Xem mẫu
- Norman Blake
It would be dangerous to look for any significance in this switching.
The same applies to Gawain at the castle of Hautdesert. Generally, he
addresses his host with 7,eforms,though occasionally he relapses into a
pou form. Although in the temptation scenes the lady uses pou forms to
Gawain, he uses 7,eformsto her except for an occasional lapse. Shortly
before she offers Gawain a ring in the third temptation scene, he relapses
into using one py:
'Now iwysse,' quo)? J>at \vy3e, 'I wolde I had here
J>e leuest )?ing for )>y luf )>at I in londe welde.'
(Tolkien & Gordon 1967: 1801-2)
('"Now truly", said that knight, "I wish that for your sake I had the
most valuable thing on earth that I own here".')
It has been suggested that his lapse 'makes the mental turbulence
dramatic - is a concrete indication of how far the lady has driven him'
(Evans 1967: 42), though this is perhaps to read too much into a single
example which may have arisen for a variety of reasons including scribal
corruption.
6.4.6 Other fourteenth- and fifteenth-century authors, such as
Chaucer and Malory, are able to exploit this difference between the two
forms. By Malory's time it may well be that the thou of intimacy was
becoming less common, for it has been largely abandoned by the
Pastons in their correspondence, and this has left it as a mark of
contempt or as a social marker. Field has pointed out a notable use of
thou forms in Le Morte Darthur (Field 1971: 101). When Gareth first
appears in Arthur's court the king is sufficiently impressed by the young
man to useje forms, which would be the polite form of address in any
case. When Gareth asks for food and drink for a year, the king switches
to thouforms,for the request seems to suggest someone of a low social
rank who has come to beg. Later, when it is revealed that Gareth is his
nephew the king reverts to ye forms. This switching of forms indicates
that Malory was aware of the potential for social and other implications
of the thou and you forms and was prepared to exploit them.
As for the Canterbury Tales Keiko Shimonomoto has examined the
relationship oiyou and thou forms with both forms of address and forms
of the imperative (Shimonomoto 1986). The use of the polite you forms
appears to be a characteristic of genre in so far as they occur regularly
in courtly romances. In the Franklin's Talejo# forms are used by all the
characters among themselves, though they use thou forms to gods and
538
- The literary language
goddesses and the clerk uses them in his final speech to Aurelius (CT 6:
897-902 [V.1613-18]). Since the clerk is not an aristocratic character,
his use of thou forms may indicate his status. He may behave
magnanimously by waiving his fee, but his language reveals him to be
of a different class. Terms of address also have a significance in
indicating the genre of the tale. Ladies in courtly romance can expect to
be called madame, whereas those in other tales are apt to be addressed as
dame. The latter is not used in tales like the Knight's Tale, but it is found
in the Reeve's Tale, where it is said of the miller's wife 'Ther dorste no
wight clepen hire but dame' (CT 1: 3948 [1.3956]). The forms of the
imperative, whether they have the base form or the ending -eth, seem to
be interchangeable in Chaucer. What is more significant in requests is
the overall syntax rather than the form in which the verb itself appears.
In more courtly language commands become concealed as gentle
requests or wishes, as is found in the Host's request to the Prioress to
tell a tale (CT 10: 447-51 [VII.447-51]). Inevitably there is switching
between you and thou forms, and in most cases it can be attributed to the
author attempting to manipulate an effective response on the part of the
reader. Because courtly romances use thejou forms as a marker of genre,
there are not many examples where a courtly speaker switches to thou
forms. When this happens, it may be interpreted usually as a sign of high
emotional tension. Arveragus switches briefly to thou forms in the
Franklin's Tale when he forbids Dorigen to reveal to anyone that he has
instructed her to keep her word to Aurelius (CT 6: 771-6 [V.1481-6]).
The switching is probably more characteristic of other tales such as the
Nun's Tale in which Cecilia at first addresses the judge Almachius with
you forms, but then switches to thou forms as her contempt for him
grows. Such large shifts in attitude are not characteristic of courtly
romances and hence shifting is less likely to be found there. Once again
one cannot be certain that all these switches are significant, and the
forms are liable to corruption in scribal transmission, but one can often
provide some check on the reasons for the change by considering forms
of address and the syntax of commands.
Both these examples indicate that writers were becoming more
sophisticated in their exploitation of linguistic features at the end of the
Middle English period. To some extent this was because linguistic
conditions were becoming more stable and also because increasing
attention was being paid to English style, which it was increasingly felt
needed to be elevated to make it a suitable vehicle for literary expression.
The Middle English period represents a series of adjustments to changes
539
- Norman Blake
taking place in the language and society generally. It is hardly surprising
that the literary language should appear to be somewhat fragmented and
that stability and cohesiveness should appear only towards the end of
the period.
FURTHER READING
Useful background studies may be found in Curtius 1953, Minnis 1984 and Eric
Auerbach, Literary Language and its Public in Late Latin Antiquity and in the
Middle Ages (London: Routledge and Kegan Paul, 1965) who explain some of
the features of literary language and the reception of literature from a European
standpoint, dealing particularly with the Latin background. H. J. Chaytor,
From Script to Print (1945), tackles vernacular, particularly French, literature,
and Atkins (1943) and Pamela Gradon, Form and Style in Early English Literature
(London : Methuen, 1971) deal more specifically with the English background.
A work dealing only with the end of our period is A. J. Gilbert, Literary
Language from Chaucer to Johnson (London and Basingstoke: Macmillan, 1979).
The continuity of prose style was first proposed by R. W. Chambers, On the
Continuity of English Prose from Alfred to More and his School (London: Oxford
University Press, 1932), but this approach was questioned by N. Davis,
'Styles in English prose of the late Middle and early Modern period', Langue et
litte'rature (1961: 165-84) and by R. M. Wilson 'On the continuity of English
prose', Melanges de Hnguistique et de Philologie (1959: 486-94). More recent work
on Middle English religious prose includes Riehle (1981) and Elizabeth Salter,
Nicholas Love's ' Myrrour of the Blessed Lyf of Jesu Christ' (Salzburg: Institut fur
englische Sprache und Literatur, 1974). A more general book on prose style is
Robert K. Stone, Middle English Prose Style (The Hague and Paris: Mouton,
1970). On Malory there are good books in M. Lambert, Style and Vision in Le
Morte Darthur (New Haven, CT: Yale University Press, 1975), P. J. C. Field,
Romance and Chronicle (1971), and Larry D. Benson, Malory's Morte Darthur
(Cambridge, MA, and London: Harvard University Press, 1976). On the
Lollards and the Wycliffite translation of the Bible one may consult Anne
Hudson, Lollards and their Books (1985) and C. C. Butterworth, The Literary
Lineage of the King James Bible 1340-1611 (Philadelphia: University of
Pennsylvania Press, 1941). On Caxton see Blake (1968).
On the continuity of the Old English style through to Middle English see
Tatlock (1923) and Blake (1969b). For the language of La3amon the more recent
study of Iwasaki (1986b) supplements earlier studies such as Tatlock (1923),
H. C. Wyld, 'Studies in the diction of Layamon's Brut', Language 9 (1933):
47-71,171-91 and 10(1934): 149-201, and D. Everett, Essays on Middle English
Literature (Oxford: Clarendon, 1955, 28-45). A major study of Lajamon's
language and style remains an important desideratum. Several essays in David
540
- The literary language
A. Lawton (ed.), Middle English Alliterative Poetry and its Literary Background
(Cambridge: Brewer, 1982), deal with style, as does T. Turville-Petre, The
Alliterative Revival (1977). The style of Sir Gawain and the Green Knight is dealt
with in Marie BorofT, Sir Gawain and the Green Knight: a Stylistic and Metrical
Study (New Haven, CT: Yale University Press, 1962). Also important is R. A.
Waldron, 'Oral formulaic technique and Middle English alliterative poetry',
Speculum 32 (1957): 792-804, and Crosby (1936). The formulas used in
alliterative poetry are collected in J. P. Oakden, Alliterative Poetry in Middle
English (1930-5).
For the influence of translation the standard work is still Workman (1940),
though it now needs updating. Work on the influence of features of Latin and
French style may be found in Richardson (1984), Bornstein (1978) and Burnley
(1986). The use of glosses to help appreciate the register of words is dealt with
in Blake (1988). Works which try to focus on the colloquial elements of style
are Schlauch (1952) and Clark (1978c). Many of these matters are treated also
in N. F. Blake, The English Language in Medieval Literature (1977). The question
of applying some modern linguistic advances, such as politeness theory, to
medieval texts is looked at in Sell (1985a and b).
The language and style of Chaucer is examined in Burnley (1979), J. D.
Burnley, A Guide to Chaucer's Language (1983) and R. W. V. Elliott, Chaucer's
Language (1974). Gregory Roscow, Syntax and Style in Chaucer's Poetry
(Cambridge: Brewer, 1981) is largely concerned with syntactic features; for
vocabulary see J. Mersand, Chaucer's Romance Vocabulary (1937). Two books by
Michio Masui can also be consulted with profit: one a collection of essays,
Studies in Chaucer's Language of Feeling (Tokyo: Kinseido, 1988), which deal with
various aspects of Chaucer's style; and the other his The Structure of Chaucer's
Rime Words: an Exploration into the Poetic Language of Chaucer (Tokyo:
Kenkyusha, 1964, but frequently reprinted) remains valuable for showing how
Chaucer exploited rhyme. For Chaucer's followers one may consult Norton-
Smith (1966) for Lydgate and J. C. Mendenhall, Aureate Terms: a Study in the
Literary Diction of the Fifteenth Century (1919) for Chaucer's influence on the
fifteenth century.
Features of romance style including traditional formulas and expressions are
treated in Wittig (1978) and Baugh (1959). For particular studies, that by
Burrow (1984) shows how Chaucer exploited the romance style in Sir Thopas.
For a study of the language of the Middle English lyric the two fullest studies
are still D. Gray, Themes and Images in the Medieval English Religious Lyric
(London: Routledge and Kegan Paul, 1972) and Rosemary Woolf, English
Religious Lyric in the Middle Ages (Oxford: Clarendon, 1968). For the relation of
alliteration to the lyric see M. Fifield, '13th century lyrics and the alliterative
tradition', JEGP 62 (1963): 111-18.
- 7 ONOMASTICS
Cecily Clark
7.1 Sources and methodology
7.1.1 General principles
Names, whether of places or of people, have by definition a distinctive
standing vis-a-vis the language at large. Although ultimately derived
from elements of common vocabulary (not necessarily that of the
language they currently grace), they have become emptied of their
original etymological denotation; and this is true even for those whose
form still coincides with that of the related lexical items: no-one expects
to find cattle wading across the river at Oxford and, should a Mr
Butcher actually be in the meat trade, the coincidence almost excites
mirth.
On the one hand, this semantic detachment promotes cross-cultural
survival: some Present-Day 'English' place names are traceable to
Celtic forms at least two millennia old, a few even suspected of going
back to pre-Celtic times; some 'English' baptismal names have Hebrew
origins. On the other, it lays names open to phonological attrition, for
no more of any form need survive than is required for acting, in context,
as an unambiguous signal or pointer. Name compounds are thus subject
to early obscuration, to having their unstressed syllables reduced more
drastically than similar ones of analogous ' meaningful' forms, and to
being 'folk-etymologised' (Lass 1973; Coates 1987; Colman 1989a and
b; Clark 1991). As well as complicating the etymologising process, this
makes name material an unreliable guide to the incidence and the
chronology (though not the nature) of general sound changes; it raises,
indeed, a possibility of there having been specifically onomastic changes,
related to the general ones but carrying them further (see further below,
pp. 593-4).
542
- Onomastics
Name studies are also distinctive in that they stand upon what is - in
conventional terms — the boundary between 'linguistics' and 'history'.
Onomastic source-materials are mainly ones otherwise associated with
socio-economic and administrative history. Likewise, the aims and the
findings of onomastics bear at least as closely upon cultural, social and
economic circumstances, often, indeed, on settlement history, as upon
linguistic developments: a reminder of how artificial the conventional
compartmentalisation of disciplines is (see further Clark 1990).
7.1.2 Source materials
For onomastics, and especially for anthroponymy, placing the notional
break between 'Old' and 'Middle' English at 1066 corresponds with a
certain reality. What it means for the specific sorts of name will be
considered under the relevant heads. Contrasts bearing upon general
methodology involve the range as well as the volume of source material
available. Whereas study of Old English naming, personal naming in
particular, is hamstrung by dearth of material, for the Middle English
period there is so much documentation extant — much of it as yet neither
published nor onomastically searched - that work can proceed only
selectively, and therefore provisionally. As always, each type of source,
and often each individual document, requires separate assessment.
The sources for name study of all kinds normally consist of
administrative records; and for the Middle English period this means
that most name forms found are at least perfunctorily Latinised (for
some problems of interpretation that this raises, see below, pp. 548-9).
The vast bulk of records surviving from this period is in itself of
moment, for it reflects a growth of bureaucracy at all levels from central
government to manor court (for the period before 1300, an excellent
general introduction is given by Clanchy 1979; detailed references will
be given under the separate heads).
For the opening of this new period, the chief group of sources is the
same as for the close of the Old English one: Domesday Book (DB) and
its 'satellites', the fruits of the great inquest initiated by the Conqueror
in 1086 (vol. I, pp. 453-4; for recent scholarship see Sawyer 1985 and
Holt 1987, and for further reading Bates 1986). The focus of study has,
however, shifted. The fact that DB offers the earliest extant record of
many place names ceases to be central, in so far as most 'major' English
place names had long since become fixed (see below, pp. 588-91).
Personal naming was, by contrast, at this very time beginning to
543
- Cecily Clark
undergo its most far-reaching changes, those that were to result in the
present-day system. In this context, the evidential value of DB is
enhanced by its consisting of two strata: one representing 1086 and the
other tempus regis Edwardi, that is, 1066 or shortly before. For each date
the principal landholders, under-tenants as well as tenants-in-chief, are
specified; and DB thus provides, for the socio-economic classes in
question, two contrasting yet comparable name corpora. Despite the
scope thus offered, only the earlier corpus has so far enjoyed intensive
analysis (Feilitzen 1937; cf. Dodgson 1985a; for the 1086 stratum
Hofmann 1934 is interesting as a pioneer effort but its premises, and in
consequence its conclusions, are now outdated).
Interpreting DB material is never easy. The two main volumes —
known as 'Great' and 'Little' DB - represent, in varying degrees,
recastings of the preliminary returns made by the several panels of
circuit commissioners. Recasting meant risk of scribal error, and some
forms are now explicable (if at all) only by comparison between the main
texts and corresponding passages in the various 'satellites', such as
Liber Exoniensis, Inquisitio Comitatis Cantabrigiensis, Inquisitio E/iensis, t he
Domesday Monachorum of Christ Church, Canterbury, and others (see,
e.g. H. B. Clarke's paper in Sawyer 1985). Errors apart, orthography is
a problem here. Old English spelling traditions, although to be
maintained in some quarters for a further thirty-five years or more, were
largely disregarded by the DB scribes, who, engaged as they were in
compiling a Latin record, usually adopted Latin conventions ill-suited
to the English sound system (see, e.g., Clark 1984a: 100-3). That the
difficulties are best regarded as orthographic must be emphasised:
approaching DB name forms, or those of any other post-Conquest
records, through speculation as to 'Anglo-Norman sound-substitution'
imports a needless additional level of uncertainty (see below, pp. 548-9,
and 592-4; also Clark 1991).
DB represents only the first fruits of the new bureaucracy. Many
series of governmental records followed, including numerous quasi-
national ones, such as the Hundred Rolls of 1279 (see Cam 1930 and
Kosminsky 1956: 7—46), the many sets of Lay Subsidy Rolls and
especially those running from 1290 to 1332 (see Beresford 1963: 1—7;
also Willard 1934) and the Poll Tax Rolls of 1377, 1379 and 1381
(Beresford 1963:19-29). In principle, such records not only covered the
whole country in uniform style but did so with a socio-economic scope
far exceeding that of any pre-Conquest ones. No series, however,
survives in its entirety. All are, besides, circumscribed by their
544
- Onomastics
compilers' aims, listing only individuals liable to tax and mainly
therefore heads (90 per cent of them male) of prosperous households
(see, e.g., Ekwall 1951-.passim);even the Poll Tax Rolls, which do often
name servants and other subsidiary adult members of households, omit
not only children under fourteen, beggars and successful tax evaders but
also wives (see, Owen 1984: 221—34). Consequently, no tax-roll
furnishes a fair sample of women's names, and few embrace the least
prosperous members of society. Explicit genealogical information is
sparse. What such records do offer are localisations — to a village or, in
large towns, to a particular ward (they therefore give forms of all
relevant place names) — and, except with the flat-rate Poll Taxes, fiscal
assessments based on relative degrees of prosperity, together with
occasional indications of trades practised. Within its limitations, each
roll, or local section of a roll, provides a cross-section of personal-name
patterns, analysable in geographical and/or socio-economic terms (the
latter dimension thus far largely ignored by modern scholars); when, as
with the Lay Subsidy Rolls, there survives a chronological run of similar
lists for the same localities, the cross-sections can be built up into
diachronic patterns. In this chapter, two Lay Subsidy Rolls for London
that are fortunately available in an excellent edition (Ekwall 1951) will
be used to illustrate the name fashions current among well-to-do
burgesses of ca 1292 and ca 1319.
The Middle English period is thus the earliest for which docu-
mentation is ample enough to permit of viewing personal naming in a
socially stratified geographical and chronological framework: a possi-
bility that gives anthroponymics primacy among sociolinguistic studies.
For, alongside the governmental records, there also survive others
which, less systematic though they are in coverage, put flesh on the
statistics, sometimes indeed allowing identification of particular tax-
payers and the compilation of thumbnail biographies for them (see the
notes to Ekwall 1951: passim).
Thus, there survive from ca 1200 onwards voluminous records -
each sort again in its own way selective — from the various types of law-
court: the King's Bench, the county assizes and the periodical eyres
dealing with the graver crimes, the coroners' courts, the manorial courts
and also the church courts responsible for matrimonial causes and some
other kinds of case (editions of some of the earliest of these records have
appeared among the publications of the Selden Society and in some of
the county record series). The collections oimiracula appended to saints'
Vitae also in a sense belong under this head, in so far as constituting
545
- Cecily Clark
evidence for the canonisation process. The onomastic information
afforded by judicial records is, in comparison with that from tax-rolls,
sparse as well as random; but it is better backed by evidence of familial
links and social standing.
Most valuable of all are the many local archives, civic and manorial as
well as monastic, that survive, some going back to ca 1100 or even
earlier (for early borough records, see G. H. Martin 1960-4 and 1963;
and for examples, see, e.g., Bateson 1899 and Owen 1984; for the sorts
of material to be expected from a religious house, see, e.g., Owen 1976;
J. Martin 1978; Thomson 1980). For the twelfth century, there survive
schedules of peasants dwelling on the estates of abbeys such as Bury St
Edmunds (Douglas 1932: 25-44; Davis 1954; cf. Clark 1987a), Burton
(Bridgeman 1916), Glastonbury (Jackson 1882) and Ramsey (Hart &
Lyons 1884-93: III, 218-315) and others listing the urban tenants of
houses like Battle (Searle 1980: 52-9; cf. Clark 1980a) and Christ
Church, Canterbury (Urry 1967: 221-382; cf. Clark 1976a). After
ca 1200 borough records grow to rival the monastic ones: there are, for
instance, extensive gild-rolls surviving, together with supporting
information, from towns like Leicester (Bateson 1899: 12-35) and
King's Lynn (Owen 1984: 295-313; cf. Clark 1982a and 1983a). Also in
the thirteenth century there begin many series of manor-court rolls
detailing routine village business (Harvey 1984, and for examples, see,
e.g. Holt 1964 and Harvey 1974; DeWindt & DeWindt 1981 and Raftis
1982, although not themselves onomastic studies, make clear the
anthroponymical potential of the types of document used). Substantially
antedating as many of them do the main series of tax-rolls, these local
records afford the earliest extensive evidence extant for non-aristocratic
English personal naming (Clark 1987a; cf. vol. I, pp. 461—2); and at the
same time they offer authentic forms of place names, especially those of
the 'minor' kinds (see below, pp. 595-604). Again coverage is selective,
being centred on individuals (under 10 per cent of whom were, on
average, women) responsible for property and the obligations stemming
therefrom. At their fullest, on the other hand, these local materials
allow of studying names in the context of neighbourly and familial
relationships. Among sources of this kind often to be cited here are the
earlier of the two Bury St Edmunds estate surveys already mentioned
(Clark 1987a) and also the late-thirteenth-century Carte Nativorum or
'peasants' charters' from Peterborough Abbey (Brooke & Postan 1960;
cf. King 1973: 99-125 and Clanchy 1979: 34-5).
Choosing, from among this embarras de richesses, which type of source
546
- Onomastics
to investigate depends upon the questions to be asked; for each has
shortcomings as well as strengths. Thus, the Lay Subsidy Rolls cover
large parts of the country systematically and at precisely datable
intervals but, indications of relative prosperity apart, give little
background information. Their textual reliability is, besides, like that of
all the other sorts of record that survive mainly in copies at several
removes from the original, sometimes suspect (McClure 1973; Rumble
1980: xiv-xvi, xxv, and 1984: 42-4; cf. McKinley 1988: 22; Hunnisett
1971 on the similar unreliability of coroners' rolls). The London Lay
Subsidy Rolls have none the less been chosen as illustrative material to
be cited here, mainly because of their availability in an edition whose
rich annotation compensates for the documents' own bareness. In
general, the records likely to be soundest as well as most forthcoming on
matters of detail will be either originals, vi^. rough rather than fair
copies, or at least copies made by scribes familiar with the localities
concerned rather than by ones serving a remote bureaucracy. Such local
records, on the other hand, lend themselves less well than do the more
stereotyped materials, such as tax-rolls, to comparison with ones from
other localities or even from the same locality but of different date.
7.1.3 Statistics
The variability as well as the bulk of medieval onomastic source material
has long made the applicability to it of statistical analysis a vexed
question, and the increasing use of computers now makes it a pressing
one. Some scholars have doubted, mainly on the grounds that no two
samples are truly comparable, whether any such analysis of medieval
personal-name material can ever be valid (e.g. Michaelsson 1947 and
1954). Records do, however, as indicated, fall into groups based upon
similar types of selectivity, listing gildsmen, or prosperous burgesses, or
customary tenants, or people allegedly involved in crime, and so on;
and this seems to authorise either some cautious comparison of like with
like for different areas or different dates or, alternatively, for a single
place and date, a no less cautious contrasting of different socio-economic
groups (see Clark 1990: 62-4).
That said, all personal-name records — modern censuses not excluded
— carry an irreducible randomness due to the perpetual fluctuations of
population, and therefore of name patterns, caused by deaths and births.
To calculate percentages, as is sometimes done, to several decimal places
therefore gives only a spurious exactitude. Indeed, although a wisely
547
- Cecily Clark
chosen sample ought to be broadly representative of usages at the place
and time in question and among the social class concerned, to stand
upon small differences in name ratios must always be risky.
Name statistics can, besides, be based in at least two different ways:
either upon the number of individuals named or else upon the stock of
forms represented; and each mode has its own appropriateness.
Reckoning by individual name bearers can show, for instance, how
eagerly a new fashion is being taken up; but it does require accurate
distinction among those individuals and that, as long as by-name usage
remains thin or capricious, poses problems. Analysing the name stock is
simpler, but represents the swings of fashion less well; where this is
valuable is for revealing, say, the long-term contribution made by a
spent cultural influence to local cultural patterns.
The imbalance between the names of the sexes already noted as
characterising most types of medieval record means, furthermore, that
all name patterns may initially be best assessed in terms of men's names
alone (see below, pp. 583-7).
7.1.4 Languages of record
For this period even more than for the Old English one, name studies
depend mainly upon Latinised materials. From Domesday Book
onwards, some administrative currency also of (Anglo-)French may at
times need to be taken into account; but it must be borne in mind that
- except in the generations immediately following the Norman settle-
ment, and even then only among a small group of families, mostly
aristocratic ones, of immigrant stock - French was no less than Latin a
language deliberately learnt, not a cradle-tongue (Shelly 1921; Wood-
bine 1943; Roth well 1968, 1975-6, 1978 and 1983; Richter 1979 and
1985; cf. Clark 1991). To speak, therefore, as some do, of 'Anglo-
Norman scribes ignorant of English' is likely to be untrue, except when
some specific document of early post-Conquest date is involved (see
Clark 1987c). What constantly complicates modes of interpretation are
scribal intentions of writing in languages other than English.
Orthography is, as always in historical linguistics, a basic problem,
and one that must be faced not only squarely but in terms of the
particular type of document concerned and its likely sociocultural
background. Thus, to claim, in the context of fourteenth-century tax-
rolls, that 'OE p, 6 are sometimes written /, lowing to the inability of
the French scribes to pronounce these sounds' (Hjertstedt 1987: 45)
548
- Onomastics
involves at least two unproven assumptions: (a) that throughout the
Middle English period scriptoria were staffed chiefly by non-native
speakers; and (b) that the substitutions which such non-native speakers
were likely to make for awkward English sounds can confidently be
reconstructed (present-day substitutes for / 0 / and / 6 / , for what they
are worth, vary partly according to the speakers' backgrounds, often
involving, not / t / and / d / , but other sorts of spirant, e.g., / s / and / z / ,
or /f/ and /v/) Of course, i f - a s all the evidence suggests - (a) can
confidently be dismissed, then (b) becomes irrelevant. The ortho-
graphical question that nevertheless remains is best approached in
documentary terms; and almost all medieval administrative documents,
tax-rolls included, were, in intention, Latin documents (see Ekwall
1951: 25-6, 29—31). A Latin-based orthography did not provide for
distinction between / 9 / and / t / or between / 6 / and /d/, and so
spellings of the sorts mentioned may reasonably be taken as being, like
the associated use of Latin inflections, matters of graphic decorum
rather than in any sense connected with pronunciation.
Indeed, Latinate graphic decorum masks vernacular usages in many
ways. Baptismal names, as well as being equipped with Latin inflections,
were sometimes represented by conventional archaisms, such as
Radulfus / Raditlplms, abbreviated as Rad\ for the name spelt in the
vernacular first as Ran//, later as Raul or Rafe (cf. PDE Ralph,
traditionally [reif]). The descriptive phrases constituting early forms of
personal by-names and of 'minor' place names (w^. field names and
street names) were likewise commonly Latinised. In so far as it
underlines the artificiality of the record and also the pre-onomastic
status of the forms concerned, such translation can be salutary; but it
does create uncertainty. The naming of a man, as say, Robertas tincter
conceals not only the colloquial form of the baptismal name but also
whether the local term for 'dyer' was ME deter, dextere or litestere
(Fransson 1935: 104-6; cf. OE deagian ' to dye' and ME liten < Scand.
lita ' to dye'). It further leaves it unclear whether the phrase means
' Robert, who works as a dyer' or ' Robert Dyer/Dexter/Lister': a point
crucial to the history of family naming. Similarly, the Latinising of an
early street name as vicus tinctorum conceals the local term for 'street' as
well as that for 'dyer'. Because modern translators and indexers often
decide quite arbitrarily how to render such forms, onomastic work must
always be based on the original texts themselves. Even a printed edition
of the Latin text may prove a trap, if by arbitrary capitalisation it implies
onomastic status for phrases not yet possessing it (for further comments
549
- Cecily Clark
on some dangers of relying on printed sources, see Lofvenberg 1942:
xx v).
With Anglo-French materials, problems can be more subtle. From
about the mid-thirteenth century onwards, French (by then current in
England almost wholly as a second language) functioned as a secondary
language of record, sometimes interchanging with Latin within one and
the same document (Ekwall 1951: 26—8, 31—3). It was also the source of
many loan words widely current in Middle English. Judging the status
of French forms figuring in documents is thus a delicate business. Some
Gallicisation is patently superficial, as when a vernacular personal by-
name atte grene ' dwelling beside the village green' is rendered as de la
grene (Lofvenberg 1942: 82—3); even the more consistent de la verteplace
betrays its own artificiality. With French terms that had provided loan
words into Middle English, uncertainties grow: for instance, current
Middle English terms for 'blacksmith' included not only the native
smith but also the more specialised ferrour < OF ferreour and marshal <
OF mareschal, both meaning 'farrier' (see Fransson 1935: 142—5); and,
in so far as Present-Day English family names perpetuate Middle
English personal by-names (see below, pp. 577—83), the existence of
Farrer and Marshall alongside Smith implies that both loan words did
figure as vernacular by-names. So, forms such as le ferrour and le
mareschal found in English medieval records might either have been
reflecting colloquial usage or else have represented scribal rendering of
'the (shoe)smith'. On the other hand, some Present-Day English family
names certainly represent Old French terms that were apparently
unknown, or at most very rare, in the Middle English current
vocabulary: e.g. Rous(e) < ME (le) Rus (Mod.F Leroux), the Middle
English lexical equivalent of which was rare indeed; and this suggests
some specifically onomastic transmission of French by-names (see Clark
1985). Interpretation of French items figuring in the records thus
involves judging between multiple possibilities, with certainty seldom
attainable unless the records themselves happen to reveal relationships
between bearers of variant forms.
7.1.5 Caveat
The accounts to be given here of Middle English onomastic history will
represent only overcompressed summaries of what is accepted in the
early 1990s. Vast masses of records remain unanalysed, and findings
therefore remain provisional. This is one of the fundamental differences
550
- O nomastics
between the Old English situation and the Middle English one:
although analysis of the extant materials is not yet complete for the Old
English period either, those materials are so limited as, on the one hand,
to allow of one day being fully exploited yet, on the other, to put a
complete survey for ever out of reach; but for the Middle English
period documents are so plentiful that - except perhaps for the personal
names of the very poorest people, who escaped record unless accused of
crime — a truly comprehensive survey might well be possible, but only
in the very long run indeed. Development of family naming could, for
instance, be clarified by establishing tens of thousands of late-medieval
and early-modern genealogies. Early forms of field names and street
names could be collected and collated to form the basis of a systematic
history, interlinked with that of rural and urban developments in the
wider sense. No findings described here are to be taken as definitive. A
vast uncultivated tract of name material still lies available for the
breaking in.
7.2 Anthroponymy
7.2.1 A. change of system
Neither the structure nor the content of the Present-Day English
personal-name system owes much to pre-Conquest styles. The typical
Old English personal designation consisted of a single distinctive name
(or' idionym'), such as Dudda, Godgifu or Wuljstan; and only occasionally
was this supplemented by a qualifying by-name (usually postposed),
such as se'o dxge ' the dairymaid' or se hwita ' the white(-haired man)' (see
vol. I, pp. 469—70). A Present-Day English 'full name', on the other
hand, necessarily involves two components: the second denoting a
patrilinear family group and the first (which may consist of one unit or
of several), an individual within that group. In Present-Day English
usage, moreover, the familial, hereditary component is the crucial one
for close identification, whereas in Old English usage, as in early
Germanic ones generally, the idionym was central, any addition being
optional. Beside this total change of structure, it is trivial that, out of the
hundreds of Old English idionyms, only a handful, and those mainly
ones which, like Edith, Edmund and Edward, were associated with widely
venerated saints, are today represented among Present-Day English
first-names.
The change of system demands a change of terminology. The special
- Cecily Clark
term 'idionym', no longer appropriate, will be replaced by 'baptismal
name' ('Christian name' is needed for a more specific sense, neither
'forename' nor 'first name' is appropriate until family naming is well
established, and ' font name' is unidiomatic). For an optional identifying
component, the term 'by-name' will be retained. Only when continuity
between generations is demonstrable will the term 'family name' be
used ('surname' is rejected as insufficiently precise).
This twofold shift in English personal naming was a specifically
Middle English process. Among the mass of the population, the name
system of ca 1100 was still virtually the classic Late Old English one,
modified only by somewhat freer use of ad hoc by-names (Clark 1987a);
but by ca 1450 a structure prefiguring the present-day one had been
established, with hereditary family names in widespread use, though not
yet universally adopted. As for the ousting of pre-Conquest baptismal
names by what were virtually the present-day ones, that had been
accomplished by ca 1250. This series of changes involved several
convergent processes.
The shift away from single idionyms to combinations of family name
with baptismal name affected most of western Europe. One cause of it
may have been the general decline, in some areas manifest well before
1000, in the old Germanic custom of permutating the conventional
name elements (or 'themes': see below, p. 554) in such ways as
continually to create fresh idionyms. What was crucial, however, was
the subsequent reliance not merely on a finite stock of set forms but
largely on a very few disproportionately favoured ones. This latter
practice seems to have arisen spontaneously in many areas, for it shows
itself in the native name patterns of late-eleventh-century England as
well as, for instance, in those of pre-Conquest Normandy (Clark 1987a;
Fauroux 1961: index). A corollary of this was that, from the early
eleventh century onwards, by-names came everywhere to be more and
more needed for distinguishing between individuals of like idionym
(see, e.g., Aebischer 1924: 120-41; Beech 1974: 85-7; Clark 1987a).
Then practical convenience, essential to the growing bureaucracy and
sometimes abetted by family pride, soon led to passing of the distinctive
by-names from generation to generation, albeit at first mainly from
father to heir.
In England, the Norman Conquest complicated these processes.
Within a few years of it all but a few of the pre-Conquest landholders
had, as Domesday Book shows, been replaced by foreigners, partly but
by no means exclusively 'Normans'. New appointments of bishops and
- Onomastics
abbots likewise brought foreigners into positions of prestige. Some
English towns experienced influxes of merchants from Normandy,
France and Flanders (it was from such stock that St Thomas Becket
sprang). All these immigrants brought with them name fashions current
in their homelands; and these fashions so swiftly captivated the native
English that within hardly more than a generation baptismal names
characteristic of the settlers were beginning to appear among towns-
people and peasantry alike (Ekwall 1947; Clark 1987a). By ca 1250
these borrowed names had ousted virtually all the pre-Conquest ones;
and thenceforth the typically ' English' baptismal names were reflexes,
not of Old English ones like JElfgifu, Godivine, Le'ofpryd, Wulfstdn and so
on, but of typically 'continental' ones like (to give them in their PDE
forms) Alan, Alice, Christine, Geoffrey, John, Maud, Robert, Stephen, Susan,
Walter and William (for a classification of the sorts of name involved,
see below, pp. 556-8). Adoption of so many fresh forms brought,
however, no increase in variety. Already existing tendencies towards
disproportionate reliance on a very few of the many names available
were indeed reinforced, with the result that, for instance, the London
sections of the Lay Subsidy Roll for 1292 show John as accounting for
over a sixth of the men's names recorded and William for over a
seventh, the two together thus making up almost a third of the total
(Ekwall 1951: 35).
Patently, such baptismal-name patterns were scarcely adequate for
social identification, far less for administrative and legal purposes. By-
naming - that is, supplementation of baptismal names by phrases
specifying their bearers in genealogical, residential, occupational or
characteristic terms - became, in England as elsewhere, a general
necessity. Such specifying phrases had been in occasional use among
English people since well before the Conquest (see vol. I, pp. 469-70),
and all signs are that shrinkage of the name-stock would in any case have
soon compelled their general adoption. The Norman Conquest may
well, however, have acted as a catalyst to the process, in so far as it
brought in a new aristocracy among whom, as is clear not only from
Domesday Book but also from the early records surviving from the
settlers' homelands in Normandy, France and Flanders, use of by-
names, and of territorial ones especially, was already widespread, with
even some tentative movements towards family naming (see further
below, pp. 580-1). The prestige thus accruing to use of a by-name
would hardly have hindered wider adoption of them among the general
native population. Exactly how colloquial usages may have developed
553
- Cecily Clark
in this respect we cannot know. In the sorts of record that constitute our
main source material, use of an unqualified baptismal name had at all
events become by the early thirteenth century rare even in reference to
the peasantry. Nor is this development hard to understand: where rents,
taxes and duties were concerned it was in the lord's interest to ensure
that the individuals concerned were specified precisely; and, where
holdings, rights and inheritances were concerned, it was greatly in each
individual's own interest to be so specified. The processes of change
then spiralled: just as disproportionate favouring of a certain few
baptismal names had evoked recourse to by-naming, so in its turn the
universalisation of by-naming reduced checks upon the disproportion-
ate favouring of just a few baptismal names.
7.2.2 Insular name styles and their post-Conquest survival
The pre-Conquest English personal-name stock (to which the term
'insular' will be applied) had incorporated various strains reflecting
aspects of the country's social and political history (see vol. I,
pp. 456-68; also Clark 1987a, 1987b: 33-40, and Feilitzen 1937). The
fifth-century West Germanic settlers had brought with them a stock
that consisted, not of set names, but of name elements or 'themes',
mostly carrying heroic meanings, the permutation of which could
produce both simplex (' monothematic') and compound ('dithematic')
idionyms in endless variety. The Vikings who from the late ninth
century on settled north and east of Watling Street, in what became
known as the Danelaw, observed analogous yet distinctive North
Germanic naming customs, soon widely adopted in the districts where
they settled. The whole pre-Conquest period also saw continual, though
light, influence from continental West Germanic styles, which were
introduced by visiting clergy, by merchants and, during the Confessor's
reign, also by immigrant nobility. Along the borderlands, in Cornwall,
in the Welsh Marches and in Cumbria, Celtic styles partly survived. The
basic geographical patterns are clear, even though much remains to be
discovered about the distributions of particular items. Much less clear is
what forms, if any, social stratification of personal naming might have
taken (see vol. I, pp. 461-2).
Even apart from the outside influences just mentioned, pre-Conquest
usages had never been uniform or static. Sometimes it is asserted that,
except among the peasantry, simplex names and 'short-forms' of
dithematic ones lost favour after ca 900 (see vol. I, pp. 461-2); but such
554
- Onomastics
a generalisation now seems too sweeping. As already pointed out, the
crucial development, in England no less than on the Continent — and in
all areas partly, no doubt, as a consequence of familial theme
permutation — was that there grew up during the eleventh century a
grossly disproportionate favouring of just a few out of the many name
forms current. Thus, English personal naming had on the eve of the
Conquest been variegated, long since receptive of outside influences and
approaching a crisis of its own.
The post-Conquest swamping of native styles by foreign ones,
although virtually complete in under two centuries, was far from being
catastrophically sudden. Indeed, some of our knowledge of pre-
Conquest uses comes by extrapolation from post-Conquest materials,
these being geographically more representative as well as more plentiful
and socially more comprehensive than most earlier ones. Twelfth-
century records offer many previously unrecorded forms, feminines
especially, of authentic Old English kinds (Reaney 1953). How far any
of these late-recorded forms represented fresh creations can hardly be
determined; but certainly twelfth-century familial name patterns show
theme permutation as still practised to a limited extent. Moreover, the
new wealth of local records gives retrospective insight into localised
fashions, such as the currency at Canterbury of JElfbfah and Dunstan
(Urry 1967: 459, 457) and the favouring in Cumbria of names associated
with the former earls (Insley 1982). For Anglo-Scandinavian styles in
particular, the dearth of pre-Conquest records from the Danelaw means
that most relevant ones date from the twelfth century. Here, even more
than with the native Old English tradition, the most enlightening
studies are proving to be localised ones (e.g. Fellows-Jensen 1968;
Insley 1979, 1982, 1985a and b, 1987; cf. Clark 1982a: 52-5). These
show that average levels of Scandinavian influence varied from district
to district, being graded from 15-20 per cent in Bedfordshire and in
Suffolk, through 35-40 per cent in the east midlands and in Norfolk, to
60 per cent and more in Yorkshire and in north Lincolnshire: figures
which roughly correlate with the durations of Scandinavian lordship
over the districts concerned and also with the general levels there of
Scandinavian linguistic and cultural influence (how far they might
correlate with densities of settlement is a question that here is best left
unbroached).
Medieval baptismal-name usages and distributions can be further
investigated through those of patronymic by-names (see below,
pp. 568-9), with the advantage of then being able to exploit the
555
- Cecily Clark
plentiful documentation surviving from the later thirteenth century.
With caution, modern family names, many of which retain to this day
geographically circumscribed distributions, may also be pressed into
service; doing so does, of course, entail assuming in advance of the
discussion below (pp. 568—9) that family names of apparently patro-
nymic form adequately reflect early medieval styles of baptismal naming.
If that be granted, then fair survival of native Old English names into
at least the early thirteenth century, the date when fixing of family names
seems likely to have begun, seems implied by modern surnames like
Aylmer ( < O E JEdelmair), Edrich (Eadric), Goodwin {Godwine), Wooldridge
(Wulfrtc) and so on. Similar survival of Anglo-Scandinavian ones is
implied by Arkell ( < Scand. Arnke[ti]ll), Brand (Brandr), Gamble
{Gamall), Grimes (Grfmr), Thorburn {Porbjorn) and so on. Admittedly, the
distributions of the Present-Day English names give only the roughest
indications of what the medieval patterns might have been, and serious
study must therefore aim at working back - perhaps by genealogical
methods — to groups of clearly localised medieval materials. Of these,
an excellent example is given by the late-thirteenth-century Peter-
borough Carte Nativorum, which show that on the abbey's estates, for
which the tenth-century list of festermenn offered a name stock
comprising over 40 per cent of Scandinavian-derived items, the
patronymics of medieval peasants also included numerous Anglo-
Scandinavian forms such as Arketyl, Brand, Game/, Gubbe ( < Scand.
Gubbi), Harold (Haraldr), Teyt (Teitr) and so on (Brooke & Postan 1960:
passim; cf. vol. I, pp. 465—8).
7.2.3 The new baptismal-name stock
The name styles favoured by the Norman duke's followers are recorded
not only in the 1086 stratum of Domesday Book and in post-Conquest
English charters and chronicles but also in records concerning
Normandy itself and the other regions, including Brittany, Picardy and
Flanders, from which settlers were drawn (this wide background makes
the widespread use of 'Norman' as a code term for such names
misleading).
Throughout those regions, the dominant eleventh-century styles
were of West Frankish origins. Structurally, the names concerned partly
resembled the native Old English ones, in so far as likewise created from
a battery of'themes' that had at one time been fairly freely permutated.
Although not all themes were common to all the Germanic onomastic
556
- Onomastics
dialects, some were widely enough current to make a good few names
ambivalent, at least in their documentary forms: e.g. Wimund, formally
attributable equally to OE Wlgmund, Scand. Vigmundr or CWGmc
Wigmund. Moreover, CWGmc names like Wilhelm possessed a general
familiarity of structure that perhaps eased their adoption by English-
speakers. In the main, however, CWGmc names were distinctive. Partly
this was because certain of them were based on themes, such as Grim-/
-grim and Rod- < Common Gmc Hrod-, absent from the original Old
English system, although sometimes known through the Scand. as well
as the CWGmc forms borrowed from the late ninth century on. It was
due also to differentiation through the normal phonological processes of
those themes that were held in common, so that CWGmc Ans-
corresponded to OE 6s- (Scand. As-), CWGmc Aud- > Od-/Ot- to
OE Ead- (Scand. And-) and so on. Distinctiveness was also due to the
Continental West Germanic favouring of a wide range of hypocoristic
suffixes, some of which were foreign to native Old English usages (see
especially Marynissen 1986). With the disuse of theme permutation, the
set names thenceforth current evolved, no matter what their original
structure, as single units, subject to the various local sound laws.
There were, in addition, certain specifically 'Norman' types of name,
chiefly the Scandinavian ones handed down by the early-tenth-century
Viking colonists (see Adigard 1954). Some of these are — at all events, in
their conventional documentary forms — not easy to tell from the
corresponding West Frankish ones; and, in England, some are hard to
tell from their Anglo-Scandinavian equivalents. Although there are
philological rules to go by — such as that Ansketil, with Arts- by analogy
with the Frankish reflex of the theme, is Norman, whereas Osketel/-cjtel
is Anglo-Scandinavian - allowance has to be made, when dealing with
medieval English materials, for scribal (or even colloquial) substitution
of the more prestigious Norman variant. (In the early post-Conquest
period, on the other hand, substitution sometimes went the other way,
so that, for instance, the name of the Norman Turstein, archbishop of
York 1119-40, often appears under the Anglo-Scandinavian spelling
Purstan.) A minor, and historically related, group of' Norman' names
consisted of Irish forms brought by Vikings who had reached
Normandy via Dublin: e.g., Brian, Murdac, Muriel and Neil— the latter
Latinised as Nigellus, which formally was a diminutive of Latin niger
(Musset 1975: 48). Then, too, Normandy, like many parts of western
France as well, saw some currency of names, such as Alan, that had been
borrowed from the neighbouring Bretons.
557
nguon tai.lieu . vn