Scientific Journal of Eduardo Mondlane University, Serie: Arts and Social Sciences

Editorial

Aidate Mussagy — 2025-08-08

CITSHWA PHRASAL TONOLOGY

Charles W. Kisseberth — 2025-08-08

In this paper we examine the two principal rules of phrasal tonology in Citshwa language of Tsonga group (Zone S50 in Guthrie’s classification of the Bantu languages). Using verb-noun sequences, this paper argues that a verb that ends in a H tone is separated from a following initial H noun by downstep (“!”). However, if the following noun has initial 0H, the H of the verb spreads to the initial toneless syllable and induces the Downstep of the following H. If the noun has two or more initial toneless syllables preceding a H, the verb H spreads to the initial syllable but no further. The H in the noun is automatically downstepped by virtue of the toneless syllable(s) intervening between it and the preceding H tone. A significant complication arises when the noun is modified, as the verb’s H spreads to an initial toneless syllable of all nouns, but in the case of a toneless noun does not spread any further. However, in Citshwa there is another phrasal tone rule that deletes a phrase-final H tone. A phrase-final word that ends in a H tone(-span) deletes this H tone obligatorily when it immediately precedes a main verb that has a H tone in its pre-stem structure, and optionally deletes even when it is separated from the verb by another phrase. The syntactic role of the phrase is irrelevant. The present paper sketches these dual phrasal principles: the spreading of a H tone as well as the deletion of a word-final H tone.

CHANGANA PHONOLOGY (S53)

Maria Mendes Cantoni — 2025-08-08

This paper examines the phonetic and phonological characteristics of Changana, as spoken in Mozambique, with a focus on understanding the regular patterns of this linguistic system. Changana boasts a rich phonetic inventory, with over one hundred documented consonants, including both simple and modified consonants (i.e., those with secondary articulations). These form series of pre-nasalized, aspirated, and labialized consonants. The aim of this study is to describe Changana's system of distinctive segments and its sound inventory. In order to achieve this, the study draws on descriptive works related to the language's sound features, particularly the proposals of Ngunga and Simbine (2012) and Langa (s.d.). The discussion focuses specifically on the phonetic inventory, its organization, and its classes, with the goal of proposing a phonological system. Special attention is given to the consonant system, considering documented phonological processes such as epenthesis, deletion, and assimilation. Based on the analysis of the consonant system, we also propose a syllabic structure for the language. The analysis of Changana consonants is particularly significant for phonological studies, as its extensive consonant inventory, traditionally interpreted as contrastive, is typologically rare in terms of both complexity and size. Another key contribution of this work is providing a general overview of the language's phonetic inventory, as the limited descriptions currently available primarily focus on analyzing complex and rare sounds, such as aspirated and labialized pre-nasalized consonants.

SOME ASPECTS OF NOUN CLASS 9/10 PREFIX NASALISATION IN SHONA AND BARWE

Esau Mangoya — 2025-08-08

Generally, most Bantu Languages have nouns that are constructed by way of a combination of a noun prefix and a noun stem. Taking a Shona example of the noun class 1 noun “mukomana” (boy) it is composed of a noun prefix mu- plus the noun stem –komana. Equally, the plural class 2 noun for the same noun is “vakomana” (boys) is constructed by way of noun prefix va- that is added to the noun stem –komana. In the same manner we can also have a Barwe classs 1 noun “muthu” (person) as being composed of the noun prefix mu- plus a noun stem -thu. The plural for the same noun falling into noun class 2 is wanhu (people) also composed of noun prefix wa- plus a noun stem –nhu. However, a different phenomenon occurs with some class 9 and 10 Shona and Barwe nouns where in place of a stand- alone prefix there is some phonemic process that occurs at the noun initial consonant onsets. This is different from other noun classes where a noun prefix is placed before a noun stem. In the present paper it is observed that there are instances where some class 9 and 10 nouns go through the phonological processes that play substitute to the noun prefix. In the class 9 and 10 nouns the process targets the word initial consonants where some morphophonological processes take place in the Shona and Barwe languages. The research is corpus based as the researcher used Shona and Barwe corpora to retrieve the class 9/10 nouns. The analysis is based on “Lexical Phonology Theory”

TENSE AND ASPECT IN CITSHWA

Lucério Gundane — 2025-08-08

The according of Kiparsky's Lexical Morphology and Phonology (1982), the study, based on Meeussen (1967), analyzes the distribution of the tense and aspect constituents in Citshwa (S.51 according to GUTHRIE, 1967-71). The problem that arises lies in the analysis of the distribution of these constituents in the structure of the verb, as well as the way in which they are expressed. The study is exploratory and qualitative and during data collection, the methodological triangulation involved bibliographical research, qualitative interviews and the questionnaire. The Fieldwork took place in the city of Maxixe from 2014 to 2015. From a universe of 300 informants, were selected 70 Citshwa speakers, using variables such as age, sex, educational level and mother tongue. The Time was described in the category’s simple past, compound past, habitual present, punctual present and simple future, in affirmative and negative forms. Regarding aspect, the study described the 4 aspectual categories (COMRIE, 1976): perfective, habitual imperfective, punctual imperfective and progressive imperfective, having shown that these morphemes are distributed in different positions in the structure of the verb and, in some cases, as is the case with the perfective aspect, its position coincides with that of the simple past. Thus, it is concluded that in Citshwa the constituents tense and aspect are attached to the verb and are distributed within the 11 positions in the internal structure. These constituents are expressed by segmental and suprasegmental morphemes.

THE EXPRESSION OF THE COMPOSITIONAL ASPECT THROUGH EXTENSION MORPHEMES IN CINYANJA

Geraldo Macalane — 2025-08-08

Aspect is one of the most complex and controversial domains of modern linguistics. This fact is justified by the existence of several perspectives of approach to this category, from the traditional ones, centered on Aktionsart, or the inherent meaning of the verb, to the most current, oriented towards compositionality. This study, based on compositional theories from Verkuyl (1972) and Moens (1987), aims to understand the role of verbal extensions in the expression of aspectual values in Cinyanja. Specifically, the research is conducted to (i) identify the extension morphemes that act as aspectual value marking operators in Cinyanja; and (ii) explain the transitions that occur from certain aspectual categories to others, as a result of the incorporation of extension morphemes in verb forms. The study is descriptive, based on linguistic items from Cinyanja, extracted from a bibliographic research. For data treatment, content analysis was privileged. The research concludes that some extension morphemes in Cinyanja, when incorporated into the verb structure, trigger modifications, not only at the syntactic level, but also in the semantic-aspectual domain, where (i) Culminated Processes, Culminations and States are converted into Processes; and (ii) Culminated Processes, Processes and Culminations are converted into States.

STRATEGIES FOR VERBAL NOMINALIZATIONS OF NON-HUMAN CLASSES IN CINYANJA

Ronaldo Rodrigues de Paula — 2025-08-08

This article aims to describe the processes of verbal nominalization in Cinyanja relating to nominal classes distinct from classes 1 and 2. The derivation of nouns from human classes is a productive process in the language for the formation of agentive nouns from verbs, through from the affixation of prefixes of the aforementioned classes and the final vowel –i as in m-phunzits-i (teacher) derived from the verb ku-phunzíts-a (teach) (MCHOMBO, 2017). Nominalizations constructed from prefixes of nominal classes distinct from classes 1 and 2 do not form agentive nouns, but denote the very eventivity expressed by the verb or the way in which such eventivity is performed. Take as an example the name ‘dream’ lot-o derived from the verb ‘to dream’ ku-lot-a, formed from class 5, a null prefix, and the final vowel – o. Also the name 'praise' derived from the verb 'to praise': ku-tamand-a, constructed with the prefix of class 7: ci-tamand-o, or even the name 'cure' n-ciz-o of class 3, derived from the verb 'to heal/to cure' ku-ciz-a. We also found another form, built from the class 14 prefix with the final vowel -o, u-phunzíts-o ‘way of teaching’. We note that, regardless of the singular deverbal nominalization used, the plural will be the prefix of nominal class 6 “ma-“, thus we have ma-lot-o, ma-tamand-o, ma-ciz-o and ma-phunzits-o, respectively. The introspective method was used, since the co-author is a native speaker, and the bibliographic method was used to search for available material. The main objective of this article is to describe the aforementioned nominalization processes and their relevant contexts of use.

NOUN CLASSES AND PREFIXES IN BANTU

David Alberto Seth Langa — 2025-08-08

Bantu languages exhibit a complex verb structure, not only because of their positions available to receive the verb's internal constituents, but, above all, because of their relationship with syntax. More recently, linguists have also turned their attention to the (re)analysis of nominal classes and prefixes. The literature tends to consider, using a functionalist analysis of its constituents, that the nominal prefix licenses grammatical gender information (singular and plural), thus following the line of pioneers in these studies. Because these languages have a number of nominal classes that vary between 1 and 10, and are interpreted, roughly speaking, as marking the singular/plural opposition, which is considered reductive for languages that have such complex verbal morphology. Hypotheses suggest that nominal classes and prefixes encode lexical information (Bleek, 1862). Therefore, it is suggested that nominal prefixes be (re)analyzed as a derivation, in the traditional sense of the term, and not an inflection as they have commonly been classified. With the aim of testing these hypotheses, through a comparative approach, supported by interviews, philology and introspection, this article argues in favor of the lexical coding of nominal classes and prefixes in Changana (S53), corroborating the hypothesis advanced by Taraldsen et al. (2018) of the reanalysis of nominal prefixes in silent names.

MORPHOSYNTACTIC ANALYSIS OF THE MORPHEMES -UR- AND -AM- IN CIWUTEE

Joaquim João Razão — 2025-08-08

Morphemes are minimal units of morphological analysis. Taking this assumption into account, it is clear that, in order to capture the parametric features of morphological units, it is necessary to explore their larger field (morphology). The focus of the research is on morphemes that can occur before and/or after the base. The research aims to analyse, test, characterise and distinguish the morphemes -ur- and -am-, from a morphosyntactic and lexico-semantic perspective. To this end, the distribution test, the Baker Mirror Principle (1995) and Baker Incorporation (1988) theories were used to assess the existence of some linearity and possible lexico-semantic restrictions in the concatenation of derivational units which, in turn, can establish a relationship with the extralinguistic world. With these theoretical approaches, we intend to assess the real category of morphemes -ur- and -am- that, in some Bantu languages, function as fossilized verbal extensions. The objectives listed refer us to the question “do the morphemes -ur- and -am- behave as verbal extensions and what is the real morphosyntactic and lexico-semantic category of the bases that aggregate them in Ciwutee? The possible answers that arise from the above question are: a) the morphemes -ur- and -am- behave as extensions in Ciwutee; b) the morphemes -ur- and -am- establish distinct morphosyntactic and lexico-semantic restrictions on the bases that aggregate them; and; c) the morphemes -ur- and -am- are fossilized extensions in Ciwutee. Using the qualitative approach, the bibliographic method and Introspection, as well as the distribution test, it was possible to conclude that -ur- is a verbal extension that joins the verbal bases with cyclic movement semantics and -am- is not an extension, but rather a verbalizer that joins only the ideophonic bases. This general conclusion was given by 84.7% of informants from the 100% sample universe.

THE SYNTAX OF NON-SEGMENTAL RELATIVE CLAUSES IN CINYANJA

Mário Biriate — 2025-08-08

The present research presents a proposal for analyzing non-segmental relative clauses in Cinyanja, a language from the Bantu group (N31). The data analysed were selected by the author, using the linguistic competence of native speakers of the language. Before analysis, the sentences were also submitted to other speakers and teachers of the language to check their grammaticality and acceptability. In Cinyanja, in addition to segmental resources, which involve the marker – mene and the relative suffix –o, placed to the right of the last word of the sentence, relative sentences can also be formed using a high tone (with or without the relative suffix -o). In this last relativisation strategy, the subject necessarily appears in a post-verbal position. Thus, non-segmental relative markers are analysed as structures in which C+wh has a status identical to that of clitic relative markers that appear incorporated into the structure of the verbal complex in other Bantu languages (DEMUTH and HARFORD, 1999), which typically trigger the subject’s inversion. As the relative marker is an element without a phonological matrix, it behaves like a clitic form and must be affixed to T. Assuming the Distributed Morphology model (Halle and Marantz, 1993), it is argued that the affixation of C to T It occurs at the post-syntactic level of grammar, under adjacency. The subject remains in situ, in Spec-VP, obtaining the same word order patterns as simple sentences with the subject in post-verbal position, namely VOS or, in the presence of the agreement mark with the object, VSO.

AFTER ALL, HOW MANY WH-INTERROGATIVE STRATEGIES OCCUR IN KIMWANI, A BANTU LANGUAGE SPOKEN IN NORTHERN MOZAMBIQUE?

Calawia Salimo — 2025-08-08

This article with the theme “After all, how many Wh-interrogative strategies occur in Kimwani, a Bantu language spoken in Northern Mozambique” describes and analyze the pattern of Wh-interrogatives in Kimwani. According to Braga, Kato and Mioto (2009), Wh-interrogatives are part of a set of constructions called Wh-constructions and are divided into three subgroups: (i) relative constructions, (ii) cleaved constructions (iii) Wh-interrogatives. The article, describes the last subgroup, the typology of Wh-questions. Methodologically, it is a qualitative approach centered on content analysis based on grammar theory. The data analyzed in this article were collected, through interviews, in 2015 and in 2020 with native speakers of Kimwni as part of the production of the master's dissertation and doctoral thesis. Therefore, a while later, the same data is reanalyzed in this article for a Wh-interrogative approach in Kimwani. Priority was given to content analysis of the data aided by introspection, phenomenological methods and acceptability and grammaticality judgment testing with native speakers of Kimwani. The latter made it possible to access the intuitive knowledge of speakers with greater fidelity. The result of the data analysis showed that in Kimwani three Wh-type interrogative formation strategies occur, namely: Wh in situ, Wh-totally moved, Wh-partially moved. The latter is marked by not occurring more frequently and considered a bit strange by speakers.

OBJECT MARKING IN BANTU

David Alberto Seth Langa — 2025-08-08

Object marking is a common linguistic phenomenon in Bantu languages, in which the object of a sentence is referenced in the verbal structure through a prefix. The aim of this article is to systematize the descriptions of the phenomenon of object marking in Changana. To this end, we compiled the morphosyntactic and semantic-pragmatic contexts in which object marking emerges in the language, based on a bibliographic review of studies by Chimbutane (2002), Ngunga and Simbine (2012), Langa (2013), and Ngunga, Duarte, and Quesler (2016). The descriptions of the phenomenon indicate that object marking serves to encode topical interpretation, as the argument marked on the verb represents given information in the discourse. In this context, the referenced object may have optional morphophonological realization in the sentence as a full NP, since the object marker makes it possible to recover the reference of the marked NP. This phenomenon occurs in transitive or transitivized sentences and in double object constructions. Furthermore, object marking allows flexibility in the order of constituents, as the marker's coreferentiality ensures the interpretation of the marked NP as the object of the sentence. Finally, we identified a divergence regarding the grammatical status of the object marker, with two proposals: one characterizes the marker as an incorporated pronoun, while the other considers it as concord marker. Thus, the systematization of the information proposed in this article serves as a basis for further studies on object marking.

APPLICATIVE CONSTRUCTIONS IN KIMWANI

Ana Clara Passoni Moraes — 2025-08-08

This article presents a formal analysis of the applicative constructions in Kimwani, which is morphologically realized by the morpheme {-ir-} and its allomorphs. This investigation is justified by the lack of detailed studies on the syntactic behaviour of the applied and direct object. In this sense, this research aims to examine if Kimwani is a symmetrical or asymmetrical object language. To implement our proposal, we assume generative hypotheses in its most recent version, the Minimalist Program, and in other works, regarding the syntax of the applicative constructions in bantu languages. We provide an analysis of the contexts in which an applied object is introduced by the Appl^o head in inaccusative, inergative and ditransitive verbs. Our hypothesis is that Kimwani exhibits grammatical properties that are typical of asymmetrical applicative language, insofar as only the applied object can be passivized and can trigger the object marker on the verb stem. We also assume that, owing to the fact that applicative head does not open an extra specifier position, the direct object is retained inside the VP, thereby being sent to Spell-Out very early during the syntactic derivation. In this regard, we conclude that only the applied object can participate in the syntactic operations such as passivization and object agreement.

WHO SAID CHIKUNDA WAS NOT A LANGUAGE?

Rozenn Guérois — 2025-08-08

Chikunda is a Bantu language classified N42 by Guthrie (1971), and spoken at the junction of three southern African countries: Mozambique (Zumbo), Zambia (Luangwa) and Zimbabwe (Kanyemba), as well as in southern Malawi in the districts of Mwanza and Chikwawa. The Chikunda socio-cultural group has received much attention from historians due to their function of slave soldiers during the “prazo” system in colonial Mozambique. From a common social identity, an ethnic identity emerged that involved the creation of a language, Chikunda, resulting from the mixture of dialects and intra-Bantu languages. The unusual history of the Chikunda raises fascinating questions from a linguistic point of view, with possible important repercussions for the historical narrative of this community. This article presents the preliminary results of the project OriKunda: On the origins of Chikunda, “a language without a land” (April 2023-March 2027, French ANR-22-CE54-0009), led by Rozenn Guérois (CNRS LLACAN), whose objective is to review the unique history of the socio-cultural group and the Chikunda language, from its genesis to the present day, through historical linguistics, anthropological linguistics and sociolinguistics. Based on recent fieldwork surveys, the present article describes preliminary sociolinguistic and linguistic elements of Chikunda.

THE LINGUISTIC LANDSCAPES AND SOUNDLANDSCAPES OF THE CITY OF MAPUTO AS LANGUAGE POLICIES IN MOZAMBIQUE

Ezra Alberto Chambal Nhampoca — 2025-08-08

This paper explores the language policies of Mozambique, with a particular focus on Maputo City, through the concepts of linguistic landscapes and sounds capes. These landscapes are understood as manifestations of implicit multilingual language policies that interact, both directly and indirectly, with explicit language policies. Language policies play a pivotal role in the dynamics of language use, especially as they serve as pedagogical tools, such as educational policies promoting bilingualism. The study examines the linguistic and sound landscapes of the city, which serve as identity and symbolic expressions, encompassing a range of meanings and functions, including connections to ancestry, ways of life, economic survival, rituals, and ceremonies. It is argued that linguistic landscapes carry significant cultural, identity, and political implications through a semiotic analysis of the landscape that includes written texts, art, graffiti, and other visual and auditory forms. . Similarly, the soundscape encompasses the lived experiences of individuals within the sound environment, examining their relationship with sounds and noise. Data was collected through photography and ethnographic methods, guided by the following research question: What languages and sounds constitute the linguistic and soundscapes of Maputo? The choice of Maputo as the focus of this study is grounded in its significance as the political capital of Mozambique, a central hub for institutional and cultural activities, and a city characterized by its rich and complex multilingual environment.

BILINGUAL EDUCATION AND GENDER INEQUALITIES IN MOZAMBIQUE:

Crisófia Langa da Câmara — 2025-08-08

This paper analyses the role of Bilingual education in retaining girls in school and in gender inequalities reduction in Mozambique, by comparing the number of girls who start grade 1 and those who reach at grade 6. The methodology that underpinned the construction of this paper consisted of bibliographical research and documentary analysis. In effect, a literature review was carried out in papers which discuss bilingual education (CHIMBUTANE, 2009; PATEL, 2012 e CHIMBUTANE, STROUND 2011), initial literacy in mother tongue (CHIMBUTANE, 2011, BENSON 2005), language policy and linguistic rights (UNESCO, 1996 2000a e 2002; BENSON 2002; CALVET 2007). Likewise, data were collected from various documents such as the 2023 MINEDH statistical survey, Final Report on Bilingual Education (BENSON 1997; INDE 1997; INE 2019, 2023) and Mozambican 1^st to 3^rd Cycle Basic Education Programs (2003, 2015). Our interest arises from the fact that studies have shown that the inclusion of Mozambican languages as a medium of instruction not only solved the problem of the high rates of school wastage (dropouts and repeaters) between 1984 and 2000, which were believed to be related to the language used as a medium of instruction, but also to the retention of girls in school. The role of women is associated with domestic activities and the family, which means that they have little exposure to the languages of communication used in this context, including school. As a result, school ceases to be a welcoming space and becomes a place where gender inequalities multiply. Although we believe that more studies are needed to explain the reasons for school dropout, the 2023 data analysed in this paper show that bilingual education offers solutions for this problem, since the figures for monolingual education lead us to argue that the language of instruction may be one of the obstacles faced by students. In fact, we saw that in both monolingual and bilingual, there is more wastage of girls in the monolingual version, with 214,029 more compared to the bilingual version.