‘He was a good hammer, was he’: Gender as Marker for South-Western Dialects of English. A Corpus-based Study from a Diachronic Perspective

This chapter investigates assumptions that the gender system peculiar to present-day Southwest English might have its origins in similar patterns in that area in Middle English. The present-day dialect uses masculine pronouns as the general reference for most nouns denoting inanimate and countable referents, so that it is not the default gender as in the standard. On the basis of all the textual files specifically localised as Southwest in the relevant subsections of LAEME, the Helsinki Corpus of English Texts (HC) and the Middle English Grammar Corpus (MEG-C), the author investigates whether the seeds of these systems might already have been present in the ME ancestors of those dialects, but concludes that this is not the case – in the Middle English Southwest texts, it can already be considered as the default gender for all nouns denoting non-living things (barring a small number of exceptions discussed in detail). What this investigation ultimately demonstrates is that traditional dialects are not living fossils, and have had their own share of extra-linguistic circumstances to affect them in their long histories.

A Parsed Linguistic Atlas of Early Middle English

This chapter reports on the construction of a new resource, the Parsed Linguistic Atlas of Early Middle English (PLAEME). Prose is underrepresented in the period 1250-1350, which is why this period is also underrepresented in the Penn Parsed Corpora of Historical English (PPCHE). This data gap is unfortunate, as we know that the period is important for morphosyntactic change. PLAEME addresses that data gap by transforming material from the Linguistic Atlas of Early Middle English (LAEME) into the same format as the PPCHE. The chapter present a detailed account of its construction, as well as three case studies replicating three recent studies of Middle English syntax: the establishment of not as the expression of sentential negation (Ecay and Tamminga 2017), the fixing of the syntax of the dative alternation (Bacovcin 2017), and the introduction of argumental headed wh-relative clauses (Gisborne and Truswell 2017). These case studies show that PLAEME allows these changes to be charted in much greater detail, and hence demonstrates how PLAEME fills an important data gap.

Old and Middle English Spellings for OE hw-, with Special Reference to the ‘qu-’ Type: In Celebration of LAEME, (e)LALME, LAOS and CoNE: In Memoriam Angus McIntosh

This chapter demonstrates how the four main electronic resources created in the same tradition as A Linguistic Atlas of Late Mediæval English (LALME), i.e. LAEME, LALME itself (and its electronic version eLALME), A Linguistic Atlas of Older Scots (LAOS) and A Corpus of Narrative Etymologies from Proto-Old English to Early Middle English and accompanying Corpus of Changes (CoNE) can be used in tandem to support an investigation into the initial wh-cluster in words such as when, where, what, who, which. No fewer than 57 different spellings are found for this cluster, from the earliest attested Old English to ca 1500. The authors show how LAEME, eLALME, and LAOS provide the data that allow this spelling variation to be analysed as reflecting various scribal choices, whether determined by orthographic variation (including traditional contextual rules for the use of <v> or <u>), phonological variation, geographical variation, and/or diachronic variation. The final section showcases CoNE, and reconstructs a diachronic account on the basis of these spellings, revealing a coherent, if extremely complex, picture of lenitions, fortitions, and reversals.

Early Spelling Evidence for Scots L-vocalisation: A Corpus-based Approach

This chapter showcases the From Inglis to Scots (FITS) Project database, which comprises texts from the Linguistic Atlas of Older Scots (LAOS), of the period 1380-1500. This new resource for historical dialectology makes it possible to test earlier assumptions about phonological changes that are characteristic of Scots and not shared with Southern English. This chapter uses LAOS to test the claim that L-vocalisation, which entails the loss of coda-/l/ following short back vowels with concomitant vocalic lengthening or diphthongisation (as in OE full > OSc fow), was completed by the beginning of the sixteenth century. Based on attestations of <l>-less forms and reverse spellings, including /l/~ø alternations in borrowed items from (Norman) French (as in realme~reaume ‘realm’), the chapter maps the spread of <l> loss in different phonological contexts over time and space, and presents evidence of <l> loss in less than 1% of relevant environments. The final position of <l> is an important locus, but there is no evidence of a spread.

The Development of Old English eo/ēo and the Systematicity of Middle English Spelling

This chapter uses a new resource, the Middle English Grammar Corpus (MEG-C), a corpus of 14th and 15th Century English texts, to answer an old question: it is possible to find traces of a systematic distinction between the reflexes of Old English e/ē and eo/ēo in Middle English? An investigation into the spelling variation found in 27 lexical items that contain a vowel representing Old English eo/ēo as well as the equivalent Old Norse element jó throws up a wide range of spellings, the vast majority of which show <e>/<ee>. Spellings that might suggest a rounded pronunciation are also fairly robustly present, however, particularly <eo>, with the Southwest Midlands as its core area. The second part of the investigation retrieves all words that were spelled with the digraph <eo>. The vast majority of these turn out to be reflexes of Old English eo/ēo, and almost all of them are localized to the Southwest Midlands. They occur either as reflexes of OE y/ȳ, or in unstressed syllables, or in words where <eo> follows <w> – three groups for which a rounded pronunciation would be plausible.

Historical Dialectology and the Angus McIntosh Legacy

This chapter provides an overview of the historical text corpora and digital repositories hosted by the Angus McIntosh Centre for Historical Linguistics and created by its predecessor, the Institute of Historical Dialectology: A Linguistic Atlas of Late Middle English (LALME), and its remodelled electronic version eLALME; A Linguistic Atlas of Early Middle English (LAEME), A Linguistic Atlas of Older Scots (LAOS) and The Corpus of Narrative Etymologies from Proto-Old English to Early Middle English (CoNE). The chapter also highlights related resources created at the University of Stavanger, most prominently the Middle English Scribal Texts programme (MEST), and its offshoot, The Middle English Grammar Corpus (MEG-C), which provides tagged and annotated diplomatic transcriptions of 410 LALME texts; and the Corpus of Middle English Local Documents (MELD) which comprises transcriptions of over 2000 fifteenth-century documents.

Examining the Evidence for Phonemic Affricates: Middle English /t͡ʃ/, /d͡ʒ/ or [t-ʃ], [d-ʒ]?

Affricates represent an analytic challenge, as a category intermediate between simple stops and a sequence of a stop and a fricative. The paper traces the historical evidence for the development of OE [c], a single segment, to palatal [cj], assibilated [tʃ], the sequence [tʃ], and back to a single segment contour /t͡ʃ/, building on diagnostics like the blocking property of medial clusters versus singletons in resolution in OE verse, alliteration, metrical treatment in terms of syllable weight, data from language acquisition, phonetics in terms of durational properties, the interaction with Middle English sound changes, as well as the early neutralization of the singleton-geminate contrast. Further support comes from spelling, including a possible Celtic origin for OE <cg>, and <ch> spellings in LAEME as evidence supporting Orthographic Remapping of Palatal c. Finally, the author considers the impact of Old French loanwords, where the simplification of affricates in Anglo-Norman is argued to be delayed compared to Central French due to the existence of the sequences [tʃ] and [dʒ] in Middle English.

An East Anglian Poem in a London Manuscript? The Date and Dialect of The Court of Love in Cambridge, Trinity College, MS R.3.19

This chapter investigates the text The Court of Love (Cambridge, Trinity College, MS R.3.19, folios 217r-234r). It was long time held to be a poem by Chaucer, a notion debunked by Skeat as a neo-medieval fabrication dating from the fourth decade of the sixteenth century at the earliest. This chapter presents a new appraisal of the date, which is tentatively pinpointed as the middle of the 15th century, and the localization of the poet, who is localized in East Anglia based on the evidence of rhymes and some spellings. Frequent failures of rhyme point to the linguistic differences between the poet and the scribe, as reconstructed from rhyme, metre and possible “relicts”. The hypothesis that the poet was from East Anglia and the scribe from London is confirmed by evidence from eLALME dot maps, and shows that instances in the poem that were identified by Skeat as “false grammar” are in fact examples of syntax that is true to the poet’s own dialect.

The Development of Old English ǣ: Middle English Spelling Evidence

This chapter presents the range of spellings for the reflexes of ǣ1 and ǣ2 in ME dialects, as found in SED, LAEME and LALME. Old English ǣ appears to have raised early in Middle English, as the dominant spelling is <e(e)>; this is further supported by the fact that <a/ǣ/ea> spellings are more frequent in the early LAEME texts than in the later ones. The spelling variants show geographic variation in Old English, with ǣ1 and ǣ2 appearing to have merged in some dialects but kept apart in others. Their reflexes are not kept apart in spelling in any systematic fashion in any ME dialects, but their distribution is certainly are not random. As the sound-changes affecting the two ǣ’s took some time to reach completion, they overlapped in time with the early stages of the Great Vowel Shift; the author argues that they must be seen as part of that shift, rather than as similar but unrelated changes.

Approaching Transition Scots from a Micro-perspective: The Dunfermline Corpus, 1573–1723

This chapter describes the compilation of a new digital resource for historical dialectology: The Dunfermline Corpus, from the Late Middle Scots period (c. 1550–1700), recently relabelled as “Transition Scots” (Kopaczyk 2013). Transition Scots is the outcome of a contact situation of two written varieties – Scots and Southern English – that are both on the verge of standardisation. A diachronic analysis of five linguistic variables in The Dunfermline Corpus that are known to be distinctive features of Older Scots as opposed to Southern English usage confirm that Anglicisation proceeded at a faster page at supralocal levels than at local levels. Using a sociolinguistic, paleographic micro-approach, the author reconstructs the “community of practice” of town clerks that produced the local records. The findings suggest that the town clerks were slow to adopt Southern English forms because many clerks and scribes were trained by their own fathers, almost as if the clerkship was a family-run business. It is only when the transmission of the orthographic idiolect of this community was disrupted by a new clerk from outside the immediate scribal network that we see bursts of change towards the English forms.