The semantics of verbal morphology in central Tanzanian Bantu languages: a comparative study
In most Bantu languages, tense, aspect and mood [TAM] systems are typically very complex, covering numerous morphosyntactic categories. The lexical semantics of these TAM systems, especially in combination with verb properties, is an understudied area. The individual studies that do exist do not easily allow for cross-linguistic comparison even within the Bantu family. I will conduct a thorough analysis of the semantic construal of TAM notions and their grammatical encoding on the verb in the East Ruvu Bantu languages (Kagulu, Kami, Kwere, Kutu, Luguru and Zalamo). The project will examine the TAM systems of the sample languages by studying the forms, their basic meanings, extended functions and distribution. The language data and analyses will enhance our understanding of Bantu languages and constitute a valuable cross-linguistic documentation of this particular area of grammar; they will also feed into the ongoing debates in theoretical linguistics on TAM, and the lexicon/grammar interface both within and beyond Bantu studies. Moreover, the analytical tools that I will develop for the relevant types of language data (e.g. semantic verb property lists and their interaction with TAM markers) could serve as standard instruments to be integrated into future work on other languages, fostering Bantu comparative work at a larger scale. Finally, we do not know what lesser-studied languages can tell us about TAM in general and what implications there may be for linguistics more broadly.
Final report
The purpose of the project was to describe and analyse the semantic construal of (universal) tense, aspect and mood (TAM) notions and their grammatical encoding on the verb in the East Ruvu Bantu languages (Kagulu, Kami, Kwere, Kutu, Luguru and Zalamo). The project examined the TAM systems of the sample languages by documenting foremost the tense and aspect markers, their basic meanings, extended functions and distribution, as well as the interaction of lexical semantics of the verb stems.
There were two new developments that unfolded during the project. One is the collaboration with Leora Bar-el from the University of Montana. While their research backgrounds focus on different language families, Bar-el and Petzell’s research interests intersect in two important ways: (i) their interests in the documentation and analysis of aspectual classes and tense and aspect categories, and (ii) the methods used to uncover these subtle distinctions that rely on native speaker intuitions and creative field methodologies.
The other new development is the collaboration with a data manager who has developed an annotated database of the project data, as well as preparing the data for deposit with the Swedish National Data Service. This collaboration has also resulted in a publication (Jordan & Petzell. In press.); a welcome result.
A mid-evaluation reviewer suggested that it would be fruitful to write an article about the challenges of fieldwork. A chapter on precisely this issue will appear in an Oxford anthology of work on Bantu languages (Bloom Ström & Petzell. In press.)
Implementation
The proposed theoretical framework, i.e. Botne’s cognitive model (Botne & Kershner 2008; Botne 2012, 2014)) continues to be used to analyse the data, following the suggestion by one external reviewer of the project application. We are in communication with Robert Botne about the project data and we are discussing potential collaborations in the near future. Since this project ties into our newly awarded project (by The Swedish Research Council) on valency, we will continue conducting research on the verb in the East Ruvu languages. We will continue to acknowledge Riksbankens Jubileumsfond as the original funder in future publications and presentations.
The first Tanzania fieldwork trip was conducted in 2016. Petzell returned to Tanzania again in April 2017 to teach at the University of Dar es Salaam under the auspices of the Linnaeus-Palme exchange programme, and was fortunate to be able to organize a side fieldwork trip and meet with speakers to conduct a small pilot survey with a newly developed questionnaire. The third field trip took place in November-December 2018 and the final one in November 2019.
The project's three main results
One major result was uncovering the reduced morphological systems of these languages. Bantu languages are known for their complex tense and aspect morphology in general. Consequently, languages with reduced tense and aspect morphology within a language family known for complex tense and aspect systems raise a variety of interesting questions. Some of the languages under investigation only exhibit a past and a non-past morpheme (i.e. no distinct present or future tenses), and the past also doubles as the perfective (and anterior). This reduced system, especially with the neutralized past/perfective, had not been documented in the literature prior to this project. What is more, the past/perfective usually carries null realization (i.e. no overt marking). We are still researching how this interacts with the aspectual categories and vice versa, and how much temporal and aspectual information is conveyed through other constituents such as temporal adverbials, and periphrastic verbs (including auxiliaries and defective verbs).
Another result was further documentation and analysis of the unique and complicated marker tsa. This marker encodes some type of shared knowledge or shared reference in Luguru, with meanings stretching from ‘at a specific time’, ‘at that place’ to ‘as we know’ or even ‘for that reason’. Not only is the function of tsa new in Luguru (it is not attested in older sources), it is not documented for other Bantu languages and it cannot be traced back to Proto-Bantu. Our data demonstrate that tsa is a marker of what the speaker takes to be shared knowledge in a wide sense. It refers to a definite span of time or space, or more abstract notions, e.g. reasons and expectations. Tsa is a marker that cannot be used in out-of-the-blue contexts without some sort of anchor. This is corroborated by the fact that combining tsa with an indefinite adverbial results in infelicity.
The third finding has to do with contrasting perfectivity with imperfectivety for understanding actionality (lexical aspect of the verb). Some frameworks suggest that an inventory of actional types can be determined based on the behavior of verbs in combination with perfective and imperfective morphology, and that perfectivity vs imperfectivety can be used as diagnostics for actionality. However, there is no clear imperfective/perfective marking in the East Ruvu languages. This lack of temporal/aspectual morphology raises important questions about how temporal/aspectual meanings are conveyed in languages with reduced temporal/aspectual morphology which we are currently exploring in a paper. Also, we are looking into how cross-linguistically viable the actionality diagnostics used for other languages of the Bantu family are for the East Ruvu languages.
Dissemination of results and collaborations
Dissemination of results of this project took place in a variety of venues. The conference ‘The semantics of verbal morphology in under-described languages’, organized at GU as part of this project in June 2017, provided a forum for the discussion of the morphological encoding of TAM markers within the verb, their meaning, and how they can combine with the lexical aspect of the verbal stem. The conference received 40 abstracts out of which we chose 24. The speakers came from 12 different countries in Europe, North America and Africa. The conference brought together researchers working on these topics in under-described languages to share their work and provided a forum to create new partnerships and collaborations between an international body of researchers. The event was funded by RJ’s ‘Forskningsinitiering’.
The results of the conference are being disseminated through the thematic volume ‘The semantics of verbal morphology in under-described languages’. We have chosen to publish the volume in the journal StOrE (https://journal.fi/store) since it publishes papers that are of high quality, online, open-access and peer-reviewed. Lotta Aunio (University of Helsinki) is the editor of the journal, and will also serve as one of the three editors for this volume, alongside Malin Petzell and Leora Bar-el.
Moreover, September - November 2019 Leora Bar-el was a guest researcher at the department within the project. She is specializes in describing the lexical semantics of under-described languages.
Additionally, we participated in a number of conferences disseminating the results of the project, such as the International Bantu Conference in Helsinki, Finland in June 2016, the World Congress of African Linguistics, hosted by Mohammad V University of Rabat in Morocco in August 2018, and the Conference of the Association for Linguistic Typology, University of Pavia, Italy in September 2019. Other conferences and workshops include Approaches to morphosyntactic variation in Bantu Dar es Salaam, Tanzania in July 2016, Workshop on Language Contact, Uppsala University, Sweden in June 2018, Workshop on Fieldwork: methods and theory, University of Gothenburg, Sweden in December 2018.
Finally, the project has resulted in a small but strong team of researchers working on the verbal morphology of Tanzanian Bantu languages affiliated with the department. Ponsiano Kanijo, a doctoral student whose dissertation focuses on the lexical semantics of Nyamwezi (co-supervised by Petzell) obtained his PhD in October 2019. A project assistant during the first project field trip to Tanzania (Lengson Ngwasi - supervised by Petzell) is now a doctoral student at GU. Ngwasi is working on verbal morphology (voice) in four Tanzanian Bantu languages. Kanijo and Bar-el, together with post doc Sebastian Dom, are part of the newly awarded project To break or be broken ? A study of valency-decreasing alternations in East Ruvu Bantu languages (from the Swedish Research Council) which forms a natural continuation of this project. Having native Swahili speakers at the department (Kanijo and Ngwasi) also sparked a spin-off effect: the starting of an introductory university course in Swahili - the first one at the University of Gothenburg.
There were two new developments that unfolded during the project. One is the collaboration with Leora Bar-el from the University of Montana. While their research backgrounds focus on different language families, Bar-el and Petzell’s research interests intersect in two important ways: (i) their interests in the documentation and analysis of aspectual classes and tense and aspect categories, and (ii) the methods used to uncover these subtle distinctions that rely on native speaker intuitions and creative field methodologies.
The other new development is the collaboration with a data manager who has developed an annotated database of the project data, as well as preparing the data for deposit with the Swedish National Data Service. This collaboration has also resulted in a publication (Jordan & Petzell. In press.); a welcome result.
A mid-evaluation reviewer suggested that it would be fruitful to write an article about the challenges of fieldwork. A chapter on precisely this issue will appear in an Oxford anthology of work on Bantu languages (Bloom Ström & Petzell. In press.)
Implementation
The proposed theoretical framework, i.e. Botne’s cognitive model (Botne & Kershner 2008; Botne 2012, 2014)) continues to be used to analyse the data, following the suggestion by one external reviewer of the project application. We are in communication with Robert Botne about the project data and we are discussing potential collaborations in the near future. Since this project ties into our newly awarded project (by The Swedish Research Council) on valency, we will continue conducting research on the verb in the East Ruvu languages. We will continue to acknowledge Riksbankens Jubileumsfond as the original funder in future publications and presentations.
The first Tanzania fieldwork trip was conducted in 2016. Petzell returned to Tanzania again in April 2017 to teach at the University of Dar es Salaam under the auspices of the Linnaeus-Palme exchange programme, and was fortunate to be able to organize a side fieldwork trip and meet with speakers to conduct a small pilot survey with a newly developed questionnaire. The third field trip took place in November-December 2018 and the final one in November 2019.
The project's three main results
One major result was uncovering the reduced morphological systems of these languages. Bantu languages are known for their complex tense and aspect morphology in general. Consequently, languages with reduced tense and aspect morphology within a language family known for complex tense and aspect systems raise a variety of interesting questions. Some of the languages under investigation only exhibit a past and a non-past morpheme (i.e. no distinct present or future tenses), and the past also doubles as the perfective (and anterior). This reduced system, especially with the neutralized past/perfective, had not been documented in the literature prior to this project. What is more, the past/perfective usually carries null realization (i.e. no overt marking). We are still researching how this interacts with the aspectual categories and vice versa, and how much temporal and aspectual information is conveyed through other constituents such as temporal adverbials, and periphrastic verbs (including auxiliaries and defective verbs).
Another result was further documentation and analysis of the unique and complicated marker tsa. This marker encodes some type of shared knowledge or shared reference in Luguru, with meanings stretching from ‘at a specific time’, ‘at that place’ to ‘as we know’ or even ‘for that reason’. Not only is the function of tsa new in Luguru (it is not attested in older sources), it is not documented for other Bantu languages and it cannot be traced back to Proto-Bantu. Our data demonstrate that tsa is a marker of what the speaker takes to be shared knowledge in a wide sense. It refers to a definite span of time or space, or more abstract notions, e.g. reasons and expectations. Tsa is a marker that cannot be used in out-of-the-blue contexts without some sort of anchor. This is corroborated by the fact that combining tsa with an indefinite adverbial results in infelicity.
The third finding has to do with contrasting perfectivity with imperfectivety for understanding actionality (lexical aspect of the verb). Some frameworks suggest that an inventory of actional types can be determined based on the behavior of verbs in combination with perfective and imperfective morphology, and that perfectivity vs imperfectivety can be used as diagnostics for actionality. However, there is no clear imperfective/perfective marking in the East Ruvu languages. This lack of temporal/aspectual morphology raises important questions about how temporal/aspectual meanings are conveyed in languages with reduced temporal/aspectual morphology which we are currently exploring in a paper. Also, we are looking into how cross-linguistically viable the actionality diagnostics used for other languages of the Bantu family are for the East Ruvu languages.
Dissemination of results and collaborations
Dissemination of results of this project took place in a variety of venues. The conference ‘The semantics of verbal morphology in under-described languages’, organized at GU as part of this project in June 2017, provided a forum for the discussion of the morphological encoding of TAM markers within the verb, their meaning, and how they can combine with the lexical aspect of the verbal stem. The conference received 40 abstracts out of which we chose 24. The speakers came from 12 different countries in Europe, North America and Africa. The conference brought together researchers working on these topics in under-described languages to share their work and provided a forum to create new partnerships and collaborations between an international body of researchers. The event was funded by RJ’s ‘Forskningsinitiering’.
The results of the conference are being disseminated through the thematic volume ‘The semantics of verbal morphology in under-described languages’. We have chosen to publish the volume in the journal StOrE (https://journal.fi/store) since it publishes papers that are of high quality, online, open-access and peer-reviewed. Lotta Aunio (University of Helsinki) is the editor of the journal, and will also serve as one of the three editors for this volume, alongside Malin Petzell and Leora Bar-el.
Moreover, September - November 2019 Leora Bar-el was a guest researcher at the department within the project. She is specializes in describing the lexical semantics of under-described languages.
Additionally, we participated in a number of conferences disseminating the results of the project, such as the International Bantu Conference in Helsinki, Finland in June 2016, the World Congress of African Linguistics, hosted by Mohammad V University of Rabat in Morocco in August 2018, and the Conference of the Association for Linguistic Typology, University of Pavia, Italy in September 2019. Other conferences and workshops include Approaches to morphosyntactic variation in Bantu Dar es Salaam, Tanzania in July 2016, Workshop on Language Contact, Uppsala University, Sweden in June 2018, Workshop on Fieldwork: methods and theory, University of Gothenburg, Sweden in December 2018.
Finally, the project has resulted in a small but strong team of researchers working on the verbal morphology of Tanzanian Bantu languages affiliated with the department. Ponsiano Kanijo, a doctoral student whose dissertation focuses on the lexical semantics of Nyamwezi (co-supervised by Petzell) obtained his PhD in October 2019. A project assistant during the first project field trip to Tanzania (Lengson Ngwasi - supervised by Petzell) is now a doctoral student at GU. Ngwasi is working on verbal morphology (voice) in four Tanzanian Bantu languages. Kanijo and Bar-el, together with post doc Sebastian Dom, are part of the newly awarded project To break or be broken ? A study of valency-decreasing alternations in East Ruvu Bantu languages (from the Swedish Research Council) which forms a natural continuation of this project. Having native Swahili speakers at the department (Kanijo and Ngwasi) also sparked a spin-off effect: the starting of an introductory university course in Swahili - the first one at the University of Gothenburg.