W. W. (Walter Wilson) Greg.

The Assumption of the Virgin. A miracle play from the N-town cycle online

. (page 1 of 7)
Online LibraryW. W. (Walter Wilson) GregThe Assumption of the Virgin. A miracle play from the N-town cycle → online text (page 1 of 7)
Font size
QR-code for this ebook




Prirffiur st-.i'.h <; <>r / tirpcncf net














1. The Scribe 6

2. Dialect 9

Rime-index 22

3. Metre 26

Rime-scheme 34

4. Authorship 36

5. Source 39

Legenda Aurea, Cap. CXIIII 40




*** The Facsimiles are inserted between pages 44 and 45.



Macula, moder, was neuere in thee,
Filia Syon, thou art the flour ;
Full sweteli schalt thou sitte bi me
And bere a crowne with me in tour,
And alle my seintis to thin honour
Schal honoure thee, moder, in my blis,
That blessid bodi that bare me in bowur,
Vem\ coronaberis.

Lambeth MS. 853.


IN the Cottonian Manuscript, Vespasian D. viii, containing the
N-town cycle of miracle plays, commonly but incorrectly known
as the Ludus Coventriae, there is one play, that numbered
forty-one, treating of the Death and Assumption of the Virgin,
which stands markedly apart from the rest. It is written in
a hand which appears nowhere else in the volume, and this hand
presents certain peculiarities differentiating it sharply from that
in which, with few exceptions, the whole of the rest of the cycle
is written. The paper of the play likewise is quite distinct from
any used in other parts of the volume ; it forms a single quire
inserted in the middle of one of the other quires of the manu-
script, and could be removed without in any way interfering with
the remaining leaves. These facts necessarily raise a question as
to how far this particular play forms an integral part of the cycle
in which it appears, or at least as to whether it may not have an
origin essentially different from the rest. It is noteworthy that
the play in question is not recorded in the very explicit catalogue
which serves as a prologue to the cycle, but the same is true of
other plays which do not differ in handwriting from their com-
panions. To enter fully into this question would be to raise the
whole intricate problem of the origin and history of the N-town
cycle, which it is not my purpose to do on the present occasion.
My immediate and more modest intention is to inquire whether
any marked difference in dialect or style of composition exists
between the Assumption play and the bulk of the cycle, such as
could be adduced in support of the bibliographical evidence for
an independent origin. This limited inquiry would hardly of
itself have necessitated, or perhaps even justified, reprinting the
text in question, since the whole cycle is already accessible in
J. O. Halliwell's edition, issued by the Shakespeare Society in
1841 under the title of Ludus Coventriae. I have, however,


been Impelled to the more ambitious course by two other con-
si<5eratioft. -The first of these is that one of the most remark-
able characteristics of the play is its metrical structure, and that
this, though perfectly clear in the manuscript, is very effectively
concealed in the printed edition. The other is the opportunity
which a reprint affords of bringing the English text into close
relation with its source in the narrative of the Legenda Aurea>
and thus of drawing attention to the importance of Jacobus de
Voragine's work for the study of the religious drama. For this
purpose I have reprinted at the end of this Introduction those
portions of the legend of the Assumption upon which our present
play is based.

I owe a debt of gratitude to Dr. Henry Bradley, to whose
expert assistance is due whatever may be found of value in my
Notes, and to Mr. C. E. Stuart for help with the Latin both of
the manuscript and of the Legenda.


The bulk of the N-town cycle is written in a clear, rather
commonplace hand of the later fifteenth century. At the end
of one of the plays is the date 1468, and there does not appear
to be any reason to suppose that this is not the date of writing.
The scribe makes free use oi y (/) for th t though the latter also
appears, and likewise of the letter 3, though more often for y
than for gh. He also has the well-recognized East Anglian, or
at least East Midland, peculiarity of writing x in place of sch in
such words as shall, &c. The ink used is generally some shade
of brown.

Of the play of the Assumption Halliwell remarks (p. 417):
' The whole of this pageant is written in a more recent hand, of
the time, I should think, of Henry VIII.' I do not know upon
what ground he based this opinion, possibly upon the fact that
the special letters, or uses of letters, just mentioned as occurring
in the rest of the manuscript, are all but entirely absent from the
play in question. But whatever may have suggested Halliwell's
opinion, that opinion is wrong. There is nothing whatever in
the character of the hand, as may be seen by consulting the
accompanying facsimiles, to suggest that it is not contemporary


with that of the main scribe, and it is easy to prove that it is
so. For the whole manuscript, including our play, has been
rubricated at one time and in one manner, obviously by one
person, and a careful examination of his work throughout the
volume will show that that person was none other than the main
scribe. Moreover, at line 2,61 of our particular play, some words
that had been added in the margin and subsequently mutilated
have been written over again by a different hand, and this hand
is that of the main scribe. It follows, therefore, that the
Assumption play was written at any rate not long after the rest
of the manuscript and may even have been written before it.
The ink is black.

The main scribe we know belonged, as already said, to the
East Midlands, probably to the more restricted area to which the
term East Anglia may be applied. Is there any indication as to
the locality in which our play was written down? In it the
letter 3 does not appear,/ is used only twice, and y for th is so
rare as to suggest mere accident. Nor does x replace sch. This
obvious criterion of East Anglian writing is therefore absent.
But another recognized test exists in the dropping of the
guttural. The peculiarity is properly a scribal, not a dialectal,
one, for the sound probably went out of pronunciation in the
midlands early in the fifteenth century, but except in the extreme
eastern district it continued as a rule to be written. We start in
OE with final -h or -ht\ in ME these are represented by -^, -j/
or -gk, -ght. With the loss of the guttural -^, -gh disappear
altogether, -}t gives -/, and -ght may give -ht. In the last case,
the h having no phonetic value, the termination comes to be
written indifferently -ht or -th> and this leads to the substitution
of -th for -/ even where there was no original guttural. The
question has been discussed at some length by Furnivall in his
* Afterwords ' to the EETS edition of the Macro Plays (1904,
p. xxxv). He there cites many such forms as fite (fight), lite
(light), rith (right), sith (sight), deth (dead), kyth (cut), from
Norfolk manuscripts. Now this peculiarity, though observable
in the main body of the cycle (p. 58, syte ; p. 87, nowth ; p. 147,
sowth ; p. 1 86, rowthte, rout), is not very prominent ; we usually
find the more customary spellings. But in the Assumption play
the case is altered. Spellings with gh seem to be as completely


absent as those with j. A final guttural usually disappears, as in
hye for high (1. 33), but we once find nyhyn (1. 194). Before /,
however, the scribe does not affect mere omission. The ending
ght becomes regularly -th or -tht or, less frequently, -ht (1. 5,
tauht) 1. 146, myhtis, 1. 199, myhtys).

The only other orthographic peculiarity that seems worth
mentioning is the use of qw for wh, which occurs more than once
in our play, though by no means regularly (11. 137, 269, qwyche).
This is peculiarly though not exclusively northern ; Furnivall in
the above-mentioned ' Afterwords ' cites instances from the East
Midland play of Mankind. Anyhow it occurs in other parts of
the cycle (see Halliwell's glossary), and is, therefore, not distinc-
tive of the present piece.

There is one striking piece of evidence that the scribe had
northern leanings, to say the least of it. In line 238, namely, we
find the form skele for skill. Now this form is characteristically
northern, or even Scotch, but it will be observed that it is due to
the scribe, not the author, for the rime requires skill. Curiously
enough there is just such another isolated piece of evidence that
the main scribe of the other portions of the cycle had the same
northern tendency. We once (p. 404) find the very distinctively
northern spelling ssalte, in place of what would be the regular
xalte, shalt.

Consequently, I do not think that there is any reason to suppose
that the play of the Assumption was written down in a different
locality from the rest of the manuscript in which it is found.

A word may be added on the use of contractions by the scribe.
In the English text these are as a rule perfectly normal and
present no difficulty. It will be sufficient to remark that the
contraction for ser consists of a long / with a mark like a 7
through it. This I have represented by 'fs' (11. 194, 209, 210),
which must be regarded as a single symbol. Whether the cross
stroke of tt and ft was intended by the scribe to have any par-
ticular meaning it is difficult to say, but the fact that he avoids
adding a final e to these letters seems to indicate that the marks
were not wholly devoid of significance. As is so often the case,
difficulty is caused by the final curl over a letter which may be
either n or u. I have printed n wherever possible. But now and
then the temptation to print u has been great : for instance in


adon and mon, 11. 620, 622. But it must be observed that in the
same rime-series we have demon which requires no u> and town
where the insertion would be preposterous. I think, therefore,
that the curls are best regarded as throughout insignificant.

In the Latin portions the contractions are much more frequent
and less usual, the scribe apparently considering that a curl
might be expected to do duty for any termination, or indeed for
any part of a word which he felt disinclined to write in full.
In certain cases his forms are, to say the least, misleading. For
specific instances the notes may be consulted. I have there con-
sidered such difficulties as arise, and have also indicated the
meaning of certain normally contracted forms which may never-
theless be unfamiliar to modern readers.


The problem of determining whether the dialect of the
Assumption play differs from that of the rest of the N-town
cycle is rendered the more difficult by the fact that even apart
from this play the collection is manifestly of very complex
origin and may well include portions originally composed in
widely different localities. In what follows it must, therefore*
be borne in mind that if we succeed in showing that the one play
under review possesses dialectal peculiarities not shared by any
of its companions, we shall, of course, have satisfactorily estab-
lished its independent origin, but that if, on the other hand, we
fail to do so, it does not in the least follow that the dialect of the
whole cycle is homogeneous, nor even, strictly speaking, that the
dialect of our particular play is identical with that of any other
single member of the collection.

The language of the N-town cycle was investigated by Dr. Max
Kramer as long ago as 1892 in his dissertation on Sprache und
Heimat des sogen* Ludus Coventriae. His object was to com-
pare the dialect of the plays with that of the Coventry records,
and his result that the former, so far from being identical with
the latter, belonged in its present form rather to the northern
border of the East Midland district. He treated the dialect of
the cycle as a whole, and though recording certain anomalous
forms, made no attempt at a more individual treatment of the

separate pageants. I propose, nevertheless, to take his investi-
gation as a basis, and to examine how far, in the extensive
collections of variant forms his work contains, those from the
Assumption play can be regarded as forming a class by them-
selves. The only statement regarding the dialect of this particular
play with which I am acquainted is one by Mrs. M. H. Dodds
in the Modern Language Review for January 1914 (vol. ix, p. 90).
Speaking of the N-town cycle she says : ' All the plays are in
the dialect of the East Midlands except the addition of " The
Death, Funeral, Assumption and Coronation of the Virgin ",
which bears some traces of the northern dialect '. No evidence,
however, is adduced in support of this opinion.

In the following survey I shall adopt the order of Dr. Kramer's
paragraphs. OE a gives in ME a (sometimes d in open syllables),
but this in turn gives % in NE, and it may be asked whether
this process had already begun at the time when the N-town
cycle was written. That it had seems proved by rimes with
words of which the vowel is derived from OE e or x. But it
should be observed that the evidence for this seems stronger
in the case of the Assumption play (A) than in that of the
rest of the cycle (LC). Thus we find, A 575, save (OF salver),
have (OE habbari), riming with belave (OE belxfan}, and, A 310,
declare (OF declarer), fare (OE faran), care (OE cearu) : lare
(OE Ixran). Elsewhere we do indeed find rimes of care, fare,
spare (OE spariari), with ware, were (OE wxrori), but the latter
is open to the suspicion of having been influenced by ON wdru
(LC 47 12 , 73 10 , i83 26 : the large figures indicate the pages, the
superior figures the lines, of Halliweirs edition). One clear case,
however, is quoted, LC 3O 27 , qweke (OE cwacian) : freke (OE
freed}, breke (OE brecari). Possibly the same sharpening of the
sound may be observed in rimes with OF at, particularly the
word sertayn, of which there are several instances, LC 43 1 , 327 18 .
One case occurs in A 30, name (OE nama), same (ON samr),
defame (L defamare) : attayne (OF attaindre), but here the text
is very likely corrupt. As a rule before m, n the a seems to
remain pure or to become o. An exception is, according to
Kramer, A 45, on (OE an) : won (OE wuniari). But here it
must be observed that on being emphatic has been assimilated to
the <fo-rimeSj and that won is rather from ME wdnien, OE


wanian, or else perhaps from OE wandian. However, these words
undoubtedly show the Anglian change of a>o before a nasal.

The ME vowel derived from OE x, or by shortening from x,
rimes both with a and e. No instances from our play are quoted,
and very few appear to exist. In the rime, A 269, bad : sad, the
vowel in either case goes back to OE x, and nothing can there-
fore be inferred. There is one instance of an *-rime, A 314,
ment (OE mxned) : sent (OE sended), but this belongs to a group
which regularly has e in ME. Once we have, A 575, grave (OE
grsef) : cave (OF cave) y have (OE habbari), but these vowels,
as we saw above, must have been sharpened, for the rime
series likewise includes lare (OE Ixran). The paucity of
examples of rime words with these vowels in our play is unfor-
tunate, for the cycle as a whole shows a marked tendency to
make these rime on e, indeed whole classes, such as the preterites
and participles which in OE end in -xdan, -xpan, do so ex-
clusively. The tendency points to the southern and south-eastern

The ME representative of OE u is found riming with the
vowels derived from OE 6 (OF o\ from OE d> and once from
OE a (o) before ng. Only the second of these is illustrated by
Kramer from our play, and this in the word won> A 45, which
he takes to represent OE wunian, whereas we have already
seen that it represents either OE wanian or else OE wandian.

The weakening of OE i to e does not seem to be illustrated in
our play ; on the other hand, the parallel weakening of OE y to
e is common, as throughout the cycle. Rimes with short e are
A i%4,mende (OE gemynd), kynde, kende (OE cynd) : hende (OE
gehende), pretende (Q pretendre) ; A 660, mend, kend : ende (OE
ende), wend (OE wendan) ; A $<)6,felthe (OEfyty?) : helthe (OE
hxlf), welthe (ME wele, OE weld), the vowel in each case being
probably shortened ; A 614, kerne (OE hyrne),brenne (ON brenna).
Rimes with ^ are only found in the case of the word meche (OE
my eel), as in A 326, where it rimes with speche (OE spxc), teche
(OE txcari), preche (OF precher). It may be remarked that
of course the representative of OE y also rimes with *', as in
A 232, hyllys (OE hyll) : wyll is (OE gewill), though indeed both
may have become dulled.

That OE d had in the great majority of cases developed into


8 is shown by the frequent rimes with OF o and the representa-
tive of OE 6. Instances from our play are, for the former, A 333,
gon (OE gdn) : iron (OF trone) ; A 430, alon (OE dn) : tron ; for
the latter, A 250, sone (OE sona), done (OE ddn) : one, none (OE
dn), gone*(OE gdn} ; for both, A 608, do (OE ddn) : go (OE gdn) :
harro (OF haro). But in a certain number of cases the northern
tendency to retain d manifests itself, as appears from rimes
between the representative of OE d and OF, ON, or OE a.
Examples from the cycle are, LC 5 9 , more (OE mdrd), sore (OE
sdr) : war (OE wxr, ON var), Abyacar ; LC 43 1 , gan (OE gdn) :
man (OE mann^ monri)> tan (ON taka), certayn (OF certain) \
from our play, A 310, ever mare (OE mdra) \fare (OE far an),
care (OE carian), declare (OF declarer) : lare (OE tierari). It
will be observed that not only does the Assumption agree with
the rest of the cycle in sometimes retaining OE d, but likewise in
indicating a certain sharpening of this sound as shown by the
rimes with certayn and lare respectively.

As one would expect, the open and closed values of 6 rime
freely together throughout the cycle, though their different
development in NE (to d and d respectively) shows that they can
never have been identical. It is also to be observed that our
play agrees with the rest in riming the representative of OE dw
with that of OE, ON ag\ for instance, LC S; 12 , knawe (OE
gecndwan) : lawe (OE lagu), withdrawe (OE dragari)> awe (ON
agi) ; A 44, knawe, blawe (OE blow an) : lawe, awe.

The OE gave in ME both open and closed ^ according as it
was derived from Germanic ai or not. In Chaucer the two
values are distinguished : not so in any part of the N-town cycle.
We may instance, LC 4o 22 , sprede (OE sprdan, Gm. *spraid-
jari) : indede (OE dxd, Gm. *ddt*) ; LC 9i 28 , lede (OE l&dan,
Gm. *laidjan) : sede (OE sd t Gm. *sdom) ; and, A 389, sprede :
sede. In the same way ME ^ (open) from OE from Gm. ai
rimes with the closed ^ from OE <? ; and ME ^ (closed) from OE
* not from Gm. ai rimes with open e from OE ea. Examples of
the former are, LC 132, arere (OE drran, Gm. *raizjan) '
here (OE Mr) ; LC tftjft lere (OE l&ran, Gm. *laiajan) : here,
fere (OE gefera) ; and A 542, lere : bere (OE b&r, Gm. *brd),
here ; for the latter, LC a; 5 , reed (OE rd, Gm. *rdoz), dred
(OE andrddn> Gm. *drdan) : ded (OE dead), sted (OE stede) ;


LC I57 21 , methe (OE mxp, Gm. *mxpiz], brethe (OE brxp, Gm.
*brxpoz) : dethe (OE deaf) ; and A 492, red (OE ^dk, Gm.
*rxdari), blede (OE J/Afo) : <fe*.

We have already, in considering OE a, found an instance in
our play of the representative of OE x riming with d, namely,
A 310, tare (OE Ixran, Gm. *laizjan) : evermare, fare, care,
declare. A similar instance is, A 575, belave (OE belxfan, Gm.
*bilaibjan) : grave, cave, save, have. No such rimes are quoted by
Kramer from the rest of the cycle. They are, however, to be ex-
plained less as broadening of the ^than as sharpening of the d, #,and
of this, as we have seen, there is evidence elsewhere in the cycle.

Beside the common forms there, were (from be) the N-town
cycle also presents instances in which both the vowels a and o
appear: LC 73, thare, ware : are (OE Merc, earun, North.
aron), bare (OE bxr) ; LC 47 12 , were : care (OE cearu},fare (OE
farari), bare; LC in 24 , thore : bore (OE boreri), before (OE
fore), restore (OF restorer) ; LC 146 24 , wore : bore, beffore. In
the Assumption the form there only occurs once in a rime,
A 542, there : bere (OE bxr), here (OE h$r), fere (OE fxr), kre
(OE Ixran), while the form thore occurs twice, A 164, thore:
evermore (OE mdra), lore (OE Idr), before ; A 266, thore : more,
bore. Neither thare nor any of the forms of were appears
in a rime. Kramer derives the forms thore, wore from ON
par, waru (vdru). In the case of the latter this is no doubt
correct, but as regards there the ON form is par not f dr. This,
however, raises no difficulty, for in OE the forms par, para are
well-established variants of pxr. As regards the forms thare,
ware Kramer points out that the vowels may be instances of the
retention of a, a tendency we have already observed, or that they
may really represent ^-sounds riming with a sharpened a, and he
decides for the latter alternative on the ground that the forms
also appear in texts which show no other evidence of the reten-
tion of original OE, ON d. This, then, affords further evidence
of the sharpening of the #-sound outside the Assumption play.
Kramer further remarks that in some cases OE mxnan has given
the form mone in the N-town cycle, and he quotes LC 98 7 , 346 14 .
But in both these cases mone is the substantive, moan, which
must go back, not to OE mxnan, but to the unrecorded OE
*man from the same prehistoric stem *main-. The word does not


occur in the rimes of the Assumption unless, A 319, ment is an

In a few cases, according to Kramer, the representative of OE t
rimes with ^-sounds from OE /, , &, OF e. He quotes LC 349 19 ,
lyff (OE ///): greff (OF gref\ theff (OE }tof)\ and A 3 4 3>
teme (OE tima) : sweme (OE aswxman), queme (OE civeman),
seme (OE seman). The first of these appears to be correct, the
second is not. The word teme is not OE tima, time, but OF
(*teme) tesme, theme. Similar to the ^: t rimes are those of 6 : A.
Of these may be quoted, LC I9O 22 , book (OE boc) : sowke (OE
sucan) ; LC I46 15 , boun (ON bon) : downe (OE dune) ; LC 315",
don (OE don) : mon (OE mugon). In our play we find the
strange assortment of rimes, A 615, preso(ii)n (OY prison, prisun) :
demon (OF demon, med.L demon) : so(u)n (AF soun, OF son ;
OE sdn from L .waitf probably did not survive) : ado(ii)n (OE of
dune) : mo(u)n (OE mugon) : town (OE tun), but it may be ques-
tioned whether any of these were strictly ^-sounds.

The Assumption play agrees with the rest of the cycle in
riming the word here, hear, with / only: LC 79 9 , here : appere
(OF stem aper-, apareir) ; LC 1 14 15 , here \fere (OE gefer), perse-
vere (F persMrer), dere (OE dfore) ; and A 3, here : lere (OE
Ixran), clere (OF cler),yere (OE gear). This shows derivation
from the Anglian hran and not WS hyran.

Throughout the cycle the words frende &h&fende (QfLfreond,
feond) are found among the ordinary rimes in -end, and no indica-
tion appears of any such difference of quantity as appeared later
and lead to different developments in NE. Thus, LC I36 29 ,
frende : wende (OE wendan) ; A 660, frend(e) : ende (OE ende),
mend (OE gemynd), kend (OE cynd), wend\ and LC 276 2 ,
ffende : mende\ A 184, /<?<& : mende,kynde, hende (OE gehende),
pretende (OY pretendre), assende (L ascendere).

Discussing the diphthong ay with its variants, the representative
of OE xg, eg both long and short, Kramer attempts to show
that the word again rimes in the N-town cycle with the repre-
sentative of OE ^, &. We have, LC i68 4 , a}en : qwen (OE
cwen) ; LC i6q**,ageyne : quene ; LC I77 28 , a$en : ben (OE beon) ;
LC 379*, ageyn, serteyn (OF certain) : seyn (OE s/on). He also
quotes, A 659, ageyn \greyn. He does not say what origin he
proposes for the latter word, but his quoting it in this connexion


implies that he takes it as OE grene, green. It is, however,
certainly OF grain in the sense of dye, stain. The word again
occurs three times in our play, but always riming with true diph-
thongs. Moreover, it does not appear to be correct to say that

1 3 4 5 6 7

Online LibraryW. W. (Walter Wilson) GregThe Assumption of the Virgin. A miracle play from the N-town cycle → online text (page 1 of 7)