Language change in its socio-historical context

On 16 November, a panel of three Early Career Researchers will convene to present their research on language change in its socio-historical context; the presentations will be followed by a round table discussion chaired by Ranjan Sen (Sheffield). The speakers are: Christien Wallis, Claire Childs, and George Bailey; abstracts of their talks can be found below.

The presentations and round table discussion will take place at the University of Sheffield, Humanities Research Institute, 4.15pm.


Standardisation and the Old English Subjunctive

by Chrstine Wallis (University of Sheffield)

Traditional accounts of Old English (‘OE’) (Campbell, 1959; Hogg, 1992) often focus on early or otherwise dialectally marked manuscript texts for evidence of the history of the language.  Such manuscripts are chosen as the basis of this evidence because they are closest to the original author’s or translator’s work, and are felt to reflect ‘real’ OE in a way that later copies do not (Miller, 1890: v-vi).  Where more than one manuscript of a text exists, those which diverge most from the most conservative versions are rarely discussed in detail in general histories.

This paper presents an alternative way of viewing the development of OE, through the more sociolinguistically-orientated lens of scribal copying.  A text with several surviving manuscript witnesses allows us to see what linguistic forms were deemed acceptable to individual language users/writers (i.e. features which were copied literatim), and which were not (i.e. those emended or updated by later copyists) (cf. Laing, 2004).

The OE translation of Bede’s Historia ecclesiastica is one such text, surviving in four main copies, whose scribes diverge to varying degrees from the Mercian dialectal character of the translator’s (now lost) original text.  The paper focuses on one case study, that of plural, preterite subjunctives, which in the earliest manuscripts commonly appear with denasalisation (e.g. hie wolde instead of hie wolden ‘they wanted’).  A range of strategies is used by the scribes studied and this talk will show examples of the different responses of the Bede scribes to this feature when they copied the text.

This paper shows how evidence not normally considered in larger histories of the language can usefully be brought to bear on ideas of standardisation in the pre-Conquest period.  In the absence of direct metalinguistic comment, the actions and decisions of copyists and correctors have much to tell us about attitudes to correctness and linguistic norms in the period.


The present-day interaction of longitudinal changes: Stative possession and negation

by Claire Childs (University of York)

This talk will focus on the modern-day variation between stative possessive HAVE and HAVE GOT in negative contexts, which is the result of two intersecting historical changes. Firstly, DO-support as a means of expressing negation arose in English around the 15th century, but not immediately in stative possessive contexts (Warner 2005). Secondly, stative HAVE GOT came to be used as an alternative to HAVE in around the 16th century (Lorenz 2016: 489). It was not until the 19th century that DO-support became possible with stative HAVE (Hundt 2015: 70). Contemporary studies of British English have indicated that HAVE GOT is becoming increasingly used for the expression of stative possession in affirmative contexts (Tagliamonte 2003), but DO-support is also thought to be rising (Trudgill et al. 2002: 6). With these two tendencies seemingly pulling in different directions – since HAVE GOT is incompatible with DO-support (*I don’t have got any money) – how does this manifest itself in present-day British English?

To answer this question, I will present initial findings from a quantitative variationist analysis of HAVE (GOT) in negative contexts in British English, based on a 2.5-million-word sample of conversational speech from the British National Corpus 2014 (Love et al. 2017). The results reveal that while HAVEN’T GOT was the favoured way of negating a stative among speakers aged 60+, this decreases in apparent-time to the extent that DON’T HAVE becomes the majority form among younger speakers. Although British English is thought to be more variable in terms of the syntactic status of HAVE – i.e. it can behave like an auxiliary or a lexical verb – HAVE is actually rarely contracted and thus has the syntactic properties of a lexical verb, just as in Canadian English (D’Arcy 2015). My findings allow two independent observations of subject-type constraints on contraction (McElhinny 1993) and stative possession variation (Tagliamonte et al. 2010) from the literature to be reconciled. More broadly, my analysis shows how insights gained from separate analyses of single linguistic variables can be explained as part of a larger system within the grammar.


When sound change isn’t led by social change: The case of Northern English (ng)

by George Bailey (University of Manchester)

Incorporating sociolinguistic evaluation into explanatory models of language variation and change has become increasingly popular in recent years (e.g. Eckert 2000; Zhang 2005; Podesva et al. 2015), dating back to Labov’s (1963) influential study of Martha’s Vineyard. However, not all objects of linguistic variation can accrue social meaning (Eckert & Labov 2017), and there remain a number of apparent limitations relating to its role in the incrementation and propagation of sound change (Bermúdez-Otero forthcoming). This paper bears directly on this debate by reporting on a recent change in Northern English /ŋɡ/ clusters, which sees increasing post-nasal [ɡ]-presence in words like wrong and hang when in pre-pausal position (Bailey 2018). Post-nasal /ɡ/-deletion progressed along a systematic pathway of change throughout the Modern English period, following an ordered set of stages laid out by the life cycle of phonological processes (Bermúdez-Otero & Trousdale 2012). However, this new pre-pausal behaviour does not represent the next natural stage along the same pathway of change laid out by the life cycle, but is rather an entirely separate and unpredicted innovation. As such, it is amenable to an analysis in which external factors – such as sociolinguistic evaluation – play a central role.

Independent evidence from a matched-guise task reveals another source of apparent time change: the indexical strength of [ŋɡ] as a feature of northern dialects is increasing over time. However, this does not translate to uniform evaluation, with no evidence of a shared evaluative norm among these subjects. Furthermore, despite the change in production being restricted to pre-pausal contexts, this change in the social meaning of (ng) is not concentrated on any particular environment, suggesting that the two are operating at different levels of granularity and that there is no causal link between them. Consequently, these results cast further doubt on the extent to which social meaning is involved in producing macroscopic patterns of sound change.

The diachrony of initial consonant loss in Cape York Peninsula (Australia)

by Jean-Christophe Verstraete (University of Leuven)

This paper revisits the diachrony of initial consonant loss, a type of sound change that is found in several areas in Australia (Hale 1976a, Alpher 1976, Blevins 2001), but is rare from a world-wide perspective (Blevins 2007). So far, the literature has mainly analysed initial loss as the outcome of a gradual process of initial weaking, caused by a shift of stress away from the initial syllable (Alpher 1976, Blevins & Marmion 1994). In this paper, I use data from a set of eight Paman (Pama-Nyungan) languages of Cape York Peninsula (Australia), which illustrate not just loss of initial consonants, but also initial consonant lenition and the loss of entire initial syllables. Using these data, I argue that (i) the classic model of gradual initial weakening needs to be supplemented with more abrupt mechanisms, specifically analogy-driven loss based on synchronic alternations, and contact-induced loss, and (ii) the causal link with stress shift needs to be refined, and in some languages loss of initial consonants in part of the lexicon may itself cause changes in the stress system.

Cape York Peninsula, in Australia’s northeast, has several ‘hotspots’ of initial loss (Alpher 1976, Sutton 1976), e.g. in the north and on the central east coast. This paper focuses on the eastern hotspot in the Princess Charlotte Bay area, specifically eight languages from three different subgroups of Paman, viz. Middle Paman (Umpithamu, Yintyingka, Umpila), Lamalamic (Lamalama, Umbuygamu, Rimanggudinhma) and Thaypanic (Kuku Thaypan, Aghu Tharrnggala). The Middle Paman languages show a combination of retention, lenition and loss of initial consonants, as shown in (1) for Umpithamu, while the Lamalamic and Thaypanic languages show systematic loss of initial consonants, as shown in (2a) for Umbuygamu, and/or of entire initial syllables, as shown in (2b) for Lamalama (Proto-Paman reconstructions from Hale 1976b).

(1) a. kuwa ‘west’ ~ *kuwa b. ya’u ‘foot’ ~ *caru c. aangkal ‘shoulder’ ~ *paangkal
(2) a. agarr ‘flesh’ ~ *pangkarr b. karr ‘flesh’ ~ *pangkarr

When confronted with these data, the classic model of gradual initial weakening only works for Umpithamu, which shows phonologically systematic patterning of initial consonant loss, lenition and retention. The other languages deviate in two ways, suggesting two further pathways to initial loss. In Umpila and Yintyingka, lenition, loss and retention of initial consonants do not show any phonological systematicity: their patterning can only be explained in terms of contact-induced change, specifically borrowing from neighbouring languages that do have gradual initial weaking. In Lamalamic and Thaypanic, loss of initial consonants is complete, but initial vowels are retained or lost. Against expectations in the literature (e.g. Blevins & Garrett 1998, Sommer 1976), initial vowels in these languages do not show any signs of weakening, but are in fact in a strong position, with a large number of contrasts. Instead, initial loss in these languages can be related to specific phrasal structures that induce construction-specific loss of the initial vowel of the first lexeme, which creates a regular synchronic alternation between forms with and without an initial vowel, and can serve as an analogical model driving the systematic loss of initial vowels.

The classic model of stress shift towards the second syllable is equally problematic when confronted with these data. Lamalamic languages show the predicted pattern of stress on the first consonant-initial syllable in the root, but crucially Middle Paman languages do not. In Umpithamu, for instance, stress placement can be generalized in terms of a right-aligned system of moraic trochees, which crucially allows initial stress for some types of vowel-initial roots. This suggests that the classic model of linear stress shift causing initial loss does not seem to work. Instead, initial consonant loss in part of the lexicon, as observed in Umpithamu, may itself be a crucial factor leading to a shift in stress alignment from left to right edge (compare Lahiri 2015 on reanalysis driving changes in stress alignment in the history of English). This still leaves the root causes of initial loss to be addressed, but at least it shows that stress shifts in these languages are not always simple linear shifts, and that they are not necessarily the cause of patterns of initial loss but can also be an effect.


This paper will be read at the Philological Society meeting at the University of London, Senate House (Room G3), London WC1E 7HU, on Friday, 19 October, 4.15pm.


References:

Alpher, Barry. 1976. Some linguistic innovations in Cape York and their sociocultural correlates. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 84-101.

Blevins, Juliette & Doug Marmion. 1994. Nhanta historical phonology. Australian Journal of Linguistics 14: 193-216.

Blevins, Juliette. 2001. Where have all the onsets gone? Initial consonant loss in Australian Aboriginal languages. In Simpson, Nash, Laughren, Austin & Alpher, eds. Forty years on: Ken Hale and Australian languages. Canberra: Pacific Linguistics. 481-492.

Blevins, Juliette. 2007. Endangered sound patterns: Three perspectives on theory and description. Language Documentation and Conservation 1: 1-16.

Blevins, Juliette & Andrew Garrett. 1998. The origins of consonant-vowel metathesis. Language 74: 508-556.

Hale, Ken. 1976a. Phonological developments in particular Northern Paman languages. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 7-40.

Hale, Ken. 1976b. Wik reflections of Middle Paman phonology. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 50-60.

Jolly, Lesley. 1989. Aghu Tharrnggala. A Language of the Princess Charlotte Bay Region of Cape York Peninsula. BA Hons thesis, UQ.

Lahiri, Aditi. 2015. Changes in word prosody: Stress and quantity. In Honeybone & Salmons, eds. The Oxford handbook of historical phonology. Oxford: OUP. 219-244.

Rigsby, Bruce. 1976. Kuku-Thaypan descriptive and historical phonology. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 68-77.

Sommer, Bruce. 1976. A problem of metathesis. In Sutton, ed. Languages of Cape York. Canberraː AIAS. 139-143.

Sutton, Peter. 1976. The diversity of initial dropping languages in southern Cape York. In Sutton, ed. Languages of Cape York. Canberra: AIAS. 102-123.

Thompson, David. 1988. Lockhart River ‘Sand Beach’ Language. An Outline of Kuuku Ya’u and Umpila. Darwin: SIL.

Parallel evolution of relative clauses in Indo-European

by Nikolas Gisborne & Robert Truswell (University of Edinburgh)

The Indo-European indefinite/interrogative pronouns *k wi-/k wo- are the source of relative pronouns in several daughter languages, including varieties of Romance, Slavic, and Germanic among others. These pronouns did not head relative clauses in PIE, and so their presence in the relative clauses of the daughter languages is a result of processes of historical evolution which have recurred in different subfamilies. However, this recurring parallel process is by and large confined to Indo-European. Comrie (1998) claims instead that the interrogative relative pronoun strategy is a European areal phenomenon, because it is also found in neighbouring languages such as Hungarian and Georgian. However, there is ample evidence that endogenous innovation gives rise to interrogative relativizers in English and several other Indo-European languages. This suggests that such endogenous processes may be wholly or partly responsible for the emergence of interrogative relativizers across Indo-European. However, these processes are not the same across daughter languages: there appear to be several meandering paths from the same start point to similar endpoints.

In this talk, we establish a framework for describing both the parallel diachronic pathways and the dimensions of variation around those pathways. The broad outline of the parallel developments can be established by combining a typological perspective on Indo-European indefinite/interrogatives with results from Haspelmath (1997) on the relationship between interrogative and indefinite pronouns, from Belyaev & Haug (2014) on the typology of correlatives and conditionals, and from Haudry (1973) on the relationship between correlatives and headed relatives. At the same time, the behaviour of individual lexical items within this typological space is less predictable, accounting for the variation around this broad pathway.

This paper was read at the Annual General Meeting of the Philological Society in Oxford, Somerville College, on Saturday, 16 June, 4.15pm.

An audio recording and screencast of the paper can be found below and on the Society’s YouTube channel. A PDF version of the presentations is also available.

Bashkardi – a language by convergence?

by Agnes Korn (CNRS, Paris)

Bashkardi, spoken in Southern Iran inland from the Strait of Hormuz, is a very little known language. The dialects differ on all levels of grammar and show strong influence of Persian. This talk will present some salient features of the phonology and morphology of Bashkardi and compare them to other Iranian languages to shed light on the development of the grammatical structures. I will examine the hypothesis that Bashkardi is not a genetic entity, but a group of Iranian dialects of diverse origin which developed common traits by a process of convergence, having found themselves next to each other in a small region that remains remote even today.

This paper will be read at the Philological Society meeting in London, SOAS, Brunei Gallery building, first floor, room B104, on Friday, 11 May, 4.15pm.

Mechanising historical phonology

by Patrick Sims-Williams (Aberystwyth University)

The neogrammarian approach to historical phonology involves propounding sound-change laws and explaining exceptions by means such as sub-laws, rearranging the relative chronology, and appeal to special factors such as analogy, borrowing, incomplete lexical diffusion, and sporadic phenomena like metathesis. Progress is mostly made manually, but in the second half of the 20th century some linguists looked forward to the ‘triumph of the electronic neogrammarian’. Although this hasn’t been realized yet, I’ll argue that there are opportunities to make important advances with comparatively little effort.


This paper will be read at the Philological Society meeting in Cambridge, Selwyn College, Diamond Room, on Saturday, 10 March, 4.15pm.

Language, learning and usage-based theory: tackling nominal and verbal morphology in Slavic

by Dagmar Divjak (University of Sheffield)

Usage-based theories of language are built on the assumption that our ability to extract and entrench the distributional patterns available in the input enables learners to build a grammar from the ground up. This circumvents the needs for an innate universal grammar. But it does not tell us which patterns are relevant. And it remains customary for linguists to approach the data using linguistic categories—such as Case or Tense, Aspect and Mood—categories that were never intended to reflect the workings of the mind. In this talk, I will argue that it might be better to take the input as starting point and derive categories that resemble those native speakers might derive. Models from Learning Theory can help with this. I will present two case studies that capitalize on a merger of cognitive linguistics and cognitive psychology, and aim to infuse Usage-Based linguistics with insights from Learning Theory … with a little help from computational engineering.

The first case study uses insights from Learning Theory to challenge the idea that theoretical linguistic constructs such as tense, aspect and mood (TAM) predict best how native speakers of Russian read sentences containing verbs meaning to try in real time. Discrimination learning, as implemented in the NDL algorithm, proposes simple 3-letter usage-patterns and predicts the time it takes subjects to read and integrate these verbs into a sentence significantly better than all TAM markers combined.

table_divjak

Contrary to what mainstream (psycho)linguistic models assume, speakers do not (and do not need to) analyse verb forms in terms of abstract linguistic concepts such as tense, aspect and mood when they process language. Instead, they can rely on simple letter sequences that are linked directly to an experience and embed crucial information about that experience (i.e., is it over, ongoing, or coming up; was it something that they completed, or simply did for a while; was it an order). This demonstrates that honouring parsimony (naivety and simplicity) in the structures that are hypothesized to exist, and in the way in which behaviour is explained, is a powerful research stance, in particular for designing cognitively realistic accounts of language knowledge and representation.

The second case study demonstrates how biologically inspired machine learning techniques can pinpoint the essence of native speaker intuitions. Polish boasts fascinating examples of seemingly unmotivated allomorphy, and the genitive singular of masculine inanimate nouns (which can be -a or -u) is its prime example. Criteria for choice have been proposed that are semantic, morphological or phonological in nature, but most of these are unreliable, yielding conflicting predictions (Dąbrowska 2005). Furthermore, although -u occurs with at least twice as many nouns, -a is the default ending for new words entering the language. The NDL algorithm, that implements discrimination learning, predicts the choice between -a and -u better using simple sequences of 3 letters (letter triplets or trigraphs) than models running on richly annotated corpus data. In addition, it explains the unexpected preference of -a as genitive ending for new words in terms of the learnability of words taking the -a ending, their phonological predictability and their contextual (semantic) typicality.

On their own, linguists and psychologists would have approached these questions rather differently and, from within their disciplinary cages, would have arrived at answers that would necessarily have remained partial. Integrative interdisciplinarity, on the other hand, relies on a simultaneous, interspersed methodological endeavour to arrive at more encompassing answers that combine depth of analysis with breadth of explanation. It presupposes mutually complementary theories, shared testable hypotheses as well as compatibility of research methodologies. But what wins the game is a good dose of willingness to question your customary ways of doing things.


A video of the talk can be found below.

This paper was read at the Philological Society meeting in London, SOAS Main Building, Room 116, on Friday, 9 February, 4.15pm.

Syntactic microvariation in Romance – bridging synchrony and diachrony: the case of SI

by Sam Wolfe (University of Oxford)

Major syntactic differences between the medieval Romance languages and their modern counterparts have been noted for well over a century (Tobler 1875; Diez 1882; Thurneysen 1892; Meyer-Lübke 1889), with a body of more recent work highlighting important synchronic variation amongst the medieval languages (Vance, Donaldson & Steiner 2009; Wolfe 2015, forthcoming), and diachronic variation observable in texts from different stages of the medieval period (Ledgeway 2009; Labelle & Hirschbühler 2017; Galves forthcoming). In this talk, I focus on a particular aspect of the syntax of Medieval Romance: the grammar of the particle SI, which abounds across the early textual records, but eludes a satisfying analysis.

Based on a new hand-annotated corpus of seven Old French texts, I show that the numerous and frequently contradictory claims in the literature regarding SI (Marchello-Nizia 1985; Reenen & Schøsler 2000; Ledgeway 2008) can often be reconciled under an account where its formal characterisation, discourse-pragmatic value, and interaction with other areas of core clausal syntax varies markedly, both synchronically and diachronically, within the period conventionally referred to as ‘Old French’. Specifically, I sketch a grammaticalisation pathway where SI becomes progressively bleached through a process of upwards reanalysis (Roberts & Roussou 2002). This entails a change from SI (>SIC) as an adverbial encoding temporal succession, to topic continuity marker (Fleischman 2000), then two distinct expletive stages, where SI acts as a last-resort mechanism to satisfy the Verb Second constraint. The core empirical observation is that there is large-scale variation between SI in 12th-century and 13th-century texts and, furthermore, small-scale variation in the syntax of SI across texts which are conventionally considered contemporaneous.

In the second part of the talk I bring in data from a range of Medieval Italo-Romance varieties, showing that SI in Sicilian, Florentine, Piedmontese and Venetian texts mirrors almost exactly the distribution of SI in 12th-century French, but does not show the distributional properties of the highly grammaticalised element found in 13th-century French.

The core intuition behind the analysis of Medieval Romance SI is that the element in question can occupy distinct positions within an articulated left periphery (on which see Rizzi 1997, Benincà & Poletto 2004 and Ledgeway 2010) during different stages of the grammaticalisation process. Furthermore, throughout its history, SI cannot be understood in isolation from ongoing changes in the Medieval Romance Verb Second property and its correlates (Wolfe 2016), but may also have a previously overlooked role in shaping a number of the morphosyntactic isoglosses observable within Romance-speaking Europe today. In particular, I suggest that differences in the syntax of Old French SI and its Old Italo-Romance counterparts may account for major contemporary Italo- vs. Gallo-Romance differences in the syntax of topicalisation, focus and the null subject property.

Overall, although SI may seem like a small and parochial area of Medieval Romance syntax, its synchronic and diachronic significance for an understanding of the evolution of Romance grammar cannot be underestimated.


References

Fleischman, Suzanne. 2000. Methodologies and Ideologies in Historical Linguistics: On Working with Older Languages. In Susan C. Herring, Pieter Th. van Reenen & Lene Schøsler (eds.), Textual parameters in older languages. Amsterdam; Philadelphia, Pa.: John Benjamins. 33–58.

Galves, Charlotte. Forthcoming. Partial V2 in Classical Portuguese. In Theresa Biberauer, Sam Wolfe & Rebecca Woods (eds.), Rethinking Verb Second. (Rethinking Comparative Syntax). Oxford: Oxford University Press.

Labelle, Marie & Paul Hirschbühler. 2017. Leftward Stylistic Displacement in Medieval French. In Eric Mathieu & Robert Truswell (eds.), Micro-change and Macro-change in Diachronic Syntax. Oxford: Oxford University Press.

Ledgeway, Adam. 2008. Satisfying V2 in early Romance: Merge vs. Move. Journal of Linguistics 44(2).

Marchello-Nizia, Christiane. 1985. Dire le vrai: L’adverbe «si» en français médieval: Essai de linguistique historique. (Publications Romanes et Françaises CLXVIII). Geneva: Droz.

Roberts, Ian & Anna Roussou. 2002. Syntactic change a minimalist approach to grammaticalization. Cambridge: Cambridge University Press.

Vance, Barbara, Bryan Donaldson & B. Devan Steiner. 2009. V2 loss in Old French and Old Occitan: The role of fronted clauses. In Sonia Colina, Antxon Olarrea & Ana Maria Carvalho (eds.), Romance Linguistics 2009. Selected papers from the 39th Linguistic Symposium on Romance Languages (LSRL), Tuscon, Arizona. (Current Issues in Linguistic Theory 315). Amsterdam: John Benjamins. 301–320.

Wolfe, Sam. Forthcoming. Redefining the V2 Typology: The View from Medieval Romance and Beyond. (Ed.) Christine M. Salvesen. Linguistic Variation (Special Issue: A Micro-Perspective on V2 in Germanic and Romance).

Wolfe, Sam. 2015. The Old Sardinian Condaghes. A Syntactic Study. Transactions of the Philological Society 113(2). 177–205.


A video of the talk can be found below. The accompanying handout is available here.

This paper was read at the Philological Society meeting in London, SOAS Main Building, Room 116, on Friday, 12 January, 4.15pm.