DMs for some researchers are cases of grammaticalization, whereas they are viewed by some other researchers as elements that escape from the grammatical domain to the pragmatic and communicative domain. This paper argues with the grammaticalization of DMs. It provides evidence to this view from Jordanian Arabic. The cases under investigation in this paper are the DMs ma , hu and mahu . This paper offers synchronic evidence supporting their grammaticalization. This evidence is based on two main conclusions: (1) these DMs share common grammatical items most of the grammaticalization sub-processes that lead to their emergence, and (2) they are not dis-integrated elements from the sentence grammar. They rather belong to the peripheral domain of grammar.

Emphasis spread has been empirically investigated in a number of Arabic vernaculars. Previous accounts of its (non-)application indicate that its domain varies across dialects, ranging from the adjacent vowels to the syllable, to the morpheme, and finally to the (phonological) word. This research paper, which contains an experiment analyzing data from 14 native speakers of  Rural Jordanian Arabic (RJA) spoken in Irbid, explores the domain of emphasis spread in this sub-variety, which is an understudied dialect. The results show that emphasis spread is not sensitive to the adjacent vowels or the syllable, but rather to the morpheme type; that is, it applies to suffixes and proclitics, but not to clitics, unless the source of emphasis is very close to the target clitic. Following Selkirk (1996) and consistent with Jaradat (2018), we propose a  prosodic account to emphasize spread in RJA: proclitics and suffixes are prosodically integrated into the host stem, and therefore they allow emphasis spread. On the contrary, clitics are prosodically free. Hence, the prosodic word boundary between a stem and a clitic blocks emphasis spread from the stem onto the clitic, unless the source of emphasis is too close to the target clitic. This implies that the domain of emphasis spread is not only dialect-specific, but may also differ among the sub-varieties of one Arabic dialect.

This article, which reports on an experimental study based on data obtained from 86 native speakers of Northern Rural Jordanian Arabic (NRJA), offers empirical evidence that emphasis spread can apply bidirectionally across word edges when they are prosodically word internal. In other words, emphasis can spread from one word to a neighboring (plain) word when the two words form one prosodic word (ω). In order to support this hypothesis, we examine simple construct-state nominals (i.e., nominals consisting of two members), which are widely assumed to behave prosodically as one word in Arabic grammar (see, e.g., Borer, Hagit. 1999. Deconstructing the construct. In Kyle Johnson & Ian Roberts (eds.), Beyond principles and parameters, 3–89. Dordrecht: Kluwer). A simple construct state nominal is mapped onto one ω, as proposed in the current article. Additionally, this article examines instances of construct state nominals in which two words could be members of a ω by cliticization. In this case, a weak form (e.g., a reduced functional word such as an auxiliary verb) is attached to the ω of a lexical word, as an enclitic. The article shows that emphasis spreads across word edges within the ω mapped from a simple construct state nominal or a word along with its enclitics. In so doing, it provides evidence that the domain of emphasis spread is larger than the syllable, morpheme, or single-word form, as proposed by other studies for other Arabic varieties.

No abstract

No abstract