ICEHD 1 - Abstracts
[page under construction]
Keynote
Lectures
The first emigrant English: some Hiberno-English problems and perspectives
Ut
custodiant litteras: Editions, Corpora and Witnesshood
The arena in which historical dialectologists operate can be likened to
an archaeological dig or a crime scene. All three are venues whose detailed
history needs to be reconstructed. It is a commonplace in archaeology and
forensic praxis that as far as possible (a) the scene must not be contaminated
by material brought in by the investigator or anyone else, and (b) the chain of
custody (the sequence of provenances of all objects found on the scene) must be
immaculate. The same requirements hold for the objects of historical
dialectology, which are primarily written texts. Any interference with these
witnesses is a potential contamination: either intrusion of alien
material into the historical record, or a loss of genuine material. Both
are falsifications, both equally destructive.
This paper is a
methodological prolegomenon to certain basic aspects of historical
dialectology, rather than a substantive contribution. Its topic is the
maintenance of forensic cleanliness in a field full of potentially corrupted
information sources. Among the most dangerous of these corrupt sources are
edited texts. They are dangerous because of the degree to which they are
trusted and characteristically regarded as ‘data’, worthy of inclusion in
historical corpora. No text which shows any or all of the following ought to be
admitted as a historical witness:
1. An attempt to
reconstruct a ‘lost original’ or ‘archetype’ from a multi-source tradition, or
to produce a ‘best text’; in other words any multi-sourced or conflate
‘reading-text’, such as the standard editions of Chaucer or Shakespeare.
2. Any emendation, even of what appear to be patent errors.
3. Modernised
spelling, including the replacement of thorn, edh, yogh and wynn by
modern equivalents.
4. Any addition
of modernised punctuation of MS texts, whether by alteration of original
punctuation or by the pointing of unpointed texts.
5. Any alteration of scribal word-division or lineation.
This means that the only acceptable source for older materials is a diplomatic transcript. The first part of the paper attempts to justify the exclusion of the above source-types as data for historical dialectology (or indeed any kind of linguistic historiography). The second part illustrates one version of what a ‘clean’ historical corpus might look like, and some of the things it can do. The basis for this illustration is the corpus currently being prepared for the Linguistic Atlas of Early Middle English (LAEME).
Papers
This paper will discuss evidence for stigmatisation of regional
pronunciations in Pronouncing Dictionaries of the late eighteenth and early
nineteenth centuries, showing how these provide early evidence for some of the
most salient markers today, such as the FOOT / STRUT split and its absence from
northern varieties.
Onomasiological Variation and Change in
Historical Dialectology: Some Principles and Problems.
Onomasiology deals with the speakers’
preference for naming a concept or a class of concepts through one lexical item
rather than another (Geeraerts 1997: 43-46).
This type of lexical variation has always attracted the attention of
dialectologists, both of Present-Day English (e.g. Orton and Wright 1974) and
of earlier stages in the history of English (e.g. McIntosh et al. 1986). In this paper, rather than focusing on the
geographical distribution of independent lexical items, I will concentrate on
the onomasiological arrangement of small sets of semantically-related words and
the analysis of the processes of diatopic and diachronic variation related to
them.
In order to illustrate this, I have selected a group of
terms related to social and religious ceremonies (with special attention to marriage) in early Scots (1200-1700). My
data is based on A Dictionary of the Older Scottish Tongue (Craigie et
al. 1937-2002), which gives full semantic and grammatical information on every
lexical item recorded in early Scottish texts. The detailed definitions given
by the editors of the DOST are used to propose historical
onomasiological arrangements of the words under scrutiny, allowing an in-depth
insight into the different ways Scottish speakers conceptualised the
corresponding semantic areas. More importantly, the resulting onomasiological
reconstruction can be compared to similar classifications of the vocabulary
found in other historical varieties of English (as the ones proposed by the
members of the Historical Thesaurus of English team, such as Roberts and
Kay 1995 and, especially Coleman 1992), enriching historical dialectology with
a more social and cognitivist approach.
Coleman,
Julie M. 1992. Love, Sex and Marriage.
An Historical Study of the English Vocabulary. Unpublished PhD Thesis:
King’s College, London.
Craigie,
William et al. 1937-2002. A Dictionary of the Older Scottish Tongue. Oxford: Oxford
University Press.
Geeraerts, Dirk.
1997. Diachronic Prototype Semantics. A Contribution to Historical
Lexicology. Oxford: Clarendon.
Roberts, Jane, and Christian J. Kay with Lynne
Grundy. 1995. A Thesaurus of Old English. London: Centre
for Late Antique and Medieval Studies.
Julia Fernández Cuesta / Nieves Rodríguez Ledesma (Seville)
Northern
Features in 15th- and 16th-Century Legal Documents from
Yorkshire
This
paper, which is based on an ongoing research project on northern dialects, aims to analyse to what extent northern
features are found in Yorkshire wills and testaments from the fifteenth and
sixteenth centuries, when compared with other text types from the same period
(civic records, letters, memoirs). More than 300 wills have been read from Swaledale Wills and Inventories, Testamenta Eboracensia and York Clergy Wills. All
exhibit, to a greater or lesser extent, some dialectal features in spelling,
morphology, syntax and vocabulary. Since none of these editions are diplomatic,
it has been necessary to study the original manuscripts. For this purpose a
selection of forty texts has been made from the above collections. Analysis of
the manuscripts confirms the presence of northern features, although some
editorial deviations and errors have also been detected. The following features
have been identified:
a)
spelling: <a> for <o> (knaw); <u> for <o(o)> (gud); <i> as a diacritic to
indicate length (maid); <qw->
for <w> (qwat); <k> for <ch> (kirk); <-lk> in ilk, whilk; <-f(f)> in giff, haff.
b)
morphology: plural nouns and genitive
singular in -is, -ys; third person plural personal pronouns (thai, tham thair); present indicative
inflections (first and third persons singular and plural in -is, -ys,
-es); present participle in -and;
preterite and past participle of weak verbs in -id, -yd; past participle of strong verbs in -in, -yn; plural present indicative of the verb to be.
c)
syntax: endingless genitive; first person singular
present indicative in -is, -es when
it is not immediately preceded by the subject pronoun.
d)
vocabulary: gar/ger
(‘cause to do’) vs. do, make; till
vs. to; dialectal words such as fother, stirke, stotte, twinter, reckan,
etc.
A
number of texts seem to be more standardized than
others. In the less standardized ones a series of features have a strong
tendency to co-occur. This variation may be related to sociolinguistic factors
such as the education and social background of the amanuenses or the testators/testatrixes
(in the case of holographs).
Markku
Filppula (Joensuu) / Juhani Klemola (Tampere)
Conservative BrE dialects as a source of
evidence for language contacts
Regional
dialects offer a rich but so far relatively untapped source of data for the
study of the history of English language. The reasons for the marginal status
of dialectal evidence in historical studies are manifold, an obvious one being
the unavailability of systematically collected tape-recorded data.
The Survey of English Dialects tape recordings represent the largest
systematically collected corpus of naturally occurring traditional dialect
speech. In this paper we look at a number of grammatical and lexical features
of British English dialects to be found in the SED tape recordings and Basic
Material, and in various dialect monographs and glossaries. We shall argue that
the geographical distribution of several grammatical and lexical features of
British English dialects can be interpreted as evidence for language and
dialect contacts both in the Modern and the Mediaeval periods.
Susan
Fitzmaurice (Northern Arizona Univ.)
The paper will address methodological problems in historical dialectology
research of two periods in the history of the English language (likely to be
late Middle English and the eighteenth century), with a focus on the impact
that the search for an ideal English has on the shape of the scholarship in the
period. There will also be an analysis of the social and intellectual pressures
on the dialectical relationship between the idea of a standard and varieties of
English in each period’s scholarship.
Standard Wisdoms and Historical Dialectology: The Discrete Use of Historical Regional Corpora
Just as in mainstream historical dialectology, in the diachronic study
of regional varieties of English there are many cherished ideas about how these
varieties arose and what they were like at earlier historical stages.
Especially with regional forms of English from the British Isles, these
questions are by no means trivial as they have a clear bearing on the
transportation of English overseas and what we regard as the possible
continuation of dialect features from input varieties to those locations where
English was transported.
In recent years a number of
historical regional corpora have become available and are of assistance in
determining what the shape and contours of earlier forms of regional varieties
were like. In the present instance, the author’s A Corpus of Irish English
(Benjamins: 2003) will be investigated with a view to demonstrating what
historical evidence is contained in the texts it consists of. In particular,
questions of dialect syntax (aspectual structures) and morphology (pronominal
forms for the second person) will be examined and some exemplary studies will
be presented briefly.
North Northumbrian and South Northumbrian: a geographical statement?
Since the time of Bulbring (1902) and Lindelöf (1893,1901) there has
been a recognition of two varieties of Northumbrian which appear to belong to
two distinct dialectal areas, which are described, respectively, as North
Northumbrian and South Northumbrian. Restricting ourselves to the
later Northumbrian texts, since such distinctions are far less clear for early
Northumbrian, we may follow the above scholars in assigning texts as follows.
For North Northumbrian: The Lindisfarne Gospels and The Durham Ritual;
South Northumbrian: the Northumbrian portions of The Rushworth Gospels (Ru2).
The geographical attribution
of these texts has, since that time, always seemed secure. It is known that the
northern texts were written by a scribe from Chester-le-Street, County Durham,
Aldred. Equally, it is known the Ru2 was written by a scribe called
Owun, and the usual view has long been that the whole manuscript was written at
Harewood, an insignificant village about half-way between Leeds and York.
Consequently the geographical distinction between the two varieties has seemed
particularly clear-cut.
This neat and simple
division has been seriously undercut in a recent paper by Richard Coates
(Coates, 1997), who argues that The Rushworth Gospels were written at
Lichfield but that Owun was a Northumbrian scribe who had been sent there with The
Lindisfarne Gospels. If Coates is correct, then the scribe of Ru2
was as likely to have been “North Northumbrian” as Aldred. And hence the
dialect of Ru2 was also as likely to have been “North
Northumbrian” as the dialect of The Lindisfarne Gospels.
The aim of this paper,
therefore, is to examine the plausibility of Ru2 as a North
Northumbrian text. I shall do so in two ways. Firstly I shall examine some of
the linguistic (phonological and morphological) differences between the
“northern” and “southern” texts, both in terms of quantity and substance. In
other words, are the differences sufficiently clear-cut to allow us to be able
to distinguish two different dialects? If that is the case, can we go further
and hypothesise geographical distinctions in the way that has traditionally
been done?
Secondly, if that proves
impossible, or, at least, implausible, is it possible to assess the value of
the scribal variations between, on the one hand, Aldred, and on the other,
Owun? It has to be recognised that this is a particularly difficult task, given
the known idiosyncrasies of Aldred.
It may well be the case
that it is simply impossible for us to come to any positive conclusions, other
than that the traditional views are no longer secure. But that merely
highlights the necessity for further exploration of the issues involved.
In this paper I will outline the thinking behind a larger project which
I am developing for funding, and suggest the preliminaries of a case study to illustrate
this thinking. Dialect syntax at the clausal level so far has been focussed on
differences that may be attributed to inflection-related changes in word order
(differential loss of OV word orders; differences in the constraints on Verb
Second) and have been attributed to the differentiating effects of Scandinavian
influence on the North-eastern dialects of English. But the database for this
work is as yet quite narrow, and the assumption that English in pre-Viking
times was syntactically more homogeneous seems less than warranted. We need,
apart from anything else, to look at the dialects in their own right, and to
approach the syntax with a model that allows for any amount of fine-grained
detail. Such a model can be derived from a theoretical approach in terms of
recent versions of Principles and Parameters theory, a model which apart from
its theoretical claims allows for any amount of detailed categorization as well
as establishing implicational relationships between seemingly unrelated features.
The latter is a tool that may be of considerable help in making sense
of a complex web of syntactic data. A further tool that is needed
is, of course, a dialectally responsible corpus with sufficient linguistic
enrichment to do productive research (much along the lines employed in the
LAEME corpus). I will illustrate this approach with a pilot case study on the
relation between finite verbs and adverbs in various dialect samples.
Middle
English authorial dialects have on occasion been placed using as few as three
linguistic variables plus some biographic background information. It is
reasonable to think that placings of Old English literary dialects reached by
the use of seven linguistic variables will be not less reliable, especially
where all point unambiguously in a single direction. There are good statistical
reasons to believe that where linguistic variables probative for an author (or
text) are mappable from Anglo-Saxon charter boundaries, the dialect can be
placed with high probability to within a few miles, or with a margin of error
of say 20 miles with something not far short of certainty. If true, this should
of course mean that mappable items not used in the argument should point the
same way, and so should bits and pieces not quantifiable as samples but visibly
dialectal. Of dialects for which charter-based analysis has been published,
this kind of confirmation turns up not only for Ælfric, for whom all the variables
used clearly agreed, but also to reinforce the likeliest placing for King
Alfred, which uncertainty about exact positions of isoglosses left short of a
full score. The implications need to be embraced more fully than they have been
by Anglo-Saxonists, not least in interpreting the ambivalent hints which are
the most that is known of Old English authors’ biographies.
As
the number of authors and texts so analysed increases, the probability of
correctness of individual placings will be enhanced, as each linguistic sample
located cuts down the area available for those distinct from it. So abstractly
stated, most participants know this from the “fit” technique used in the Atlas
of Late Middle English. It applies most in the West Saxon heartland, where both
charter boundaries and Old English literary texts are best represented. As in
Middle English, further north margins of error are larger. A practical
difference between the periods is that variables with large enough samples to
be mappable from charter boundaries overlap too little with those in literary
texts for identical questionnaires to be usable for all texts (or, possibly,
for any two not textually related). Relevant variables in place-names may also
be sui generis, as in the Orosius. Margins of error here depend partly
on what kind of change over time is credible in particular items.
Similar
questions can arise for more narrowly linguistic items, where a variable
striking in a text clearly is related to one(s) mappable in some sort from
other sources, but it is debatable at what level of abstraction and how
identical their distributions ought to be. Instances occur in the Vespasian
Psalter. The least favourable conditions for investigation are where, because
of possible accommodation to an external standard or otherwise, it is not
established what the linguistic variables are, let alone what they should be
related to. Some suggestions will be offered on the currently exciting case of
Ruthwell1. Charter boundaries show, at a minimum, that some
identifiably local spelling-tradition(s) existed in Northumbria as part of the
late West Saxon realm.
If time permits, speculations will
be uttered about Old English poetic manuscripts.
To the non-dialectologist, the term ‘dialectology’ usually suggests static displays of dots on regional maps, indicating the distribution of phonological, morphological or lexical features. [However] space is only one dimension of dialectology. Spatial distribution is normally a function of change over time projected on a geographical landscape. But change over time involves operations within speech communities; this introduces a third dimension — human interactions and the intricacies of language use. Dialectology therefore operates on three planes: space, time and social milieu.
(Laing and Lass: forthcoming)
For
historical dialectologists social milieu must be taken to refer to the whole historical background of the
language under study; the social variables will differ according to which
historical vernacular is being investigated and at what period. The data presented in this paper are drawn
from those collected for A Linguistic Atlas of Early Middle English (LAEME).
Problems facing the investigator of any past stage of a language:
1. Contingent
survival of text witnesses — we are confined to an accidental sample of text
languages and since all of our informants are dead, the sample is not
expandable.
2. The increasing
opacity of social milieu with the passage of time — the further back we go, the
more decontextualised our witnesses will tend to be.
3. Our witnesses are samples of written rather than of
spoken language. The ‘native speakers’
of past stages of a language are writers and copying scribes.
Early
Middle English (ca. 1150–1300) is a time in the history of the language when
scribes are beginning to use their native language in written form after an
extended period during which all new writing was in Latin or French. For the whole of the early Middle English
period, English was the least commonly used of the three written languages. At
this stage there was no established approach to the writing of contemporary
spoken English and we see a great deal of experimentation in spelling. Scribes
used their training and experience in the writing of Anglo-French and
Anglo-Latin, as well as Old English traditions, to design orthographies for the purpose. These display varying
degrees of individual inventiveness, some being economical in how they match
sound to symbol, some much more profligate and complex.
Most of the ‘social’ variables associated with an investigation of
Middle English are therefore those arising from scribal strategies: (a) the
design of individual spelling systems; (b) whether or not a copying scribe is a
‘translator’, a literatim copyist or a ‘mixer’ (McIntosh 1989 [1973]:
92). The intricate interactions between
these variables give rise to layered complexities of various kinds — the
stratigraphy of this paper’s title.
I will examine the output of a number of early Middle English scribes to
illustrate how different approaches to decoding and re-encoding from exemplar
to copy may affect the mapping of text languages in the dimensions of time,
space and scribal milieu.
References
Laing, M. and R. Lass (forthc.) ‘Early Middle English
Dialectology: Problems and Prospects’ for Handbook of the History of English
eds. Ans van Kemenade and Bettelou Los to be published by Blackwell.
McIntosh, A.
1989 [1973]. ‘Word geography in the lexicography of mediæval English’, Annals
of the New York Academy of Sciences 211 (1973), 55–66; repr. in: M.Laing,
ed. Middle English Dialectology: Essays on some Principles and Problems.
Aberdeen University Press, 86–97.
Substratum vs. Borrowing:
an assessment of the
Scandinavian Element in Shetland and Yorkshire Traditional Dialects
Shetland
dialect today must be described as a variety of Scots, yet with a substantial
component of Scandinavian, manifested at all levels of language. This component
differs from the Scandinavian linguistic heritage in other parts of Britain
(with the exception of Orkney), not only in size but also in structure and
history. The Norse invaders of Yorkshire, for example, met a native Anglo-Saxon
population with whom they - allegedly - could communicate. They influenced the
Anglo-Saxon language and some of this influence has survived, mostly in the
form of lexical borrowings. In Orkney and Shetland, on the other hand, we see
the still powerful impact of a Scandinavian substratum,
supported by positive - to the degree of romantic - feelings of affiliation
with Scandinavia.
This paper identifies and examines in some detail the Scandinavian
element in Shetland and Yorkshire traditional dialect, drawing on data from the
Survey of English Dialects, the Linguistic Survey of Scotland and my own
fieldwork in Shetland. The findings are discussed in the light of Thomason
& Kaufman (1988), with special reference to their categories of language
maintenance and the case study of the Norse influence on English.
References
Thomason, S.G. & T. Kaufman (1988) Language
Contact, Creolization, and Genetic Linguistics. Berkeley and Los Angeles:
University of California Press.
Anneli
Meurman-Solin (Helsinki)
From Inventory to
Typology. Methodological Considerations Based on the Corpus of Scottish Correspondence
The
methodological points I will make about inventories and typologies are all
directly linked with the new genre of corpora created by researchers at the
Institute for Historical Dialectology (IHD), University of Edinburgh. The Corpus
of Scottish Correspondence (CSC), 1500-1750, compiled by myself as a member
of the Research Unit for Variation and Change in English, University of
Helsinki, represents this genre of corpora by exclusively containing
diplomatically transcribed manuscripts of letters selected to create a
diatopically representative database. Part of this corpus has already been
lexico-grammatically tagged using software developed by Keith Williamson at the
IHD.
In my paper I will briefly comment on transparency, flexibility and
multidimensionality as criteria in evaluating corpora as tools in linguistic
research and then discuss the genre
‘correspondence’, highlighting features that make letters a relevant source in
morpho-syntactic study. The main questions can be summarized as follows: What
are the theoretical and methodological implications of using linguistic inventories
based on large philologically digitized and diatopically, diachronically and
diastratically representative tagged databases in the production of
typologically relevant new knowledge about variation and change in text
languages? What new insights does the data-driven and data-oriented
variationist approach give into how quantitatively and qualitatively improved
linguistic data affect our conceptualization of categories, taxonomies and
typologies? How does a tagging system developed for analysing overlapping and
intersection between categories sharing the function of clausal connective, for
instance, help us in the identification of the typological implications of
various degrees of fuzziness? How can we create a variationist typology?
I will try to suggest answers to these questions by examining the system
of clausal connectives in the CSC data. What kind of criteria can be developed
for categorizing chunks of text as ‘clause’ or ‘sentence’? What is the typology
of clause-combining devices in the data? How does data of this kind change our
views on subordination, for instance? How are the varying degrees of linguistic
and stylistic competence of the letter-writers reflected in their use of
connectives? Can we identify dialect-specific patterns of usage? An important
motivating factor in my present work is that less integration seems to have
taken place between the corpus-based variationist approach and descriptive work
aiming at the reconstruction of grammatical systems in terms of typologies than
could perhaps be expected, given that linguists, including typologists, have
had a wide range of corpora available for them for quite a long time now. It is
not difficult to find examples to illustrate this. The typology of
clause-combining devices in English as well as other European languages has
been quite extensively discussed in recent literature by Devriendt & al.
1996, Kortmann 1997, and van der Auwera 1998, for instance. However, in these
studies typologies have been construed by chiefly using secondary sources such
as dictionaries and grammars (Kortmann 1997: 53).
Auwera, Johan van der (ed.) 1998. Adverbial Constructions in the Languages of Europe. Berlin and New
York: Mouton de Gruyter.
Devriendt, Betty, Louis Goossens and Johan van der Auwera
(eds) 1996. Complex Structures. A
Functionalist Perspective. Berlin and New York: Mouton de Gruyter.
Kortmann, Bernd 1997. Adverbial Subordination: A
Typology and History of Adverbial Subordinators Based on European Languages.
Berlin and New York: Mouton de Gruyter.
Meurman-Solin, Anneli 2002. Simple and complex
grammars: The Case of Temporal Subordinators in the History of Scots. In Helena
Raumolin-Brunberg, Minna Nevala, Arja Nurmi and Matti Rissanen (eds), Variation Past and Present. VARIENG Studies on English for Terttu
Nevalainen.
Helsinki: Société
Néophilologique, 187-210.
Williamson, Keith 1992/93. A Computer-aided Method for
Making a Linguistic Atlas of Older Scots, Scottish Language,11/12:138-173.
Williamson, Keith 2000. Changing spaces: Linguistic
relationships and the dialect continuum. In Irma Taavitsainen, Terttu
Nevalainen, Päivi Pahta and Matti Rissanen (eds), Placing Middle English in
Context. Berlin and
New York: Mouton de Gruyter, 141-179.
History
in the making: The Standardisation of the Mauritian Vernacular
This paper draws from my PhD
thesis entitled Creative Writing in Mauritian Creole: The emergence of a
literary language and its contribution to standardization (Leeds, 2000). My
undergraduate study of the history of English drew my attention to the more
recent historical development of Mauritian Creole (MC). Although interesting
parallels can be drawn between the rise of national languages in Europe
(Scaglione 1984) and the language of my study, the socio-political and
historical contexts behind the former and the latter are in sharp contrast.
The (corpus and
status) development of MC from language of orality to language of literature
and national language de facto has taken place within a compressed period of
time in Mauritius - within 35 years since independence. In fact, the Mauritian
linguist and creative writer Dev Virahsawmy, referred to as ‘the cultural
one-man movement’ in the promotion of MC (Eriksen 1998: 21), has recently
pointed out that MC should be referred to as Morisien since it is a
post-creole, moving away from its creoles structures while pursuing an
independent evolution from its lexifier (Tranquille 2000). The development of
this vernacular is the more phenomenal since the Mauritian State persists in
its position of non-intervention vis-à-vis MC. While MC remains officially
unacknowledged, its development is in the hands of the civil rather than
the political society.
I assess the
factors which have led to the rapid elaboration of MC and which are leading to
its standardisation: from the early postcolonial nationalist sentiments to the
current pragmatic and urgent need for literization in the vernacular, and the
outburst of creative writing. I also consider, through the literary uses of MC
– which encourage the processes of selection, codification, elaboration,
acceptance (Haugen 1966) –, the changes being brought to its linguistic
features mainly in terms of orthography, lexicon and syntax.
References
Adone, D. and Plag, I. (eds) (1994) Creolization
and Language Change. Tübingen: Niemeyer.
Eriksen, T.H. (1998) Common
Denominators: Ethnicity, nation-building and compromise in Mauritius. Oxford:
Berg.
Haugen, E. (1966) Language Conflict and
Language Planning. Cambridge, Mass.: Harvard University Press.
Mooneeram, R. (2002) Creative Writing in
Mauritian Creole: The emergence of a literary language and its contribution
towards standardization. Unpublished PhD thesis, University of Leeds.
Scaglione, A. (1984) The Emergence of
National Languages. Ravenna: Longo.
Sebba, M. (1997) Contact Languages:
Pidgins and Creoles. London: Macmillan.
Tranquille,
D. (2000) Interview with Dev Virahsawmy
(http://pages.intnet.mu/develog/, 2000)
Mieko Ogura (Yokohama) / William S-Y. Wang (Hong Kong)
Ogura (1990) and
Ogura, Wang & Cavalli-Sforza (1991) proposed a method of dynamic
dialectology which unifies the study of language in its temporal and areal
aspects, and showed how diffusion from word to word spreads spatially through
time. Ogura & Wang (1998) defined
it as a 2-dimensional diffusion model: diffusion from word to word and
diffusion from speaker to speaker through time.
Complex systems are made up of a large number of entities that
by interacting locally with each other give rise to global properties that
cannot be predicted or deduced from an even complete knowledge of the entities
and of the rules governing their interactions.
In many cases they are adaptive systems, that is, they tend to change in
ways that depend on the particular environment in which they exist.
Gell-Mann (1992) points out that the selective effect is the
central feature of complex adaptive system, and selection provides a bottleneck
that induces adaptation. In the Origin
of Species, Darwin noted that natural selection cannot directly promote
altruistic acts. Yet cooperation is
abundant in nature. Evolutionary game
theory (Maynard-Smith 1982) has allowed biologists to analyze such dynamics. Linguistic selection is unconscious
functional selection between available variants by the learners. Languages become adapted to the productive,
perceptual and cognitive abilities of human beings in the transmission across
generations. The changes arising from
random variation or social factors spread by the cooperation in the repeated
pairwise interactions of the individuals.
Both linguistic selection and language games are important mechanisms in
language evolution.
In this paper we would like to synthesize linguistic selection
and language games in the complex adaptive system of dynamic dialectology. Our discussion is based on the simulation
and historical data from English: evolution of word order, Great Vowel Shift,
simplification of the inflectional endings of nouns and verbs, the development
of -ing in the present participle for linguistic selection, and
vocabulary emergence, the development of -s in the third person singular
present indicative, the development of West Germanic *a before nasals in
the Mercian speech community for language games.
We will show that the dialect differences emerge in both types
of language evolution. A change may
start in a particular location and spread out from there to cover neighboring
areas. Some changes may spread so much
that they eventually cover the whole country.
Others will only spread locally. In linguistic selection type of
evolution, languages tend towards uniformity rather than diversity. In language game type of evolution, the size
of the neighborhood determines the number of the individuals that interact, and
the socially influential people have an increased probability of being imitated
by their neighbors. Hence, successful
changes spread locally.
Finally we will explain why the change that started slowly in
a few words spread through more and more words in a rising S-curve from
Kolmogorov Complexity.
Gell-Mann,
Murray. 1992.
“Complexity and Complex Adaptive Systems”, The Evolution of Human Languages
ed. by John A. Hawkins and M. Gell-Mann. Reading Mass: Addison-Wesley, 3-18.
Maynard-Smith,
John. 1982. Evolution and the Theory of Games. Cambridge: University
Press.
Ogura,
Mieko. 1990. Dynamic Dialectology: A study of language in time and space.
Tokyo: Kenkyusha.
Ogura,
Mieko, William S-Y. Wang & L. Luca
Cavalli-Sforza. 1991. “The Development of Middle English i: in England: A study
in dynamic dialectology”. New Ways of Analyzing Sound Change ed. by
Penelope Eckert, 63-106. New York: Academic Press.
Ogura,
Mieko & William S-Y. Wang. 1998. “Evolution Theory and Lexical Diffusion”. Advances
in English Historical Linguistics. ed. by Jacek Fisiak & Marcin
Krygier. Berlin:
Mouton de Gruyter, 315-344.
Graham Shorrocks (Memorial University of Newfoundland, Canada)
Robert Southey’s A True Story of the
Terrible Knitters e’ Dent ---:
A Linguistic
Analysis of a Neglected Dialect Specimen
In a previous conference presentation on “The Dialects
of Early Modern English and Their Study”, I reviewed some of the problems
surrounding Early Modern English dialectology, and the sources for the study of
the Early Modern dialects.[1] The dialects of English of the
sixteenth-eighteenth centuries emerged as perhaps the most neglected and poorly
researched of all in the history of the language. Under such circumstances, it is important to identify and then
analyse both MSS. and printed documents that contain examples of dialectal
forms and constructions from the period.
Although published repeatedly in the nineteenth and twentieth centuries,
Robert Southey’s A True Story of the Terrible Knitters e’ Dent -- (posthumous,
1847) has remained an essentially-neglected source, not least — one imagines —
because of the fact that it was not republished by the English Dialect Society
in their important series of dialectologically-significant works, because the
editions in which it was published were somewhat obscure, and perhaps
also because it has not attracted comment from mainstream literary critics
either. Though well-known to readers
and writers of Cumbrian-dialect literature, it seems to be completely unknown
to professional dialectologists post-Wright.[2] In my paper, I will offer an analysis of a
sample of the text at all linguistic levels, from the discourse-analytical and
stylistic through the orthographical-phonological to the grammatical and
lexical. To facilitate this explication
of the text in the time-frame allowed, a detailed hand-out will be provided in
advance. It will be seen that,
despite certain difficulties, Southey’s text gives us a considerable amount of
linguistic information, and that it is especially valuable by virtue of its
being one of a group of relatively-early texts devoted to the dialects of the
North-West of England. Further, given
that the informant is relating events that she had experienced in the 1770s,
that the transcribers had had close contact with this dialect and their
informant for many years before they took down the piece, that society was
still relatively static at the time, and that their probable antiquarian agenda
was conservative, it is not unreasonable to imagine that this text offers us
some insights into the Westmoreland dialect of the later and even
mid-eighteenth century. It is an
example of what needs to be done, of the sources which remain to be fully
exploited, if we are better to illuminate the dialects of Late Modern English.
The
text is also of interest if we look beyond purely-linguistic matters to the
social, political, and literary contexts.
It provides a possible argument that Romantic authors such as Wordsworth
and Southey had indeed judged dialect speech in a positive way; the full title,
--- Which will be Read with Interest by Humane Manufacturers, and by Masters
of Spinning Jennies with a Smile, suggests a socio-political dimension;
while the publication of a text completely in a non-standard dialect might
arguably challenge the received wisdom amongst most scholars to the effect that
the English Romantics never used dialect in their works.
References
Denwood, Marley & T.W.
Thompson, comps. A Lafter o’ Farleys
| in t’ Dialects o’ Lakeland | 1760ÿ1945 | Compiled and Introduced | by — and
—. Carlisle: Charles Thurnam & Sons Limited,
Printers, for the L.D.S., 1950.
[Southey, Robert.] “A True Story of the Terrible Knitters e’
Dent Which will be Read with Interest by Humane Manufacturers, and by Masters
of Spinning Jennies with a Smile.” The
Doctor, Etc. Vol. 7. Ed. the Rev. John Wood Warter. London:
Longman, Rees, Orme, Brown, Green & Longman [sic.], 1847. 78-89.
Republished
in The Doctor, Etc. B/by the Late
Robert Southey. Edited by His
Son-in-Law, John Wood Warter, B.D. New
Edition, Complete in One Volume.
London: Longman, Brown, Green,
& Longmans, 1848.
Repr.
numerous times in Specimens of the Dialects of Westmoreland and Specimens
of the Westmoreland Dialect ---.
[See below.] Anthologised in Denwood & Thompson (1950: 48–53).
Specimens of the Dialects of
Westmoreland. By the Rev. Thomas Clarke,
Rector of Ormside; William Bowness, Esq.; and Robert Southey, Esq., LL.D. Part First. Kendal:
Atkinson & Pollitt, 1877.
[In effect, another edn. of Specimens of the Westmoreland
Dialect. Many further edns.]
Specimens of the Dialects of
Westmoreland. Parts
1, 2 & 3. Kendal: Repr. by Atkinson & Pollitt, 1924.
Specimens of the Westmoreland
Dialect. By the Rev. Thomas Clarke,
Rector of Ormside; William Bowness, Esq.; and Robert Southey, Esq., LL.D. Kendal: Atkinson & Pollitt, 1872. [T’ Terrible Knitters
e’ Dent (Repr. by Permission from The Doctor) 47-52.]
Specimens of the Westmoreland
Dialect; Consisting of T’ Reysh Bearin, and Jonny
Shippard’s Journa to Lunnan. By the
Rev. Thomas Clarke. Reprinted from The
Westmoreland Gazette. Also,
T’ Terrible Knitters e’ Dent. By Robt. Southey. Reprinted by Permission from The
Doctor. Kendal: Printed by Thomas Atkinson, 1865.
Specimens of the
Westmoreland Dialect; Consisting of T’ Reysh Beearin, and Jonny Shippard’s
Journa ta Lunnan. By the Rev. Thomas
Clarke. Reprinted from The Westmoreland
Gazette. Jimmy Green at Brough Hill
Fair. By W. Bowness. From a Series of
Sketches in the Westmoreland Dialect. Also, T’ Terrible Knitters e’ Dent.
By Robert Southey.
Reprinted by Permission from The Doctor. Kendal: Atkinson & Pollitt, 1867. [At least two further
edns.]
Wright, Joseph, ed. The English Dialect Dictionary. Being the Complete Vocabulary of All Dialect
Words Still in Use, or Known to have Been in Use During the Last Two Hundred
Years. Founded on the Publications of
the English Dialect Society and on a Large Amount of Material Never Before
Printed. 6 vols. London & Oxford: Henry Frowde, Publisher to the English
Dialect Soc.; New York: G.P. Putnam’s
Sons, 1898–1905.
One of the more notorious footnotes
in Alistair Campbell’s classic Old
English Grammar reads as follows:
I merely suggest the probable approximate position of
the [Old English] vowels, and do not attempt to decide if they were tense or
lax. It is fundamental to the history of the English vowels that the long
and short vowels were practically identical in quality till about 1200, and
that afterwards they became distinguished by the short sounds becoming more
open or more lax than the long sounds to which they had previously
corresponded. (Campbell 1959: 14,
fn.2: my italics.)
It is clear that this axiom was, for Campbell, obvious: so obvious,
indeed, that he does not explain the basis upon which it rests. But it is not so obvious for other
scholars. Thus, for instance, Roger
Lass, in his major survey of Middle and Early Modern English phonology in the Cambridge History of the English Language,
argues on evidential grounds that the qualitative distinction between Middle
English i and iµ did not emerge until the middle of the seventeenth century (see
most fully Lass 1999: 87-88).
There would seem therefore even now to be some unresolved
issues to do with the quality of the Middle English short vowels. The present paper attempts to deal with some
of the problems of interpretation and explanation to which the history of these
vowels gives rise. Its main focus is on
the reflexes of Old English i and u, but other Middle English short
vowels are also brought under review.
The main argument of the paper is that Campbell and
Lass are both correct – a seeming paradox which is resolved once variation in
space and time is considered.
Gjertrud F. Stenbrenden (Oslo)
On the inception and spread of early ME £ to Î: some observations
Conventionally,
certain ME long-vowel changes are grouped together under the label
‘the Great Vowel Shift’ (‘GVS’) and are assigned to the period ca. 1400-1750.
Two other ME long-vowel changes are assigned to the period ca. 1250-1350, and
are set apart from the ‘GVS’, although they are of fundamentally similar kind.
These are the Southumbrian backing and raising of eME £ to Î , and the Northumbrian fronting and raising of eME Í to [ü:]. This paper reports on aspects of a doctoral project concerned with the
interpretation of these sound-changes – their inception, chronology, and
regional spread.
Importantly, data abstracted from three linguistic atlases of Middle
English, i.e. LALME, A Survey of Middle English Dialects
1290-1350 (Kristensson 1967, 1987, 1995, 2001), and the forthcoming LAEME,
seem to run counter to the conventional account. More precisely, material from
these sources indicates that some of the core ‘GVS’-changes may have started
100 years or more prior to the date traditionally set for the beginning of the
‘Shift’. Moreover, the material indicates that eME started to round and raise
as early as 1150, but that this presumably unrelated change was still in
progress after 1450 (cf. Stenbrenden 1996, 1999, 2001). The latter seems
to be the case also with the Northumbrian change of Í to [ü:]. There is thus a
temporal overlap between the ‘GVS’-changes and the earlier changes, which makes
it difficult to treat them as belonging to different sets of changes.
This paper treats material for eME £ from the entire ME period,
looking at how the data are to be interpreted. Further, my paper investigates
the consequences of my interpretation for the entire concept of ‘the Great
Vowel Shift’.
References
Kristensson, G. (1967), A Survey of Middle English
Dialects 1290-1350: the Six Northern Counties and Lincolnshire, Lund: CWK
Gleerup.
----- (1987), A Survey of Middle English Dialects
1290-1350: the West Midland Counties, Lund: Lund University Press.
----- (1995), A Survey of Middle English Dialects
1290-1350: the East Midland Counties, Lund: Lund University Press.
----- (2001), A Survey of Middle English Dialects
1290-1350: the Southern Counties, Lund: Lund University Press.
LAEME, see Laing, M.
Laing, M. (in prep.), A Linguistic Atlas of early
Middle English, University of Edinburgh.
LALME, see McIntosh et
al. 1986.
McIntosh, A., M.L. Samuels, M. Benskin et al. (1986), A Linguistic Atlas of late Mediaeval English, Vols. I-IV,
Aberdeen: Aberdeen University Press.
Stenbrenden, G.F. (1996), The Great Vowel Shift:
problem of reification and analysis, (unpublished MA thesis), University of
Oslo.
----- (1999), ‘A reassessment of certain English
long-vowel changes, ca. 1250-1500’. In: López, A.B. et al. (eds.), ‘Woonderous
Ænglissce’. SELIM Studies in Medieval English Language 14,
Universidade de Vigo: Servicio de Publicacións.
----- (forthcoming), ‘On the Interpretation of Early
Evidence for ME Vowel-Change’, paper read at the 15th International Conference
on Historical Linguistics, Melbourne, 13-17 August 2001.
During
the fifteenth century, the general picture of linguistic variation in written
English undergoes an important change.
While linguistic variation in the Middle English period is mainly geographically
conditioned, towards the end of the period the patterns of variation become
increasingly non-regional. The
Linguistic Atlas of Late Mediaeval English (= LALME; McIntosh et al.
1986) includes localized texts from the late fifteenth century, as well as the
occasional text from the early sixteenth; however, texts containing a strongly
regional language are becoming exceptional by this time (cf Benskin 1992:
71-72.).
The spread of the so-called "Chancery Standard" during this
period is a complex matter, and the limitations of "Chancery
Standard" as a concept have been emphasized by Michael Benskin (e.g.
Benskin 1992). However, other processes
during the same period also tend towards rendering regional differences less
marked. These processes include, above
all, the use of "colourless" types of language. Increasing colourlessness in texts written
during the fifteenth century may be taken to reflect both the increased
mobility of writers and texts and the new national function of written English;
a degree of colourlessness may, indeed, be seen as the inescapable result of
the copying of texts across dialect areas.
Another, parallel, process is the development whereby certain forms lose
their strongly regional character and come to be used as dialectally
"neutral" variants (cf Samuels 1981 [1988: 87]). With the growth of text production,
originally regional forms could also be attached to a particular type of text
or a centre of production, and come to be identified with those instead of, or
as well as, their original regional context.
In the late fifteenth century, written English language is still highly
variable; however, by this time, the continuum of regional dialects is being
replaced in importance by a continuum between the more and less regional. At the same time, genres, text traditions,
centres of book production and networks of various kinds may be expected to
play an important role in the written variation. To study the ways in which such non-regional patterns map onto
the geographical framework established in LALME is a major challenge facing the
historian of Late Middle English.
It is one of the goals of the Middle English Grammar Project (Glasgow
and Stavanger) to attempt to recover some of these non-regional patterns, by
making use of the framework and materials contained in LALME together with
search methods made possible by present-day technology. The present paper will suggest some
approaches that appear promising. In
particular, reference will be made to text types as an important category for
the study of the spread of spelling conventions. The general view taken is that the de-regionalisation of written
English during this period follows naturally from the contemporary developments
in literacy and society.
References
Benskin, M. (1992), 'Some new perspectives on the
origins of standard written English' in J.A. van Leuvensteijn and J.B. Berns
(eds), Dialect and Standard Language in the English, Dutch, German and
Norwegian Language areas, Amsterdam: Royal Netherlands Academy of Arts and
Sciences, North-Holland, 71-105.
McIntosh, A., M.L. Samuels and M. Benskin (1986), A
Linguistic Atlas of Late Mediaeval English, 4 vols. Aberdeen: University
Press.
Samuels, M.L. (1981), 'Spelling and Dialect in the
Late and Post-Middle English Periods' in M. Benskin and M.L. Samuels (eds), So
meny people longages and tonges: philological essays in Scots and mediaeval
English presented to Angus McIntosh, MEDP: Edinburgh, 43-54. Repr. in J. Smith (ed) (1988), The
Language of Chaucer and his Contemporaries, Aberdeen: University Press,
86-95.
Late Middle English Verb
Second Order: Establishing and Interpreting a Dialectal Database
It
seems clear that early Middle English continues the Old English system, whereby
there is a contrast between V2 contexts with nominal subjects and V2 contexts
with personal pronoun subjects (Kohonen 1978, Kroch and Taylor
1997). After tha/tho ‘then’ and a small group of other adverbs, inversion
is normal with both types of subject. After other fronted or initial elements
inversion is common with nominal subjects, but usually fails with a ‘weak’
pronominal subject (van Kemenade 1987, Haeberli 2001, 2002). I shall
investigate the dialectal distribution of this contrast in later Middle
English. This is of particular interest since it has been claimed that Northern
Middle English lacks this contrast and is thus typologically distinct from
Southern Middle English (Kroch and Taylor 1997, Kroch, Taylor and Ringe 2001).
First, then, I need to establish a reasonable sampling of texts whose
syntax is dialectally representative,
and I have constructed a database of prose texts which seem likely to retain
the dialectal syntax of the author.
Secondly,
I have established a distinction between types of inversion in Middle English,
setting aside contexts in which the distribution of inversion with pronominal
and nominal subjects is apparently controlled by the informational status of
the subject.
It is then possible to establish a clear distinction between northern
and southern systems, of the overall type suggested by Kroch and Taylor, so
that there is a contrast between inversion with nominal and with pronominal
subjects in the south, but no such contrast in the north. The dialectal
distribution of the northern system is however much better characterized and
evidenced than in the account of Kroch and Taylor (1997), Kroch, Taylor and
Ringe (2001), since they relied on a very narrow database. This points to
the potential value enhancing the PPCME2 for late Middle English, in particular
by adding material which is dialectally differentiated. This is something I and
my colleagues hope to undertake.
Haeberli, E. 2001 Adjuncts and the Syntax of Subjects
in Old and Middle English. In S. Pintzuk, G. Tsoulas and A. Warner (eds.) Diachronic
Syntax: Models and Mechanisms, Oxford: Oxford University Press.
Haeberli, E. 2002. Observations on the Loss of Verb
Second In the History of English. In D. Lightfoot (ed.) Syntactic effects of
Morphological Change. Oxford: Oxford University Press, 88-106.
Kemenade, Ans van 1987 Syntactic Case and
Morphological Case in the History of English. Dordrecht:
Foris.
Kemenade, Ans van 1997. V2
and embedded topicalization in Old and Middle English. In A. van Kemenade and
N. Vincent (eds.) Parameters of Morphosyntactic Change. Cambridge:
Cambridge University Press, 326-351.
Kohonen, Viljo. 1978. On the Development of English
Word Order in Religious Prose around 1000 and 1200 A.D.: a Quantitative Study
of Word Order in Context. Åbo: Research Institute of the Åbo Akademi
Foundation.
Kroch, A and Ann Taylor. 1997. Verb Movement in Old
and Middle English: dialect variation and language contact. In A. van Kemenade
and N. Vincent (eds.) Parameters of Morphosyntactic Change.
Cambridge: Cambridge University Press, 297-324.
Kroch, A and Ann Taylor.22000. Penn-Helsinki
Parsed Corpus of Middle English. Philadelphia: Department of Linguistics,
University of Pennsylvania.
Kroch, A, Ann Taylor and Don Ringe. 2001. The Middle
English Verb Second Constraint: a case study in language contact and language
change. In S. Henning (ed.), Parameters in Older Language,
Amsterdam: Benjamins.
McIntosh, Angus, M. Samuels, Michael Benskin.
1986. A Linguistic Atlas of Late Mediaeval English. Aberdeen
University Press.
Synchrony, System(s) and Continua
In this paper I examine the
notions of synchrony, system and continuum in Historical
Dialectology, as conceived for the Edinburgh mediæval linguistic atlas projects
and their corpora of lexico-grammatically tagged texts – i.e., A Linguistic Atlas of Early Middle English
(LAEME) and the Linguistic Atlas of Older Scots (LAOS).
The synchronic–diachronic
dichotomy is a keystone of the Structuralist tradition in Linguistics. A synchronic description of a
language excludes any reference to the language’s past states as irrelevant to
what its speakers “know” about their language (their langue / competence). Dialectology, which has traditionally been
concerned with spatial distribution of variant linguistic phenomena, falls
under Synchronic Linguistics while diachrony is the preserve of
Historical Linguistics — how (and why) languages have changed through time.
In mainstream
Linguistics the notion of synchrony is an abstract ideal — the absence
of the Past in relation to some état de la langue. However, rather than defining synchrony
negatively, as the absence of a diachronic dimension, I shall argue that
in any kind of Dialectology the notion of synchrony, if it is to be
used, should be defined positively, and then only sensibly on an empirical
basis.
The data that we
deal with in reconstructing the patterns of linguistic variation in late
medieval English and Scots have time-depth as well as geographical spread:
variation is the result of diffusion of change. In Historical Dialectology synchrony and diachrony
meet. Saussure was well aware of the
interdependence, indeed inseparability, of the two dimensions of Space and Time
in relation to Linguistic Geography:
Geographical
diversity should be called temporal diversity […] geographical differentiation
is pictured completely only when projected in time” (Saussure [ed. Culler] 1974: 198).
We can capture this idea by
conceiving of our data as existing in Spacetime. I shall present a methodology for presenting
and analysing data in a unified Spacetime continuum. This will also include discussion of the
notion of continuum and a practical characterization of it for use in
the atlases.
Variation is an inherent part of language and it is variation
that interests us in the text languages which provide the data for the atlas
corpora. But how can we conceive a
linguistic system and parts of systems at different levels of linguistic
analysis (orthography, phonology, morphology, syntax, semantics) in language
communities which exhibit considerable degrees of variation? I take it, albeit perhaps unconventionally,
that the linguistic knowledge (langue / competence) of language
users should account for a capacity both to produce and understand
variation. Since history has bequeathed
us an accidental sample of linguistic witnesses in the form of texts, the
witnesses more often than not offer incomplete parts of systems and subsystems. Using the methodology devised for a computer
version of the fit-technique (Benskin 1980, 1991; Williamson 2000), I
will suggest a reconstructive method that can supply lacunæ in paradigms and
linguistic data sets in an “enriched” dialectal Spacetime matrix.
Benskin,
Michael 1980. An analytical technique for computational placing of Middle
English dialect material on a map.
Middle English Dialect Project Working Paper (unpublished MS).
-- 1991. ‘The “fit”-technique explained’. In:
Felicity Riddy (ed.) Regionalism in Late Medieval Manuscripts and Texts
(Cambridge: D.S. Brewer), pp. 9–26.
Williamson, Keith 2000 ‘Changing Spaces. Linguistic Relationships and
the Dialect Continuum’. In: Irma Taavitsainen, Terttu Nevalainen, Päivi Pahta
and Matti Rissanen (eds.) Placing Middle English in Context, Topics in
English Linguistics series (Berlin: Mouton de Gruyter), pp. 141–79.
back to ICEHD 1 preliminary programme
back to ICEHD 1 homepage