Content uploaded by Elzbieta Adamczyk
Author content
All content in this area was uploaded by Elzbieta Adamczyk on Jun 21, 2020
Content may be subject to copyright.
T D A O E
S-S D1
Abstract
The present study investigates the fate of one of the minor consonantal inectional
types in Old English, the s-stem paradigm, which is a continuation of the PIE neuter
*-es-/-os- stem declension. The focus of the study is on the developments, both
phonological and morphological, which contributed to the later gradual restructuring
and eventual demise of the s-stem declension in English. Crucial to the present
investigation is a diatopic perspective which is to offer a more transparent, detailed
and thus adequate picture of the realignments within the s-stem paradigm. Using the
evidence of corpus data from the dialects of Old English: Mercian, Northumbrian,
Kentish and West-Saxon, the analysis explores the tendencies characteristic of
the nascent restructuring process in each of the investigated dialects, with a view
to accounting for the most signicant disparities between them. The ndings of
the present qualitative and quantitative investigation, carried out on the available
textual material, strongly demonstrate that the Old English s-stem paradigm was
on the verge of collapse, with individual dialects testifying to different stages of
this process.
1. Preliminary remarks
The present paper focuses on the developments which affected the early English
nominal inection, bringing about far-reaching changes to the system inherited
from Proto-Germanic and occasioning the eventual loss of the original declensional
diversity. An appreciable degree of instability of the English nominal system is
attested in its earliest stage (i.e. the Old English period), and is well manifested in
the paradigms of minor, unproductive declensional types. The extent of the erosion
of these paradigms seems to vary, ranging from almost complete or advanced (as
in the case of u-stems) to moderate (as in the case of nouns of relationship) (cf.
Adamczyk 2009, 2010). One of the minor consonantal classes which suffered a
substantial degree of decomposition is the Old English s-stem nominal paradigm2,
1 An earlier version of this paper was presented at the First Triennial Conference of the
International Society for the Linguistics of English (ISLE) in Freiburg, 2008. I am very
grateful to the Hertie Foundation (Gemeinnützige Hertie-Stiftung) for its generous support
of the research for this paper. I would also like to express my deep gratitude to Professor
Arjen P. Versloot for his invaluable help with the data analysis, in particular, with preparation
of the corpus for a dialectal investigation, as well as for his constructive comments on the
content of the present paper. I would also like to thank Professor Marcin Krygier for his
insightful and detailed commentary on an earlier version of the paper. Finally, I am indebted
to the anonymous reviewer for his/her critical and valuable suggestions which certainly
helped improve the quality of the paper.
2 From the synchronic, Old English point of view, the class can be referred to as r-stems,
due to the regular operation of rhotacism in North-West Germanic, whereby *-es-/-os- >
Gmc. *-ez-/-az- > -Vr-. The group is not to be confused with the original (PIE) r-stem
declension, comprising kinship terms (nouns of relationship).
388 Elżbieta Adamczyk
containing a small group of neuter nouns, viewed as unproductive in the historical
stage of most of Germanic (cf. footnote 3). With a single exception of the fossilised
plural form children, where the original -r- (OE cildru) has been preserved till
present-day English, this inectional type was lost completely.3 Traces of the
process of the gradual decay of this inectional pattern are to be seen in the
frequent uctuation of the nouns between the inherited s-stem inection and the
novel productive inection of a-stems (masculine and neuter), best attested in the
texts of West-Saxon provenance.
The present study offers a comprehensive investigation of the pattern of retention
of the original s-stem inection in the Old English material, attempting to ascertain
the extent of the dissolution of this inectional type, engendered by the analogical
spread of the innovative inection at this early stage. Central for the present study
is the diatopic dimension, which is to afford insight into the niceties of the process,
and is, in fact, expected to have had a more general bearing on the pattern of
preservation of the original s-stem inection in Old English.4
2. The origin of the Germanic s-stem inection and its shape in Old English
The Germanic s-stems constituted a very small group of neuter nouns which can be
viewed as vestiges of the original PIE *-es/-os inection. This originally numerous
and important group of nouns (possibly still in Proto-Germanic (Prokosch 1939:
256)) must have been signicantly reduced by the time of the rst attestation of
Germanic. In stark contrast to the Gothic or Old Scandinavian pattern of preservation,
where traces of this inectional class are non-existent, Old English managed to
retain this declensional type, despite its rather scant attestation, vulnerability to the
working of extensive analogical processes and thus “imperfect” shape. Since the s-
stems were originally declined according to the pattern of other consonantal stems,
the Proto-Germanic *-ez-/-oz- stem paradigm can be reconstructed as presented in
Table 1 (after Lehmann 2005–2007, §3.2.6; cf. Ringe 2006, 278).
3 Cf. the pattern of development of this inectional type in Old High German, where it
turned out to be remarkably productive, with an early extension of the -ir- element, as a new
expansive plural marker, to masculine nouns of various inectional classes (for details see
e.g. Harnisch (2001), Kastovsky (1995), Wurzel (1989, 1992)).
4 As regards the nomenclature used in the present paper, two terms need to be claried
at this point in order to avoid ambiguity or confusion. Firstly, the term innovative, as used
in the present study, refers to the presence of traces of other inectional types, in particular,
of the a-stems, in the inherited paradigm. Accordingly, the “innovativeness” of the s-stem
paradigm is conned to the effects of interparadigmatic rather than intraparadigmatic
realignments, brought about by the working of analogical processes. Secondly, the paper
employs the term productivity of an inectional class in line with the denition provided by
Wurzel (1989, 149), whereupon a productive inectional class is characterized by: (a) ability
to acquire new words (borrowings and neologisms), (b) ability to attract words from other
inectional classes, and (c) resistance to losing words to other inectional classes. In contrast,
an unproductive inectional class is one which does not meet any of the abovementioned
criteria.
Towards a Diatopic Approach to the Old English s-Stem Declension 389
Table 1. A reconstructed Proto-Germanic *-ez/-oz stem paradigm56
singular plural
N. *lambaz6
G. *lambezaz
D. *lambizi
A. *lambaz
N. *lambazō
G. *lambizōn
D. *lambizumiz
A. *lambazō
Formally, the class of the Old English s-stems can be divided into two subgroups:
(a) nouns in which the attested nominative and accusative sg. appear without the
r-stem formative (e.g. lamb, cealf, dæg, hrēð), and (b) nouns in which the sufxal
-r is attested consistently in all cases, including the nominative/accusative sg. (e.g.
ēgur, hrīðer, ēar, wildor). In the latter case, their identity as *-es/-os stems can
be recognised essentially on the basis of their relationship to the forms which
did not preserve the -r formative, and the occasional consonantal inection of
the dative sg. (Brunner 1965, 244, cf. Boutkan 1992, 12).7 This subdivision may
be of signicance when seen from the point of view of the restructuring process:
while the rst group can be expected to exhibit some synchronic variation in the
paradigm (between the inherited and innovative inection), the second, according
to standard historical grammars (e.g. Brunner 1965; Campbell 1959; Hogg and
5 The interpretation of some of these inectional endings as they developed in Old English
seems fairly problematic. Such is the case with the genitive/dative sg. and nominative/
accusative pl. pre-Old English ending *-ur (OE -or > -er), where -u can (arguably) be
interpreted as a parasite vowel, inserted in place of -e in the sequence *-er, which, once in
the nal position, was subject to the process of apocope (cf. Campbell 1959, 258, where the
archaic forms such as calfur or lombor are deemed to contain an original West Germanic -ur
< *-ar). The development may well have involved the process of vowel harmony, whereby
the quality of the vowel of the inectional ending was harmonised with that of the root vowel
(for details on the patterns of vowel combinations in West-Germanic see Versloot 2008,
190ff.). For a very informative overview of alternative opinions and an elaboration on this
hypothesis, see Boutkan (1992, 16-18) and for a most recent summary of the problem see
Hogg and Fulk (2011, 59-60).
6 Since the form of the nominative/accusative sg. was the only o-grade form in the whole
paradigm (PIE *-os), it must have been subject to the analogical pressure of other cases and
hence it was subsequently replaced by the e-grade sufx *-ez (> *-iz) (Boutkan 1992, 14;
Campbell 1959, 258; cf. Weyhe 1906, 89).
7 The nominative/accusative sg. of the second subgroup (dōgor, hālor (< *dōgr, *hālr))
is believed to have been an effect of intraparadigmatic levelling and, accordingly, the
nominative form owes its existence to the presence of analogical forms in the oblique cases
(e.g. dōgore, dōgores), which appeared due to the inuence of a-stems, serving as a template
for the reanalysis, cf. footnote 6 above (Boutkan 1992, 15; Wright and Wright 1908, 213;
Hirt 1932, 59; cf. Casaretto 2000, 219).
390 Elżbieta Adamczyk
Fulk 2011; Wright and Wright 1908), almost entirely follows the productive
inectional pattern of a-stems.
The Old English s-stem inection was characterised by a combination of two
signicant features which render it distinct from other inectional types. One of
them is the presence of the vestigial -r- in cases other than the nominative/accusative
sg., which constitutes a regular development from the PGmc sufxal *-z (< PIE
*-s) (e.g. the nominative/accusative pl. calfur, dōgor, lombur, or the genitive pl.
calfra, hrōðra, lombra, etc.; see the paradigm in Table 2).8 The other notable trait
is the presence of an umlauted vowel in the nominative and accusative sg. (as in
cælf, lemb, hrēð, dæg), resulting from the impact of the original inectional ending
containing a high front vowel -i (PGmc *-iz: *calz, *lambiz9). Since the mutated
base form could be extended to other cases in the paradigm, forms such as the
genitive sg. celfes, cælfes, etc. are to be sporadically found in the available material.
It must be noted, however, that the mutated vowel, while typical of Anglian texts,
tended to be levelled according to the pattern of the oblique cases in the texts of
West-Saxon provenance. The signicance of the activity of i-umlaut in the context
of the restructuring process is particularly transparent in light syllable stems, where
the retention of the original desinential vowel (-e < -i) may have facilitated their
early transfer to (masculine) i-stem declension (e.g. hete ‘hatred’, sige ‘victory’ vs.
masculine i-stem wine ‘friend’) (cf. section 3 below).
Finally, as has been mentioned, the original pattern, as presented in Table 1,
is not very consistently preserved in the Old English material, since many nouns
initially belonging to this class lost their inherited sufxal *-s and, in consequence,
were ready to transfer to other declensional types.
3. Paradigmatic restructuring in the Old English s-stem inection
The process of gradual morphological restructuring of the s-stem inection must
have occurred relatively early in Germanic, as the Gothic and Old Scandinavian
materials testify to a complete or almost complete demise of this inectional type.
In both, the original *-es/-os stems must have transferred to the productive a-stem
inection prehistorically, since no synchronic alternation is attested in the available
material (Bammesberger 1990, 208; Prokosch 1939, 256; Schenker 1971, 51-
52).10 In Old English, however, the process of morphological reanalysis can still
8 All the Old English s-stems appearing in this section are glossed in section 5.1 where
the details of the quantitative analysis are discussed.
9 The nominative sg. *lambiz has been explained as an analogical formation on the pattern
of the genitive and dative sg. The preservation of -i- in the Gothic hatis, sigis (in place of the
expected *hats, *sigs) is adduced as a corroboration of this assumption (Wright and Wright
1908, 213).
10 For details concerning this class in Gothic see Braune (2004, 108-109), and Gordon
(1981) for an overview of inectional patterns in Old Norse, where the original s-stems are
not treated as a separate declensional type.
Towards a Diatopic Approach to the Old English s-Stem Declension 391
be captured in the making, and the synchronic vacillation between archaic and
innovative forms is, at least to some extent, attested in the extant records.
With reference to the Old English pattern of preservation of the original *-es/-os
inection, Krahe (1969), adducing the attested r-less forms, such as lomb, lombes,
lombe, asserts that the Old English, non-Anglian s-stems followed pre-eminently
the paradigm of a-stems (which he views as directly linked to the elimination of
the effects of i-umlaut in the nominative/accusative sg.).11 It is indeed the Anglian
material where the original shape of the s-stem paradigm is more or less consistently
retained; the later, West-Saxon data can be expected to exhibit a substantial
departure from the archaic pattern.
Granted that a number of the Old English historical s-stems (whose afliation
with s-declension can be conrmed by the existing cognates in the other Germanic
languages) do not testify to any vestiges of the original s-stem inection in the
attested material, the process of internal restructuring of the s-stem paradigm must
be dated relatively early, i.e. earlier than the rst attestations of English. These early
transfers to other declensional types entailed shifts following two major paths: 12
(a) to the large i-declension (e.g. æsc ‘esh’, īes (ēos) ‘eece’, geban(n)
‘summons’, gedyre ‘door-post’, gefōg ‘joint’, geheald (gehield)
‘keeping, guard’, gehlyd ‘noise’, gehnāst ‘conict, clash’, geswinc ‘toil,
effort’, gewēd ‘rage, madness’, gewealc ‘rolling’, gehield ‘guard’, ofdele
(ofdæle) ‘descent’, oferslege ‘lintel’, orlege ‘fate’, sife ‘sieve’, spere
‘spear’; masculine light-stems: bere ‘barley’, ege ‘terror’13, hete ‘hatred’,
mene ‘necklace’, sele ‘hall’, sige ‘victory’; as well as heavy-stemmed
gæst (attested alongside the a-stem gāst14) ‘guest’, hæl ‘health’, hilt (helt)
‘handle’, hlæw ‘mound’, hræw ‘body’, læn ‘loan, grant’, sweng ‘blow’)
(cf. Hogg and Fulk 2011, 61)
11 ‘Außerhalb des Anglischen hat auch das Ags. normalerweise die Flexion nach den
ă-Stämmen (ausgehend vom Nom. Akk. Sg., in welchem der i-Umlaut analogisch beseitigt
wurde), daher ohne das r-Element, z. B. lomb, lombes, lombe, usw.’ (Krahe 1969, 44).
12 The classication of these nouns is somewhat problematic, since standard historical
grammars of Old English, as well as the major dictionaries, provide quite discrepant
information as to the synchronic membership of particular s-stems. The reason for such a
state of affairs is a large scale migration of the minor nouns to the productive inectional
types. The trend is to be observed in both u-stems (where e.g. feoh ‘cattle’ shows no vestiges
of the original paradigmatic pattern) and i-stems (where the confusion with strong ja-stems
is enhanced by the presence of a mutated vowel in the stem). The present list of original s-
stems was compiled on the basis of the classication found in Brunner (1965), but compare
Campbell (1959), Wright and Wright (1908), Bosworth and Toller (1898).
13 Cf. OE n-stem egesa (egsa) ‘terror’ (OHG, OS egiso).
14 The existence of two parallel stems is a characteristic of other Indo-European languages,
too, e.g. Lat. pecu alongside pecus, Skt. janu ‘birth’ (feminine) alongside janas ‘sex, race’
(neuter) (Unwerth 1910, 15). In fact, for many of these nouns in Old English parallel a- (-ja-)
stem forms without a mutated vowel are attested, e.g. gāst, sæl, hlāw, hrāw.
392 Elżbieta Adamczyk
(b) to the synchronically unproductive u-declension (feoh ‘cattle’, liðu-,
liðu ‘member’, lim ‘limb, part of body’, sidu, seodu ‘custom’, sceadu
‘shade’15) (cf. van Helten 1911, 502).16
With reference to the gender of s-stems, Schenker (1972, 53), among others,
claims that the change of paradigmatic afliation is often accompanied by gender
reassignment. Accordingly, as will become evident in the present investigation, the
majority of the s-stems lose their original neuter gender, becoming masculine, and
thus following the new masculine pattern of inection. Such a situation seems to
be well justied by the fact that the potential target group, i.e. the class of neuter
i-stems, constituted a small group of nouns (only light syllable stems are attested17),
whereas the neuter u-stems were basically non-existent, and, therefore, the words
newly entering these declensional types were likely to change their inherited gender,
being readily absorbed by a much better represented masculine subtype. The
prehistoric shift of the original s-stems to i-stem declension has been explained in
terms of the generalisation of the PIE *-es- formative in the nominative/accusative
sg., whereby this case fell together with the nominative/accusative sg. of the i-
stems (cf. sige ‘victory’ < *segiz and (i-stem) wine ‘friend’ < *weniz) (Casaretto
2000, 218). The interparadigmatic realignments which followed were apparently
triggered by this formal similarity of the nominative sg. of the i- and s-stems. On
top of that, the analogical transference of the original s-stems to the numerous
i-declension must have been facilitated by the fact that these forms, just like
the i-stems, were subject to i-mutation (Unwerth 1910, 2). A similar factor may
have played a role in triggering the transference to the ja-stem declension, whose
members regularly underwent the process of i-mutation, tending thus to attract the
unstable s-stems.
Given that the Old English s-stem declension comprised only neuter nouns, the
synchronically attested process of morphological reanalysis may be expected to
have involved the redesigning of the inherited paradigm on the pattern furnished
by the neuter a-stems. At the same time, in the light of the expansiveness of the
masculine a-stem pattern, with its pervasive nominative/accusative pl. -as marker,
the pressure of this most productive type on the neuter paradigm of the s-stems is
only naturally to be envisaged. Potentially, members of the s-stem declension could
15 Cf. the Gothic wulþus ‘glory’, sidus ‘custom’ which consistently follow the pattern
furnished by the u-stems.
16 The fact that the two minor inectional types, i-stems and u-stems, were able to attract
nouns from other declensional classes proves that the productivity of these two otherwise
unproductive patterns must have lasted for some time before it was eventually lost.
17 There is no group of heavy syllable neuter i-stems attested in the Old English material;
nevertheless, a number of nouns deriving from other inectional types can be viewed as
neuter heavy stems, since they show an umlauted vowel and terminate in a consonant in the
nominative/accusative sg., complying thus with the pattern of the heavy-stemmed masculines
and feminines (Brunner 1965, 224).
Towards a Diatopic Approach to the Old English s-Stem Declension 393
evince also some inclination towards the (gradually expanding) weak inectional
pattern, shown primarily in the adoption of the endings of the genitive pl. -ena
and the nominative/accusative pl. -an (occasionally -on/-en/-æn). Accordingly, the
following innovations can be expected to appear in the paradigm of the neuter s-
stems (Campbell 1959; Brunner 1965; Wright and Wright 1908):
- the genitive. sg. -es marker in place of a historical zero ending
- the dative sg. -e ending in place of a historical zero ending
- the neuter zero ending (without r-stem extension) in the nominative/
accusative pl.
- the masculine -as ending in the nominative/accusative pl.
- the dative pl. -um ending in place of the expected -rum marker
- the genitive pl. -a or -ena (weak) ending in place of the original -ra
- the weak -an ending in all cases in the singular except the nominative,
and in the nominative/accusative pl.
The new analogical endings were essentially attached to the forms without the
r-stem extension, e.g. calfum, celfes, lambum. Frequently, however, they could
be joined to the formations where the -r- formative was preserved throughout the
entire paradigm, as in dōgores, sigores, ehras (cf. examples in section 5.3). At the
same time, the original stem formative -r- was more consistently retained in the
paradigm of the plural, where, in contrast to the singular, the inectional endings
were typically attached to the stem extension. In order to account for the systematic
preservation of the stem formative in the plural, Kastovsky (1995, 234) resorted to
semantic considerations, evoking the functional-semantic status of the plural and
asserting that the category of number ‘was semantically more salient, dominant,
or marked and therefore it took precedence over case’. Consequently, the process
of analogical levelling is more expected to operate within one number category
(spreading across cases) rather than within one case category (spreading across
number).
Table 2 below depicts the two competing paradigms of the Old English s-stems:
the archaic one, continuing the Proto-Germanic heritage, and the innovative one,
redesigned under the pressure of the productive inection (a-stems).
394 Elżbieta Adamczyk
Table 2. The competing inections in the Old English s-stem paradigm18
archaic innovative
singular plural singular plural
nominative
genitive
dative
accusative
cælf calfur, -eru
calfur calfra
calfur calfrum
cælf calfur
cealf cealf, -as
cealfes cealfa
cealfe cealfum
cealf cealf, -as
The Old English endings of the genitive and dative pl., whether attached to a bare
stem (e.g. lamba, lambum), or to the stem formative -r- (e.g. lambra, lambrum),
can be ascribed to the early, prehistoric (pre-Old English) pressure of the inuential
a-stems. Prehistoric inuence can also be detected in the OE -u ending of the
nominative/accusative pl. (e.g. cildru), which presumably derives from *ō, extended
from the productive thematic inection. In the present investigation, however, the
forms of the genitive and dative pl. which preserve the r-stem formative (-ra, -
rum) will be considered archaic and, accordingly, counted against the plain forms
without the stem formative (-a, -um). Likewise, the -ru ending in the nominative and
accusative pl., though it could arguably be interpreted as an innovation, extended
from the strong neuter paradigm, is here viewed as archaic. Such an approach to the
abovementioned pattern of inections seems legitimised, given the special status of
the -r formative as a hallmark of the West Germanic s/r-stem inection.
Finally, representing one of the minor declensional types, the class of the
Old English s-stems was synchronically unproductive, yet the pattern seems to
have been attractive to some nouns, which did not originate in *-es/-os stems,
e.g. brēadru (brēad) ‘bread crumbs’, hæmedru (hæmed) ‘married relationships’,
hæteru ‘garments’, hālor (hæl, hāl) ‘salvation’, lēower (lēow) ‘ham, thigh’,
mædrum (**mæd, *mæder?) ‘measures’, salor ‘hall’ (sæl or sele), scerero ‘shears’
(scear), speldra (speld) ‘torches’, stæner ‘stone’ (stān) (Brunner 1965, 244,
cf. Campbell 1959, 259). The following examples ((1) – (12)), culled from the
Old English electronic corpus (Healey 2004), can serve as an illustration of this
tendency. Most of them are attested in early Mercian material, dated to the 8th c.,
18 The opposition æ/a : ea reects the dialectal distribution of these vocalic elements,
representing, respectively, Anglian and West-Saxon forms; it will have no bearing on the
results of the quantitative investigation of the interparadigmatic realignments.
Towards a Diatopic Approach to the Old English s-Stem Declension 395
including the earliest glossaries (Épinal, Erfurt and Corpus Glossary19), noted for
their conservative nature.20
(1) He sent gicelstan his breadru hlafes beforon ansyne cyles his hwylc
standeþ (Mittit cristallum suum sicut frusta panis ante faciem frigoris
eius quis subsistet.) [PsGlD (Roeder)]
(2) frusta panis breadru. [OccGl]
(3) lenocinia hæmedru forspenningce. [AldV 1]
(4) Se hund þa ða he ne moste his lichaman derian. totær his hæteru
sticmælum of his bæce. and hine draf geond þa weallas: [ÆCHom I, 26]
(5) we soðfæstra þurh misgedwield mod oncyrren, ahwyrfen from halor, we
beoð hygegeomre, forhte on ferðþe. [Jul]
(6) (mee gibre pernas omnes libera tuta pelta protegente singula) mines
lichoman lewera alle alæs gesundum plegscylde gescyldendum anra
gehwylc. [LorGl 1]
(7) (Mei gibrae pernas omnes libera tuta pelta protegente singula ut non
tetrae demones in latera mea librent ut solent iacula) mines lichoman
leower ealne gefria ðine plæg sceldæ gescyldendum anra gehwylc þæt
þa sweartan diou on minre sidan cueccen swa swa gewuniað scytas.
[LorGl 2]
(8) Gif ðeor sy in men, wyrc drænc, nim þas wyrte, nyoþowearde nul &
bisceopwyrt, æscðrote, ealra efenmicel, þyssa twega mæst ufonwearde
rudan & betonican, ofgeot mid III mædrum ealoð... [Med 3]
(9) Forces scerero. [CorpGl 2]
(10) forces sceroro. [ErfGl 1]
(11) Malleoli þara spaca speldra. [ClGl 1]
(12) (qui autem supra petrosa seminatus est hic est qui uerbum audit et …)
seðe soðlice ofer stæner vel on staener sawende wæs ðis is vel wæs seðe
word geherde... [MtGl (Li)]
4. Dening the Old English dialects: Some methodological considerations
The present study takes the diatopic aspect as an important criterion in investigating
and describing the gradual process of reanalysis in the class of s-stems. It is aimed
19 The Épinal and Erfurt glossaries have traditionally been dated to the early 8th century,
whereas the Corpus Glossary to the late 8th century (Hogg 1992, 5, Toon 1992, 427, but
compare Campbell 1959, 7, who dates both Épinal and Erfurt to the 9th c.).
20 The use of the abbreviations in the present section as well as in section 5.3 follows the
practice of Cameron (1973), who provides a complete list of texts cited in the Dictionary
of Old English corpus, with short titles and detailed bibliographic information. An updated,
informative and easily accessible list of texts is also available on the homepage of the
Dictionary of Old English project at: http://www.doe.utoronto.ca/st/index.html.
396 Elżbieta Adamczyk
at shedding some more light on the disintegration process by using the data from
texts traditionally classied as West-Saxon, Mercian, Northumbrian and Kentish.
Such an approach may potentially raise questions about the precise identication of
the dialect features in texts which are internally inconsistent, or texts of uncertain
or much disputed origin, and calls, therefore, for a brief consideration of some
methodological problems involving dialect mixture, scribal practices, or intrinsic
limitations of the corpus studied. The dialect mixture problem concerns primarily,
though not exclusively, Kentish and Mercian dialects, which among the established
varieties of Old English, appear most difcult to dene. Some of the texts ascribed
to these dialects are linguistically inconsistent, showing much dialectal mixture,
whilst others, though consistent linguistically, are poorly localised. Kentish texts,
for instance, display a substantial admixture of West-Saxon as well as Mercian
forms. The heterogeneous linguistic character of Kentish, probably best manifested
in charters, has been viewed as directly correlated to the Mercian political supremacy
in the 8th and early 9th century. The Kentish charters of the Mercian period, i.e.
produced under Mercian domination, tend to display some unambiguously Mercian
characteristics, such as the uctuation of <a> vs. <o> before nasals, to mention
but one of these traits (Toon (1992, 450); cf. Coleman (2004, 191ff.); Kitson
(1995, 106); for details on problems with identifying the Kentish material see
Lowe (2001)). A similar interpretative problem is posed by the presence of West-
Saxonisms in the Mercian texts of Vespasian Psalter and Rushworth Gospels (Ru1),
such as the appearance of Saxon wæron vs. Anglian weron, or the inconsistent use
of <æ> vs. <e> for reexes of the WGmc *a and *e, the interpretation of which can
be problematic (Kitson 2004, 227). A closely related limitation regarding the nature
of the available dialectal data is the scribal interference and scriptorial convention,
and the fact that the standard written languages ‘were sometimes used by clerks far
beyond the bounds of the spoken dialects’, with, for instance, Mercian written at
Canterbury or Late West-Saxon used at York (Crowley 1986, 99).
In the present study, the traditional stance (or the majority view) was consistently
followed when it came to assigning the dialect label to the texts which have commonly
been considered of mixed origin, combined with the dialect information which is
available in the Dictionary of Old English corpus (with the rough subdivision into
Saxon, Anglian and Unknown, cf. section 5.1). The Corpus Glossary may serve
as an example here: it appears in the appendix section in the Mercian sources,
on the authority of Campbell (1959) and Brunner (1965). Its dialectal status is,
nonetheless, rather ambiguous (for details see Kuhn (1939)) and consequently the
DOE assigns it a label Unknown. Accordingly, the few instances of s-stem forms
found in this text, albeit cited in the illustration section (examples 9 and 28), were
not included in the quantitative analysis.
A further important facet of the diatopic approach is related to the scope of the
investigated material. An investigation carried out on the complete corpus of Old
English texts, much as it can be exhaustive, is not free from limitations inherent in
Towards a Diatopic Approach to the Old English s-Stem Declension 397
the corpus, namely, the disproportions in the amount of available material deriving
from individual dialects. Accordingly, the extensive material of West-Saxon
provenance must be juxtaposed against the relatively modest corpus of non-West-
Saxon texts. It must be emphasised, however, that it is the state of affairs as attested
in all available extant material that was of primary concern to the present author,
and hence the study aims at a comprehensive analysis, encompassing all dialectally
identiable texts rather than a representative sample of texts from individual
dialects, roughly adjusted size-wise.
Given the scope of the present investigation, it becomes apparent that the study
will not be entirely devoid of the difculties any diatopic account unavoidably needs
to face. Some of the limitations of the traditional approach to dialectal variation (e.g.
treating Old English dialects as ‘independent and discrete linguistic varieties’ (Hogg
1988, 187)) cannot be escaped in a study of this nature, and this very fact needs to be
taken into account when interpreting the ndings. At the same time, it must be reiterated
that the rationale behind the present enterprise is to explore the general tendencies per-
taining to the process of restructuring in individual dialects of Old English as they
are known or dened in traditional accounts. Therefore, endeavours aimed at seeking
strict demarcation lines between the areas where the process of disintegration took
place, or exploring the details of the process in texts whose dialectal status is not
clearly dened, remain beyond the scope of the present analysis.
5. Qualitative and quantitative analysis
5.1. The corpus and the results of the investigation
The present investigation was carried out on the complete body of the Dictionary
of Old English electronic corpus (Healey 2004) and entailed all dialectally
identiable texts. The size of the corpus enabled a detailed diatopic investigation,
covering the material deriving from the four major dialects of Old English. The
study of texts of various geographical provenance was essential for the analysis,
since individual varieties of Old English were expected to evince divergent patterns
in the preservation of the original s-stem inection. As the present investigation
into the s-stem declension is a part of a larger project, aimed at a thorough and
systematic examination of analogical restructuring in all minor declensional
classes, the present quantitative analysis was conducted on the Dictionary of Old
English corpus adjusted to the needs of this integrated undertaking. The preparation
of the corpus for the purpose of this study entailed a major renement, namely, the
addition of precise information on the dialect distribution to the already existing one,
whereby all the texts were classied as Saxon, Anglian and Unknown. In the present
investigation, the group of texts classied as Unknown was left out of consideration
altogether, whereas the two remaining subgroups were further subdivided, drawing
the traditional distinction between the West-Saxon and Kentish material, on the
one hand, and Mercian and Northumbrian, on the other. The corpus was further
rened for the purpose of investigating the dynamics of the process in the noun cild,
398 Elżbieta Adamczyk
where the diachronic perspective was added. The existing subdivision into Early,
Intermediate and Late material was supplemented, or more precisely, replaced with
more exact dating based on the century from which the identied texts derive.
It must be noted that it is consistently the date of the manuscript rather than the
assumed date of composition of a given text that was taken into consideration when
dating the phenomena under discussion.
The list of nouns to be examined was compiled on the basis of the information
found in the standard historical grammars of Old English (Brunner 1965, Campbell
1959, Hogg and Fulk 2011), the Anglo-Saxon Dictionary by Bosworth and Toller
(1898), and was subsequently checked against a list of Proto-Germanic stems
provided by Bammesberger (1990, 208-214), and veried etymologically by recourse
to the information found in the Oxford English Dictionary (1989).21 Since the nouns
subject to investigation were also found in compound formations (when used as the
second element of the compound), these forms were included in the analysis of the
material and a complete list of simplexes comprised of the following nouns: (a) cealf
(cælf) ‘calf’, cild ‘child’, lomb ‘lamb’, dæg (dōgor) ‘day’22, æg ‘egg’, hrēð ‘glory’;
(b) ægor/ēgur ‘sea’, alor (alr) ‘alder-tree’, ēar (eher, æhher) ‘ear of grain’23, gycer
‘acre’24, hōcor ‘mockery’, hrīðer, hryðer ‘horned cattle’, (masculine) nicor ‘water-
monster’25, sigor ‘victory’26, salor ‘hall’, wildor ‘wild animal’.
The interpretation of some of the forms turned out to be rather problematic.
An example here is dōgor ‘day’ which is directly related to a heavy-syllable stem
21 Cf. Kluge 1926, 44-45, 76.
22 Bammesberger (1990, 213) points to a PGmc. stem *dōg-uz- as a point of departure
for the OE dæg, admitting that the word is closely related to the PGmc. *dag-a- (a-stem).
An unambiguous identication of the original s-stem seems to be rather problematic in the
Old English electronic corpus, as the original length of the vowel is not marked. According
to Weyhe (1906), it is only the text of the Lindisfarne Gospels that preserves a vestige of
the original nominative/accusative sg. *dōgiz, whereas in the remaining material, including
other Northumbrian texts, the expected heavy-stemmed form is consistently ousted by the
light a-stem dæg (Weyhe 1906, 85).
23 According to Ross (1937, 97), the forms ehher and æhher continue into the later stage of
English in forms of the pl. echirris in Douglas and Modern Scots dialect icker (< OE ehher)
and acher, acre (< æhher). The noun uctuated between the neuter and masculine gender,
presumably due to the analogical impact of the masculine tahur ‘tear’ (tēar) (van Helten
1911, 499) (cf. the feminine gender of present-day German Ähre < OHG neuter ehir, ahar).
24 The word is apparently singly attested in the Old English material (Leiden Glossary:
iugeres gycer), presumably in the singular (cf. Sauer 1917). Bammesberger (1965, 418),
following Noreen’s (1894) and Streitberg’s (1963) interpretation of a related Gothic jukuzi,
rejects the s-stem theory of its origin and interprets the Old English form as an original
active perfect participle. A corollary of such an interpretation is the reassignment of gender,
since given that the Gothic form was feminine, the Old English cognate must also be viewed
as such upon this interpretation.
25 The Old High German nichus (nihhus) ‘crocodile’ seems problematic in this context
since it constitutes a counterevidence to the origin of the Old English nicor as an s-stem (but
ON nykr). See Classen (1915, 85-86) for an alternative explanation of the Old English nicor.
26 The Old English sigor is attested as a masculine noun, which can be attributed to the
inuence of the related masculine i-stem sige ‘victory’.
Towards a Diatopic Approach to the Old English s-Stem Declension 399
dæg - a form which merged at some point with a parallel light syllable a-stem
dæg, rendering unambiguous interpretation impossible (cf. footnote 22). Following
standard historical accounts and relying on the opinions quoted there (Ross 1937,
Brunner 1965), the present quantitative analysis took into account only those forms
of dæg which are attested in the Northumbrian dialect of Old English, in particular,
in the Lindisfarne Gospels, where their origin as s-stems is reected in the spelling
variant doeg.27 Another qualication bearing on the analysis involves the forms of
the nominative and accusative sg. of the s-stems: since they cannot testify to any
synchronic alternation, they are of no direct relevance for the present quantitative
investigation (they are italicised in the tables below).
The results of the analysis are demonstrated in the tables below, and accordingly
Tables 3 and 4 present the distribution of s-stem and a-stem inection in the
respective Mercian and Northumbrian material, whereas Tables 5 and 6 offer a
summary of the distribution in the texts of the Kentish and West-Saxon provenance,
respectively.
Table 3. The distribution of archaic and innovative inection in the texts of Mercian
provenance
archaic
innovative
singular
plural
singular
plural
nominative
genitive
dative
accusative
(5)
(2) 40%
(2) 100%
(4)
(2) 100%
(5) 100%
(2) 66.7%
(9) 100%
–
(3) 60%
(0)
–
(0)
(0)
(1) 33.3%
(0)
TOTAL (22) 84.6% (4) 15.4%
Table 4. The distribution of archaic and innovative inection in the texts of
Northumbrian provenance
archaic
innovative
singular
plural
singular
plural
nominative
genitive
dative
accusative
(9)
(0)
(1) 5.3%
(22)
(0)
(2) 100%
(4) 100%
(11) 91.7%
–
(3) 100%
(18) 94.7%
–
(0)
(0)
(0)
(1) 8.3%
TOTAL (18) 45% (22) 55%
27 Interestingly, several isolated instances of the nominative/accusative sg. form dōgor,
found in the investigated material (late West-Saxon) alongside the Northumbrian doeg,
testify to the persistent spread of the -r- element from the oblique cases throughout the
singular.
400 Elżbieta Adamczyk
Table 5. The distribution of archaic and innovative inection in the texts of Kentish
provenance
archaic
innovative
singular
plural
singular
plural
nominative
genitive
dative
accusative
(3)
(0)
(0)
(8)
(0)
(0)
(0)
(2) 100%
–
(0)
(0)
–
(1) 100%
(2) 100%
(0)
(0)
TOTAL (2) 40% (3) 60%
Table 6. The distribution of archaic and innovative inection in the texts of West-
Saxon provenance
archaic
innovative
singular
plural
singular
plural
nominative
genitive
dative
accusative
(69)
(4) 3.5%
(5) 12.5%
(119)
(18) 62.1%
(59) 45.4%
(23) 69.7%
(32) 39%
–
(111) 96.5%
(35) 87.5%
–
(11) 37.9%
(71) 54.6%
(10) 30.3%
(50) 61%
TOTAL (141) 32.9% (288) 67.1%
For the purpose of greater transparency, the ndings of the investigation for individual
dialects are juxtaposed in Table 7 and presented graphically in Figure 1.
Table 7. The distribution of the archaic and innovative forms in individual dialects
of Old English
archaic
innovative
Mercian (22) 84.6% (4) 15.4%
Northumbrian (18) 45% (22) 55%
Kentish (2) 40% (3) 60%
West-Saxon (141) 32.9% (288) 67.1%
Towards a Diatopic Approach to the Old English s-Stem Declension 401
Figure 1. The distribution of the archaic and innovative forms in individual dialects
of Old English28
What becomes evident when interpreting the data, apart from the fact that the s-stem
class is rather scantily attested in the Old English corpus, is the striking discrepancy
in the pattern of restructuring attested in the Anglian vs. non-Anglian material.
While the former dialects tend to be more resistant to the inuence of productive
inection, the latter display remarkable vulnerability to analogical pressure, readily
adopting the novel inectional endings. As could be expected, the most archaic
paradigm is found in the Mercian prose texts (glossaries), which retain the archaic
inection in nearly 85 per cent of all the attested forms. The relatively balanced
percentage of archaic and innovative forms in the Northumbrian material (45% vs.
55%), on the other hand, testies to a stage of transformation in which the process
of morphological restructuring can be captured in the making. The synchronic
alternation between archaic and novel forms in Anglian is attested only in forms of
the genitive and dative sg., and, to some extent, in the accusative pl. The retention
of the inherited inection in Northumbrian is essentially conned to the plural (e.g.
lombor, lomboro, lombro), whereas the singular turns out to be rather progressive,
showing the mutated forms without r-extension. Finally, the West-Saxon material,
where the diffusion of the innovative forms was most advanced and reached over
67 per cent of all the attested forms, testies to extensive synchronic variation in
28 When interpreting the data from Figure 1, one needs to bear in mind the discrepancies in
the number of attested forms for individual dialects, in particular the scarcity of the attested
forms in Kentish (cf. Table 5).
402 Elżbieta Adamczyk
all cases, with the genitive and dative sg. taking precedence over other cases in
terms of their innovative shape. Irrespective of the dialect, the advancement of the
disintegration process is best attested in the genitive sg. which appears to have been
most innovative, evidently favouring the a-stem inection.
Interestingly enough, hardly any inuence of the weak productive inection has
been found in the Old English material, which seems unexpected in view of the
extensively attested Middle English forms of cildren, calveren or lambren, ægeran,
eiren. The rare exceptions here are the genitive pl. forms (e.g. cildena), found in
the Northumbrian text of Durham Ritual, stēopcildena (late West-Saxon), as well
as the accusative pl. cildræn (see section 5.3 below).
The overall distribution of the archaic (s-stem) vs. innovative (a-stem) inection
in the paradigm of s-stem nouns, irrespective of the dialect, is presented in Table 9
and Figure 2 below.
Table 9. The overall distribution of s-stem inection vs. a-stem inection (ir-
respective of the dialect) in the analysed material
s
-
inflection
a
-
inflection
singular
plural
(14) 7.6%
(169) 53.5%
(170) 92.4%
(147) 46.5%
TOTAL
(183) 36.6% (317) 63.4%
Figure 2. The overall distribution of s-stem inection vs. a-stem inection with
respect to the number
Towards a Diatopic Approach to the Old English s-Stem Declension 403
The discrepancy between the singular and the plural with respect to the retention
of the s-stem inection in Old English is remarkable. While the former turned out
to have been very susceptible to the pressure of the innovative inectional pattern,
retaining the original inection in no more than 7.6 per cent of forms, the latter
seems to have been moderately innovative, with 46.5 per cent of forms adopting
the innovative pattern. A factor which may be of signicance here is certainly the
fact the r-stem formative, albeit never generalised as a plural marker, did have a
distinctive function at some stage, marking plurality in this group of nouns. As
such, it managed to be preserved longer in the plural paradigm than in the singular.
Another crucial facet is related to the predominance of the rapidly spreading and
super-stable genitive sg. -es marker (cf. Wurzel 1989), which effectively eliminated
any chance for the retention of the relic endingless form in the singular.
Another notable characteristic of the restructuring process in the group of s-
stems becomes evident when the data for the plural are compared from a dialectal
perspective. Remarkably, there is a clear borderline between the Anglian and non-
Anglian material: while the former dialects display minor traces of the impact of the
new inectional pattern, the latter testify to a fairly extensive spread of the a-stem
inection in the paradigm. The marked congruence of Mercian and Northumbrian
data in this respect is certainly worth noting. It seems that the time dimension is
of little bearing here: even though the dates of attestation of Northumbrian on the
one hand and (early) Mercian on the other are discrepant, the restructuring process
follows an identical path of development in both dialects. Most importantly,
disregarding the results for the singular paradigm and interpreting the ndings for
the plural, changes considerably the perception of Northumbrian as representing an
intermediate stage which manifested a relative balance between archaic and novel
inection (cf. Fig. 1). It seems that the r-stem formative must have prevailed there
as a marker of plurality, and was apparently resistant to the encroaching impact of
the a-stem inection.
With respect to the plural, it is the diatopic rather than the diachronic contrast
that is essential to the pattern of distribution of the competing forms in the s-stem
paradigm, with a borderline drawn roughly between the South and the North. The
details of this dialectal distribution of the plural forms are presented in Table 10 and
graphically in Figure 3 below.
404 Elżbieta Adamczyk
Table10. Dialectal distribution of the s-stem and a-stem inection in the plural
forms of s-stems
s
-
stem
a
-
stem
Kentish 40% 60%
West-Saxon 48.2% 51.8%
Mercian 94.7% 5.3%
Northumbrian 94.4% 5.6%
Figure 3. Dialectal distribution of the s-stem and a-stem inection in the plural
forms of s-stems
When approached from a broader perspective, however, i.e. considering the data
from both the singular and the plural, the following pattern (not totally unexpected)
emerges from the attestation of the original *-es/-os stems in individual Old
English dialects. In the light of the available data, three stages of the development
of the Old English paradigm can be distinguished (cf. Schenker (1972, 55) for Old
Towards a Diatopic Approach to the Old English s-Stem Declension 405
High German), which largely correspond to the dialects of Old English: Mercian,
Northumbrian (late), and West-Saxon and Kentish.29
I (Early Mercian) II (late Mercian, Northumbrian) III (West-Saxon, Kentish)
N. cælf N. cælf N. cealf
G. calfur G. calfes/cælfes G. cealfes
D. calfur D. cealfe D. cealfe
A. cælf A. cælf A. cealf
N. calfur, -eru N. cealfru, calfero N. cealfas
G. calfra G. calfra G. cealfa
D. calfrum D. cealfrum D. cealfum, -on
A. calfur, -eru A. cealfru, calfero A. cealfas
The paradigms presented above clearly indicate that the mechanism of the
process of morphological restructuring in Old English must have involved three
developments: (1) a gradual restriction of the forms containing the r-stem formative
to the plural (Northumbrian) and (2) an accompanying elimination of the mutated
forms in the singular (partly Northumbrian, West-Saxon). These two developments
were followed by (3) a total eradication of the original stem formative from the
plural (West-Saxon, Kentish).
It must be emphasised at the same time that the elimination of the mutated
forms from the paradigm of the s-stems, though certainly an innovation, is not to be
viewed as an innovative feature in the present investigation (cf. footnote 4). It seems
feasible that the new, unmutated forms, undoubtedly analogical, appeared due to the
intraparadigmatic rather than interparadigmatic inuence, i.e. the pressure of other
case forms, especially those of the genitive and dative sg. within one paradigm.
The overall distribution of the productive inection relative to the inherited
pattern clearly indicates that the class of the s-stems, as attested in the Old English
material, was in the course of extensive restructuring which eventually led to a total
demise of this declensional type in English.
29 The word cealf serves here only as a representative of the s-stem inection, illustrating
the overall pattern of preservation of this declensional type. It must be noted, however,
that not all of the forms of cealf presented in the paradigm are attested in the Old English
material.
406 Elżbieta Adamczyk
5.2. A comment on cild.
The noun cild, due to its controversial status with respect to its origin, was analyzed
separately and the results of the investigation for individual dialects are presented in
Tables 10 and 11. According to the OED, cild (< PGmc *kilþom) did not originate as
an s-stem, but was towards the end of the Old English period partially assimilated to
the inection of neuter s-stems. This is testied by the appearance of the nominative
pl. cildru, genitive pl. cildra, widespread in the later West-Saxon texts. Alternatively,
the noun has occasionally been classied among the s-stems and derived from the
PGmc. *kelþ-ez- (cf. Go. kilþei ‘womb’, inkilþo ‘pregnant’) (notably Bammesberger
(1990, 211)). Given the uncertain origin of cild, the present quantitative analysis
examined the incidence of the s-stem vs. a-stem inection rather than the degree of
archaism vs. innovation present in the paradigm. Importantly, it is only the paradigm
of the plural which is of relevance for the present quantitative investigation, since
the singular consistently displays the features of a regular a-stem, with the -es and -e
markers in the genitive and dative, respectively. At the same time, such a distribution
of forms, i.e. the absence of any traces of the s-stem inection in the paradigm of the
singular, may be indicative of the fact that (1) the noun did belong originally to the a-
stem declension, showing no departure from the original inection whatsoever, and,
implicitly, (2) that the s-stem pattern must have been an innovation in the paradigm
of cild rather than a relic inection.
Tables 11 and 12 present the incidence of s-stem vs. a-stem inection in the
paradigm of cild with respect to individual dialects, whilst Table 13 provides an
overall percentage distribution irrespective of dialect. Finally, Figure 4 sheds some
light on the dynamics of the process, showing the diachronic vacillation of the noun
between the two declensional types.
Table 11. The distribution of archaic and innovative inection in the paradigm of
cild in the texts of Mercian and Northumbrian provenance
plural
Mercian
Northumbrian
s
-
stem
a
-
stem
s
-
stem
a
-
stem
nominative
genitive
dative
accusative
(0)
(1) 25%
(0) 100%
(0)
(2) 100%
(3) 75%
(1) 100%
(1) 100%
(0)
(1) 33.3%
(0)
(0)
(2) 100%
(2) 66.7%
(1) 100%
(4) 100%
TOTAL 12.5% 87.5% 10% 90%
Towards a Diatopic Approach to the Old English s-Stem Declension 407
Table 12. The distribution of archaic and innovative inection in the paradigm of
cild in the texts of Kentish and West-Saxon provenance
plural
West
-
Saxon
Kentish
s
-
stem
a
-
stem
s
-
stem
a
-
stem
nominative
genitive
dative
accusative
(14) 17.7%
(28) 35.4%
(8) 6.6%
(11) 14.9%
(65) 82.3%
(51) 64.6%
(114) 93.4%
(63) 85.1%
(0)
(1) 33.3%
(0)
(0)
(0)
(2) 66.7%
(1) 100%
(3) 100%
TOTAL
17.2%
82.8%
14.3%
85.7%
Table 13. The overall distribution of s-stem and a-stem inection in the paradigm
of cild
plural
s
-
stem
a
-
stem
(64) 16.9% (315) 83.1%
Figure 4. The diachronic distribution of the competing inectional types in the
plural paradigm of cild
408 Elżbieta Adamczyk
The investigated material allows one to conclude that the noun belonged primarily
to the a-stem inection which appears in over 80 per cent of forms and is attested
even in the early sources (e.g. Mercian Vespasian Psalter (early 9th c.)). The
distribution of the forms in individual dialects implies that cild was certainly not an
original s-stem, but rather gradually adopted the s-stem inection. The introduction
of the -r- stem formative in the plural paradigm appears to have been denitely a
southern feature, found primarily in West-Saxon texts belonging to the 10th and
11th centuries. Hardly any traces of the s-stem inection are found in the Anglian
material (cf. Table 11). Such a distribution is markedly distinct from the distribution
of plural forms in other (i.e. authentic) s-stems, where the incidence of archaic
forms in Anglian dialects reaches almost 95 per cent (cf. Table 10).
When seen from the diachronic perspective (Fig. 4), the data testify to no
discernible tendency whereby the noun might go in either direction, i.e. favour the
s-stem or a-stem inection; instead it remains relatively stable throughout the Old
English period, with no peaks or declines in its rate of vacillation between the two
declensional types. Consequently, no unambiguous starting point for the adoption
of the s-stem inection could be identied.
As far as case distribution is concerned, it is the nominative and accusative pl.
which turned out to be most resistant to the s-stem inection in non-West-Saxon
dialects, favouring either zero or the masculine -as ending. Both the genitive and
dative testify to vivid synchronic alternation between these two types of inection
and a gradual increase in the forms with the r-stem formative throughout the Old
English period.
Finally, the noun testies to an occasional lack of stability with respect to gender,
manifested in an apparent vacillation between the neuter and masculine paradigm:
accordingly, alongside the expected accusative pl. neuter cild, the masculine forms
cildas and cildes are found (attested only in Northumbrian). The signicance
of these attestations lies in the fact that they may be viewed as a presage of the
tendencies to come, whereby the masculine -as marker will remain the prevailing
plural ending in the North (15th century childes; cf. a related Scottish chield ‘chap,
lad’ where the regular plural form is preserved as chields).30
Certainly worth noting is the fact that hardly any instances of weak forms of
cild were found in the analysed material, which again may be surprising given their
popularity in the Middle English period, especially in the southern dialects (cf.
section 5.3, examples 32-33). The evident dearth of such forms in the Old English
corpus suggests that the emergence of the weak inection in the paradigm of cild can
be dated no earlier than the Middle English or possibly Early Middle English stage.
30 It must be noted, however, that the attestation of -as as a plural marker is certainly a
reection of a general tendency, not necessarily involving cild; the regular Middle English
Northern and North Midland plural forms were childer, childre (OED s.v. child; cf. Wełna
1996, 86-87).
Towards a Diatopic Approach to the Old English s-Stem Declension 409
5.3. Illustration of the pattern of analogical restructuring in the s-stem paradigm
The tendencies discussed in the foregoing section are illustrated in the following
set of forms found in the investigated corpus (examples (13) - (43)). The most
extensive traces of the impact of productive inection found in the genitive sg. are
illustrated by examples (13) - (17):
(13) (tertia uituli qua euangelistam lucam a zacharia sacerdote sumsisse
initium prægurauit) ðirdda celfes of ðon vel of ðæm ðe godspellere lucas
from zacharia meaessapreoste infeing vel ingann frumma foregebecnade.
[LiProlMt]
(14) (…et uultus eorum facies hominis et facies leonis et facies uituli et facies
aquilae)…& onsion hiora ondwlita vel hioful monnes & ondwlita vel
onsion leas & ondwlita cælfes & ondwlita earnes. [LiProlMt]
(15) & hy awendon wuldur his on gelicnesse calfes etendis heg (et mutauerunt
gloriam suam in similitudinem uituli manducantis foenum). [PsGlD
(Roeder)]
(16) broð’ wvtas gie þæt ne sceondlicvm seolfre l golde gileseno aron gie of
idlvm ivere wosa l fadorlices giselenisses ah diorwyrðvm blode svoelce
lombes vnawidlades & vnawoemmed’ crist’. [DurRitGl 1 (Thomp-
Lind)]
(17) ðas aron ða ðe gicvomon of costvnge micle & aðvogon gigerela hiora
& gihvidadon hia in blode lombes f’ðon aron bif’a hehsedle godes &
giherað him... [DurRitGl 1]
Examples (18) - (20) demonstrate the pressure of the innovative inectional pattern
in the dative sg.:
(18) (…cum omnia uidissent quae fecerat hierosolimis in die festo et ipsi
enim uenerant in diem festum) miððy allo gesegen ðaðe geuorhte
hierusolimiscum on doege halgum & hia vel ðailca forðon gecuomon on
halgum doege. [JnGl (Li)]
(19) Soðlice sylfwilles seo eorðe wæstm berað ærest gærs syððan ear, syþþan
fullne hwæte on þam eare. [Mk (WSCp)]
(20) & eall his cynn mon ofslog, þy læs hit monn uferan dogore wræce, oþþe
ænig oþer dorste eft swelc anginnan. [Or 4]
The dissemination of the innovative inection, both masculine and neuter, in the
nominative pl. and accusative pl. is shown in sentences (21) - (22) and (23) - (31),
respectively.
(21) (Tunc oblati sunt ei paruoli ut manus eis imponeret et oraret discipuli
autem increpabant eos) ða gebroht werun him lytla cnæhtas l cildas þæt
hond him gesette he & gebede ða ðegnas uutedlice geðreatadon hia.
[MtGl (Li)]
410 Elżbieta Adamczyk
(22) Ne onfo of huse ðinum calfas ne ne of eowdum ðinum buccan (Non
accipiam de domo tua uitulos neque de gregibus tuis yrcos). [PsGlF]
(23) (Angelo praemonente ioseph cum christo fugit in aegyptum et herodes
occidit infantes) ðe angel foregelærde oððe foregetahte mið crist geeh
in egipt & ofslog ða cildes. [MtHeadGl (Li)]
(24) Moyses sende cnihtas, ðæt offrodan twelf cealfas. [Exod]
(25) (Factum est autem in sabbato secundo cum transirent per sata uellebant
discipuli eius spicas et manducabant confricantes minibus) aworden wæs
ðonne on ðone æfterra daeg miððy oferfoerdon ðerh gecoecton ðegnas
his ða croppas l ehras & eton gebrecon mið hondum. [LkGl (Li)]
(26) Se Hælynd for on restedæge ofyr æcyras; Soþlice hys leorningcnihtas
hingryde & hig ongunnun pluccian þa ear & ætan. [Mt (WSCp)]
(27) Eft wæs geworden þa he restedagum þurh æceras eode, his leorningcnihtas
ongunnon þa ear pluccigean. [Mk (WSCp)]
(28) (Spicas) ear. [CorpGl 2]31
(29) Farað nu, nu ic eow sende swa swa lamb betwux wulfas. [Lk (WSCp)]
(30) Abraham ða gesette seofan lamb on sundron. [Gen]
(31) Þu dest ælce dæg on þæt weofod twa enetre lamb, An lamb on morgen,
oþer on æfen... [Exod]
Finally, examples (32) – (33) present the extension of the weak inectional pattern
to the genitive pl. form:
(32) þætte hia gifoega ðv gmeodvmia bitvih ðæm feoero & feortigvm &
hvnd’ ðvsenda cildena ða ðe hehstaldo ðerhwvnedon & hia mið vifvm
ne giwidladon on ðara mvðe facen gimoetted nis svæ... [DurRit]
(33) He sceal beon bewerigend wydewena & steopcildena, & stale alecgen,
& forliger gewitnigen…[ÆAbusWarn]
The nal portion of material, i.e. sentences (34) - (43), illustrate the archaism of
the Old English inection, manifested primarily in the preservation of the inherited
pattern in the nominative and accusative pl. and the genitive pl. forms.32
(34) …& godcvnd’ godes god to s’iðre engel to vynst vitge ivih gifylga
ðrovres f’e gæ ivih hiordo æc gifylga ivih gihalda driht’ scip’ & exen
calfero hors… [DurRit]
31 The single attestation found in the Corpus Glossary was not included in the nal
quantitative distribution for the reasons discussed in section 4 above.
32 In general, it is the compound formations, where the s-stem constitutes the rst element,
that tend to be much more conservative than simplexes, often preserving the inherited pattern,
as in: æg which appears as æger- in aegergelu ‘yolk’, or calfur (cylfer) in cylferlamb.
Towards a Diatopic Approach to the Old English s-Stem Declension 411
(35) Ic ne onfoo of huse ðinum calferu ne of eowdum ðinum buccan (Non
accipiam de domo tua uitulos, neque de gregibus tuis hircos). [PsGlA]
(36) & dydun cælf in choreb & weorðadun greftas & onwendun wuldur his
in gelicnisse calfur eotendes heg (Et fecerunt uitulum in choreb, et
adorauerunt sculptile et mutauerunt gloriam suam in similitudinem uituli
manducantis faenum). [PsGlA]
(37) (…dicentem ego dissoluam templum hoc manu factum et per triduum
aliud non manu factum aedicabo)…. ł cuoeðende ic undoe ł ic toslito
tempel ðis mið honde aworht & ðerh ðreo dogor oðer ne mið honde
aworht ic getimbro willo. [MkGl (Li)]
(38) (Holera et oua, pisces et caseum butirum et fabas et omnia munda
manduco cum gratiarum actione) Wyrta & æigra, sc & cyse, buteran &
beana & ealle clæne þingc ic ete mid micelre þancunge. [ÆColl]
(39) (ultro enim terra fructicat primum herbam deinde spinam deinde plenum
frumentum in spica lustum) forðon eorðo wæstmiað ærist gers æfterðon
ðone ðorn soðða full hwæte in eher. [MkGl (Li)]
(40) (In illo tempore abiit iesus sabbato per sata sabbato discipuli autem eius
essurientes coeperunt uellere spicas et manducare) in þa tid eode se
h[e]lend þurh acras on ræstedæge leorneras þa his hyngrede ongunnon
hriopan æchir & eton. [MtGl (Ru)]
(41) (Ite ecce ego mitto uos sicut agnos inter lupos) gað heono ic sendo iowih
swa lombor bitwih wulfum. [LkGl (Ru)]
(42) Muntas for hwon uphofun ge swe swe rommas & hyllas swe swe
lomberu scepa (Montes quare exultastis ut arietes et colles uelut agni
ouium). [PsGlA]
(43) (...diligis me plus his dicit ei etiam domine tu scis quia amo te dicit ei
pasce agnos meos) ...lufæstu mec suiður from ðissum vel ðisra cueð him
to gee drihten ðu uast þætte ic lufo ðec cueð him foed vel lombor mino.
[JnGl (Li)]
6. Concluding remarks
The picture emerging from the investigation of the fate of the Old English s-stems
presents an exceedingly divergent development of this inectional type in individual
dialects. The process of disintegration ranges alongside a continuum where at one
extreme is the West-Saxon dialect, followed closely by the exiguous Kentish, both
having a highly innovative pattern, testifying to a large-scale transference of the
original s-stems to the vocalic inection, where the productive inectional features
appear as early as in the 9th c. At the other extreme is the rather limited set of Mercian
archaic forms, evincing the original pattern of inection, with old mutated forms of
the nominative sg. (caelf, ceolborlomb) and endingless nominative/accusative pl. or
genitive and dative sg. (calfur). The space between these two poles is occupied by
the Northumbrian material, where members of the s-stem paradigm display some
412 Elżbieta Adamczyk
synchronic alternation between the inherited and innovative inections (albeit only
when both the singular and the plural are taken into account). Such a distribution
seems to a large extent justied and expected given the dates of attestation of
individual dialects and the fact that the earliest attested Mercian glosses, dating
back to the 8th and early 9th c., are of necessity juxtaposed with the somewhat later
Northumbrian and West-Saxon material, going back to the 10th and 11th centuries.
The attested greater archaism of Anglian seems to accord with some other archaic
features the dialect displays, such as the preservation of reduplication in class
VII of strong verbs or the lack of distinction between the two variants of the long
vowel ē (ē1 and ē2). At the same time, the substantial conservatism may be to some
extent attributed to the character of the investigated material and, in particular, to
the text-type discrepancy between the Anglian and non-Anglian data: while the
former are exclusively glosses (including very early texts), the latter, though still
translations, are somewhat more ‘vernacular’. The greater dynamics characteristic
of translations may have contributed to the scribes’ inclination to use novel forms
which reected, or at least were closer to, the spoken form of the idiom, which
tended to be more innovative.
Looking at the set of examples presented in the illustration section, it becomes
clear that the s-stem paradigm stayed under the inuence of both neuter and masculine
productive inection, and accordingly, alongside the expected neuter endingless
nominative/accusative pl., the masculine -as plural forms are occasionally attested.
This inuence appears to have been still sporadic and rather irregular at this early
stage, but it is indicative of an ongoing reassignment of gender in this group of
nouns.
It seems feasible that the impact of the endingless neuter a-stem inection may
have been enhanced by semantic factors, namely, the fact that the endingless plural
was a salient feature of the neuter a-stems denoting animals (e.g. dēor ‘deer’,
scēap ‘sheep’, etc.). Accordingly, s-stems such as lamb, cealf, hrīþor, constituting
a semantically homogeneous group, may have been receptive to the impact of
their a-stem (semantic) counterparts. Another characteristic which may have had
some bearing on the pattern of analogical restructuring is related to the concept of
plurality which seems to be inherent in some of the nouns. Consequently, the fact
that some of the attested s-stems had the nature of pluralia tantum (e.g. ēar ‘ear of
grain’, ægor ‘sea’, hrīðor ‘horned cattle’ or hōcor ‘mockery’) may account for the
notable lack of an explicit plural marker in this group.
An early extension of the e-grade (originally attested in the oblique cases only)
to the nominative/accusative sg., characterized by the o-grade, seems to be crucial
to an adequate interpretation of the restructuring process in Germanic s-stems. As
a result of this extension, the shape of these two cases became identical with that
of i- and a-stems. This development must have considerably disturbed the stability
of the s-stem inection, which opened the way to subsequent reshaping of the s-
stem paradigm on the pattern of the more inuential inectional types. In effect,
æ y
æ y
Towards a Diatopic Approach to the Old English s-Stem Declension 413
the class of s-stems began to shrink dramatically as its members drifted away to
the i-stem inection (as was the case in prehistoric Old English), or to the a-stem
declension (as was the case in the documented stage of Old English).
Finally, what is probably most intriguing about this declensional type when seen
from a broader Germanic perspective, is the subsequent divergent development
of this inectional pattern in English and German, and the fact that while English
lost this archaic inection completely (save for the isolated children), German
extended it to many other nouns which did not originate as *-es-/-os- stems. To
account for the divergent development of the original s-stem declension in English
and German, and, in particular, for the fact that the r-stem formative was never
generalised in English as a plural marker, the overall state of Old English nominal
inection must be taken into consideration, namely, the fact that in consequence
of a complex interplay of phonological and morphological factors, the English
inection relatively early became “monoparadigmatic, stem-invariant and word-
based” (Kastovsky 1985, 102), leaving thus no chance for an extension of -r- as
a plural marker. An exploration of this aspect of the development of the s-stem
declension, necessitating a wider comparative approach, would certainly offer
some further insight into the fate of this declensional type in Germanic.
Adam Mickiewicz University E
ż
A
R
Adamczyk, Elżbieta. 2009. Evolution of Germanic Nominal Inection: The Case of West
Germanic Kinship Terms. Sprachwissenschaft 34/4, 399-433.
Adamczyk, Elżbieta. 2010. Morphological Reanalysis and the Old English u-Declension.
Anglia. Zeitschrift für Englische Philologie 128/3, 365-390.
Bammesberger, Alfred. 1965. Old English gycer and Gothic jukuzi. Language 41, 416-419.
Bammesberger, Alfred. 1990. Die Morphologie des Urgermanischen Nomens. Heidelberg:
Winter.
Bosworth, Joseph, ed. 1898. An Anglo-Saxon Dictionary. (Supplement by T. Northcote
Toller, Oxford, 1921; addenda and corrigenda by A. Campbell, Oxford, 1972).
Oxford: Clarendon Press.
Boutkan, Dirk. 1992. Old English -ur/-or in the r- and s-Stems. North-Western European
Language Evolution 20, 3-26.
Boutkan, Dirk. 1995. The Germanic ‘Auslautgesetze’. Amsterdam-Atlanta, GA: Rodopi.
Braune, Wilhelm/Heidermanns Frank. 2004. Gotische Grammatik. Tübingen: Max Niemeyer
Verlag.
Brunner, Karl. 1965. Altenglische Grammatik, nach der angelsächsischen Grammatik von
Eduard Sievers. (3rd ed.). Tübingen: Niemeyer.
Cameron, Angus. 1973. A List of Old English Texts. In Roberta Frank and Angus Cameron
(eds.), A Plan for the Dictionary of Old English, 25-306. Toronto: University of
Toronto Press.
Campbell, Alistair. 1959. Old English Grammar. Oxford: Clarendon Press.
Casaretto, Antje. 2000. Korpussprachen und Produktivität. Einige Überlegungen zu den
Gotischen s-Stämmen. Historische Sprachforschung 112, 210-238.
414 Elżbieta Adamczyk
Classen, Ernest. 1915. O.E. ‘Nicras’ (‘Beowulf,’ 422, 575, 845, 1427). Modern Language
Review 10, 85-86.
Coleman, Fran. 2004. Kentish Old English <b>/<B>: Orthographic ‘Archaism’ or Evidence
of Kentish Phonology? English Language and Linguistics 8, 171-205.
Crowley, Joseph P. 1986. The study of Old English dialects. English Studies 67, 97-112.
Gordon, Eric V. 1981. An Introduction to Old Norse. Oxford: Clarendon.
Harnisch, Rüdiger. 2001. Grundform- und Stamm-Prinzip in der Substantivmorphologie des
Deutschen. Heidelberg: Universitätsverlag Winter.
Healey, Antonette di Paolo, ed. 2004. The Dictionary of Old English Corpus in Electronic
Form. Toronto: Toronto University Press.
Helten, W. van. 1911. Grammatisches. Beiträge zur Geschichte der Deutschen Sprache und
Literatur 36, 435-515.
Hirt, Herman. 1932. Handbuch des Urgermanischen II. Heidelberg: C. Winter.
Hogg, Richard M. 1988. On the Impossibility of Old English Dialectology. Luick revisited:
Papers read at the Luick-Symposium at Schloß Liechtenstein, 15-18 September 1985,
ed. by Dieter Kastovsky and Gero Bauer, 183-203. Tübingen: Gunter Narr.
Hogg, Richard M. 1992. A Grammar of Old English. Vol. 1: Phonology. Oxford: Black-
well.
Hogg, Richard M. and R. D. Fulk. 2011. A Grammar of Old English. Vol. 2: Morphology.
Oxford: Blackwell.
Kastovsky, Dieter. 1985. Typological Changes in the Nominal Inectional System of English
and German. Studia Gramatyczne 7, 97-117.
Kastovsky, Dieter. 1995. Morphological Reanalysis and Typology: The Case of the German
r-Plural and Why English did not Develop it. Historical Linguistics: Selected Papers
from the Eleventh International Conference on Historical Linguistics, Los Angeles,
16-20 August 1993, ed. by Henning Andersen, 227-238. (Current Issues in Linguistic
Theory 124). Amsterdam: Benjamins.
Ker, Neil Ripley. 1957. Catalogue of Manuscripts Containing Anglo-Saxon. Oxford:
Clarendon Press.
Kitson, Peter. 1995. The Nature of Old English Dialect Distributions, Mainly as Exhibited in
Charter Boundaries. Medieval Dialectology, ed. by Jacek Fisiak, 43-135. Berlin/New
York: Mouton de Gruyter.
Kitson, Peter. 2004. On Margins of Error in Placing Old English Literary Dialects. Methods
and Data in English Historical Dialectology, ed. by Marina Dossena and Roger Lass,
219-241. Bern: Peter Lang.
Kluge, Friedrich. 1913. Urgermanisch. Vorgeschichte der Altgermanischen Dialekte.
Strassburg: Verlag von Karl J. Trübner.
Kluge, Friedrich. Nominale Stammbildungslehre der Altgermanischen Dialekte. Max Nie-
meyer Verlag: Halle.
Krahe, Hans. 1969. Germanische Sprachwissenschaft II. Formenlehre. Berlin: Walter de
Gruyter.
Kuhn, Sherman M. 1939. The Dialect of the Corpus Glossary. PMLA 54, 1-19.
Lehmann, Winfred P. 2005-2007. A Grammar of Proto-Germanic. University of Texas at
Austin, available at http://www.utexas.edu/cola/centers/lrc/books/pgmc00.html
Lowe, Kathryn, A. 2001. On the Plausibility of Old English Dialectology: The Ninth Century
Kentish Charter Material. Folia Linguistica Historica 22, 67-100.
Noreen, Adolf. 1894. Etymologisches. Indogermanische Forschungen 4, 320-326.
Prokosch, Eduard. 1939. A Comparative Germanic Grammar. Philadelphia: University of
Pennsylvania.
Ringe, Don. 2006. From Proto-Indo-European to Proto-Germanic. A Linguistic History of
English. Vol. 1. Oxford: Oxford University Press.
Ross, Alan S. C. 1937. Studies in the Accidence of the Lindisfarne Gospels. Leeds: Kendal.
Sauer, Romuald. 1917. Zur Sprache des Leidener Glossars. Augsburg: P. J. Pfeiffer.
Schenker, Walter. 1971. -es/-os-Flexion und -es/-os- Stämme im Germanischen. Beiträge
zur Geschichte der Deutschen Sprache und Literatur 93, 46-59.
Towards a Diatopic Approach to the Old English s-Stem Declension 415
Schlerath, Bernfried. 1995. Bemerkungen zur Geschichte der -es- Stämme im West-
germanischen. Verba et Structurae. Festschrift für Klaus Strunk zum 65. Geburtstag,
eds. Heinrich Hettrich et al., 249-264. Innsbruck: Institut für Sprachwissenschaft.
Simpson, John A. and Edmund S. C. Weiner, eds. 1989. Oxford English Dictionary (OED),
available at: http://www.oed.com/
Streitberg, Wilhelm. 1963. Urgermanische Grammatik. Carl Winter Universitätsverlag:
Heidelberg.
Toon, Thomas E. 1992. Old English Dialects. Cambridge History of the English Language,
ed. by Richard M. Hogg, 409-451. Cambridge: Cambridge University Press.
Unwerth, von, Wolf 1910. Zur Geschichte der Indogermanischen es/os-Stämme in den
Altgermanischen Dialekten. Beiträge zur Geschichte der deutschen Sprache und
Literatur 36, 1-42.
Versloot, Arjen P. 2008. Mechanisms of Language Change. Utrecht: LOT.
Wełna, Jerzy. 1996. English Historical Morphology. Warszawa: Wydawnictwo Uniwersytetu
Warszawskiego.
Weyhe, Hans. 1906. Beiträge zur Westgermanischen Grammatik. Beiträge zur Geschichte
der Deutschen Sprache und Literatur 31, 43-90.
Wright, Joseph and Elisabeth Wright. 1908. [1967]. Old English Grammar. London: Oxford
University Press.
Wurzel, Wolfgang Ullrich. 1989. Inectional Morphology and Naturalness. Berlin: Aka-
demie-Verlag.
Wurzel, Wolfgang Ullrich. 1992. Morphologische Reanalysen in der Geschichte der deut-
schen Substantivexion. Folia Linguistica Historica 13, 297-307.
APPENDIX
As the present investigation was conducted on the sizeable corpus of Old English
texts, compiling a comprehensive list of the texts studied, and enumerating all of
them here was found rather impractical - given the limits of space. This applies
in particular to the texts of West-Saxon origin. Since the size of the West-Saxon
material exceeds considerably that of the non-West-Saxon corpus, the s-stems were
found, quite expectedly, in an incomparably larger number of texts (over 350 texts,
including some longer ones divided into chunks in the DOE corpus). Consequently,
the present appendix section is conned to a list of texts (or fragments of texts)
of non-West Saxon provenance, in which the s-stems were actually found in the
course of the present investigation. The forms of s-stems which happened to occur
in texts dialectally unidentiable, classied as Unknown in the DOE corpus, were
dismissed from the quantitative investigation altogether, and these texts are not
mentioned here either (cf. section 4).
The relevant information on the dating and dialectal afliation of individual
texts presented below comes from standard historical grammars, including Brunner
(1965), Campbell (1959) and Hogg (1992), supplemented by the data from Ker’s
authoritative Catalogue of Manuscripts Containing Anglo-Saxon (1957). Detailed
information about the manuscripts and editions used in the DOE corpus can be
found in the ‘List of Texts’, appended to the corpus, containing all texts cited in the
dictionary (cf. footnote 20). For the purpose of greater transparency and veriability
of the investigated corpus, the short title as used in the DOE and the corresponding
Cameron number were added to each of the mentioned texts.
416 Elżbieta Adamczyk
Kentish
Kentish Hymn (10th c.) KtHy, A26
Kentish Psalm (10th c.) KtPs, A25
Kentish Glosses (Glosses to the Proverbs of Solomon) (10th c.) OccGl
49, C49
Kentish charters (9th c.) S1482, B15.6.1; S1188, B15.2.1; S1195,
B15.2.2, S1482, B15.6.1, S1197, B15.2.4,
S1510, B15.6.27
Canterbury Psalter (c. 1150)33 PsGlE, C7.3; PsCaE, C11.2
Mercian
Corpus Glossary (8/9th c.): CorpGl 1, D4.1
Erfurt Glossary (8th c.): ErfGl 1, D36.1; ErfGl 3, D36.3
Vespasian Psalter (mid-9th c.): PsGlA, C7.7
Rushworth Gospels (Ru1) (10th c.): MtGl (Ru), C8.2.1; MkGl (Ru),
C8.2.2, JnGl (Ru), C8.2.4
Mercian charters (9th and 10th c.): Ch218, B15.1.8, Ch1327, B15.3.24;
Ch1362, B15.3.35, Ch1380, B15.8.572, Rec
23.6, B16.23.6)
Northumbrian
Durham Ritual (10th c.): DurRitGl1, C21.1; DurRitGl2 C21.2
Lindisfarne Gospels (10th c.): MtHeadGl (Li), C21.3; MkHeadGl
(Li), C21.5; JnHeadGl (Li), C21.9; LkGl
(Li) C8.1.3; MtGl (Li), C8.1.1; MkGl
(Li) C8.1.2; JnGl (Li); C8.1.4; LiProlMt
(Skeat), C20.3, MkFestGl (Li), C21.6;
MtFestGl (Li) C21.4
Rushworth Gospels (Ru2) (10th c.): LkGl (Ru), C8.2.3; JnGl (Ru),
C8.2.4
33 It must be noted that the extensive Dictionary of Old English corpus covers the period
c. 600-1150, and therefore, contains also some fairly late texts, which can be viewed as
belonging to at least the transitional period between Old and Middle English, dated as late
as to the 12th century.
