The Cambridge History of the English Language Volume 3 part 8 pptx

(Modern German has seen the re-institution of the /e/ v. /ε/ contrast on
the basis of <
ee> v. <a> spellings, at least for many speakers.)
The continuing importance of Latin and the absence of a well-defined
norm for English up to the eighteenth century meant that ‘good education’
became closely connected with ‘proper language’ comparatively late in the
social history of English. True enough, Thomas Elyot referred to the impor-
tance of choosing the right nurse to provide a pattern for proper pronunci-
ation as early as 1531 – but he stressed the importance of good Latin even
more (1531: 131–2). In the seventeenth century, lexicographers advertised
their books as guides to good English, and Locke stated that proper English
was a necessary part of the education of a gentleman. However, it was only
in the times of Lord Chesterfield that this concern became dominant. From
1737 on, he wrote a long series of letters to his son of which at least fifty deal
with questions affecting the English language (Neumann 1946):
The language of the lower classes is, of course, to be avoided because it
is full of barbarisms, solecisms, mispronunciations, and vulgar words and
phrases, all of which are the marks of ‘a low turn of mind, low educa-
tion, and low company ordinary people in general speak in defiance
of all grammar, use words that are not English, and murder those that
are’. (466, quoting Chesterfield’s Letters 701, 729)
It was only through works like Johnson (1755) and Lowth (1762) that
proper guidance could be provided on lingustic law and order. However,
writers of guidebooks realised that their efforts might well be thwarted by
the neglect or inability of the users. Trusler admits (1766: 18):
Though Humoursom, instead of Humorous, be chiefly heard among the low
People, (none of whom, in all Probability, will ever study this Book, to
learn good English) yet, as there are few bad Expressions used by the
Vulgar, but that sometimes make their Way into better Company, it is
proper to take Notice that the Word, which implies Comical, is Humorous,
and not Humoursom; the Signification of which last Word is Peevish,

Froward, Hard to please.
6.1.4.3 Demographic facts
Although there is of course no straight correlation between the currency
of dialect and its evaluation on the one hand, and urbanisation and density
of population on the other, a look at changes in demographic patterns
between 1800 and 1900 can serve to throw into relief the sociolinguistic
conditions that underlie my discussion. To the facts represented in figures
6.3 and 6.4 (from Darby 1973: 393, 676) should of course be added the
Regional and social variation
465
increase of mobility (aided by modern developments in transport), educa-
tion and communication.
The maps also indicate that in 1800 there was little chance for lower-class
urban dialects to develop outside London (if we assume that a population
of a certain size is necessary for such varieties to emerge), but that the sit-
uation had drastically changed by 1900.
Manfred Görlach
466
0 100 km
Persons per square mile
800 and over
400 – 799
200 – 399
100 – 199
Under 100
Population 1801
by registration districts
Figure 6.3 Population of England, 1801. Based on H. C. Darby (ed.) A New
Historical Geography of England, Cambridge University Press, 1973
6.1.5 The geographical scope: England and the problem of Wales, Scotland,

Ireland and America
At the beginning of the Early Modern English period variation in English
was a problem confined to England. Harrison, in his introduction to
Holinshed’s Chronicle, carefully distinguished between England and
Scotland, attributing three languages to each: English, Welsh and Cornish
Regional and social variation
467
0 100 km
Persons per square mile
800 and over
400 – 799
200 – 399
100 – 199
Under 100
Population 1901
by registration districts
Figure 6.4 Population of England, 1901. Based on H. C. Darby (ed.) A New
Historical Geography of England, Cambridge University Press, 1973
as against Scots, Gaelic and Norn (1577; text in Görlach 1991: 233–6), and
Mulcaster (1582: 256) has a similar (much-quoted) remark:
Our English tung is ofsmall reatch, it stretcheth no further then this
Iland of ours, naie not there ouer all.
In fact, the effective anglicisation of Wales did not start until the sixteenth
century, and Wales was still predominantly Welsh-speaking in the nineteenth
century, and Cornish survived until the eighteenth. From Early Modern
English onwards, the range of varieties of English therefore expanded in
coherent areal speech communities which had English as a second language
(ESL), with a gradual shift to native-language status (ENL) around the
fringes first, and (in Wales) a speedier change from the period of early indus-
trialisation onwards, i.e. after the end of Early Modern English. Where the

shift to English was completed, local forms of it may still be characterised
by accent, but have not developed into broad dialects, the language having
been transmitted in its standard form, through schools and books.
Scotland had developed a semi-independent standard before 1603, in the
times of the independent kingdom, on the basis of educated Edinburgh
usage. The question whether sixteenth-century Scots should be considered
as a language, or rather as a dialect of English (and therefore part of this
chapter) is impossible to decide unambiguously. When I had to decide
whether or not to include Scots in my Introduction to Early Modern English,I
tried to summarise the pros and cons as follows (Görlach 1991: 22):
On the one hand, Scots fulfilled the critera usually assumed to be consti-
tutive for a language:
1. It was a national language whose use coincided with the political bounda-
ries of the Scottish kingdom.
2. It had developed a literary/written standard.
3. The court at Edinburgh and the University of St Andrews provided a
norm of written (and presumably also of spoken) Scots.
4. There are several statements extant indicating that some users considered
Scots an independent language (cf. Bald 1926).
On the other hand, the weight of these criteria is diminished by the
increasing convergence of Scots with English in the course of the period;
and there are other factors which argue against independent language
status:
1. The reciprocal intelligibility of Scots and English was not seriously
endangered even when the two were furthest apart (in spite of the
remarks made above).
Manfred Görlach
468
2. Structural differences were most marked in phonology/orthography and
– in some texts – in lexis, but much less so in inflexion and syntax.

3. Educated speakers remained conscious of the common descent of Scots
and Northern English, and of the close historical relationship between
Scots and English in general.
It can therefore be argued that Scots is and always has been a sub-system
of English, whose incipient separation from Early Modern English was
slowed down as a consequence of political, economic and cultural factors
in the sixteenth century and finally blocked by the adoption of English as
the written (and, later, the spoken) language of higher prestige (cf. McClure
1994).
Ireland had an old (medieval) English-speaking community, which sur-
vived into the Early Modern English period (and right into the nineteenth
century) mainly in ‘The Pale’ just north of Dublin and in County Wexford,
where its archaic character was noted as early as 1577 when Stanihurst com-
mented upon it in his contribution to Holinshed. Further dialects (mainly
Western English and Scots) were transported with the settlers of the Ulster
Plantation, where they are still distinct as Mid-Ulster English and Ulster
Scots. Later anglicisation of Ireland, mainly from Cromwell onward, had a
non-standard English input, but without any regional bias; the more typical
features of Hiberno-English are due to a combination of incomplete
second-language acquisition by speakers of Irish Gaelic and the settlers’
and administrators’ dialects. Although the Irish brogue became a butt of
irony for educated London society in the eighteenth century (including
comment by Irish emigrants such as Swift) and thus came to be a stylistic
variety within British English (cf. the texts assembled in Bliss 1979), it is
not included in my discussion. (Also compare earlier stage Irish English
spoken by Macmorris in Shakespeare’s Henry V, 6.3.4.2 below.)
Dialects were also transported to the North American continent.
However, even where settlers came from mainly one area (the ‘Pilgrim
Fathers’ came predominantly from the Midlands) there was the expected
‘colonial levelling’ (Trudgill 1986 and cf. the early statements collected in

Matthews 1931) so that only in very isolated pockets did British dialects
have a chance to survive (such as Southwestern English dialects in out-of-
the-way Newfoundland fishing communities). In a few other places, the
provenance of the input may still be detectable (Scots and Irish in the
Appalachians/Ozarks, disputed dialect and Hibernian English features in
Caribbean creoles), but these components were fused with other elements
in the proverbial ‘melting-pot’ so that it now takes a historical linguist to
identify them.
Regional and social variation
469
Apart from Ulster, only Orkney and Shetland saw the expansion of the
Scots language, but even here this was replaced by school English in the
eighteenth century, as also happened in the Gaelic-speaking Highlands
after the abortive 1745 rising.
The spread of English within Britain and further abroad is, then, quite
different from the situation in extraterritorial German settlements – all
dialect-speaking, even though Standard German is normally available
through the church and the schools. The reason for the greater homoge-
neity of spoken English around the world is certainly partly that English
had achieved greater unity, at least as a formal written language, by the time
it came to be transplanted so that dialect speakers had a common denom-
inator of ‘correctness’ if they wished to conform linguistically.
6.1.6 Historical sociolinguistics and the problem of sources
Much of the best tradition of historical linguistics has always taken the
social and political realities of earlier stages of the language into account.
In this respect, books like Wyld (1936), Horn & Lehnert (1954) and
Jespersen (1909–49) are relevant to our topic. However, there has only
recently been a group of studies that actually claim the title of sociohistor-
ical linguistics; three of these studies cover our period and are at least partly
in our field:

Romaine (1982) is an attempt to correlate linguistic variables (indicative
of ‘anglicisation’) with sociolinguistic factors, here represented by four text
types in sixteenth-century Scots, her main concern being to account for
different distributions of relative pronouns. While the study is of impres-
sive depth and rigour, it fails to do sufficient justice to some sociohistori-
cal factors: for one thing, the Early Modern English ‘input’ is not analysed,
and further, it remains open what social distinctions the four text types are
taken to represent since we do not learn much about authors and their
intentions, addressees and patrons, formal restrictions deriving from text-
specific decorum or about the relevance of sources (in the case of the
translation here analysed).
Devitt (1989), on a quite similar topic, is an advance over Romaine, since
Devitt takes into account more linguistic variables and more text types,
which are interpreted as specimens of written communication within social
frameworks and functions as far as these can be reconstructed. Both
authors have, significantly, chosen a field where, with two related and lin-
guistically similar standard languages clashing under quite well-known con-
ditions, and amply documented, individual features can be plausibly
Manfred Görlach
470
ascribed to one of the two systems and ‘interference features’ be easily
detected, counted and interpreted with regard to the writer’s motives, and
possibly correlated with what is known about both writer’s and addressee’s
social status features.
Tieken (1987) was in a more difficult position when investigating the
social relevance of ‘empty’ do in eighteenth-century texts. As school-book
knowledge has it, the feature ought to have been dead by the end of the
seventeenth century, and it generally was in respectable prose. (Pope
objected to it even in verse, where it served metrical convenience.)
However, while we can certainly agree with the author that do in affirmative

non-question sentences is an indicator of informality, it is difficult to pin
down its social relevance. (Compare, for lexis/phrasing, Wyld’s (1936: 22)
remarks on the surprising outspokenness and absence of genteel diction in
many upper-class women in the early eighteenth century.)
The number of smaller studies illustrating the impasse of sociolinguis-
tic interpretation of historical data – even for quite well-documented com-
munities – could be multiplied. One of the more impressive ones is Labov’s
claim to have identified possible mergers, semi-mergers or non-mergers of
vowel phonemes in sixteenth-century educated London English (1975; the
topic is taken up in Harris 1985 and in Lass this volume). We are forced to
admit that there cannot have been general mergers of, say, ea [
ε] and ai [
1
]
in the sixteenth century, if the two sounds have separate histories in the
later standard. However, it is quite a different matter how this non-merger
is to be interpreted in social terms. Hart, one of the astutest observers of
the emerging standard and certainly aware of sociolinguistically relevant
distinctions, does have this merger (if we can trust his painstaking
transcription) and claimed it was part of ‘the best English’, possibly
becoming entrapped in a self-introspective (dialect-based?) fallacy – or that
there was still more than one form of ‘best English’.
But even where the evidence is very clear, its social interpretation may not
be. How much tolerance is there towards linguistic variation in a given society,
and can we assume that there are universal or common regularities in
degrees of acceptance, or must the choice offered within a system charac-
terised by variability sooner or later lead to functional differentiation – how
long can variation be neutral? And how much credit can we give to the state-
ments of language-conscious participant observers, many eager for linguis-
tic law and order, and some coming to the battlefield with axes to grind?

Generations of schoolmasters and orthoepists insisted on a phonic rep-
resentation of written <
h
>, even when the majority of speakers had /
ø
/
or /
f
/ in word-final position or before /
t
/ (type high: height, laugh: laughter).
Regional and social variation
471
Now Elizabeth I, not only competent in at least five languages, but also edu-
cated by Roger Ascham as her tutor, spells the word rhymes as righmes in her
translation of Boethius of 1593 (text in Görlach 1991: T20/99) – an indi-
cation that she could not have pronounced words like night with /
x
/, or else
she would not have put gh in where it did not belong (taking igh as an unam-
biguous spelling for the diphthong developed from ME /
i
/). What do we
make of the evidence that the queen herself did not pronounce her /
x
/s
properly (cf. Lass this volume: 3.5.1)? What of the fact that the very unusual
spelling cannot have seemed correct even at a time when spelling was much
more variable than later? (We do in fact have indications that spelling did
not matter as much, as a sociolinguistic indication of proper education, as

it did from the eighteenth century on with people like Lord Chesterfield.)
Phonology provides a long list of features that were stigmatised in
certain speech communities and periods but are not so now, or vice versa.
Instances are:
/h/-dropping, which came to be discredited only from the late eight-
eenth century onwards (Milroy 1983)
the /oi ϭ ai/ merger in noise: nice, which was apparently common in edu-
cated speech in the eighteenth century, but became provincial in the
nineteenth (cf. Lass this volume: 3.4.2.6);
/-n/ for -ng, which was common in unstressed syllables, and remained so
in conservative RP as late as about 1900, but is now a highly stigma-
tised feature in most formal varieties of English.
On the other hand, pronunciations that led to modern standard British
English (RP) great and dance were strongly disliked when first used.
All this serves to show that scholars can go badly astray if they extrapo-
late uncritically from their own speech to describe earlier phases of English
– or other geographical varieties.
6.1.7 Reconstruction
There is nothing wrong, in principle, with using diachronic evidence to
reconstruct earlier dialects (although the linguist cannot hope to reconstruct
full systems of subvarieties – let alone their social and stylistic ‘meaning’).
It will be helpful to show a few cases in which the principle has been use-
fully applied (or is awaiting judicious application) to
Early Modern English
varieties:
(a) The publication of LALME (McIntosh et al. 1986) seems to cry out for
a scholar to relate the Middle English data to the nineteenth- and twen-
tieth-century data collected in EDD and modern atlases, fitting in the
Manfred Görlach
472

evidence we have of Early Modern English regional differences. (But
note that LALME concentrates on spelling to the virtual exclusion of
other linguistic levels.)
(b) The stability of dialect boundaries, or rather, the shift of the ‘isophones’
of individual features has been adequately treated for ‘the Southwest’,
‘Watling Street’, the ‘Humber–Ribble line’ and the English–Scottish
border. For periods less well documented, as Early Modern English often
is, we can extrapolate the movement of receding or advancing features.
(c) Transported varieties of Early Modern English, spreading first to
Scotland and Ireland, later to America (both to New England and to the
Caribbean), have developed into new varieties of English whose features
– through all the haze of language contact and colonial levelling – throw
some light on varieties of Early Modern English, especially where the
geographical and social provenance of settlers, and their educational and
religious backgrounds are fairly well known.
(d) Structural insights derived from regional and social variation in English of
various periods can not only supplement our data, but also interpret them.
Whether all this should be subsumed under sociohistorical linguistics is a
matter of label. One of the most convincing illustrations of the principle
appears to be Lass’s conclusion, based on the development of the long
high vowels in northern dialects, that the Great Vowel Shift cannot be
explained by means of a drag-chain hypothesis (Lass this volume: 3.3).
All evidence of this kind has to be handled with very great care, but it seems
that the chances of successful reconstruction are much better for Early
Modern English than for other periods, since so much more linguistic and
sociohistorical data are available and can be correlated.
6.1.8 The contribution of Early Modern English dialects to the standard language
There is no comprehensive study of the topic, so any account must be
incomplete and partly conjectural. The following generalisations would
seem to need verification very urgently, but they can still contribute towards

the setting up of hypotheses for comprehensive investigations.
The processes by which the English standard came to be established at
a very early date (compared with other northern European countries)
suggest that the ‘fusion’ happened in the fifteenth century, and that regional
features had no great chance of being accepted into the standard after
1500; such ‘influences’ are rather to be expected, especially as far as pro-
nunciation and syntax are concerned, in ‘vertical’ diffusion, i.e. they reflect
an interchange of coexisting social and stylistic varieties within London
English.
Regional and social variation
473
Lexis is slightly different. An individual item (or a variant pronunciation
of an existing word, say kirk for church) can easily be adopted from a dialect
if there is some justification for it, e.g. in the designation of a local object
or custom. However, the number of such internal loanwords is very low in
English: this is certainly a consequence of the way standardisation pro-
ceeded, and the scant evidence is therefore in stark contrast to the great
number of regional items in Modern German. Three types of such bor-
rowings can be distinguished:
(a) A few words were restricted to Early Modern English dialects, but later
lost their regional flavour, apparently via adoption into the supraregional
language: clever, tidy.
(b) Other words came through literature where they were often used to des-
ignate dialect (e.g. ‘northernisms’ in Spenser), but when adopted into the
common language shed both their regional and literary connotations: hale
(from Spenser), weird (through Shakespeare), glamour, gruesome, raid (from
Scott – the richest source).
(c) Finally, there were a great number of words referring to plants, tools, etc.
in the language of farmers, artisans and sailors. Although most of these
were not accepted into Standard English, there are quite a few that

remained in use in the special jargon of the trades, with or without addi-
tional regional restrictions (cf. expressions for ‘vessels’: fat/vat, keg, keeve
‘tub’, South West).
The apparently very limited interchange (in contrast to the vast influence
of Standard English on the dialects) is an important indicator of the
inequality of the standard and various forms of non-standard language in
Early Modern English times, and also of the circumstance that other focal
areas – comparable, for instance, with the capitals of small dukedoms in
Germany – were lacking in England after 1400. Note that the interchange
was much more frequent between sociolects (their speakers being in more
frequent contact); for instance, words might become acceptable when they
lost their stigma through the rise to power of the speakers with whom they
were associated (cf. 6.4).
6.2 Attitudes
6.2.1 Introductory comments
How did people react to variation in Early Modern English, and how far
did they correct their speech, selecting from the varieties available the
ones most appropriate to situation and purpose? (At least in written
usage we must take into account that the educated were guided by the
Manfred Görlach
474
rules of classical decorum as formulated by handbooks on rhetoric
(6.2.3).) The surviving sources are, however, scant, and many of the
statements are vague or ambigious. An instance is the anecdote about Sir
Walter Raleigh, who is reported not to have accommodated to courtly
London speech patterns. The account is worth quoting in full (from
Wyld 1936: 109):
he heard from Sir Thomas Malet, one of the justices of the King’s Bench,
who had known Sir Walter, ‘that notwithstanding his so great mastership
in style, and his conversation with the learnedest and politest persons, yet

he spoke broad Devonshire to his dyeing day.
However, this statement is found, some fifty years after Raleigh’s death,
in Aubrey’s Short Lives (not published until 1898). Was it really a dialect,
or just a regional accent? Did he, for instance, use South-West dialect
words and inflections, or broad ‘Zummerzet’ pronunciation, or was it
only the non-London quality of his vowels that made his speech remark-
able? And can we assume that Raleigh did not bother to conform
because he was too powerful, whereas, by contrast, all the others did?
How well attested is the claim that Gabriel Harvey ‘took speech lessons
to acquire a more elegant pronunciation’ (Holmberg 1964: 11, who inter-
prets this as seeming ‘to imply that people were conscious of educational
or social differences in pronunciation’). What does James VI’s linguistic
conversion mean in sociolinguistic terms when he published his works
written after 1603, as James I, in impeccable English – and what was his
spoken English like? Linguistic misdemeanour was criticised often quite
severely, and sometimes by colleagues taking offence at each others’
usage; Gil was not in agreement with Hart, and Nashe and Harvey
fought vigorously over alleged inkhorn terms (see Nevalaihen, this
volume). Attitudes towards correct, or rather incorrect, language did not
soften after 1660. At that time authors not only criticised their contem-
poraries, but started accusing Shakespeare and Jonson of linguistic mis-
takes – sometimes anachronistically. The peak of such efforts came,
however, in late eighteenth-century Edinburgh, when authors weeded
out each other’s Scotticisms. However, such explicit statements certainly
represent the tip of the proverbial iceberg – we can safely assume that
prescriptive attitudes in schools and the pressures that linked social
upward mobility with ‘proper’ speech were much more important. But
most of this evidence is lost, and it is therefore as important as it is time-
consuming to get as close as possible to a reconstruction of what caused
linguistic stigmatisation.

Regional and social variation
475
6.2.2 The status of Latin, French and English
6.2.2.1 Introduction
Variation in English, and attitudes towards the vernacular, cannot be seen
independently of views on Latin and French, and that in two ways: first, it
was Latin’s well-regulated system of spelling, inflection and syntax that was
looked upon as a model of elegance, and there was also the parallel that
Englishmen saw in the efforts of the French to create a national standard
language in the sixteenth century – with the model function of Latin
replaced to some extent by French after 1660. Secondly, statements about
Latin and French by native writers served as guidance when decisons about
elegance and correctness had to be taken for English. Cicero, Horace and
Quintilian were among the most-quoted authors when archaisms and neol-
ogisms, dialect or low words, debatable inflectional forms or lack of con-
gruence, stylistic adequacy, the structure of sentences and logical
arguments were discussed. Again, much of this was not explicitly stated,
since what the most eminent Latin authors had advised was common
knowledge among the educated.
6.2.2.2 Latin
Jones (1953) has provided a comprehensive account of the competition
between Latin and English in the sixteenth century. Continued use of
Latin, many renaissance writers argued, would keep English crippled with
regard to the more respectable and sophisticated registers particularly of
written uses. In order to make English into a fully functional national lan-
guage, its uses had to be extended into domains associated with Latin, such
as the sciences – against the opposition of those who, with good reason,
pointed out the risks of such a development. Mulcaster (1582), who in his
spirited plea for the vernacular asked ‘Why not all in English?’, summarised
such objections to English under the following headings (cf. Görlach 1991:

T8):
(a) English is needless (but look at the time wasted in the learning of foreign
languages!)
(b) it is coarse and uncultivated [uncouth] (but look at the state of Latin before
Cicero made an effort to polish it!)
(c) it is of ‘small reatch’ (but it is indispensable in England and a perfect
means of communication)
(d) there is not much learning preserved in it (a fact that could be changed
once scholars started using it)
Manfred Görlach
476
(e) there is no hope of ‘anie greatnesse’ (but this is partly due to the fact that
England is a ‘Moanarchie’ and ruled by Christian religion, facts that do
not encourage liberty and eloquence)
(f) the use of English will hinder international scholarly communication
(but scholars in Romance-speaking countries have begun to write in their
vernaculars) . . .
Such discussions were made possible by the increasing self-confidence of
English speakers after about 1575; they were less surprising after the emer-
gence of Britain as a world power (a development that started in 1588, with
the defeat of the Armada), and they became unnecessary with the com-
pleted emancipation of English in the seventeenth and eighteenth centu-
ries.
With Latin’s change of status from a second to a foreign language after
1660, its use became a badge of humanist education in the arts, a
qualification that had lost its pragmatic functions and which was cultivated
for its own sake.
The legacy of Latin, as far as variation in Early Modern English was con-
cerned, was tremendous (as it was for other European vernaculars, too):
(a) Adoption of Latin-based words made possible full terminologies for

scholarly disciplines, the sciences and technology
(b) Transfer of Latin syntactical patterns created respectable varieties of
written English that were capable of higher degrees of abstraction and
complexity for registers which became increasingly dominated by the
written or printed word. (For an identification of such syntactic calques
see Görlach 1991: 126–30; and see Adamson, this volume.) The proper
mastery of these styles became the object of language education and
thereby a sociolinguistic mark of the well-bred in contradistinction to the
less educated, the slow reader, bad speller and clumsy user of syntax.
Latin (on top of developments that would have happened as a result of
the functional expansion in any case) thus also helped to distance written
from spoken English, that is define more clearly the most important
functional divide among the varieties according to use.
(c) The study of Latin (as stated above) provided the pattern of what a well-
ordered standard language should be like: a system with no alternatives
left in spelling, with clear (and, if possible, rational) rules in syntax and
with a vocabulary that showed a clear distinction between the ‘nice’
words on the one hand, and the colloquial, low, technical and dialectal on
the other – types that only rarely found their way into the writings of
Cicero – or of Addison and Steele. This corrective function of Latin did
not end with its dominance in the grammar schools, but the nature of its
impact changed: with the Age of Reason, Latin was increasingly seen as
Regional and social variation
477
deficient itself, and so its rules were copied only where they agreed with
the demands of logic. After 1660, grammarians started objecting to
contradictions and redundancies more than in earlier periods, so that
double negation and double comparison, redundant pronouns and lack
of concord came to be stigmatised, and unambiguous marking of
adverbs, and distinctions between who/which/that; will/shall; past/perfect;

simple and aspectual forms came to be demanded. A comparison with
Latin structures will easily show that most, but by no means all of these
developments had an equivalent in Latin.
The high prestige of Latin made misuse possible and indeed likely:
Latinate syntax and vocabularly came to be the hallmark of writers unduly
stressing their classical education, sometimes bordering on unintentional
parody. When Day (1586: 38) wanted to illustrate excesses of such a style
(‘A ridiculous maner of writing’) he did not invent a specimen (as Wilson
in 1555 had done with his inkhorn letter), but went straight to a medical trea-
tise, A. Boorde’s Breuiary of Helthe of 1547 (text in Görlach 1991: T44):
Egregiouse doctors, and maysters of the eximiouse & Archane Science
of Phisicke, of your Vrbanlyte Exasperate nat your selues against mee,
for makyng of thys little volume of Phisicke. Considering that my pre-
tence is for an vtilitie and a common wealthe. And this not onely, but also
I doe it for no detriment, but for a preferment of your lawdable science,
that euerie man shoulde esteeme, repute and regard the excellent facul-
tie. And also you to bee extolled and highly preferred, that hath and doth
studie, practise and labour this sayd Archane science, to the which none
inartious persons, can nor shall attayne to the knowledge: yet nothwith-
standing fooles and insipient persons, yea and manie the whiche doth
thinke themselues wise (the which in this facultie be fooles in deed) will
enterprise to smatter &c.
On the other hand, incompetent handling of Latinisms became a distinc-
tive feature of the sociolect of those who had ‘small Latin and less Greek’.
The use of inkhorn terms in the speech of students and other more educated
persons is contrasted with the bungling malapropisms characteristic of the
lower sociolects (6.4.1).
6.2.2.3 French
French had lost its second-language functions as the medium for the law,
the higher administration and much of written everyday communication

including private letters in the course of the fifteenth century, but its use
and its prestige as the most important modern foreign language remained
Manfred Görlach
478
unaffected in the sixteenth and seventeenth centuries (cf. Kibbee 1991).
After 1660, when the court returned from exile in France, and when the
cultural dominance of la grande nation left hardly any European nation
unaffected, English added ‘courtly language’ to its repertoire of registers
– and its fashionable misuses. Complaints about unjustified influence of
French on English last from the 1660s to the 1750s, and they range from
indirect comment in the form of satire on language in Restoration
Comedy to Dr Johnson’s outspoken warning that English would become
a dialect of French if this influence continued (see also Nevalainen, this
volume).
6.2.2.4 Purism
Purism, defined as the deliberate attempt at reducing the number of
foreign words or avoiding their use altogether, is not a modern phenome-
non. Renaissance authors could point to Cicero’s and other Latin authors’
objections to a too liberal use of Greek where Latin expressions were avail-
able – or could be coined. Purism does depend on a certain degree of lin-
guistic nationalism or at least a conviction that unrestricted borrowing
reduces to some extent the expressiveness of one’s language, a potential
which ought to be cultivated by writers, teachers and other linguistic pace-
setters.
There was little of such feeling in the sixteenth century; or at least there
were not many people who flatly rejected the borrowing of foreign vocab-
ulary. The more thoughtful users of the language (such as Sir Thomas
Elyot) tried to use foreign words only where they could not do without
them – according to Ciceronian precepts. One wonders why so few
attempted to translate (or paraphrase) terminologies into English (Golding

for medicine, Lever for logic, Puttenham for rhetoric), and why Cheke, who
so vociferously demanded an ‘unmixed and unmangled’ vernacular, was
himself so inconsistent. There was certainly no institution that could have
implemented a puristic language policy, should it have ever been formu-
lated, and there do not seem to have been many who saw the sociolinguis-
tic consequences of excessive borrowing, as Wilson did as early as 1553
when he warned against a division of English:
Therfore, either we must make a difference of Englishe, and saie some is
learned Englishe, and other some is rude Englishe, or the one is courte
talke, the other is countrey speache, or els we must of necessitee, banishe
al suche affected Rhetorique, and vse altogether one maner of language.
(1553: 87
r
= Görlach 1991: T4/76–81)
Regional and social variation
479
However much attention the unjustified overuse of loanwords in inappro-
priate situations found in many circles, lexical expansion by borrowing
from Latin was not affected in general. Puttenham (1589: 120) warned
against the diction of the universities ‘where Schollers vse much peeuish
affectation of words out of the primatiue [classical] languages,’ and the
story of the student who went to the shoemaker to have ‘two tryangyls and
two semy cercles’ put on his ‘subpedytals’ (text in Görlach 1991: T 51)
neatly illustrates the situation:
Of the scoler that bare his shoys to cloutyng
In the vnyuersyte of Oxonford there was a skoler y
t
delytyd mich to speke
eloquent english & curious termis/ And cam to y
e

cobler wyth hys shoys
whych were pikid before as they vsyd y
t
seson to haue them cloutyd &
sayd thys wyse/ Cobler I pray the set me .ii. tryangyls & .ii. semy cercles
vppon my subpedytals & I shall gyue the for thy labor/ This cobler
because he vnderstode hym not half well answerid shortly & sayd/Syr
youre eloquence passith myne intelligence/ but I promyse you yf ye
meddyll wyth me/the clowtyng of youre shone shall coste you .iij. pence.
¶By thys tale men may lerne y
t
it is foly to study to speke eloquently
before them that be rude & vnlernyd. (1526)
(Also note the mother-wit of the shoemaker, whose status is characterised
by his dialectal plural shone.) Moreover, the frequency of malapropisms
appears to indicate that even lower-class speakers loved the sesquipedalian
word. Since most of the evidence occurs in literary texts one could rightly
question its authenticity, but Cockeram’s dictionaries (1623), meant for
simple speakers and offering refined equivalents for short Saxon words,
point to the same conclusion – this was no period, obviously, for puristic
achievements.
Which individual language was objected to apparently depended on the
conspicuousness of the imports. Although Wilson (1553: 86r) ridiculed the
use of ‘oversea language’, which included ‘Angleso Italiano’, by those
returning from the Continent, his main concern (and that of his contem-
poraries) was with Latin – contrast, two hundred years later, the obsession
with ‘Gallic’ loans harboured by Dr Johnson, whose own style is an
epitome of Latinate diction.
All this shows that it is not enough to count tokens of loan words but
that stylistic appropriateness and correct use of foreign words can indeed

be pointers to idiolects of social significance. Since the function of lan-
guages and the social structures correlated with their use changed so dra-
matically in the period under discussion, the situational context must in all
cases be very carefully interpreted.
Manfred Görlach
480
6.2.3 Classical views of correctness: elegance and decorum
Grammar, according to the classical distinction, took care of the ars recte
dicendi (correctness), while the ars bene dicendi, the art of beautiful speech,
was the field of rhetoric. The system was highly formalised and was made
teachable in many handbooks of Latin and of English, in works by, for
instance, Peacham (1577) and Puttenham (1589). According to these rules,
an expression could not be correct in all contexts, but only appropriate for
a certain function, the correlation being established by decorum. The prin-
ciple is not confined to literary language, or to written uses, but it was nor-
mally discussed in books devoted to ‘The Arte of English Poesie’. Scholars
comparing the state of English with that of classical Latin necessarily
found that English was deficient on many counts. There were gaps in lexis
(and in syntactic possibilities), but stylistic flexibility was sorely lacking as
well. It is interesting to see that the legitimacy of loan words was ‘proved’
by the arguments that they added synonyms to the language (creating copi-
ousness of speech; see Adamson this volume) and that they sounded better
than native words (adding euphony and metrical or rhyming possibilities).
Obviously, to overcome the inelegance of the vernacular was considered
as urgent by writers of literary texts, as was the need to fill lexical gaps for
writers of expository scientific prose. Such problems are even more
evident when poets used the vernacular for a particular genre for the first
time, that is, could not follow in the footsteps of a predecessor in accor-
dance with the principle of imitatio. Edmund Spenser, who was the first to
write pastoral poetry, made his shepherds speak a new language composed

of archaisms, dialect and classical allusions – which provoked Ben Jonson’s
remark that ‘Spenser, in affecting the ancients, writ no language.’ And he
did not, but it was not his intention to use a form of English that had ever
been in use: certainly broad dialect, however sociolinguistically realistic for
shepherds, would not have fulfilled the tenets of decorum for pastoral
Poetry. Even Milton, as late as 1667, had no epic style to fall back on for his
Paradise Lost, so he tackled, in an English style modelled on Virgil and
Horace, ‘Things unattempted yet in Prose or Rhime’ (Paradise Lost I. 16).
Like writers of medical handbooks in the vernacular, who stressed how
much easier it would have been for them to write in Latin, Milton would
have had an easier time using Virgilian epic Latin – and would not have
incurred Addison’s scathing remark that he had built a temple of brick.
Milton possibly illustrates, in his verse and prose, the effect of decorum
most convincingly: a large proportion of his vocabulary is either restricted
to his poems (e.g. ‘hard words’; the use of existing lexemes with meanings
Regional and social variation
481
adapted from Latin equivalents) or found only in his prose writings (e.g. the
‘low’ words used for the political and religious polemics of the Civil War).
It is perhaps correct to say that Milton and the Civil War period represent
the decisive stage in the process leading to the sharp distinction between
poetic diction and non-literary language, thereby anticipating the tenets of
classicism (Davies 1970).
Such hard-won stylistic expansion, which made possible the correlation
with levels of formality, stylistic sophistication and appropriateness for
individual genres was utilised in the eighteenth century. There is probably
no period in the history of the English language when the influence of ‘the
best writers’ on what is considered correct and appropriate has been so
great as it was between 1660 and 1760 (cf. Collins 1972, Görlach 1990b:
31). Although the influence of prescriptive grammarians was also consid-

ered (cf. 6.2.4), their main impact came rather after 1750. Moreover, the
influence exerted by the writings of Dryden, Swift, Addison and Steele
before 1750 was different since they provided models to be imitated rather
than rules to be followed.
6.2.4 Prescriptive and descriptive attitudes – reason and usage
Grammar, it was widely held in the Renaissance, was an attribute of Latin;
English, lacking norms in spelling, pronunciation, morphology and
syntax, was considered to be largely irregular and, many would have
claimed, incapable of being reduced to a proper system and orderliness. If
it ever were, this would have to be on the basis of the established rules of
Latin. This conclusion was natural for grammarians who believed not only
in the superiority of classical Latin but also that the structure of all lan-
guages was, ideally, identical. It does not come as a surprise, then, that the
beginnings of grammar teaching in English are characterised by a tradition
based on Latin and with a strong prescriptive slant. This tradition lasted
well into our own days – Sir Winston Churchill still remembered his school
grammar describing nonexistent English ‘cases’ arranged in paradigms
according to Latin models. But even where the match with Latin was less
close, the prescriptive attitude remained: the increase in the number of
English grammars between 1600 and 1800 used in Michael’s (1970) study
is impressive: only thirty-four works date back to the seventeenth century
(and many of these were in Latin), but nine, seventeen, thirty-five, eighty-
one and ninety-three, a total of 235, come from the five twenty-year
periods of the eighteenth century respectively. It is safe to say that the
majority of these books tended to become shorter and more prescriptive
Manfred Görlach
482
all the time, obviously satisfying the need of many readers for unambigu-
ous advice.
The other, more or less complementary, tendency was to follow usage. In

order to do this, it had to be clear which variety should be chosen as a model.
The formula of classical Latin was to be guided by the consent of the
learned (Ben Jonson’s translation of Quintilian’s consensus eruditorum), and it
was clear from early on that educated London use was the only possible
choice. Although some might argue in favour of the greater Germanic
purity of northern English, ‘yet it is not so Courtly nor so currant as our
Southerne English is’, as Puttenham summarised common opinion in 1589
(121). He was even more explicit in stating which sub-types of Southern
English ought to be avoided (cf. Görlach 1990: 99), namely the language of:
(a) the people in the ‘marches and frontiers’ and ‘port townes’ (because of
language mixing)
(b) the universities (because of Latinate diction)
(c) rural areas
(d) the lower classes (‘of a craftes man or carter, or other of the inferiour
sort’) regardless of region
(e) the old poets (‘for their language is now out of vse with vs’)
(f) ‘Northern-men . . . beyond the riuer of Trent’ (because even the language
of the well-educated of this region shows some interference from the
northern dialect).
However, it was not at all easy to establish a consensus, and the later history
of English grammatical thinking shows that most authors came to accept
usage only grudgingly, including those who paid lip-service to Horace’s
famous dictum about ‘vse and custome’, which are (in Puttenham’s trans-
lation, 1589: 123) ‘onely vmpiers of speach’. Ben Jonson was one of the
few early grammarians who not only included a section on syntax in his
Grammar (posthumously printed in 1640), but also diverged from Latin
rules when describing English structures.
However, if we review the major developments of English syntax in the
seventeenth and eighteenth centuries, we find that most have no parallel in
Latin structures (cf. Knorrek 1938, and Rissanen this volume):

(a) the completion of functional (fixed) word order (notably free in Latin)
(b) the regulation of the syntactical uses of do (which has no equivalent in
Latin at all)
(c) the semantic distinction between past and present perfect (tense distinc-
tions in Latin are completely different – in fact almost contrary)
(d) the consolidation of aspectual distinctions (there is no formally equiva-
lent aspect in Latin).
Regional and social variation
483
All these developments began before 1660, and the first two were almost
complete by that time. We can only state that they ‘happened’ against the
rules of Latin-based grammars, whereas many syntactic transfers from
Latin (such as sentence linking with initial which and some types of parti-
cipial constructions; cf. Rissanen this volume) had all but disappeared by
the Restoration – use and custom, only umpires of speech, appear to have
won, at least in these respects, and as regards the first half of the period
here discussed.
The eighteenth century inherited a largely ordered grammar from
Early Modern English – but the system did not always agree with logical
premises. In cases of unsettled usage, writers like Johnson would have
preferred to apply reason, but he, too, realised how harmless a drudge he
was: in a much-quoted passage (1755) he pontificated on lesser: ‘A barbar-
ious corruption of less, formed by the vulgar from the habit of terminat-
ing comparatives in er; afterwards adopted by poets, and then by writers
of prose’, to which he added in later editions: ‘still it has the authority
which a mode originally erroneous can derive from custom’. Lowth’s dis-
approval of the irregularity of good and bad is even more strong-worded:
‘They are in general words of most frequent and vulgar use, in which the
caprice of Custom is apt to get the better of analogy’ (1769: 59, 104;
quoted from Leonard 1929: 141–2 – not in the first edition of 1761). In

fact, almost all the important eighteenth-century writers reflecting on the
state of the English language discuss the problem of the two opposed
principles of reason and usage, preferring the one or the other, or
looking for compromises between them (cf. the informative chapter
‘The appeal to usage and its practical repudiation’ in Leonard 1929:
139–65).
6.2.5 Views on the vernacular
It is common in times when a standard language is being established for
dialects to be stigmatised as the speech of those who cannot do any
better. England was no exception, and since the standard came early and
was quickly implemented, the discrediting of dialect use was quick and
dramatic. It is significant that the proportion of dialect texts compared to
what was intended as standard writing cannot be more than one in a thou-
sand before 1660, and mentions of dialect are also quite few. Most of
these warn against the use of dialect words or grammatical structures, but
there is Gil’s (1621: 19) remarkable statement that dialect is admissible in
poets:
Manfred Görlach
484
The use of dialects is permitted of all writers only to poets, who refrain
from employing it in general, except when they infrequently use north-
ern speech for the sake of rhyme (? rythmi) or euphony, because it is
sweetest, oldest and purest, since closest to the language of our ances-
tors. (translation M.G.)
Negative attitudes changed very slowly. No help was to be expected
from Scots: whereas the language of official documents changed slowly
to English forms by 1660 (Devitt 1989), literary texts such as James I’s
or Drummond of Hawthornden’s were almost completely anglicised.
However, there must have been the beginnings of a change of attitude
from the early seventeenth century onwards regarding the dialects of

northern England, which is likely to be connected with the appreciation
of their Germanic character (cf. the Puttenham and Gil quotations
above, and the attitudes of ‘Saxonist’ scholars such as Camden,
Verstegan and Lisle who praised the great age and the Germanic roots
of English).
It is therefore no coincidence that John Ray (cf. 6.3.3) found so much
support for his collections of dialect words in the north and that the first
‘gentlemanly’ dialect poem, George Meriton’s of 1683, comes from
Yorkshire (cf. 6.3.4); both Ray’s scholarly interest in and Meriton’s use of
dialect for ‘literary’ purposes are very early compared to other European
countries.
In the eighteenth century interest in, or at least tolerance of, dialect
appears to have further increased – if it conformed to the Augustan
idea of decorum. The so-called revival of Scots as a literary medium
by Allan Ramsay and others happened after the shock of the union of
the parliaments in 1707, i.e. the loss of what was left of Scottish polit-
ical independence; but it also filled a niche of pastoral, satiric and
comic poetry, allocated to an ‘anti-standard’ in the Augustan frame-
work of decorum. Whether the great number of dialect poems of
eighteenth-century England were stimulated by the Scottish model is
impossible to say. However, it is quite clear that the underlying attitude
was that dialect ought not to interfere with the standard: in conse-
quence, dialect words were normally excluded from Johnson’s diction-
ary of 1755 (or were clearly marked if they are there); Grose’s Provincial
Glossary of 1787 is not a proper ‘dictionary’ and it is also a complement
to his collection of cant (Dictionary of the Vulgar Tongue, 1785). Also, the
mention of dialect in grammars, most detailed in Adams (1799), is nor-
mally in the form of a warning against the use of these forms (cf. 6.3.2,
6.3.3).
Regional and social variation

485
6.2.6 Views on low speech (including occupational jargon and cant)
It is no surprise that attitudes towards the lower sociolects are even more
critical than towards dialect – which could at least have a nostalgic touch of
the Golden Age and rural simplicity about it. Puttenham’s phrasing is
remarkably straightforward when warning the prospective writer against
following ‘the speach of a craftes man or carter, or other of the inferiour
sort . . . for such persons doe abuse good speaches by strange accents or ill
shapen soundes, and false ortographie’ (1589: 120). With such views cer-
tainly dominant at the time, it is quite remarkable that Thomas Harman, a
member of the landed gentry, as early as 1567 undertook to collect speci-
mens of cant (cf. Görlach 1991: T9) – but also note the Elizabethan fasci-
nation with the underworld that is evident from cony-catching pamphlets
and similar texts. However this may be, it is remarkable that many diction-
aries continued the tradition by including Harman’s material until this and
much additional lexis were combined in Grose (1785).
Occupational jargon is different; there is a practical need for it, and its
evaluation very much depends on how narrowly the idea of a standard is
considered to be confined to the ‘respectable’ words of the liberal arts: the
first dictionaries of specialised language (such as Manwayring 1644)
appeared in the seventeenth century, at a time when this type of diction was
also being praised by Sprat (1667, cf. Görlach 1991: T17), whereas Dr
Johnson was more sceptical about such vocabulary (1755) – all depends,
obviously, on how deeply an observer is steeped in the tradition of litera-
ture and the humanities.
6.3 Evidence of geographical variation
6.3.1 The evidence on individual linguistic levels
In the section below, data on variation in Early Modern English are criti-
cally reviewed; much of the evidence discussed is, however, not unambig-
uously dialectal in the narrow sense. Just as Wilson (1553) and Puttenham

(1589) combined their warnings against broad dialect, lower-class speech,
inkhornisms and archaisms as varieties not to be used by the poet, and just
as Spenser mixed regionalisms, sociolectal and archaic elements in the
speech of his shepherds, so writers of grammars and dictionaries were not
clear about what should be marked as ‘dialectal’ and what as ‘low’. Probably
their indecision was justified: since the view of dialect as a non-standard
variety had become common after the establishment of the new standard,
it is needless to argue whether Henry Machyn’s ‘written Cockney’ should
Manfred Görlach
486
be filed under the one or the other, for it is the uneducated and regional
speech of a sixteenth-century Londoner. My section, then, focusses on
region, but does not exclude social class and stylistic features, when they
cooccur with dialect or represent the functions that regional differences
served in communication.
6.3.1.1 Spelling
Correct spelling has, at least since the eighteenth century, assumed great
sociolinguistic importance, but there is considerable evidence that the stig-
matisation of spelling mistakes is a new development. True enough, if
uneducated speakers like the undertaker Henry Machyn spelt according to
their lower-class pronunciation, the spelling would be stigmatised too; his
Diary of 1557 (cf. the excerpt in Görlach 1991: T40) is riddled with exotic
spellings which must have given him away as uneducated even in this early
period (whent, cronnacull, howsswold; cf. the discussion in Wyld 1936: 141–7).
Puttenham must have had people like him in mind when warning against
‘ill shapen soundes and false ortographie’. However, Queen Elizabeth’s
spelling righmes mentioned above indicates that rather unpredictable
spellings were found in educated writers; and letters and diaries of the six-
teenth and seventeenth centuries are still full of them – and they were often
used by female writers. Many people must have thought that to spell cor-

rectly was the secretary’s job, and not necessarily a badge of liberal educa-
tion. This attitude changed only in the eighteenth century when many of
the educated became obsessed with orthography.
The case is, of course, different with dialect words. Since no ortho-
graphic norms existed for such words, writers had to make the spelling up,
and since there is no unambiguous correlation between sounds and letters
in English, the results are not always satisfactory. In consequence, the faulty
or misleading notation of dialect evidence is one of the major concerns of
every historical dialectologist.
6.3.1.2 Pronunciation
Pronunciation has always been the most conspicuous marker of regional
or social varieties, as comments from antiquity onwards show. In English,
too, one of the first extended comments, in Trevisa’s translation (1387) of
Higden’s Polychronicon, notes regional differences in ‘sownynge of speche’,
and enlarges on the unpleasant pronunciation found in York in particular,
which also affects intelligibility (cf. the excerpt printed in Wakelin 1977: 34;
Regional and social variation
487
Lass this volume). Puttenham, although concerned with the written lan-
guage appropriate for the poet, mentions pronunciation with regard to
lower-class London dialect. Not that the educated agreed on what was ‘the
best English’: and many of those who pontificated about correctness were
indeed hampered by their own dialect backgrounds, as Dobson (1968) in
particular was careful to point out when discussing the reliability of the
statements of individual orthoepists.
Hart, who very carefully observed and described what he put forward as
the norm for others to imitate, had in his ideolect a merger of mead:maid
words which is only occasionally attested for sixteenth-century London –
and which was certainly not the basis of the later standard which has the
merger of mead:maid and maid:made (cf. Görlach 1991: 69 and the discussion

in Lass this volume).
Other writers on
Early Modern English
norms who were possibly
influenced by their native dialects include: Peter Levins (1570; there is ‘abundant
evidence of Northernisms’, though his speech ‘was certainly accommodated in
many ways to the language of the South’, Dobson 1968: 21, 24); in John Cheke’s
speech there seem to be some traces of his Cambridge origins (45); William
Bullokar’s language seems to reflect his Sussex origins (105–8); and Richard
Mulcaster has a full chapter (125–7) devoted to the Northernisms in his speech.
In an age when learning standard pronunciation for non-standard speak-
ers meant paying for an elocution teacher – a phase that lasted right into
the late eighteenth century, when the method was adopted by well-to-do
Edinburgh citizens – the norm spread but slowly and not very efficiently.
It is compelling diachronic evidence that Puttenham’s localisation of the
best English (found in London and sixty miles around it) is almost identi-
cal with the area of present-day Southern English (the area of /
ba␪ tb/
south of Northern and Midland /
ba␪ tυb/); cf. figure 6.5.
Such accommodation to a prestige pronunciation was difficult because
the correlation between graphemes and phonemes was not perfect (and still
is not): Hart (1569) had adduced various reasons for a reform of the orthog-
raphy, one of them being ‘for straungers or the rude countrie English man,
which may desire to read English as the best sort vse to speake it’ (Görlach
1991: T6: 30ff.). Even when dictionaries became more common, they did
not, before Walker (1791), include information on correct pronunciation.
6.3.1.3 Morphology
In a language with an inflectional morphology so greatly reduced as was
that of English after 1450, inflection would not seem to be a field in which

Manfred Görlach
488
dialect or other non-standard features will be prominent. Deviation from
the standard can, however, be expected where
(a) historical forms were retained in dialect, unaffected by correctness as
taught in the schools or as found in books: and where,
(b) by contrast, regularisations were generalised in dialects where the fossil-
ising influence of the written tradition, with the authority exerted by texts
Regional and social variation
489
0 100 km
northern / /versus southern / /
in butter, cousin, cucumber, cut
northern front versus southern back vowel
in swath, wasp, water, wash, quarry
northern short versus southern long vowel
in chaff, grass, last, laugh, shafts
northern short versus southern long vowel
in broth, cough, cross, frost, off
related to Puttenham's statement about
'London and 60 miles around'
York
Manchester
Bristol
Cambridge
London
Trent
Figure 6.5 The geographical basis for ‘good’ English. Transition area between
southern England and northern England, based on Puttenham (1589) and
Glauser (1991)