Previous Seasons Meetings

Synchronic and Diachronic Complex Tone in Endangered Tenyi Languages
Savio Meyase (University of York)

The languages in the northeast of India in the eastern Himalayan range are rather poorly documented and the availability of linguistic studies for these languages is next to nothing compared to many other languages of the world. The main reasons for this are geographic isolation and economic backwardness, and historic as well as extant abandonment from authorities in control. Apart from digitally archiving endangered language data, I also study the diachronic changes in both the tones and non-tonal elements of the related languages of Tenyidie and reconstruct a proto-Tenyi language with an attempt to establish the relationship of the languages and variants into a traditional language family tree.

The Tenyi languages are tonal languages employing at least four lexical tones. Any language having more than two tones in the inventory is considered a complex tone system, and only very few languages have been documented to have more than three tones. The phonological study of complex tones itself is still at a nascent stage. My previous work (2021, 2022) showed that Tenyidie tones, while appearing complicated and unpredictable, can be neatly studied with hitherto available phonological tools with the proposition that tones can (and should) be split into smaller units. The tonal model I proposed with these tonal units is shown in the PDF of this abstract (see below).

This proposal, and the model that I used, is also being borne out by the newer languages that I am looking at the moment, with support especially from Sopvoma [Mao] where the patterns of tone change are different but still bears out the tonal model used for standard Tenyidie. The comparison of tonal data between Tenyidie and Sopvoma also provides evidence of how tone in these languages could have evolved from a simple two-tone system with cues again supporting the atomisation of tones into smaller features.

The comparisons, reconstructions and the archival of these languages are done with a curated version of the Leipzig-Jakarta list (Haspelmath and Tadmor 2009).

References
Haspelmath, Martin and Uri Tadmor (eds.), 2009. Loanwords in the World’s Languages: A Comparative Handbook. Mouton de Gruyter.
Meyase, Savio, 2021. ‘Polarity in a four-level tone language’. Phonology (38).
Meyase, Savio, 2022. Tenyidie Tone. PenThrill Publications.

Comparative Constructions at the Interface between Syntax and Semantics: Lessons from Malay
Eve Suharwardy (University of Manchester)

The syntactic realisation of the standard of a comparative is either phrasal or clausal. In the case of English, it is typically assumed that both are possible, see the contrast between (1a) vs (1b/c).

1. a. Jon is taller than [Laura].
   b. Jon is taller than [Laura is].
   c. Jon owns more video games than [Laura owns videogames].

However, the availability of both comparative standards is in fact a point of variation crosslinguistically. Where Russian and Greek display the same pattern as English (Pancheva 2006; Merchant 2009), there are many languages which lack clausal standards altogether, e.g. Hindi, Turkish and Samoan (Bhatt & Takahashi 2011; Hofstetter 2009; Hohaus 2015).

In order to determine the syntactic status of the standard phrase, we can use various diagnostics. For example, the availability of constructions subcomparatives (e.g. ‘the desk is longer than the door is wide’) indicates clausal standards, whilst the availability of a reflexive remnant in the standard (e.g. ‘no star shines brighter than itself’) indicates phrasal ones. In this talk, I present original fieldwork data regarding the application of these diagnostics to the Austronesian language, Malay, the results of which have significant implications for the semantic analysis.

In this talk I will be introducing the concept of lexical sociolinguistics: as a word, or a new meaning of a word, or a new soundshape of a word, enters the language, it always does so in the language of a speaker anchored in space, time, and in a social situation, talking to another person similarly sited.  For a new word or pronunciation to spread, the innovation has to move from the initial group of similarly-sited speakers to speakers in other places and other social situations.  Therefore, all word-change has the potential to become sociolinguistically marked – that is, to gain the quality of being associated with the kind of person who first or typically used it, or went on to use it – and it is sometimes possible to recover what these sociolinguistic qualities might have been.  The research question is thus ‘what type of person used this word when, where, and in what kind of social situation’, and the sociolinguistic focus is on recovering historical social situations and affiliations.  I will introduce the concept of communities of spatial practice, and I will demonstrate with some words that historical dictionaries usually omit, such as streetnames, brand-names and numbers.

The work discussed in this talk is part of the Linguistic DNA research project (linguisticdna.org), whose principal aim was to identify the cultural and intellectual concepts marking early English modernity. To enable the search for an innovative bottom-up method for identifying concepts in discourse, the project developed an automated processor for generating concepts from a corpus of early modern English discourse, Early English Books Online-Text Creation Partnership. The resulting process, concept modelling,  generates billions of ‘quads’ (four lemmas that co-occur within a span of 100 tokens of text) (Mehl, 2022). The immediate problem was how to interpret the strong association between lemmas in a quad; work thus far has focused on developing a theory of discursive meaning and using analytical techniques to map conceptual meaning onto the quads. Although close semantic-pragmatic analysis is a thorough and nuanced approach to  identifying the structure of concepts, it is time-consuming and impractical when the datasets are so large.  Distant reading, using lexical co-occurrence data and visualisation techniques, has the potential to help us see patterns in the data, to form hypotheses about conceptual structures, and thus dramatically enrich the close semantic-pragmatic inspection of quads. In this talk, we zoom back out from the inspection of manageable sets of quads (as explored in Fitzmaurice 2021, 2022) to tackle quad constellations–namely all of the quads associated with a particular node word–to explore how data visualisation techniques might assist in revealing their conceptual meaning.

References:

Fitzmaurice, Susan. 2021.  Looking for Concepts in Early Modern English: Hypothesis building and the uses of encyclopaedic knowledge and pragmatic work. Journal of Historical Pragmatics. 22:2 (2021) 282-300.

Fitzmaurice, Susan. 2022. From Constellations to Discursive Concepts; or: The historical pragmatic construction of meaning in Early Modern English. Transactions of the Philological Society 120:3 (2022) 489-506.

Mehl, Seth. 2022. Discursive quads: New kinds of lexical co-occurrence data with linguistic concept modelling. Transactions of the Philological Society 120:3 (2022) 474-488.

This talk will looks as some of the challenges and discoveries associated with using social media (Twitter/X) as a source for examining dialect variation and change. Messages from social media constitute a fantastic source of evidence for linguistic diversity, much of which is otherwise inaccessible, allowing us to see patterns of linguistic variation across thousands, sometimes even millions, of people. We will look at some of the results of the Tweetolectology project, which has been mapping linguistic variation across various countries, with case studies from Welsh, English and Haitian Creole framed around key research or methodological issues of broad general interest

The Ethiopian Linguistic Area at the Horn of Africa comprises languages of different branches of the Afroasiatic macro-family and, to a lesser extent, of the disputed Nilo-Saharan phylum. The linguistic area was established by Charles A. Ferguson as early as the mid-seventies (Ferguson 1976). The boundaries of the contact zone and the criteria used to define it have ever since been disputed, but it is commonly agreed that languages spoken in the Ethiopian highlands show many signs of convergence. After a brief introduction to the languages in Ethiopia (classification, sociolinguistic situation and interesting typological features), some of the defining criteria that have been proposed (and questioned or refuted) by different authors are critically assessed. While most authors have concentrated on features of the phonology and morphosyntax, I will discuss, in the central part of my presentation, how language contact has influenced the organization of the lexicon of languages in the Ethiopian highlands. I will start with a review of Richard Hayward’s influential work (1991; 2000) and then discuss recent and ongoing research on shared polysemy and shared lexicalization patterns in selected semantic fields. My focus will be on two topics: (i) the use of verb ‘know’ as a means to express the experiental perfect, i.e. ‘have (n)ever verb-ed’, and (ii) the similarities in the inventories of interjections, especially for animal-directed commands, across languages. My talk is based on my own field research on Kambaata (Cushitic) and on published data for other Ethiopian languages.

References
Ferguson, Charles A. 1976. The Ethiopian language area. In M. Lionel Bender, J. D. Bowen, R. L. Cooper & C. A. Ferguson (eds.), Language in Ethiopia, 63–76. London: Oxford University Press.

Hayward, Richard J. 1991. À propos patterns of lexicalization in the Ethiopian language area. In Daniela Mendel & Ulrike Claudi (eds.), Ägypten im afro-orientalischen Kontext. Aufsätze zur Archäologie, Geschichte und Sprache eines unbegrenzten Raumes. Gedenkschrift Peter Behrens (Special Issue of Afrikanistische Arbeitspapiere), 139–156. Cologne: University of Cologne, Institute of African Studies.

Hayward, Richard J. 2000. Is there a metric for convergence? In Colin Renfrew, April McMahon & R. L. Trask (eds.), Time depth in historical linguistics, vol. 2: Papers in the prehistory of languages, 621–640. Cambridge: The McDonald Institute for Archaeological Research.

Iambic metrical systems, which have weak-strong feet in contrast to trochaic strong-weak ones, are rare. They represent under 10% of the World Atlas of Language Structures sample and are concentrated in the Americas (Goedemans & van der Hulst 2013). They are generally under-described, and little diachronic research has been conducted on iambic systems. Algonquian, a family of languages stretching over much of northern North America, is one of very few families with a large number of iambic daughters. We provide evidence from this family that can refine our typology of iambic languages. After arguing that Proto-Algonquian was iambic, we investigate how Algonquian languages behave in ways at odds with typological claims about iambic systems. First, iambic lengthening is claimed to be characteristic of iambic systems, but few Algonquian languages have it, while diametrically opposed processes like iambic shortening and change toward typologically dispreferred foot structures are widespread. Second, iambic systems are associated with duration as a cue to prominence while pitch and intensity are typically associated with trochaic systems. However, in Algonquian pitch is a common cue to prominence, which helps motivate the fact that numerous daughters have undergone tonogenesis. Algonquian metrical phonology, diachronic and synchronic, can sharpen our typology of iambic languages in general.

The purpose of this talk is to celebrate the scholarship of Dr Margaret (Meg) Laing (1953–2023). Meg spent most of her academic career as a Research Fellow in the University of Edinburgh. She and I carried out our activities in a research unit (currently the Angus McIntosh Centre for Historical Linguistics) dedicated to carrying on work begun by Angus McIntosh and M.L. Samuels relating to the dialectology of written Middle English. This had resulted in A Linguistic Atlas of Late Mediaeval English 1350–1450 (LALME), published first in 1986. Meg had contributed to LALME through her PhD thesis on the mediaeval dialectology of Lincolnshire, acquiring knowledge and skills on which she built for her subsequent work in Historical Dialectology. This was centred on A Linguistic Atlas of Early Middle English 1150–1325 (LAEME), published (2008, 2013). This was followed up by the Corpus of Narrative Etymologies (CoNE, 2013), compiled in collaboration with Professor Roger Lass (University of Cape Town) and Dr Rhona Alcorn (University of Edinburgh). CoNE is aimed at explaining the etymology of every orthographic form recorded in the lexico–grammatically tagged corpus which forms the data-base for LAEME. These large-scale works served as the well for a large number of articles dealing with problems of scripts, orthography and textual transmission in manuscripts containing Early Middle English. With Roger Lass, Meg developed the concept of Litteral and Potestatic Substitution Sets for interpreting complex orthographic systems of scribal languages. Meg and Professor Michael Benskin (University of Oslo) also edited a revision of LALME to create a freely available on-line version (eLALME, 2013). She was also an enthusiastic and valued consultant to the Oxford English Dictionary, which work she carried on after her retirement in 2013.

I will outline the concepts and methodologies of LALME, LAEME and CoNE, especially how Meg developed the methodologies for the Historical Dialectology of a mediaeval vernacular in the making of LAEME and for interpretation of complex orthographic systems. I will offer two examples of how the LAEME data might be further investigated and exploited.

North-Eastern Neo-Aramaic (NENA) is the largest remaining branch of the Aramaic language family (Semitic/Afroasiatic), although most surviving dialects are now endangered. The immense dialectal diversity, documented with increasing urgency in the last three decades, allows one to trace the emergence of new grammatical forms at different stages of grammaticalization.

One of these involves a prefix (QAM-, dialectal variants qam-, k?m-, q?m-, g?m- etc.) which apparently converts a present subjunctive into a past perfective. It exists alongside another, older past perfective form, but in many dialects the newer form with QAM- has become specialized for pronominal object indexing (the older form usually cannot take a full set of pronominal object suffixes).

Scholars since the 19th century have offered a variety of theories as to the origin of this strange tense-aspect marker. One theory, which I will develop, is that it originates in the NENA verb-form qay?m ‘he gets up’, used as a historic present (i.e. in lively oral narrative). This was proposed by Pennacchietti (1994, 1997), supported by historical, dialectal and cross-linguistic evidence, including the Catalan go-past vaig cantar ‘I sang’. Recently Fassberg (2015) has argued against this theory and proposed that the k?m- variant originates in a metanalysis of the indicative prefix k- (also in a historic present function) before the initial m- of certain verbs, with epenthetic vowel (i.e. k?-m- > k?m-).

The present paper will evaluate the theories, as well as show that the criticisms of Pennacchietti’s proposal are not well-founded. It will further bring new NENA data (both historical textual evidence and newly documented dialectal data) and cross-linguistic evidence from Romance and other language families which support a grammaticalization path from qay?m ‘he gets up’ as part of a serial verb construction used as a narrative technique. It will also show how verbs cognate to qay?m have been grammaticalized as discourse markers in other Semitic languages, most likely through a similar scenario (though with a different end result). The Neo-Aramaic case throws light upon how such grammaticalization paths work, as well as the particular circumstances under which the gram could be recycled for an additional use in facilitating object-indexing on the verb.

References
Fassberg, Steven E. 2015.‘The Origin of the Periphrastic Preterite k?m/qam-qa??lle in North-Eastern Neo-Aramaic.’ In Geoffrey Khan and Lidia Napiorkowska (eds), Neo-Aramaic and its Linguistic Context. Piscataway, NJ: Gorgias Press. 172–186.
Pennacchietti, Fabrizio A. 1994. ‘I preverbi del passato in semitici’. In V. Brugnatelli (ed.), Sem Cam Iafet. Atti della 7a Giornata di Studi Camito-Semitici e Indoeuropei (Milano, 1 giugno 1993), Milano 1994: Centro Studi Camito-Semitici. 133–150.
Pennacchietti, Fabrizio A. 1997. ‘On the etymology of the Neo-Aramaic particle qam/kim’ [in Hebrew]. In Moshe Bar-Asher (ed.), Gideon Goldenberg Festschrift (Massorot: Studies in Language Traditions and Jewish Languages 9–11). Jerusalem: Magnes, 475–82.

How do we adapt our grammar to communicate social detail? Do all working class people have a local dialect or are we able to adapt our language to communicate more subtle things about ourselves and our utterances? This talk seeks to answer these questions. Using data from an ethnographic study of high school girls, it demonstrates that we use grammatical variation to communicate alignments and to construct our social style or social identity. However, how and why we adapt our language is governed by our place in the social order. Engaging real life examples will show that grammatical variants, like negative concord (e.g. I didn’t do nothing to mean ‘I didn’t do anything’), have multiple interactional functions, but different people are more or less able (and more or less willing) to make use of these functions. This talk argues that, to truly understand how language works, we need to examine how three types of meaning – referential, pragmatic and social – interact. Understanding this spectrum of meaning (and its role in language acquisition) has implications for linguistic theory, but it has educational implications too, given that educational policy frequently asks young people to change their language style on demand.