7
She <&iflht-t«t Titian of lite
;antei[Irm[j)
f
dfigM-toft tifditura of
<Cmitei[bunj
WITH REMARKS UPON
Cjie Classification of % Ulamisuipts tmb upon
Hadrian gftauusqipt 7334
BY THE
REV. WALTER W. SKEAT,
LiTT.D., LL.D., D.C.L., PH.D., F.B.A.
ELRINGTON AND BOSWORTH PROFESSOR OF ANGLO-SAXON IN THE UNIVERSITY OF CAMBRIDGE,
AND FELLOW OF CHRIST'S COLLEGE
LONDON:
PUBLISHED FOR THE CHAUCER, SOCIETY
BY KEGAN PAUL, TRENCH, TRUBNER & CO., LTD.,
DRYDEN HOUSE, 43 GERRARD STREET, SOHO, W.
AND BY HENRY FROWDE, OXFORD UNIVERSITY PRESS,
AMEN CORNER, E.G., AND IN NEW YORK,
1909, for the Issue of 1905
/W
Sutrib S«rits, 43.
RICHARD CLAY ife SONS, LIMITtU, LONDON AND BUNGAY
INTRODUCTION
As the present article contains results of a rather miscellaneous
character, a short description of its contents is here given.
Pp. 1-18. This portion contains nothing new, and is intended less
ibr present perusal than for permanent use. I here collect all that
is valuable in the " Eight-text " (as explained at p. 1) in the way of
Titles, Colophons, Glosses, Remarks, and Side-notes. The References
to Authorities (pp. 9-18) are, of course, of very great interest and
value ; but they have long been known and were mostly familiar to
Tyrwhitt, who made use of them, and they are largely quoted in the
Notes to my six-volume edition of Chaucer. They are here, for the
first time, carefully collected and adapted for handy reference.
Pp. 18-28. Here follows a line-by-line collation of the eight MSS.
published by the Chaucer Society, showing at a glance, by means of
Tables, the exact contents of each of these MSS. One object of making
this- comparison (besides the obvious advantage of it) was to test my
theory of " The Evolution of the Canterbury Tales," which it strongly
confirms.
Assuming that the eight MSS, can be chronologically arranged in
the following order, viz. Hn., Pt., Cp., L., HI., Cm., Dd., E., we
should naturally expect to find continual additions, so that Hn. (the
oldest) would contain the least number of lines, and E. (the latest) the
greatest. And such proves to be actually the case, when casual imper
fections of the individual MSS. have been duly considered, as they
should be. Eor example, the Table at p. 21 seems to show that Cp.
and L. have fewer lines than any, as regards Melibeus; but this is
merely due to scribal carelessness ; and the same remark applies to the
Table on p. 26, where the omission of E 117, 143, 173, and the like,
cannot possibly be taken into account as real ; for no one can suppose
that Chaucer ever wrote six lines only in a seven-line stanza! Rightly
considered, the above order of the eight MSS. is justified by actual
results.
By the phrase " chronologically arranged," I refer, of course, to the
character or type of each MS., not to the absolute date at which it was
vi Introduction.
written out. There is a wide and obvious difference between the two
conditions. If we were to come across a sixteenth-century MS. contain
ing a copy of the A-text of Piers Plowman, it would of course precede,
" chronologically," every MS. that exists of the B-text and C-text.
If the reader pleases to look upon my theory of " The Evolution of
the Tales " as being wholly imaginary, this will not affect the results
given in the Tables, the usefulness of which ought, in any case, to be
apparent. When I show, e.g., at p. 27, that the lines F 1-8 occur
in MSS. Hn., Pt., HI., Dd., and E., but are lacking in Cp., L., and
Cm., I merely tabulate a fact. And the result is just the same, if any
other order of MSS. be adopted. On the other hand, it will be found
that the order which I have adopted is by far the most intelligible and
satisfactory ; probably (I think) because it is right. I adopted it, in
the first instance, merely by way of experiment; yet it explains (in
my view) nearly everything.
Pp. 29-36. Remarks upon the famous MS. Harl. 7334, regarding
which the last word has by no means been said. I show that the
Comparative Tables (pp. 18-28) certainly help us to understand it
better, and bring into relief certain characteristics of it that will have
to be considered. I also illustrate by examples, both its excellences
and its defects. See pp. 23, 24, 32, 35.
Pp. 36-43. I here consider certain readings in MS. HI. ; with
further remarks upon some of its peculiarities.
Pp. 44-55. I here discuss the " Classification of the MSS." ; with
special reference to the order already assumed, viz. Hn., Pt., Cp., L.,
HI., Cm., Dd., E. This amounts to a re-statement of the same theory
that I have already stated rather fully in my previous essay on " The
Evolution of the Canterbury Tales," which seems, so far, not to have
been properly understood ; for which reason, I now draw attention,
once more, to some of the more important results. But the present
essay really goes a good deal further ; for I have found the theory so
illuminating that I can now explain many things that were formerly
dark, and I now find hardly any difficulty left. Every fact, however
minute, seems now to drop into its right place ; and I find everywhere
confirmation of it. The darkest point was the history of lines
E 2419-E 8 in the Petworth MS., which I did not previously under
stand, but can now explain. See pp. 49, 50, 54, 55, 62-64.
I can easily understand why my theory of the "Evolution" finds
but slow acceptance; for the facts are extremely complex, and it
requires much patience to arrange them. For this reason, I reconsider
Introduction.
vn
(pp. 46-52) the most difficult points presented by the problem of re
arrangement, and discuss each "link" separately, in the hope of
making the results clear even to those whose patience or whose time
is limited. For the same reason I here give a brief summary of the
chief results. They will bear testing ; but can only be properly
tested by such as can spare leisure for it.
1. Neglecting all the Tales but four, we find that MS. Hn. has
the following arrangement, viz. Man of Law, Squire, Merchant,
Franklin.
2. Here the Man of Law and Squire are not linked together. But
Pt. provides, as a link, the old Squire's Prologue (B 1163-1190).
This Prologue is extremely common, as it occurs in all the numerous
MSS. of the "Pt." type, and in all the black-letter editions. See
pp. 47, 48.
3. The same arrangement, with the same link, appears also in Cp.,
L. See p. 47.
4. But in the next arrangement, HI., this link was practically can
celled ; and consequently disappears from the "edited" MSS., viz.
Cm., Dd., E. See pp. 47, 48, 51, 54.
5. Next, as to the Squire-Merchant. These were linked in Hn. by
the old Merchant's Prologue (F 673-708). The same order, with the
same link, occurs in Pt. But in Cp. L. the Merchant is shifted away,
and this link wholly disappears. In HI. the link probably reappeared,
only the MS. is here defective ; it certainly reappears in Cm., Dd., E.
In these it still follows the Squire, but now precedes the Franklin ;
and the changes required by its new position are duly made. It is
now called " Words of the Franklin." See pp. 48, 49, 52, 54.
6. Next, as to the Merchant-Franklin. These were likewise linked
in Hn. by lines E 2419-F 8, the old Franklin's Prologue. In Pt.
this connexion was severed, and the link became, practically, unmean
ing ; but the scribes preserved it, in front of the Franklin, though the
references to the Merchant's Tale had become absurd. Various scribes
tried various devices for saving it ; all more or less ineffective, but all
instructive to the student. What Cp. L. did with this link I do not
know ; as these MSS. are here defective. But, in HI., Chaucer came
to the rescue himself, and settled the matter easily enough. Instead
of placing it before the Franklin, he placed it before the Squire, alter
ing the allusions accordingly. And instead of placing it after the
Clerk (as in Pt.) he restored it to its old place after the Merchant (as
in Hn.), so that all came right. See pp. 49, 50, 52.
viii Introduction.
I have now accounted for all the links in the original sequence,
viz. M;m of Law, Squire, Merchant, Franklin.
7. But there is one more link, too important to be passed over.
In L., the Merchant had been placed after the Clerk, but without a
link. In HI., Chaucer now supplied an entirely new Clerk-Merchant
link (E 1213-1244) in a masterly manner. Of course Cm., Dd., E.
carefully reproduce it. See pp. 51, 54.
These seven points give, briefly, the history of the four principal
and variable links. It is not easy, but I can make it no clearer;
though I give all necessary details below. However, at pp. 62-64,
I quote in full all that is necessary for explaining the old
Merchant-Franklin link, which became the new Merchant Squire
link (E 2419-F 8).
8. It is worth drawing attention to some of the merits of MS. Hail.
7334. See pp. 24, 32, 35.
9. I make, by the way, several Notes ; especially the following'.
On A 8 (p. 37) ; A 60 (p. 55) ; A 686, 831, 1376, 2555-6 (p. 38) ;
B 4377-80 (p. 39) ; F 1, 2 (p. 52) ; F 207 (p. 57) ; H 16 (p. 43) ;
I 5 (p. 43). On nine-syllable lines; pp. 40, 41. And I give three
new references ; viz. to Gregorius, Epist. lib. x. c. 39, for G 120 (p.
17) ; to Persius, prol. to Sat. 14, for F 207 (p. 57) ; and Eabanus
Maurus, Expositio in Prov. lib. iii. c. 29, for D 2012 I (p. 24).
Mcrf (Sfta&m of ftt Canbrfro Salts
0
toit!] tfspmxl $etaa to tjp farlwn $Sk 7334.
EUBEICS, GLOSSES, AND SIDE-NOTES.
BY the "eight-text" edition I mean the eight MSS. printed in
extenso by the Chaucer Society, i. e. the original " six-text " taken
together with the separate prints of MS. Harl. 7334 and the Camb.
MS. Dd. 4. 24.
In comparing the various MSS., it is by no means without profit
to take into account the various Rubrics (i. e. Titles and Colophons),
Glosses written over hard words, and the Side-notes which throw so
much light upon the original texts which Chaucer consulted. The
last have been, no doubt, well considered already, but I propose to
give here, once for all, a complete account of them.
In order to facilitate reference and to give more concentrated
views of the results, I shall, in general, treat each " group " separately
from the rest.
CANTERBURY TALES : TITLES AND COLOPHONS.
N.B. — In quoting Titles, the numbers refer to the lines that
succeed them ; in quoting Colophons, to those that precede.
Many Titles and Colophons occur in the printed copies of the
Eight-text edition. I denote the texts by the following abbreviations,
viz. — E., Ellesmere ; Hn., Hengwrt; Cm., Cambridge; Cp., Corpus;
Pt., Petworth ; L., Lansdowne ; HI., Harleian ; Dd., Cambridge Dd.
4. 24.
I prefer in general to quote them in the particular order following,
viz._Hn. Pt. Cp. L. HI. Cm. Dd. E. I believe this to represent the
chronological order of the types of MSS. to which they belong. No
doubt, Cp. L. really belong to the same type, but the former seems to
be, in some respects, the more antique. In like manner, Cm. Dd. E.
belong to the same type; but Dd. contains lines that are not in Cm.,
and E. is the fullest of all. Even if these assumptions are all wrong,
the following enumeration of facts remains unaffected.
EIGHT-TEXT ED. CANT. TALES. B
2 Canterbury Tales (Eight-text).
I draw attention to the fact that I wholly disregard results found
in MSS. that do not belong to the Eight-text. For example, Dd. is
frequently supplemented, in the printed text of it, from the Egerton
MS. 2726 ; and the Canon's Yeoman's Tale (not occurring in Hn.) is
supplied from the Lichfield MS. But it is obviously convenient, in
the present account, to leave MS. Egerton 2726 and the Lichfield MS.
(and other supplementary MSS.) wholly unnoticed.
After copying out in full all the Kubrics and Colophons in the
8 MSS., enough to fill some 16 pages of print, I found that not much
is to be gained from the comparison of them, and that it is hardly
worth while to give them. On the whole, I should say that, in
this respect, the MSS. show a good deal of independence of each
other.
The following is a specimen :
A3 109. Hn. The prologe of the Milleres tale.
Pt. & here bygfi]nne)> fe Prologe of }>e Milleres tale.
U The Prologe.
Cp. The Millewardes Tale.
L. Incipit prologus Melendenarij (sic).
HI. And pus bygywne]? )>e prologe of pe myllere.
Dd. The prologe of )>e Millere.
E. Heere folwen the wordes / bitwene the hoost and
the Millere.
There is a tendency towards the formation of two groups. This
is particularly marked in the side-notes to the Monk's Tale. Thus, at
B3285, we find :— Hn. Cm. Dd. E. Hercules ; but Pt. Cp. L. HI. De
Ercule. At B3333 — Hn. Cm. Dd. E. Nabugodonosor ; but Pt. Cp.
L. HI. De rege Nabugodonosor.
As in nearly all other respects, Cp. L. are thus very closely allied.
For example, at B3573, we find — Cp. Bertheun Claykyrc . Olyuer
Mawny ; L. Betelmewe Cleykeynne . Oliuer Mawnye ; where the rest
have no side-note. But the most amazing instance is at F347, where
we find — Pt. Explicit prima pars. The stag of an hert ; Cp. Explicit
prima pars. The stag of an hert. Here, as the stag of an hert has
nothing whatever to do either with the Squire's Tale or with Chaucer,
it can only have arisen from the copying of an idle scribble on the
margin of a MS. But it links together these two MSS. by a bond
that cannot be broken.
The following are some points of interest.
Titles and Colophons. 3
The Knight's Tale was at first divided into three parts. After
A1880, Hn. has — Explicit prima pars; Incipit pars secunda. After
A2742, Hn. has — Explicit secunda pars; Incipit pars tercia & vltima.
But E. divides it into four parts, as in my edition, viz. after lines
1354, 1880, and 2482.
A4422. Hn. Of this Cokes tale maked Chaucer na moore; Dd.
Sic desinit fabula Coci.
The Man of Law's Tale is divided into three parts, viz. after
B385 and B875, not only in E., but in Dd. also. Dd. and E. are
closely allied.
B1163. Pt. And here bygynnejj j>e prologe of ]?e sqwiere; L. In
cipit p?-ologus Armigeri ; Cp. (no title}. The fact is that B1163-1190
constituted a Squire's Prologue in the " Petworth " and " Lansdowne "
schemes, as already shown in my essay on " The Evolution of the
Canterbury Tales," pp. 15, 17. The same 28 lines also form the
Squire's Prologue in both editions of Caxton1 and in that by Thynne.
In the " Harleian " scheme they became useless, and should have been
suppressed; and in fact they are left incomplete. In Hn. Cm. Dd.
E. they of course do not appear at all. The assignment of this
Prologue to the Shipman (as in MS. Arch. Seld. B.14, a MS. which
misarranges the Tales more than usual) is merely a later device for
preserving the lines, and has no connexion with Chaucer.
B1190. After this line there is a distinct break in all the MSS.
of the Eight-text. Hn. places the Shipman's Tale after the Pardoner's,
which is its almost invariable place in nearly all the MSS., in both
editions by Caxton, in that by Wynkyn de Worde, 1498 (which,
as far as the Tales are concerned, follows the " Harleian " order
throughout), and in all the black-letter editions that follow Thynne.
The most explicit statement as to the position of the Shipman's
Tale is in Dd., which has — Here bigynneth the Shipmans Tale / next
folwyng the Pardoner.
B1902. Hn. Cm. E. Here bigynneth Chaucers tale of Thopas;
Pt. Here bygynnejj f>e tale of chaucere by Sire Thopace; Cp. Here
bygynneth j?e tale of Chaucer of Sire Thopas. The form in Pt. is in
striking contrast with the modern idiom, which requires " the tale of
Sire Thopace by Chaucer."
1 The arrangement in Caxton 's first edition is the same as that in the "Pet-
worth " scheme and in Thynne, as noted by Dr. Furnivall. But Caxton's second
edition has a most unusual arrangement, as it places the Merchant after the
Man of Law. Nevertheless, Bl 163-1 190 follows the Merchant's Tale as a
Squire's Prologue.
4 Canterbury Tales (Eight-text).
B3181. Dd. & bigynneth pe Monkes tale / f>at is titled de casibus
virorum Illustrium : Chaucer. Cp. L. E. also allude to this Latin
title.
D829. HI. Here makith J>e frere an inte?*pretacion (!) of j>e wyfes
tale; Dd. Heere maketh the Frere an interempcioun (!) of the wyues
tale. " Interruption " is meant, though the earliest sense of the word
in this sense is given in the K E. D. as occurring in Caxton in 1489.
Interemption, i. e. slaughter, is recorded in 1656. E. has — Biholde
the wordes bitwene the Somonowr and the Frere.
The Clerk's Tale. Divided by Hn. into five parts; see E197,
449, 610, 785. Most of the MSS. recognise a similar division.
There is no Pars Sexta in the MSS.
El 2 12. Here follows, in Hn. Cm. Dd. E. a seven-line stanza —
sometimes called the " Host-stanza " — which Pt. Cp. L. HI. omit ; and
it was never put to use as a link. E. alone has a Title and Colophon
to this stanza, and Dd. only a Colophon ; as follows :
E. Bihoold the murye wordes of the Hoost. E. Heere endeth the
tale / of the Clerk of Oxenford; Dd. Sic desinit fabula Clerici. This
stanza was superseded in HI. by the new Merchant's Prologue (E
1213-44), whereby a new and permanent link with the Clerk's Tale
was established, viz. by making 1. 1213 an echo of 1. 1212.
E2419. Hn. Here folwen the Wordes of the Worthy Hoost to
the Frankeleyn ; Pt. The prologe of the Frawnkeleyn. Dd. Incipit
p?*ologus Armigeri ; E. The prologe of the Squieres tale.
There is no real contradiction here. The lines E241 9-2440 and
Fl-8, which in the Hengwrt MS. formed a Franklin's Prologue
(because the Franklin's Tale then followed) were in the " Harleian "
scheme converted into a Merchant's Epilogue and a Squire's Prologue
(because the Squire's Tale then came next); and this "Harleian"
scheme was adopted (very nearly) by the " edited " MSS., to which
Dd. and E. belong. It is essential to notice that, in all cases, E2440
was immediately followed by Fl ; or, in other words, there neither
was, nor is, any break between Groups E and F. These two Groups
are really but one.
Gl. Here Dd. has the remarkable heading : — & bigynneth the
Secund ISTonnes Tale of Seynt Cecile with-oute a Prologe ; whereas E.
has — The prologe of the Seconde Nonnes Tale. Yet both are right ;
for " The Prologe " in lines 1-28 is only a Proem or poetical introduc
tion to the " Life " ; and by no means such a Prologue as many of the
other Tales are provided with. We are not even told who "the
Glosses and Remarks. 5
Second Nun " really was, nor does the worthy Host say anything to
introduce her.
G971. At this point Pt. L., as well as E., divide the Canon's
Yeoman's Tale into two parts.
GLOSSES AND EEMAEKS IN THE EIGHT-TEXT.
Of these I have compiled a sufficiently complete list, for future
reference. Of course many of them are of little value ; but some may
be due to Chaucer himself ; and we ought to have, in a convenient
form, everything of this kind that the MSS. can give us. It is easy
to skip them.
Group A.— 1013. Cm. Arcita. 1014. Cm. Palamon. 1061. Cm.
Emelye. 1374. Hn. Dd. E. Mania. 1466. Dd. venm est. 1472.
Hn. E. Opium Thebaicum ; Dd. Opium Thebanu7?^. 1546. E. Cad
mus. 1548. Cm. has .i. edificare over beelte [built]. 1774. Hn.
Nota; E. Rotate domini. 1887. Cm. the lystis. 1893. Cm. Porta
Marmoris. 1894. Cm. has .i. ex oposito over in the opposit. 1904.
Cm. Venus. 1907. Cm. Mars. 1912. Cm. Dyane. 1955. Hn. has
ad vid : over see. Cm. Venus. 1956. Hn. has .i. mare over see.
1985. Hn. Dd. E. have .i. impetus over veze. 2059. Hn. Cp. L. Dd.
E. Vrsa maior. 2212. Cm. palamura to venus. 2221. E. The preyere
of Palamon to Venus goddesse of loue. 2242. Dd. Nota bene.
2265. Cm. Venus. 2273. Cm. Emale. 2274. Cm. Diane. 2297. E.
The preyere of Emelye to dyane goddesse of May dens. 2298. Hn.
has .i. mare over See. 2349. E. The answere of Dyane to Emelye.
2367. Cm. Arcite. Mars. 2373. E. The orisown of Arcite to Mars
god of Annes. 2438. Cm. Murmur. 2447, 2565. Dd. Note.
2581. Hn. E. have .i. sub Marte over vnder Marte. 2601. Dd.
Note. 2689. E. Nota periculum. 2760. Dd. verwra est. 2771.
Dd. Nota bene. 2843. Hn. E. Argumentum; Dd. Nota. Argu-
mentu??z. 2921. Dd. Arbores. 2987. Hn. Nota. 3017. E. Exem-
plum. 3018. Dd. Nota bene. 3021. E. Exemplum. 3120. Dd. of
Ale, with al pale above.1 3163. Dd. Nota bene. 3233. Dd. Note.
3303. Dd. Nota bene. 3417. Dd. explains For for by quia pro.2
3611. Dd. E. Auctor. 3692. Dd. Note. 3734. Hn. Nota maluw*
quid; Dd. Note quid malum. 3884. Dd. Note bene. 3920. Cp.
Fabula. 4324. Dd. qwod the Reve. 4339. Hn. E. have hie over heere.
4340. Hn. E. have audire over heere.
1 A correction ; all the other 7 MSS. read al pale (Hn. a pale, by error).
2 Hence Pt. HI. Cm. (which read For only) are wrong.
6 Canterbury Tales (Eight-text}.
Group B (Part 1), — 185. Hn. E. have .i. ceriose over ceriously.
309. Pt. Inspirant (!) ; in margin. 358. E. Auctor. 925. E. Auctor.
Group B (Part 2).— 1404. Hn. E. qi la; Dd. qy la. 1558. Hn.
Dd. E. .i. obligacionem (explaining bond). 1643. Hn. Pt. Cp. HI. E.
Domine doniinus noster. 1647. HI. Ex ore infancium 1681. Hn.
Dd. E. turpe lucrum. 1748. E. Auctor. 1775. Hn. Dd. E. carnaliter
(explaining fleshly). 1797. E. Auctor. 1812. Dd. De puero qui can-
tauit de gloriosa virgine. 1884. Hn. has .i. Chaucer over me; E. .s.
Chaucer. 1890. E. has .i. henry Bailly against I. 1900. E. has .i.
hoost over he.
Melibeus.— 2166. HI. Dd. E. Ouidius de remedio amoris. 2174.
L. Senec; Dd. E. Seneca. 2177. HI. Dd. E. Qualiter inc christus
fleuit propter mortem lazari. 2179. HI. Dd. Apostolus ad romanos ;
E. Aposfolus Paul us ad romcmos. 2181. Cp. Senek; E. Seneca.
2185. Cp. Ihesus Cyrak ; E. Ihesus Syrak. 2187. Cp. E. Salomon.
2189. Cp. lob; E. Memorandum de lob. 2193. Cp. E. Salomon.
2235. Cp. Ihesus Cirak; E. Ihesus Syrak. 2237. Cp. E. Salomon.
2247. L. Salomon; E. Nota secundum Salomonem. 2454. Hn. Dd.
humanum (explaining mannysh). 2529, 2534, 2537. Cp. E. Tullius.
2538. Pt. E. Cassidorus. 2585. Dd. men (with clerkes above).1
Besides these remarks on Melibeus, Dd. has Nota at 2337, 2348,
2374, 2382, 2629, 2645, 2702, 2706, 2749, 2769,2829, 3056 ; and at
2605, Nota & caue. E. has a large number of references to autlwr-
ities, but they tell us no more than is in the text. / also note the fol
lowing :— 3220. E. Nota de ludicibws & eorum ludiciis. 2276. E.
Of .iij. thynges J>at dryuen a man out of his hous. 2311. E. Of .iij.
thynges pat been cowtrariouse to good conseil. 2339. E. How a man
shal tellen his corcseil. 2363. E. Of conseillowrs ]?at a man oghte to
eschue. 2392. E. How a man shal examine his cowseillowrs after the
doctrine of Tulliws. 2414. E. How a man may chaungen his conseil-
lours with-outen repreue. 2527. E. Nota / of the strongeste garnisone
that may be. 2594. E. In libro decretal^. 2801. E. Vnde
versificator.
Monkes Tale, &c. — 3466. Hn. Dd. E. have simul over in feere
[Dd. in fere]. 3683. Dd. Nota. 3693. HI. Seneca (margin). 3801.
Hn. Dd. E. have lacerauit over to-tar. 4088. Dd. Dreem. 4131. Cm.
Sompnia ne cures. 4174. Cm. Naracio ; Dd. Tullius; E. Nota de
Sompnio. 4192. E. has .i. dremed over mette. 4240. E. Auctor.
4257. E. Adhuc de sompnio. 4300. Dd. De Eege Kenulpho ; E. De
1 A correction ; the other 7 MSS. have clerkes.
Glosses and Remarks. 7
sompnio sancti kenelmi. 4317. E. Adhuc de sompnijs. 4377. Dd.
Rota. 4380. Dd. has. i. secundo die Mali.1 4395. Dd. Rota
Salamon. 4514. Cp. Rota bene. 4636. Hn. Dominus Archie-
piscopus Cantuarie^s ; Dd. Kantuar'; E. s. domimis Aichiepiscopus
Cantuarienszs.
Group C.— 91. Hn. E. Rota. 173. Hn. Dd. E. have audire as a
gloss to heere. 174. Hn. Dd. E. have hie as a gloss to heere (Dd.
here). 492. E. Seneca. 493. Hn. Rota. 590. E. Of Hasardrye.
629. E. Of sweryng & forsweryng. 895. E. Auctor.
Group D.— 11. E. In Cana Galilee. 35. E. has audi over heere.
110. E. has. i. steppes over foore. 115. Hn. questio. 193. E. Bihoold
how this goode wyf serued hir. iij. firste housbondes which were goode
olde men. 228. Dd. Verum est. 333, 336. HI. Rota. 358. Cp.
argws habuit mille oculos. 453. E. Of the condicione of the fourthe
housbonde of this goode wyf / And how she serued hym. 503. E.
Ofthefifthe housbonde of this wyf / and hou she bar hire ay ens hym.
509, 519, 608. Dd. Rota. 655. Dd. E. Rota. 780. Dd. Rota. 832.
Dd. The Sompnour speketh to the Frere. 970. Dd. Nota bene. 980.
Dd. Verum est. 1109. Hn. Rota, bene; Pt. Rota bene de nobilibws;
Cp. Rota bene de Nobilitate ; Dd. Rot& causas generositatis ; E.
De generositate. 1112. Dd. arrogance is pride. 1115. Dd. Rota
verissimam causara generositatis. 1140 (or 1146). Dd. E. Exem-
plum. 1170. Dd. Rota bene. 1177 (or 1179). Pt. Cp. Dd. note
de pauperfate; E. De paupertate. 1201. Dd. Rota bene. 1205 (or
1206). Pt. L. Rota, de senectute ; E. De senectute. 1213. E. De
twrpitudine. 1523. Dd. Sompnour. 1885. E. de omtiom'bus &
leiunijs. 2017. Cm. Of an Irous potestat ; E. de quodam potestate
Iracundo.
Group E.— 6, 14, 20, 30. Cp. E. Pausacw. 38, 46, 56. E. paus-
acio.2 887. E. has i. couere as a gloss to wrye. 995. Cm. Dd. E.
Auctor. 1048. Hn. E. have ubique over oueral. 1183. Dd. Auctor,
1314. Dd. E. dona fortune. 1317. Dd. verum. 1342. Dd. ftas or to
drynke (!) against to swynke. 1461. Hn. HI. Rota. 1475. HI.
Rota. 1478. Cm. E. Placebo. 1501. HI. uota. 1519. Cm. E.
lustinus. 1523. Cm. Seneca. 1553. HI. Rota. 1601. Dd. Rota.
1655. E. lustinus. 1671. HI. nota. 1755. Dd. Rota bene. 1783.
Hn. HI. Cm. Dd. E. Auctor. 1869. Hn. HI. Dd. E. Auctor. 2057.
1 A slight mistake ; May 3 is meant.
2 Probably once also in Cp. 38, 46, 56 ; but a leaf is cut out of Cp. after
1. 32.
8 Canterbury Tales (Eight-text}.
Hn. E. Auctor; HI. nota bene. 2107, 2125. Cm. E. Auctor.
2210. HI. nota. 2266. Dd. verum quidem. 2270. Dd. Rota bene.
Group F.— 22. Hn. Pt. Cp. L. E. centrum circuli. 115. E. Of the
vertu of the steede of bras. 132. E. Of the vertu of the Mirowr.
145. Hn. E. have hie over heere. 145. Hn. E. have audire over heere.
E. Of the vertu of the ryng. 156. E. Of the vertu of the swerde.
207. Hn. E. have .i. equs Pegaseus ; Pt. Cp. have .i. equus [Cp. equs]
pedasdus (!) ; HI. Dd. .i. equus pegasews; Dd. adds — percius 4k).1
1031. E. The compleint / of Aurelius / to the goddes and to the sonne.
1045. Hn. Dd. E. have .i. luna over Lucyna ; Cp. has .i. luna lucyna.
1252. Hn. Dd. E. lanus biceps. 1281. E. Alnath dicitwr pn'ma mansio
lune. 1283. E. In nona spera. 1354. E. The compleynt / of Dorigene
ayeyns Fortune.
Group G. — 26. Cm. .i. cecilie. 29. Hn. HI. Dd. E. Inuocacio ad
mariam. 120. Hn. Cecilia virgo clarissima. 186. Hn. Dd. E. have .i.
latitante??i over lotynge; Cp. L. have .i. latitaws. 207. Cp. Et
lamentat (!). 351. Hn. vrbanus ; Dd. E. qwi s. [scilicet] vrbanus
(referring to That). 498. Hn. Dd. E. exterioribws oculis (over outter
eyen).
Group H. — (Not in Dd.). 163. E. Exemplum. de volucre. 175.
E. Exemplum. de Murelego. 183. E. exemplum. de lupo. 223. Hn.
has sine titulo over titlelees. 256. E. Nota malum quid.
Group I. — 81. E. Nota de penitence. 84. E. What Penitence is.
89. E. Nota. 94. Hn. has tene over siker (by mistake), and certum
over wey (by mistake) ; E. has tene over taak, and certuw over siker
(correctly). 96. E. The firste acciown of penitence. 99. E. The .ijde.
acciown of penitence. 100. E. The .iijde. acciown of penitence. 102.
Hn. E. have species over speces.
[E. has a large number of side-notes and references, but they merely
repeat expressions and references that are given in the text itself, and
are therefore omitted. Cp. (which ends at 1. 290, repeats a few refer
ences given in the text. Besides these, I find only the following^
125. HI. Dilexi legem tua?w, &c. 133. HI. Sex simt cause que
mouent homme??i ad contricionem. 284. Cm. .vj.a* 588. Hn. Mathee
5to Nolite iurare omnino (here inserted in the text, but only in
the margin of E., and not in the rest at all). 592. Hn. lurabis in
uerita[te] in iudicio & in iusticia (in the text of Hn., but in the
1 The right reference is to 1. 14 of the Prologus of the Satires of Persius : —
Cantare credas Pegaseium nectar. So that Chaucer made a sb. Pegaseus out
of this adjective. See p. 57 below.
References to Authorities. 9
margin of E., and not in the rest). 628. Cm. Ex abundancia cordis
os loquitwr. 714. HI. Omnium maloium mater est negligencia.
EEFERENCES TO AUTHORITIES, ETC.
Group A. — 859. Hn. Cp. L. Dd. E. lamque domos patrias [Hn.
patria] Scithice post aspera gentis Prelia laurigero.1
1164. Hn. Pt. Cp. L. E. Quis legem det amantibus?2 [Pt. Cp. L.
dat ; wrongly.]
3382. Hn. L. Dd. E. Vnde Ouidius : Ictibus agrestis [L. agrestes] ;
Dd. adds &c.3
3598. Dd. Mitte sapientem &c.
4181. HI. Qui in vno gmuatur in alio debet releuari.
Group B (Part 1).— 161. Hn. Pt. Cp. L. Dd. E. Europa [Pt. Cp.
Europia] est tercia, pars niundi.4
197. Hn. Pt. Cp. L. Dd. E. Ceptra phoronei / fratrum discordia
Thebe Flammam Phetontis Deucalionis aque. In stellis Priami species
audacia Tumi Sensus vlixeus herculeusque vigor [Pt. pharonei — phe-
tentis — herculiesque ; Cp. L. pharonei; E. Phorenei, aud repeats
fratrum].5
295. Hn. Pt. Cp. L. Dd. E. Vnde Ptolomeus libro .i.° capitulzm
8°. U Primi motus celi duo sunt / quorwm vnus est qui mouet totum
semper ab Oriente in Occideiitem vno modo super orbes &c. / Hem
aliter vero motws est qui mouet orbem stellarww currenciim contra
motum primum videlicit ab Occidente in Orientem super alios duos
polos &c. [Hn. omits Primi and mouet totum, inserting here celum,
and omits modo ; Pt. Philomels — & (for in) ; Cp. Pholomews ; L.
Tholomeus ; Dd. x° (for 8° )].
309. Hn. Pt. Cp. L. Dd. E. Omwes concordat! sunt / qwod elec-
ciowes sint / debiles nisi in diuitibws / habent enim isti licet debili-
tentwr eorwm elecciowes radicem .i. natiuitates eorum que confer tat
omnem / planetam / debilem in itinere &c. [Pt. Cp. L. add Hec
Plu70sqp7his (or PhoZoweus = Ptholomeus) ; L. co?zfortant].
421. Hn. Pt. Cp. Dd. E. Nota de iraopinato dolore [ivhich Pt. Cp.
omit]. Semper mimdane leticie / tristicia repentina succedit / Mun-
1 Statins, Thebaid. xii. 519, 520.
2 Boethius, De Cons. Phil. lib. iii. met 12.
3 But it is not from Ovid ; nor in " Pamphilus and Galatea."
4 The -world was formerly divided into these three parts ; Orosius, lib. i. c. 2 ;
Higden, Polychronicon, lib. 1. c. 6 (following Augustine, De Civitate Dei, lib.
16. c. 8).
5 Bernardus Silvestris, Megacosmos, § iii. 1. 37 (see my note).
10 Canterbury Tales (Eight-text).
dana igitur [Pt. Cp. ergo] felicitas multis amaritudinib&s est respersa :
extrema gaudij Indus occupat / Audi ergo salubre consiliu?ft in die
bonoram ne immemor sis malorwm.1
771. Hn. Pt. Cp. Dd. E. Quid twrpius ebrioso / cui fetor in ore /
tremor in corpore / qui promit stulta / p?'odit occulta / cui mens
alienator / facies tmnsformatttr / nuUum enim latet secretu??i vbi
regnat ebrietas.2
925. Hn. Pt. Cp. Dd. E. 0 extrema libidinis turpitudo, que non
solum mente??z effeminat set eciam corpus eneruat / semper sequntw
[Dd. secuntur] dolor & penitencia post, &c.3
1132. Hn. Pt. Cp. Dd. E. A mane usqwe ad vesperam mutabitw?*
tempus / tenent tympanum & gaudent ad sonum organi, &c.4
1135. Hn. Pt. Cp. Dd. E. Quis vnquam vnicam diem totam
duxit in sua dilecckwe iocunda??i / quern in aliqua parte diei / reatws
consciencie / vel impetus ire / vel motus concupiscencie non [Pt. Cp.
inde non] turbauerit / quern liuor Inuidie [which Dd. omits] uel ardor
auaricie uel tumor superbie non uexauerit / que??z aliqua iactura vel
offensa vel passio non commouerit &c.5
Group B (Part 2).— 1643. Hn. Pt. Cp. HI. E. Domine Dominus
noster.6 1647. HI. Ex ore infanciu??!.7
1764 (rather 1770). Hn. Dd. Leganms apocalipsim lobaraiis &
ibi reperimus agnum supe?* monte?w syon & cum illo cxliiijor milia
signatorwm &c. qui cantant canticu??z nouum &c. Isti sunk qui cum
mulieribws se non coinqwinaueruwt virgines autem permanserunt / Hii
sunk qui secuntur agnum qwocunq?<e vadit &c.8
1817. Hn. Dd. Kachel plorans filios suos & noluit consolara &c.9
1828. Hn. Dd. de puero qui cantauit de gloriosa virgine.
de maria qwicquid sciuit
puer cantans enutrmit
liunc ludews nequam strauit
j- Mate?*nam inopiam
puer cantans enutnuit J
hunc ludews nsqucim strauit ^ _. _ ._.
, > Diram per Inuidiam
domo sua quern numauit J
Mater querews hurcc vocauit ^ _
, . . ., > bolita precoma
hie in terra recantauit J
1 From Innocent III., De Contemptu Mundi, lib. i. c. 23. Extreraa . . .
occupat is from Pror. xiv. 13. In die ... malorum is from Ecclus. xi. 27
(Vulgate).
2 From Innocent (as above), lib. ii. c. 19. Nullum . . . ebrietas is from
Prov. xxxi. 4 (Vulgate). See my note.
3 From Innocent III. (as above), lib. ii. c. 21. See my note.
4 A mane . . . tempus is from Ecclus. xviii. 26 ; tenent . . . organi is from
Job xxi. 12. See my note.
5 From Innocent III. (as above), lib. i. c. 22. See my note.
6 Ps. viii. J, 7 Ps. viii. 2. 8 Rev. xiv. 3, 4. 9 Matt. ii. 18.
References to Authorities. 11
Puer liber mox exiuit
,
,.. ,. . ., ludeos &c.
Mortis reos lex punmit
Melibeus.— 2166. HI. Dd. E. Ouidius de remedio amoris.1 2174.
L. Senec; Dd. E. Seneca.2 2179. HI. Dd. Apostolus ad romanos;
E. Apostolus Paiilus ad Romanos.3
3307. Hn. Dd. E. Hie vates Caldeorwn [Dd. E. Chaldeorwm]
Tropheus. 3338. HI. Danielis 4to regum &c.4 3754. Hn. HI. Dd.
E. Et fecmmt filij Israel / secun&um qwod constituerat eis sacerdos
domini Eliachym.5
3917. HI. leuitici capituJ.o .ij° de spiritu phitonisse ha&endo octos'
iiij° libro ~Regum capitulo primo [see 1. 3938].6
4174. Dd. Tullius.7 4399. Hn. HI. E. Petrus Comestor. 4635.
Hn. E. Dominus Archiepiscopns Cantuari«ww ; Dd. Kantuar'.8
Group C. — 14. Hn. Dd. E. Quere in Methamorphosios.9 16. Hn.
Dd. E. Apelles fecit mirabile opus in tumulo Darij. Vide in Alex-
andri libro .6°. [E. .i°.] de zanze in libro Tullij [Dd. E. Tulii].™
115. Hn. Dd. E. Augustinus [see my note]. 240. Hn. Dd. E. ludi-
cum cap0, xj .° fuit illo tempore lepte Galaandes [for lephte Galaadites \
Judges xi. 1], 329. Hn. Dd. E. Radix omniw^ maloram est Cupid-
itas || Ad Thimotheum .6°. [1 Tim. vi. 10]. 485. Hn. HI. Dd. E.
Nolite inebriari vino in quo est luxuria [HI. omits est luxuria. Eph.
v. 18]. 508. Hn. HI. Dd. E. leionimus contra louiniamm / Quamdiu
ieunauit Adam in paradiso fuit / comedit & eiec^ws est. statim duxit
vxorem [see my note]. 522. Hn. HI. Dd. E. Esca ventri et venter
escis deus autem & hunc & illam destruet [1 Cor. vi. 13]. 529. Hn.
HI. Dd. E. Ad Philipenses cap0. 3°. [Phil. iii. 18]. 547. Hn. Dd. E.
Qui autem in delicijs est viuens / mortuus est [1 Tim. v. 6]. 549.
Hn. Dd. E. Luxuriosa res vinum / et contumeliosa ebrietas [Prov.
xx. 1]. 584. Hn. Pt. Dd. E. Noli vimm dare, &c. [Prov. xxxi. 4].
591. Hn. Pt. Dd. E. 'Policratici libro 1°. Mendaciorwra & periuriarww
mater est Alea [Polycrat. lib. i. c. 5]. 603. Pt. Stilbon .i. Mercurius
[see my note]. 634. Hn. Dd. E. Nolite iurare orrmino Mathei 5°.
[Matt. v. 34]. 635. Hn. Dd. E. leremie .4°. lurabis in veritate in
1 Ovid, Kerned. Amoris, 127-130.
2 Seneca, Epist. 74, § 29. 3 Rom. xii. 15.
4 Alluding to vasa ; .Daniel v. 2 ; 2 Chron. (4 Kings) xxv. 14.
5 Judith iv. 7.
6 The right references are to Levit. xx. 27, and 1 Sam. xxviii. 7.
7 Proving that the reference is to Cicero, De Diviuatione, lib. i. c. 27. See
my note.
8 The reference is to the Archbishop of Canterbury.
9 Ovid, Metam. x. 247. 1Q See my note
12 Canterbury Tales (Eight-text}.
ludicio & lusticia [Jer. iv. 2]. 743. Hn. Dd. E. coram canuto capite
consurge [Lev. xix. 32].
Group D. — 12. E. Qui enim semel iuit ad nupcias / docuit semel
esse nubendura [see my note]. 23. E. JSTon est vxorwm numerus
diffinitus quia secundum Paulum / Qui ha&ent uxores sic sint tanqwam
non habentes [1 Cor. vii. 29 ; and see my note]. 28. E. Crescite &
multiplicamini [Gen. i. 28]. 31. Pt. Cp. Relinquet homo patrem et
matrem & adherebit vxori sue &c. [Cp. omits vxori sue. Gen. ii.
24 (cf. Matt. xix. 5)]. 36. Pt. Cp. Genesis ij°. Quamobrem [Gen.
ii. 24]. 46. E. Si autera non continent nubant [1 Cor. vii, 9].
47. E. Quod si dormierit vir ems liberata est / cui vult nubat in
Domino [1 Cor. vii. 39]. 51. E. Si accepe?is vxorem non peccasti /
et si nupserit virgo non peccauit / set hij qui domino se vouerunt Ik»
idem &c. [1 Cor. vii. 28 (down to peccauit)]. 52. Pt. Cp. L. E.
Meliws est nubere quam vri. [1 Cor. vii. 9]. 54. E. Lameth qui
primus intrauit bigamia??i / sanguinarius et homicida est &c. [Gen. iv.
18, 19, 23]. 55. E. Abraham triganius. Jacob quadrigamus [see my
note]. 73. E. Paulws de virginibws / preceptum non ha&eo / consi-
liu??i autem do &c. [1 Cor. vii. 25]. 75. E. Inuitat ad cursum tenet
in manu virginitatis brauium qui potest capere / capiat &c. [cf. 1 Cor.
ix. 24 ; see my note]. 81. E. Volo autem onmes ho?mnes esse sicut
me ipsum [1 Cor. vii. 7]. 87. E. Bonum est komini / muliere??i non
tangere [1 Cor. vii. 1], 103. E. Vnusqwsqwe proprium ha&et donum
ex deo ; aliws quide?ra sic / alius autem sic [1 Cor. vii. 7]. 105. E.
Qui caTztant sequentwr Agmm [c]xliiijor. Milia [Eev. xiv. 3, 4]. 108.
Pt. Cp. Mathei xix°. Dix^t &utem Ihesus vade & vende omnia, que
haftes & da pauperibz^s [Matt. xix. 21]. 147. E. Ea vocac^o/ie qua
vocati estis &c. [1 Cor. vii. 20]. 155. E. Qui vxorem ha&et & debitor
dicitwr. & esse in prepucio & seruus vxo?is & qui maloium semoium
est alligatws [see my note]. 158. Pt. Cp. L. Ad Cor. vij°. vir sui
corporis potestatem non habet [Cp. non habet potestatem] set mulier
&c. [1 Cor. vii. 4]. 160. E. Et iterum seruus vxoris es / noli
propter hoc ha&ere tristiciam. Ite-m si accepe?'is vxo?'em non peccasti
tribulaczoTiem tamen carnis ha&ebunt huiusmodi &c. [1 Cor. vii. 28].
161. E. Item vir corporis sui non ha&et potestatem set uxor [1 Cor. vii.
4]. Item viri diligite vxores vesfras [Eph. v. 25]. 180. Dd. Qui
per alios non cor[r]igitwr Alii per ipsum cor[r]igentur [see my note].1
198, 199. E. lerophancias quoqwe Atheniencium vsqwe hodie cicute
1 Fliigel has shown, in Anglia, xviii. 133, that this passage and that at
1. 327 occur in an edition of Ptolemy's Almagest printed at Venice in 1515.
References to Authorities. 13
sorbiczowe castrari [see my note]. 304. E. et procurator calamistrata^s
&c. [see my note']. 327. E. Intra omnes alcior existit / qwi non curat
in cuius maim sit rrmndus [see my note]. 341. E. si?wiliter & mulieres
in ha&itu ornato cum verecurcdia & castitate ornerct se / non in tortis
crinibws aut auro aut margaritis siue veste preciosa &c. Hec Paulus.
[1 Tim. it. 9]. 361. E. eciain odiosa vxor si ha&eat viru??& bonura
&c. [see my note, cf. Prov. xxx. 20-23]. 371. E. Amor illius
inferno & arenti te?*re & incendio comparatur. Vnde illud &c.
Infernws & amor mulieris & terra que non saciatwr aqua & ignis non
clicent satis &c. [see my note-, cf. Prov. xxx. 16]. 376. E. Sicut in
ligno vermis ita perdet virum smim vxor. Nemo meliws scire potest /
quid sit vxor vel mulier nisi ille qui passus est [see my note; cf.
Prov. xxv. 20 (Vulgate).]. 401. Pt. Cp. L. Fallere flere nere dedit
[Pt. statuit] dews in muliere.1 460. E. Valerius lilro 6°. c°. 3°.
Metellius vxorem suam / eo quod vinum bibisset / f uste percussam
interemit / [Valerius Maximus, lib. vi. c. 3]. 498. E. Apelles / fecit
mirabile opus in tumulo Darij / vnde in Alexa?zcZro. li&ro. 6°. [see my
note]. 611. E. Mansor Amphorison' 19 [read Aphorism' 14]. Curn-
que in ascendente fuerint infortune turpem nota??i in facie pacietwr.
j| In natiuitatibws mulierum cu??i fuerit ascendens aliqua de domibws
Veneris / Marte existente in eis / vel e contrario erit mulier
inpudica || Idem erit / si ha&uerit capricornu?^ in ascendente || He[c]
Hermes in libro fiducie Amphoris' 24.° [read Aphorism' 25 ; see my
note]. 643. E. Valerius, hbro 6°. folio. 19.° [Valerius Maximus, lib.
vi. c. 3]. 657 [rather 651]. E. Ne des mulieri neqwara vemam
prodeundi. ecclesiastici. 25°. [Ecclus. xxv. 25 ; xxv. 34 (Vulgate).].
692. E. Quis pinxit leonem ? [see my note]. 702. E. Vterque cadit
vbi alia exaltatii/* [see my note]. 705. E. In libro Mausor primo. \\
Vn. iplanet&Tum .4. || Exaltaceo illo in loco fore dicitw?* in quo subito
patitz^r ab alio co?itrarium &c. / Velut M.ercuiius in virgine que est
casus veneris / Alter scilicet Mercurms significat scientiam & philoso-
phisun || Alter vero cantz^s & alacritates & quicquid est sapiferum
corpori [see my note]. 733. E. Quid referam Phasifphen Clitermistram
& Eriphile??z / quaium pn'ma delicijs fluens qi^'ppe vt Eegis vxor
Tauri dicitwr adpetisse concubitus. Alia occidisse viru??i suum ob
amorem Adulte?'ij. || Tercia pe?'didisse Amphiorax & saluti viri monile
Aureum pretulisse &c. || Hec Metellius Marrio s. Valerium. [see my
note]. 775. Pt. Cp. Solofmon]. melius est ha&itare [Prov. xxi. 19].
1 A medieval riming hexameter (of a sort). See Chaucerian Pieces, ed. Skeat,
xiv. 29.
14 Canterbury Tales (Eight-text).
785. E. Circulus aureus in narib?js suis Mulier formosa & fatua .i.
iwzpudica [Prov. xi. 22]. 1183. E. Seneca in epistola. Honesta res est
leta paupertas [Seneca, Epist. ii. § 4]. 1187. E. Pauper est qui eget /
eo quod non ha&et / sed qui non ha&et nee appetit ha&ere ille diues est
/ de quo mtelligitur id Apoc&lypsis . 3°. dicis quia diues sum. [Rev.
iii. 17]. 1193. E. Cantabit vacuus coram latrone viator / et nocte ad
lumen trepidabit Arundme's vmbram. [Cf. Juvenal, Sat. x. 22, 21].
1195. E. 2M* [read Secundus] Philosophus Ii Paupertas est odibile
bonum / sanitatis mate?' / curarum remoc^'o / ssupientiQ reparatrix /
possessio sine calumpnia [see my note]. 1202. E. Vnde et Crates
ille Thebanus. || Proiecto in mari non paruo auri pondere || Abite
inqm'2 pessime male cupiditates / ego vos me?'gam / ne ipse mergar a
vobis. 1794. E. L^era occidit &c. [2 Cor. iii. 6]. 1879. E. Melius
est animani saginare quam corpus. 1881. E. Yictuw & vestimentwm
hijs contenti sunms &c. [cf. 1 Tim. vi. 8]. 1968. E. Omwis virtus
vnita / forcior est seipsa dispersa. 1973. E. Dignws est operarius
mercede &c. [Luke x. 7]. 1989. E. Noli esse si cut leo in domo tua /
euertens domesticos tuos opprimens subiectos tibi [Ecclus. iv. 35
(Vulgate); 30 (A.V.).].
Group E. — 43-50. Hn. E. Est ad ytalie latws occiduum vesulus
ex Appenini lugis mons altissimws qui vertice nubila superans liquido
sese ingerit etheri / Mons suapte nobilis natura padi ortu nobilissimws
qui latere fonte lapses exiguo Orientem contra sole?™ fertur &C.1
57. Hn. Dd. E. Inter cetera ad radicem Vesuli terra Saluciarwm vicis
& castellis [satis frequens]. 59. Hn. Dd. E. grata planicies. 86. Hn.
Dd. E. cateruatim. 92. Hn. Dd. E. Tua inquid (sic) humanitas optimo
Marchio. 197. Dd. E. Euit haut procul a palacio &c. 281. Hn. Dd.
E. Vt expedites curis aliis ad videndum [Hn. visendum] dommi sui
sponsam cum puellis comitibws properaret. 295. Hn. Dd. E. Qwwm
Walterus cogitabundus incedens [E. cedens] eamque compellans
nomine. 337. Hn. Dd. E. Et insolito tanti hospitis aduentu stupidam
inuenit. 344. Hn. Dd. E. Et patri tuo placet inquid (sic) & michi vt
vxor mea sis / & credo idipsum tibi placeat set ha&eo ex te querere
&c. 354. Hn. Dd. E. Sine ulla fronts aut verbi inpugnacione.
359. Hn. Dd. E. Nil ego vnqwam sciens nedum faciam set ecia??*
cogitabo quod contra animum iuwn sit / nee tu aliquid facies / et si
me mori iusseris quod moleste feram. 372. Hn. Dd. E. Dehinc ne
quid reliquiarwrn fortune veteris nouam inferat in domum / nudari
earn iussit [E. iusserit]. 400. Hn. Dd. E. Atque apud omnes supra
1 These Latin notes are quotations from Petrarch's version.
References to Authorities. 15
cara & vene?*abilis facta, est [Dd. cara est] / vix qwod hijs ipsis
qui illius originem nouerant / persuader! posset lanicule natam esse
tantus vite tantus moium decor ea verborem grauitas atqiie dulcedo
quibi^s omnium awimos nexu sibi magni amoris astrinxerat /. 421.
Hn. Dd. E. Sic Walterws humili quidem set insigni ac prospero
matrimonio honestatis su7?raia del in pace &c. 425. Hn. Dd. E.
Qzwdqwe eximia??^ virtute??z tanta sub inopia latitantem tarn perspi-
caciter deprendisset vulgo prudentissinms habebatwr. 428. Hn. Dd.
E. Neque vero solers sponsa muliebria tantum ac domestica / set vbi
res posceret publica eciam subibat officia, 435. Hn. Dd. E. Viro
absente lites pafrie / nobilium diseordias dirimens atqwe component
tarn grauibws responsis / tantaqwe maturitate & ludicii equitate vt
omrces ad salutem publicam demissam celo fe??imam predicarent. 449.
Hn. Dd. E. Cepit ut fit interdum Walterum cum iam ablactata esset
infarctula mirabilis quedam qwam laudabilis cupiditas s&tis experta??^
care fidera coniugis experiendi alcius & iterum atqwe itenmi retemp-
tandi. 499. Hn. Dd. E. Nee verbo mota nee vultu [E. om. mota].
540. Hn. Dd. E. Suspecta viri fama. suspecta facies. suspecta hora.
suspecta erat oracio. 603. Hn. Dd. E. Par alacritas atqwe sedulitas
[Hn. E. sedalitas] solitum obsequium / idem amor nulla filie mencio.
610. Hn. Dd. E. Transiue?'ant hoc in statu anni .4. dum ecce grauida &c.
624. Hn. Dd. E. Et olim ait audisti populum meu??i egre nostrum ferre
connubiwm &c. 659. Hn. Dd. E. Fac sen^enciam tibi placere quod
moriar volens moriar. 722. Hn. Dd. E. Ceperit sensim de Waltero
decolor fama crebescere. 1037. Hn. Dd. E. Vnu??& bona fide precor
ac moneo / ne hanc / illis aculeis agites / quibws alteram agitasti
namqi^e & iunior & delicacius nutrita est / pati quantum ego vt reor
non valeret. 1142. Hn. Dd. E. Hanc historiam stilo nurcc alto [read
alio] retexere uisum fuit non turn ideo / vt matronas nostii temporis
ad imitanda?ft huius vxoris pacienciam que michi inimitabilis videtwr/
quam vt legentes ad inritanda?^ saltern femine constancia??i excitarentur.
Vt que [Dd. quod] liec viro suo prestitit / hoc prestare deo nosfro audeat
quilibet [read audeant, qui licet] / vt lacobus ait Apostolus / Intemp-
tator sit malorwm & ipse nemiwem temptat / probat tamen et sepe nos
multis ac grauibws flagellis exerceri sinit / non vt animum nostrum.
sciat / quern sciuit anteqwam crearemwr &c. 1311. Hn. Dd. E. Vxor
est diligenda / quia donum dei est. Ihesus filiws Syrac f domws &
diuicie dantwr a parentib^^s / a domino autem proprie / vxor bona vel
prudens [Prov. xix. 14]. 1327. Hn. Dd. E. Eaciamws ei adiutorem
[Dd. E. adiutoriwm] / & extracta costa de corpora Ade fecit Euam /
16 Canterbury Tales (Eight -text).
& dixit propter hec relinquet homo patfrem & motfrem & adherebit &c.
et erunt duo in came vna [Gen. ii. 19, 24]. 1362. Hn. Dd. E. lacobus
enim [E. omits enim] per consiliim mafris sue Rebecce &c. [see my
note], 1368. Hn. Dd. E. ludith [Hn. Dd. insert &c] de manibus
Oloferni &c. [Hn. Dd. omit &c]. 1369. Hn. Dd. E. Et [E. omits Et]
Abigail per suura bonu7?i consilium / virum suu??i Nabal / ab ira dauid
[Dd. dauit] liberauit. 1371. Hn. Dd. E. Ester &c ludeos per [Dd.
inserts suwn] bonuw consiliu??z simul cwn Mardocheo in regno Assueri
[Dd. assuri] &c. [E. om. &c]. 1376. Hn. Dd. E. Seneca / sicut nichil
est superius [E. omits superius] benigna ccwiuge / ita nichil est crude-
lius infesta muliere [see my note], 1377. Hn. Dd. E. Cato / vxoris
linguam si frugi est ferre memento [Dion. Cato, lib. iii. dist. 25].
1380. Dd. E. Bona mulier fidelis custos est / et bona domus [see my
note]. 1384. Hn. Dd. E. Apostolus Paulas ad Eph. / Diligite vxores
vesfras sicut cliristus dilexit ecclesiam &c. [Eph. v. 25]. 1385.
Hn. Dd. E. Aposfolus / Ita viri debent dilige?'e vxores suas vt
corpora sua / quia qui suam vxo?*em diligit se ipswm diligit / nemo
vnqwam carnem suam odio ha&uit / set nutrit & fouet earn / et
postea / vnusqwi'sqwe s\iam vxwem sicut se ipswm diligat [Eph. v. 28,
29, 33].
Group F. — 608. Hn. Pt. Cp. Dd. E. reditu [Pt. redditui ; Cp. reditui
(both wrongly)] suo singula gaudent [Boethius, lib. iii. met. 2]. 722.
Hn. E. Vnde Persius / fonte labra prolui caballino / nee in bicipite
parnaso mei memini sompniasse [Persius, Prol. to Sat. 1-3]. 1110. E.
Pamphilus ad Galatheam / vulne?-or & clausum porto sub pectore telum
&c. [see my note]. 1252. Hn. Dd. E. lanus biceps [see my note]. 1281.
E. Alnath dicitwr prmia mansio lime [see my note]. 1369. Hn. Dd.
E. [1358, Dd. ; 1364, E.] 30a Atheniensium tiranni cum Phidonem
necassent / in comiiuio filias eius virgines ad se venire iusseru?it &
scortorw??z more nudari / ac super pauimenta patris sang?^me cruentatas
inpudicis geslibus ludere que paulisper dissimulato dolore cum temu-
lentos [E. timulentos] cowuiuas ce?'nerent quasi ad requisita nature
egredientes inuicem se co??zplexe p?*ecipitaueru?zt in puteura vt vzrgini-
tatem morte seruarerct / [see my note]. 1379. E. Cum 50a virgiwes /
laced[em]oniorwm Messeni violare temptassent [see my note]. 1387.
E. Aristoclides Orcomeni tirannws adamauit virgine?/^ Btymphalidem
que cum patre occiso ad templuw diane &c. [see my note]. 1390.
Dd. Iouinianu??^ [referring to Hieronymus contra louinianum]. 1392.
Dd. (the same as E. at 1. 1465) ; see 1465. 1395. Hn. (the sameasE.
at I. 1465); see 1465. 1399. E. Nam hasdrubalis vxor capta & in-
References to Authorities. 17
censa vrbe cum se cerneret a Romania capienda &c. [see my note].
1405. E. pn'mo ponam lucreciam / que violate pudicie nolens super-
uivere maculara corporis cruore deleuit [see note]. 1409. E. Quis valet
silencio preteme vij. Milesias virgines que Gallorum &c. [see note].
1414. E. Senapho in Ciri maioris scribit / infancia occiso Habradate
&c. [see note]. 1426. E. Democionis Ariopagitarwra pn'ncipis virgo
filia &c. [see note]. 1428. E. Quo ore laudande sunt Cedasij filie &c. [see
note]. 1431. E. Nichanor victis Thebis vnius captive virginis [amore]
superatus est [see note]. 1434. E. Warrant scriptores Grecie & a.\ia,m
Thebanam virginem &c. [see note]. 1437. E. Quid loquar Nicerati con-
iugera pie [read que] irapaciews irciurie viri morte??z &c. [see note].
1439. E. Alcebiades ille Socrakicus victis &c. [see note]. 1442. E.
Alcesten fabule feriwt pro marito Adameto sponte defunctara / et
Penelopes pudicia Omeri carmew est [see note]. 1445. E. Lacedomia
[read Laodamia] quoqwe poetarz^ra ore cantatwr occiso apud Troiam
Protheselao &c. [see note]. 1448. E. Porcia sine Bruto viuere non
potuit [see note]. 1451. E. Arthemisia quoqwe vxor Mauseoli insignis
pudicijs fuisse perhibetwr &c. [see note]. 1453. E. Teuta / Illi[ri]co-
lum Regina &c. 1455. E. Mem. strato regulus. Vidi et onmes pene
Barbaras [sic] ca. .xxvj°. pn'mi [libri]. Item Cornelia &c. Imitentwr
ergo nupte Theanam / Cleobiliam / Gorgun[tem] / Thymodiara
Claudias atqwe Cornelias / in fine libri primi [see note]. 1465. Hn.
Dd. E. [Hn. 1395 ; Dd. 1392]. Singulas has historias et plures hanc
materiam concernentes recitat beatws leronimus corctra louinianum in
pn'mo suo libro cap. 39. [i. e. all these stories and many more, are from
Jerome against Jovinian, bk. i.]
Group G. — 85. Hn. Dd. E. Interpretacio nominis Cecilie • quam
ponit f rater lacobws lanuensis in legenda aurea. 120. Hn. Dd. Gre-
gorius in registro [Epist.] li&ro. 10. Ad Eulogium pafaiarcham scribit /.
Indicamws preterea quia grauem hie interpretwtt difficultate??i patimwr /
dura enira non sunt qui sensum de sewsu exprimant / set transferre
semper verbor?<ra proprz'etatem volunt oraraem dictorwm sensum con-
fundunt &c. [Epist. 1. x. c. 39 ; see Migne, Patriologia, vol. 77, col.
1099.] 645. Dd. E. Omwe quod est nimium &c. [see note]. 688. E.
Cato; Cm. Conciws esse sibi [Dion. Cato, lib. i. dist. 17]. 745. E.
Solaciura miseriorwm (sic) &c. [see note]. 962, 964. Cm. E. Non
teneas Aurum &c. ; E. Nee pulcrura pomu??i &c. [see note].
Group H. — 147. Hn. Verura quid prodest / diligens custodia cum
vxor impudica seruari non possit, pudica non debeat / feda enira
custod[ia] est castitatis necessitas / pulcra certe adamatwr / feda facile
EIGHT-TEXT ED. CANT. TALES. 0
18
Canterbury Tales (Eight-text}.
concupiscit. Difficile custoditwr quod plures amant [see note]. 338.
Cm. In multiloquio non deest peccatum. [Prov. x. 19].
Group I. — 32. E. Paulus ad Thimotheum [1 Tim. i. 4, iv. 7 ; 2 Tim.
iv, 4]. 125. HI. Dilexi legera tuara &c. [Ps. cxviii. (cxix.) 113].
133. HI. Sex suwt cause que mouent horaiwem ad contricionera. [KB.
MS. E. has a large number of references, but they merely repeat the text,
and afford no help. I notice only the few that here follow.] 204. E.
Dauid prophefa. || Qwi diligit iniqwitatem / odit awimam suam. [Ps. x.
(xi.) 6]. 355. E. Moyses per demonem [see note]. 588. E. Math. 6°.
nolite iurare omnino [Matt. v. 34]. 592. E. lurabis in veritate in
ludicio & in lusticia [Jer. iv. 2]. 670. E. N"ota de inpaciencia cuius-
dam philosophi coniia suum discipulum [see note]. 714. HI. Omnium
malorzm mater est negligencia. 766. E. Genesis || Maledictus Canaan
seruus seruorum erit / fratribws suis [Gen. ix. 25 (Vulgate)]. 776. E.
Eadem mensura &c. [Luke vi. 38]. 959. E. Mem. mors intrauit per
fenestras.
COMPARISON OF THE EIGHT MSS., AS REGARDS THEIR CONTENTS.
Group A.
I here give some account of the lines in Group A respecting which
the MSS. show the most variation; inasmuch as they are retained
in some of them, but omitted in others. The horizontal line ( — )
denotes that the MS. retains the line or lines in question ; the blank
space denotes their omission. The square brackets [ ] denote that the
MS. is mutilated or defective. Hn. = Hengwrt MS. ; Pt., Petworth ;
Cp., Corpus ; L., Lansdowne ; HI., Harleian 7334 j Cm., Cambridge
Gg. 4. 27 ; Dd., Cambridge Dd. 4. 24 ; E., Ellesmere.
Lines.
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
252 b,c
—
637-8
—
—
—
—
—
[ ]
—
2681-2
—
—
—
—
2779-82
—
—
—
—
—
—
3155-6
—
—
—
3721-2
—
4375-6
—
—
—
—
—
—
—
I beg leave to append the following notes. 252 b, c, are also in
Thynne. It seems possible that Chaucer really meant to delete these
lines, in order to connect 1. 253 with 1. 252 more closely.
Comparison of the MSS. 19
The omission of 4375-6 in HI. must be due to some blunder.
As none of the other lines appear in the Hengwrt MS., and as
their omission does not impair the sense, it is possible that they were
additions ; indeed, 3155-6 and 3721-2 seem to be comparatively late
additions. They all appear to be genuine.
Besides the above, there are many deficiencies or omissions in
individual MSS. Several of these are obviously due to the scribe
missing his place on account of the repetition of words. I append a
list of these omissions, with some remarks. I denote defects or
mutilations (very numerous and serious in Cm. and Dd.) by the
use of square brackets ; and I purposely ignore omissions in such
additional MSS. as are used to supply gaps in the Eight-text.
[1-72]. Cp. defective. [1^36]. Cm. defective. [1-252]. Dd.
defective. 290. Cm. omits, but leaves the line blank. [505-758].
Dd. defective. [757-964]. Cm. defective. [920-1170]. Dd. defective.
1250. L. omits, by a blunder; and has distris for distresse, as if to
rime with this in 1. 1248. [1502-1931]. Dd. defective. 2012-2017.
HI. omits by some confusion, reading bores corage for beres in 2018,
as if to rime with rage in 2011. 2040. HI. omits, but leaves a line
blank. 2720-1. Pt. omits, owing to confusion of disconfttynge (2719)
with disconforture (2721). 2754-7. Dd. omits, wrongly riming
adoun (2753) with dominacioun (2758). [2927-3016]. Dd. defective.
2958. HI. omits by mistake (no rime). 3045. Cp. omits (no rime).
3063. Cm. omits most of the line (defective). [3074-3088]. Cm.
defective. [3829-3890]. Cm. defective. 4355 and 4358. HI.
omits, confusing fay with faith, and so giving the rimes faith, saith,
instead of play, fay, saith, faith. [4365-4422]. Cm. defective.
4375-6. HI. omits this couplet.
Group B; Part I. (1-1162).
The MSS. of the Eight-text all agree, except for a few defects, as
below.
[1-9]. Cm. defective. [82-133]. Cm. defective. 417 (in a stanza) ;
HI. omits by mistake. [530-710]. Dd. defective. 579 (in a stanza) ;
L. omits by mistake. 873 (in a stanza) ; L. omits by mistake.
GROUP B; PART I. (1163-1190).
1163-1190. Called the Shipman's Prologue; but not recognised
as such in the Eight-text. Hn. Cm. Dd. E. all omit it. In Pt. Cp.
20 Canterbury Tales (Eight-text).
L. it is a Squire's Prologue. The treatment of it in HI. is remarkable ;
line 1175 is omitted, but a blank space is left for it; and in 1. 1179,
Squiere has been altered to Sompnour. But as this was unsuitable for
1. 1190, the "Prologue" comes to a sudden end at 1. 1185, without
completing the sentence or providing a rime for telle. It thus became
practically useless, and should have been cancelled. But as in MS.
Arch. Selden B. 14 the Man of Law's Tale was (quite exceptionally)
followed by the Shipman's Tale, it was used in that MS. as a Ship-
man's Prologue, for which there is no authority beyond this accident.
The Shipman's Tale occurs in all the MSS. further on, and has no
Prologue.
Group B; Part II. (1191-2156).
(SHIPMAN, PRIORESS, AND SIR THOPAS.)
In these three Tales I find no characteristic differences, though the
MSS. exhibit some defects. The following are the chief points.
[1191-1252]. Cm. defective. 1220-3. Pt. omits, because 1219
and 1223 both end with goode man. [1304-9]. Cm. omits, because
1303 and 1309 both end with quod she. 1355. HI. omits by mistake.
1376-9. HI. omits, because both 1375 and 1379 end with / yow
praye. 1440. L. omits by mistake. 1468. L. omits; observe that
Cp. Pt. vary from the rest and from each other. [1563-1747]. Cm.
defective. [1584-1663]. Dd. defective. 1792. In a stanza; L.
omits. [1837-1934]. Dd. defective. 1894. L. omits; no rime.
1995. Particularly observe that this line occurs in Dd. only. 2042-4.
Pt. Cp. L. all omit ; they rime comyn with wyn, as if 2041, 2045,
2046 formed the first half of a stanza ; but this does away with the
second half! [2096-2156]. Cm. defective. 2105-8. Cp. L. omit;
as if Sir Thopas ended with 2104. 2108. Imperfect:, Pt. Cp. L. HI.
all omit', but (as it occurs in Hn.) the half-line should be retained.
GROUP B; PART II. (MELIBEUS; 2157-3078).
As the Tale of Melibeus is in prose, and abounds with omissions,
the discrepancies between the MSS. may be found to give useful hints.
But finding the omissions to be very numerous, I select only such as
seem to be most significant. Many that occur in one MS. only are of
no moment and usually arise from repetition of words or phrases.
Thus, in 2281-2, where Hn. has — to do wikkednesse / and if ye wol
werke wikkednesse — Cm. omits and if ye wol werke wikkednesse owing
Comparison of the MSS. Melibeus.
21
to the repetition of the last word; but as all the rest retain this
phrase, it is obviously a mistake that is of no moment. The most
serious cases are those marked by the use of square brackets, such as
2252-3, where we find, by comparison with the original, that all the
MSS. in the Eight- text are alike defective, as compared with the
original French text.
The most important case is in 2726, where the words for ellis
were it agayn reson were omitted in Hn., and disappear likewise
from Pt. Cp. L. HL, but were recovered in the "edited" texts, viz.
Cm. Dd. E.
The Table that here follows is probably not exhaustive (it is a
tedious business), but will suffice for practical purposes. In referring
to 2190, &c., it is not always the whole of the paragraph that is
affected.
Lines
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
2190
_
[2252-3]
[ J
2313-4
—
—
—
—
—
2315-6
—
—
—
2328-61
—
—
—
—
—
2378
—
—
2387
_
—
—
2409
—
—
2417
—
—
—
2432
—
—
—
2447-8
—
2476
—
—
—
2519
—
—
2525-6
2552
\[
"t
2586
—
[2623-4]
2646-7
—
—
—
—
—
_
2654
—
2708
—
—
—
2726
2741
__
2746-7
2815-6
—
—
—
[2854]
2963-5
—
—
2967
3034
,__
3059
—
—
—
—
—
22
Canterbury Tales (Eight-text).
GROUP B; PART II. (3079-4652).
MONK'S TALE AND NUN'S PRIEST'S TALE.
The chief variations are given below.
Lines
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
3197-3204
_
_
_
3213-20
.
3611
3764
3961-80
[ ]
4112 \
4114 /
—
—
—
-<-
—
4136-7
4233-8
4479-80
—
4637-52
—
Three of these passages require special notice. Lines 3961-80
(not in Hn. Pt. Cp.) seem to have been an after-thought, and in 1.
3972 is introduced the reference to " a cloude," which occurs in what
is now marked as 1. 3956. Observe that 1. 3956 was not the last line
of the Monk's Tale in Hn., but became so in Pt. Cp. by the insertion
of the " modern " stories at an earlier place of the Tale, which brought
1. 3956 into its marked place, although in those two MSS. the new
20 lines were not yet added. The allusion to 1. 3956 first appears,
consequently, in the Lansdowne MS., and the same allusion occurs in
the Harleian. But the three " edited" MSS., viz. Cm., Dd., and E.,
unfortunately destroyed the point of the allusion by restoring the
order of stories to their first position.
The omission of lines 4233-8 in Pt. Cp. L. was due to repre
hensible carelessness, as they are obviously essential, and appear
in Hn.
Lines 4637-52 form the Nun's Priest's Epilogue. They appear in
the Eight-text in Dd. only, and copies of them are scarce. The last
line, in which the Host " Seide vnto another as ye shuln heere " is
remarkable. It implies that Chaucer did not then know what Tale
was to come next, and the problem was never settled. And thus the
Epilogue became useless, and small heed was paid to it. It is found
in MS. Royal 17 D xv., which seems to belong to the "Petworth"
scheme, and which I suppose to give, practically, the " first arrange
ment " of the Tales. I append the following notes.
Comparison of the MSS. Doctor and Pardoner. 23
[3079-3108]. Cm. defective. 3147-8. Omitted in E., which is
surprising in the case of so excellent a MS. But the scribe for once
lost his place, as both 3147 and 3149 begin with the same four words
—This maketh that oure. 3236, 3247; Pt. omits. 3469, 3501,
3533. Cp. omits; each of the three lines begins a stanza. 3561. Pt.
omits. 3570. Left blank in Dd. [3615-52]. Cm. defective. [3957-
4048]. Cm. defective. [4000-79]. Dd. defective. 4353-6. Dd.
omits ! 4479-80. HI. omits (as noted above) because the scribe con
fused But trewely (4479) with For trewely (4481). [4581-4636].
Cm. defective.
Group C ; DOCTOR AND PARDONER.
There is not much variation here ; but the sixteen lines numbered
291-306 are troublesome and require close attention. The chief
points are given below. Lines 487& and 488& seem to be spurious.
Lines
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
291-2
297-8
—
—
?
?
?
f]
—
—
299
—
—
?
?
[ ]
—
—
300
—
—
—
—
1 1
—
—
305-6
r i
•
487J?
488&?
—
603-4
—
—
_
—
—
—
i i
[1-166]. Cm. defective. 103-4. E. omits. [243-386]. Cm. defective.
333-4. Follow 346 in Dd. 356. L. omits. 409-427. L. omits!
478-9. HI. omits ! Perhaps because 477-80 all end in -eres. 743-4.
Cm. omits ; because 742 ends with rede and 744 with reed.
Group D ; WIPE, FRIAR, AND SOMPNOUR.
In this Group there are several points of interest.
1. The six lines after 44, in Dd. only, are genuine.
2. Lines 575-84, 609-12, 619-26, 717-20 (26 lines in all) are
genuine and characteristic. But they may have been added late,
appearing only (in the Eight-text) in the "edited" MSS., viz. Cm.
Dd. E.
3. HI. preserves six lines, all (I believe) genuine, but not in the
other seven MSS., numbered 2004 1>, c, 2012 b, c, and 2048 b, c.
Though marked by Furnivall as " spurious," they scan perfectly and
24 Canterbury Tales (Eight-text).
rime correctly, though 2004 b is a " nine-syllable " line. Lines
2048 b, c end in best, arrest, precisely the same final words as in E
1281. There is, however, a scribal error in 2012 6; read — "Ire is
the gate of synne, as seith the wise." This is a very characteristic
line, and perfectly correct ; for I have found the Latin original. It
translates — " lanua omnium vitiorum iracundia est," which occurs in
Kabanus Maurus, Expositio in Proverb. Salamonis, lib. iii, c. 29 ;
Opera, ed. Migne, vol. cxi, col. 775.
4. HI. also gives us two more additional lines, numbered 2037 b, c.
These are also marked "spurious," and must indeed be deleted. All
the same, 2037 b is so far from being spurious, that it merely repeats
2030 on the same page, which is given as genuine ; and 2037 c is also
partly genuine, as it repeats a part of 2031. What has happened is
this. As both 2029 and 2037 end with be deed, the scribe lost his
place, and began copying out 2030 and 2031 over again, and then
discovered that they did not rime together ; and not wishing to show
marks of erasure or deletion in his manuscript, calmly completed
1. 2031 in a way of his own; after which he went on as if nothing
had happened !
5. HI. is the only one of the eight MSS. which gives both 1295-6
and 1307-8 in their right positions, where all the rest have blundered
sadly. This requires a great deal of explanation on the part of those
who would make out that this MS. shows no trace of ' inspiration/
On the contrary, it is obvious that the scribe had access, at this point,
to a MS. of unusual value.
I now tabulate the more important variations.
Lines
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
45 &,c, <*,*,/,#
—
222
—
?
?
?
—
—
—
—
382
—
?
?
?
—
—
—
—
575-84
—
—
—
605-8
.
609-12
619-26
717-20
—
__
1295-6
?
I
?
?
—
[ ]
?
?
1307-8
?
?
?
?
—
[ ]
?
?
1731-2
—
?
?
?
—
[ ]
—
1872-4
—
—
—
—
—
— .
2004 b, c
—
2012 b, c
—
2048 b, c
—
Comparison of the MSS. Cleric, Merchant. 25
[1-76]. Cm. defective. L. begins with four additional and obvi
ously spurious lines. 10. Cp. omits. 24. Cp. supplied in a later hand,
over an erasure. [146-217]. Cp. defective. [1265-1322]. Cm.
defective. 1491-2. Cm. omits! 1525-7. Cm. partly omits, because
1525 and 1527 both begin with My trouthe wol I holde. [1631-
1746]. Cm. defective. 1655. L. omits. 2037 b, c. HI. partly repeats
here 2030-1, by mistake. [2159-2294]. Pt. defective; four
absurdly spurious lines supplied. [2285-2294]. Cm. defective.
Group E ; CLERK AND MERCHANT.
There is little variation in the Clerk's Tale; though Cp. and L.
drop three lines.
The suggestion that the " Host-stanza," or " Clerk's End-link,"
which consists of a single seven-line stanza, was originally meant to
follow line 1162 or line 1169, is entirely baseless. The evidence
shows clearly that it was meant, of course, to follow the Clerk's
Envoy, and to connect the Clerk's Tale with another one. It occurs,
accordingly, in that position, in most of the MSS. that give it, viz. in
Hn. Cm. Dd. E. ; and in Earlow 20, Egerton 2864, Harl. 1758, HarL
7333, and Royal 18. C. 2. So also in the old printed editions. But
in MS. Trin. R. 3. 15, it is absurdly placed after the Pardoner's
Tale.
It appears from the very first, viz. in Hn. But Chaucer did not
know, at that time, what the next Tale would be. At last he decided
to link the Clerk with the Merchant, which he did by discarding this
stanza altogether, and writing a new Link (absent from Hn. Pt. Cp. L.)
which first appears in the " Harleian" type of MSS. This fact gives
an extraordinary value to HI., because lines 1213-44 are obviously
genuine, whilst at the same time they rendered the suppression of the
Host-stanza absolutely necessary, in order that 1. 1213 may be an echo
of 1. 1212.1
I add a few notes (chiefly to the Merchant's Tale), after giving
the comparative Table on the following page:
1 Chaucer once placed the Clerk before the Franklin, but did not link them
himself; see the remarks on the Petworth MS. in the Six-Text, pp. 441, 476,
and pref. 53-6, discussed at pp. 49, 50 below. This is why the Franklin's Tale
follows the Clerk's Tale in all the old black-letter editions.
26
Canterbury Tales {Eight-text}.
Lines
Hn.
Pt.
Cp.
L.
HI.
Cm.
Dd.
E.
117
__
_
_
_
__
143
173
_
1213-44 1
(new link) /
—
[ ]
—
— •
1305-6
\
—
I
i
.
1646-7
1777-8
\
?
?
.
1816
*1818Z>
1927-8
2356-7
—
[ ]
[ ]
—
—
—
2419-40
—
—
[ ]
[ ]
—
[ ]
— —
NOTES. [1-112]. Cm. defective. [33-107]. Cp. defective. 65, 74,
81. L. omits. [1170-6]. Pt. omits. [1190-1292]. Cp. defective.
1213-44. This is the new Merchant's Prologue; not found in Hn. Pt.
Cp. L. [1213-64]. Cm. defective. 1305-6. Imperfect in Hn. ; Pt.
doubtful, &c. See my note upon this couplet. 1308. Cp. omits.
1357-61. E. omits, because 1357 ends with reede (error for reed),
and 1361 ends with rede', but the lines are genuine; see my note.
1483. Cm. omits; line blank. 1527-9. Pt. omits in part, because
1527 and 1529 both begin with To whom I yeue my. 1645. Pt. Cp.
L. add here to the line, as if it rimed with dere in 1648 ; and so omit
1646-7. 1777-8. Pt. Cp. L. vary, and give spurious lines, with
likand as a present participle! 1816. Pt. Cp. L. omit, and fill up
with 1818 6, which is spurious. 2281-8. L. omits, because 2280
and 2288 end alike with man. [2319-2440]. Cp. L. very defective;
122 lines lost. [2405-2440]. Cm. defective.
Group F ; SQUIRE AND FRANKLIN.
Groups E and E were made into one group, as is well shown by HI.
and the "edited" MSS. It was easily done, by altering Frankeleyn
in F 1 to Squiere. It is obvious that F 1 and F 2 are misarranged in
Hn. Pt. ; the words com neer (which make F 1 too long) should begin
F 2 (which is too short). But in the other MSS. they really belong
to F 1 ; and F 2 was lengthened accordingly ; see pp. 49, 52.
Comparison of the MSS. Squire to Canons Yeoman. 27
The chief differences are noted in the Table below.
Lines
Hn.
Pt
Cp. L.
HI.
Cm.
Dd.
E.
1-8
[ ]
[ ]
[ ]
_
_
671-2
[ ]
[ ]
673-708
—
—
[ ]
[ ]
[ J
[ ]
[ ]
917 \
918
918
yi8
t-i
918 '
918 /
~
917
917
917
]
917
1147-8
—
[ ]
—
—
—
1191-6
—
[ ]
—
1265\
1266
1266
1266
1266J
1265
1265
1265
—
1433-4
—
1455-6
1493-8
[ ]
— '
1529\
1530
1530
1530
[T
1530/
?
?
1
—
]
—
1556)
1557
1557
1557 [
—
—
1558
1558
—
—
[ ]
—
1558)
?
|
1567-8
—
—
—
— — —
[1-22]. Cm. defective. 1-8. A Franklin's Prologue in Hn. Pt. :
missing in Cp. L. [47-65]. Cm. defective. 69-72. Dd. omits-
[615-728]. Cm. defective. [617-1223]. HI. defective (12 leaves lost).
[673-708]. Cp. L. defective. [673-728]. Cm. defective. [673-753].
Dd. defective. 679, 680. Pt. omits. 1109-1110. Cm. varies, and
inserts a spurious line after 1109. 1129-32. For these, Cm. has
1130, 1131, 1129, and a blank line. 1473-4. Cm. omits, because
both 1472 and 1474 begin with Ye. [1472-1564]. Dd. defective.
1512. L. omits. [1575-1624]. Cm. defective. [1584-1624]. Cp.
defective. 1595-1602. L. omits. 1603-4. L. transposes these two
lines.
N.B. Lines 1455-6 (2 lines) and 1493-8 (6 lines) occur in E.
only; they are certainly genuine, and 1455-6 occur in the black-letter
editions. Observe how scarce are 673-708 ; this is how they came to
be overlooked in Morris's edition.
Group G ; SECOND NUN AND CANON'S YEOMAN.
The chief points to be noted are (1) the loss of twelve essential
lines (326-337) in Pt. Cp. L. ; and (2) the total absence of any
reference to the Canon's Yeoman in Hn. This tale begins with 1. 554.
I subjoin a Table of the chief differences and some notes.
28
Canterbury Tales (Eight-text}.
Lines
Hn.
Ft.
Cp.
L.
HI.
Cm.
Dd. E.
73-4
156-9
£ 213, i 214
—
[ 1
—
—
—
326-337
—
—
—
—
432
562-3
564-5
564-5
—
562-3
1238-9
—
—
—
—
—
[ ]
[1-14]. Cm. defective. [1-36]. Cp. defective. [85-140]. Cm.
defective. 103. L. omits. 155. HI. omits. 158. Cm. omits (because
158 and 159 both begin with And) ; but this is independent of the
loss of 156-9 in Cp. L. 210-6. HI. omits; because 209 and 216
both end with where. 213-4. Pt. Cp. L. confuse thing in 213 with
thing in 214. 390. Cp. omits. 484. Pt. omits. [554-1481;
Canon's Yeoman]. Hn. does not contain this tale. [706-57]. Cm.
defective. 711. Dd. omits; " deficit versus." [856-end]. Dd. defect
ive. 1046-7. CAi. omits (confusing crepe and crede). 1057. HI.
omits (added in later hand). 1283-4. HI. omits (added in later
hand).
Group H; MANCIPLE.
Not in Dd., which is defective. No particular variation. 47-52.
Pt. omits; because 46, 47 and 52 all begin with And. 215-6. Cm.
omits ; because 214 and 216 both begin with If.
Group I ; PARSON.
Groups H and I are but one. They are only separated in Hn., in
which the Tales are but partially arranged. The whole Group (H
and I) is lost in Dd. The variations do not appear to be of any
special significance. I therefore omit details.
The Harkian Manuscript 7334. 29
mr % f arleian pS. 7334.
IT remains to summarise the results given in the above Tables, in
order that we may make a fair estimate of the Harleian MS. as
compared with the other seven MSS. in the Eight-text.
First of all, we ought to leave out of account the " edited" MSS.,
Cm. Dd. E., as they are in a different category, being compiled from
previous sources, of which I believe HI. (i. e. the " original " of HI.)
to be one.
It remains to compare it (1) with Hn. ; (2) with Pt. ; and (3) with
Cp. L. ; the close connexion between the two last is very evident from
the Tables.
Group A. — As compared with Hn., HI. has ten new lines (637-8,
2681-2, 2779-82, 3155-6); but all but the last pair had been
supplied already in Pt. Cp. L. Hence it only supplies two new lines
(3155-6) ; which were adopted by Cm. E. The only later addition
is the pair of lines in E. only (372 1-2).1 But it drops the couplet
4375-6, by a scribal error. The result is that, in Group A. and not
counting obvious blunders,2 HI. has eight lines more than Hn. and
as many as Pt. Cp. Ln. As compared with the rest, it contains
as many lines as Cm. ; at least six more than Dd. ; and two less
than E. (Of course 252 &, e, are in Hn. only.)
Group B; PART I. (1-1162). HI. omits 417 in the Man of Law's
Tale. But as this is the middle line of a stanza, it must needs be a
fault of the scribe and is of no significance. It is otherwise complete.
GROUP B; PART I. (1163-1190).— Not in Hn. First in Pt. Cp.
L. as a Squire's Prologue. (What we now call the Squire's Prologue
(Fl-8) first appeared (according to my theory) in Hn. as the latter
part of a Franklin's Prologue, but the rearrangement in HI. made it
follow the Merchant's End-link ; and this excellent change was
adopted in the "edited" MSS., Cm. Dd. E.)
It is quite clear that these lines (Bl 163-1 190) ought to have been
suppressed, and should not have appeared in HI. at all. Practically,
the scribe made this discovery for himself while he was copying out
the lines. When he came to line 1179, which begins in Hn. Pt. Cp.
L. with Sayde the Squyere, he must have found out that the Squire's
1 This couplet is in Thynne and the old editions.
2 Such as the blank at 1. 2040, and the defects in 11. 2010-7 and 4375-6.
30 Canterbury Tales (Eight-text).
Tale was really a long way off, and that five Tales still intervened in
the arrangement to which he was committed.1 But after writing six
lines more, he gave the matter up as hopeless, and left off at 1. 1185
in the middle of a sentence ; which was perhaps the best thing to be
done. He then added : — Here endith )>e man of lawe his tale !
If the lines are to be preserved, the easiest thing to do is to follow
Tyrwhitt's lead in assigning them to the Shipman, and so making a
Shipman's Prologue of them, as in the Six-text. But it ought to be
distinctly understood that this arrangement is a mere make-shift, and
absolutely contradicts the evidence. The fact that one MS. happens
to support this idea is really of no weight, as the rest of the MSS.
absolutely prove that Chaucer never thought of it.
In making my own edition, I had practically no choice in the
matter. The matter had been already decided, and a new method of
numbering the lines had been already established. I am now con
vinced that the right thing to be done, in order not to falsify the
general evidence of the MSS., is as follows, viz. to keep these lines
(1163-1190) as a Squire's Prologue, and to put a line of stars after
1190, to show that there is here a complete break. There should be
also a note to say that the lines are only preserved because they are
genuine, and had at one time formed a part of the scheme ; but must
be regarded as having been cancelled. This is no more than has to be
done with the so-called " Host-stanza" at the end of the Clerk's Tale.
The Shipman's Tale (without a Prologue) follows the Pardoner's
in all the MSS. printed in the Eight-text except Pt., where it follows
Gamelyn. It also follows the Pardoner in at least a dozen other MSS.,
in all the old printed editions,2 and even in Tyrwhitt. If we are to
regard evidence at all, there is no other place for it.
GROUP B ; PART II. (1191-2156); SHIPMAN, PRIORESS, AND SIR
THOPAS. — No characteristic difference between the eight MSS.; see
above. HI. omits 1355 by a mere blunder ; 1376-9 by the careless
ness of not observing the recurrence of / yow praye ; and the final
half -line in Sir Thopas, which the scribe wrongly disregarded. It also
omits 1995, which only Dd. preserves. This is all that is wrong as
regards its contents.
1 I suggest that in his exemplar, the lines in question had not been deleted,
though the precaution had been taken of partially erasing the word squire ; so
that the scribe, seeing an initial s, filled it up as sompnour, and then found that
it would not do.
2 Dr. Furnivall gives, in his Table, the arrangement in "Caxton," which
agrees with Thynne's. By " Caxton " is here meant Caxton's first edition ; the
arrangement in his second edition somewhat varied.
The Harleian Manuscript 7334. 31
MELIBEUS. — :The Table shows that the losses in HI., as compared
with Hn. Pt. Cp. L., are not great; 2646-7, 2708, and 2741 (all in
Hn. Cp. L.) are all wanting in Pt. As compared with Cp. L. only,
it shows well by contrast, as they have lost 2313-4, 2315-6, 2328-61,
2378, 2387, 2409, 2417, 2447-8, 2476, 2519, 2586, 2654, 2746-7,
2815-6, 2963-5, and 3059 ; all of which it preserves. But it omits
parts of 2432 and 2646-7, 2708, 2730, and 2741. HI. also omits
3034, owing to the repetition of good name. I may, however, have
missed some point or other.
GROUP B; PART II. (3979-4652).— As compared with Hn., the
losses in HI. are three, viz. 3213-20 (the second stanza on Sampson),
4136-7, and 4479-80; all apparently due to carelessness, as the other
seven MSS. preserve them. As compared with Pt., HI. restores 3764
and 4233-8. As compared with Cp., HI. restores 3197-3204, 3611,
3764, 4112, 4114, and 4233-8. As compared with L., it restores
3611, 3764, 4112, 4114, and 4233-8. Compare the remarks above.
Group C.— In 291-2, HI. does not follow Hn. Pt., but the alter
native (and possible) reading in Cp. L. It omits two couplets, viz.
299, 300 and 305-6 ; also 478- and 479 (by error). It restores
603-4, which do not appear in Cp. L.
Group D. — I have already noted above that HI. alone preserves
the lines 2004 b, c, 2012 b, c, and 2048 b, c. Also, that HI. alone puts
the couplets 1295-6 and 1307-8 in their proper places.
As compared with Hn. Cp. Pt. L., HI. omits 605-8, perhaps by
design on the part of the scribe ; see p. 34. The readings of 1731-2
in Pt. Cp. L. cannot be right, because they reduce 1730 to nonsense.
Observe that 45 b-g, 575-84, 609-12, 619-26, 717-20, are aU late
additions, in Cm. Dd. E. only.
Group E. — I have already noted the great value of 1213-44, the
new Clerk-Merchant link in HI. HI. also restores the lines dropped
in Cp. L. The chief fault in HI. (in this group) is the loss of 2356-7
(as in Pt.) ; but the reason is obvious, viz. that it was due to confusing
the ending of 2355 (agayn his sighte) with that of 2357 (his sighte
agayn).
Group F. — Here HI. has unluckily lost 8 leaves, containing 608
lines (F 617-1223), which exactly agrees with the fact that a leaf of
that MS. usually contains 76 lines. Otherwise it rightly agrees with
Hn. throughout, as against Pt. Cp.' L.
Group G. — Here HI. again rightly agrees with Hn. (as far as it
as against Pt. Cp. L. After Hn. fails, HI. agrees with Pt. Cp.
32 Canterbury Tales (Eight-text).
L., which give the Canon's Yeoman's Tale with a quite unusual
correctness and without any loss ; probably because that Tale first
appeared in Pt., being somewhat belated.
Group H. — Complete in HI. ; though Pt. omits 47-52.
Group I. — I make no remarks on this group, as the evidence
seems sufficient without it.
Collecting all the results, we see that HI. (except for careless
mistakes by the scribe) is really fuller than Hn. Pt. or Cp. L. ; and,
as compared with these, has some points of superiority. The gains are
these.
1. In Group A; lines 3155-6 : —
And euer a fousand goode agayns oon badde ;
That knowest ]>ou wel pyself, but if J?ou madde.
2. In Group D; lines 2004 6, c, 2012 b, c, 2048 b, c.
Schortly, may no man, by rym and vers,
Tellen her thoughtes, fay ben so dyuers. —
Ire is the gate of synne, as saith pe wise ;
To fle )>er-fro ech man schuld him deuyse. —
An irous man is lik a frentik best
In which )>er is of wisdom noon arrest.
3. In Group D, HI. alone puts the couplets 1295-6 and 1307-8 in
their proper places. All the rest are wrong.
4. In Group E, we have the great gain of the new Clerk-Merchant
link, which does not appear in Hn. Pt. Cp. L. ; not merely 32 new
lines, but 32 lines of especial interest and usefulness, and, apparently,
Chaucer's latest addition of any length. Thus the total gain in HI.
(in Groups A, D, E) amounts to just 40 lines, all genuine; of which
E preserves 34, Dd. has 32, and Cm. (which is deficient) only 2.
If it were only for the sake of these four gains, MS. HI. ought to
receive respect. But see further below.
I attribute the value of HI. to the excellence (in respect of
fulness) of the exemplar which the scribe had before him ; but it can
hardly be denied that the said scribe had moments of carelessness not
far removed from idiocy ; and I would attribute to him such blunders
as occur in the following examples, which are by no means
exhaustive.
A 1547. Of Cadynus / )?e which was ]?e furst man.
The scansion is hopeless ; but the carelessness consists in misreading
The Harleian Manuscript 7334. 33
Cadmus as Cadinus, and then writing it as Cadynus ; for he had just
written down the form Cadme in the preceding line ! Read Cadmus,
he for j>e, and firste, and it may pass.
A1659. That frozen white as fome froj>e wood.
No doubt he had before him for yre (as in Pt. ; most MSS.
have for ire) ; yet he was capable of turning it into fro]>e, because
he had just written frozen. One wonders what he thought it meant !
A1664-. That excused in j?e world ouer al.
Surely he had execute^ in his copy.
A1682. For after may he serue)> now Dyanc.
Of course may is miswritten for Mars!
A18£6. And fey him swore his axyng euery dele.
Of course he had euery dele in his original, only it was in the pre
ceding line ! He should not have repeated it in place of faire and
wele.
A1966. A bowe he bar and arwes fair and greene.
Surely his original had kene ! For how can good arrows be
green ]
A2011. I saugh woundes laughyng in here rage.
His copy can hardly have had that ! The fact is, that there must
have been something illegible or imperfect here, as he has omitted the
next six lines and gives 1. 2018 as — The hunt strangled with wilde
bores corage ! Here not only has he misread beres as bores, but he
has boldly added corage at the end of the line, and so secured a rime !
There is no corage anywhere near, except his own. But 22 lines
further on he has done the right thing, and left a blank line in place
of an illegible one.
A2199. The riche aray of Thebes his paleys.
Surely his original had Theseus (not Thebes his ); for the palace
was at Athens. And the same man had just written Theseus, only 9
lines above !
A 3180. And eek more ryalte and holynesse.
Here more ryalte is a playful substitution for moralitee.
It is unnecessary to pursue this painful subject. It seems fair to
conclude that these blunders were due to the scribe, as in all these
cases other MSS. are well agreed as to the right reading. If, on the
EIGHT-TEXT ED. CANT. TALES. D
34 Canterbury Tales (Eight-text).
other hand, he found them all in his copy, he was so exceptionally
unlucky that one wonders how he so often gives lines correctly.
But of course his copy went wrong sometimes. They all do so to
some extent.
The scribe does not often omit a line or lines ; in this respect he is
better than many. But I cannot help suspecting that when he did
omit a line, he sometimes did so designedly, in the interest of what he
called more ryalte, or of politeness. Thus he omits C305-6, 478-9,
and D575-84, 605-12, 619-26. For although the 26 lines in
D575-84, 605-12, and 619-26 are not in Hn., Pt., Cp., L., they
ought to have been in the copy which he had before him.
All these considerations suggest that HI. was copied from a MS.
(unfortunately lost) which was both better and fuller than HI. itself ;
and therefore contrasting favourably with MSS. Hn., Cp., Pt., L.,
which would seem to be of an older type. This is, at any rate, the
theory which I propose, viz. (1) that Hn. represents an early col
lection of Tales, at a time before the Canon's Yeoman's Tale was
written, and previous to any careful attempt at arrangement; (2) that
what I have called the "Petworth1 type" of MSS. shows us the
Tales, as they were first partially arranged. (3) Chaucer then some
what altered the arrangement, and we thus get the Corpus-Lansdowne
type. (4) He further improved this arrangement, which gave us the
Harleian type ; and there he stopped, long before his work was at all
complete.
Lastly, some later compiler or editor rearranged the Tales for the
last time, thus producing the "edited" MSS. of the " Ellesmere
type," such as Cm., Dd., and E. But the rearrangement in these last
MSS. was quite unimportant; all that was done in this respect was
to place the Second Nun and Canon's Yeoman after the Nun's Priest
instead of between the Franklin and the Doctor.
The " Petworth," or first partial arrangement, is so common that
we can hardly dissociate it altogether from Chaucer's (at least partial)
responsibility.2 It appears also in Caxton's first edition, and subse
quently in Thynne and all the editions derived from him.
1 There are many such MSS., as the type is common ; viz. Harl. 7333, Harl.
1758, Sloane 1685, Royal 18 Cii, &c. Note that the Petworth MS. itself mis
places the Shipman and Prioress by putting them next to Gamelyn ; the usual
place is after the Pardoner and before Sir Thopas.
2 I think he meant E2419-F8 (found in Hn., absent from Cp. and L.,
but revived in HI.) to be temporarily suppressed or held in abeyance. Their
appearance in Pt. after the Clerk is quite inept, and the scribes made rather
wild work of it. See this discussed below, at pp. 49, 50.
The Harleian Manuscript 7834. 35
If this can be granted, it follows that the better arrangement in
the next or Lansdowne type, is Chaucer's own work also ; because it
shows a real improvement, and places the Clerk before the Merchant,
instead of behind that Tale and with the Wife-Friar- Sompnour
intervening. Note the absence of E2419-F8.
Lastly, that the Harleian type is also due to Chaucer is probable,
for two reasons : (1) because he now or about this time wrote the
Clerk-Merchant link, thus at last joining these Tales indissolubly ;
and (2) because of the great cleverness by which he now linked on
the Squire to the Merchant, and the Franklin to the Squire, and so
reduced the Clerk-Merchant-Squire-Franklin to one coherent whole ;
which no one else could have done. This proceeding involved four
manoeuvres, viz. (1) the suppression of the old Squire's Prologue
[B1163-1190]; (2) the writing of a new Clerk-Merchant link
[El 21 3-44]; (3) the conversion of the old Franklin's Prologue
[E24 19-2240, Fl-8] into a. Merchant's end-link and Squire's head-
link, thus linking the Merchant to the Squire ; (4) the conversion of
an old Merchant's Prologue [F673-708] into Words of the Franklin,
thus linking the Squire to the Franklin.
I have little hesitation in repeating that, whatever the worth or
worthlessness of the readings in the Harleian MS. may be, we must
nevertheless regard this MS. as being absolutely the most valuable that
exists, because it gives the best and latest authoritative arrangement of
the Tales ; and the settlement of this point is quite as necessary, in
its way, as the discovery of the MSS. which give us the best readings.
It was Bradshaw . who regarded this type as being " the most
authentic ; " and he is usually right.
When we speak of "edited" MSS., such as E. Cm. Dd., we of
course imply that they were edited from something else that preceded
them. And if we enquire whence the order adopted in those MSS.
was obtained, there is only one possible answer ; viz. from the
" Harleian type " of MSS. ; and what the order was in that type, is
shown by HI. itself; nor would this result be affected if the scribe of
HI. had miscopied every line throughout.
The single change of order effected in E (Ellesmere) was, as I
have said above, to place the Second Nun and Canon's Yeoman later,
or to " shunt them down." We can see why, viz. to bring the mention
of Boghton (G556) nearer to the Blee (H3). That is all very well,
if we are to go on rearranging the Tales to please ourselves ; but it is
36 Canterbury Tales (Eight-text).
absolutely uncritical, if we really desire (as we ought) to find out
precisely the order in which Chaucer left them. It is more instructive
to know the truth than to try to better it.
It is obvious that Chaucer made no arrangement that even
approached finality, because the thing was impossible. Before that
could be done, he must complete the set ; and this, as he well knew,
would never be accomplished. All that he did was to collect from two
to six Tales into Groups ; to arrange the resulting Groups in a pro
visional order ; to alter this order twice ; to write new links or to
readapt or even suppress old ones ; and so to reduce the number of
Groups gradually. The latest, or Harleian type, exhibits, after all, no
very great advance, for there were still eight groups left. Even his
latest attempt towards this end was anything but final.
But it was much better than at first appears. Seven new Tales,
from the Yeoman, the Ploughman, the Haberdasher, the Carpenter,
the Weaver, the Dyer, and the Tapicer, each with a connecting
prologue and epilogue, would have 'filled up every gap and have given
us a completed set. Each of these Tales could have been written
separately and dropped into its place. There is thus no reason why
the Eight-group scheme should not have been converted into a
triumphant success, so far as the plan went. We can even trace a
specimen of the working of the scheme ; for, by completing the Cook's
Tale and converting " Gamelyn " into a Yeoman's Tale, it would have
been very easy to link the Cook with, the Yeoman, and the Yeoman
with the Man of Law. Perhaps this is why " Gamelyn " occurs in
this position in the MSS. Lastly, Chaucer could easily have made
some shifting of the Tales that would have brought us to Rochester
before coming to Sittingbourne. But he never got as far as that.
ON SOME EEADINGS IN THE HARLEIAN MS.
The recent essay on the Harleian MS. by Prof. Tatlock is of great
value and interest. I find, however, a few conclusions of his that I
cannot accept ; and I believe it may help a few students if I point
these out.
That Chaucer never carefully revised his work line by line is
likely enough. But I hold still, as I have already shown, and have
repeated above, that he did partially rearrange the Tales, and shifted
the order of them more than once. This circumstance enables us to
arrange the eight texts hitherto printed in full in the following order,
The Harleian Manuscript 7334. 37
viz. Hn.— Pt.— Cp.L.— HI.— Cm.Dd.E. ; where I place Cp. L. to
gether, and call Cm. Dd. E. edited MSS. This I take to be the
reason why Hn. contains the fewest number of lines, and E. the most.
This is why, when anything goes wrong in Hn. (such as sertres for
sterres in A2037), the same thing has a tendency to go wrong in
MSS. of a later type. An easy example is the following. It is well
known that in no conplet is there a greater variety of readings and
greater general uncertainty than in El 305-6 (see my note). Turn to
Hn.j and what do we find there ? The reading is : —
And if thow take a wyf she wole destroys
Thy good substance, and thy body annoye.
And here the words printed in italics were added in a later hand,
and are of no authority whatever. It is surely remarkable that,
whilst the other MSS. differ as to the words italicised, they all agree
that the words " And if thow take a wyf " must begin the sentence. I
have adopted the reading of Cm. E., chiefly because it is reasonably
good, scans correctly, and makes sense ; and I may here say, once
for all, that the compiler of E. (who was not Chaucer) is usually
successful (1) because he took pains, and (2) because he came last,
and had the pick of all that preceded him.1
The reading in E. is : —
And if thou take a wyf vnto thyn hold,
Eul lightly maystow been a cokewold.
The reading in HI. shows a daring variation : —
And / if J)at J>ou take a wif, be war
Of oon peril, which declare I ne dar !
It is not without humour, as it glances at the name of cokewold as
being a word that should not be fully expressed. It is, however,
suspicious, as the former " nine-syllable " line is not happily stressed,
and the accents in the second line are unusual.2 And so I leave it.
The best way to examine the readings of HI. is (I believe) to
compare it with Hn. Pt. Cp. L., rather than with Cm. Dd. E.,
because there is always a chance that a reading in the latter set may
be derived from HI. itself. A curious example occurs in A 8 ; where
Hn. Pt. Cp. L. have :-—
Hath in the Earn his half cours y-ronne.
1 MS. E. is early, and there are many later ; but he had a sufficient choice,
as he had access to samples of all Types.
2 Yet not (perhaps) impossible ; cf. departinge, B 260, 293 ; desiring, A 1922 ;
desirous, F 23 ; d&tourbing, Compl. Venus, 44 ; redoutynge, A 2050.
38 Canterbury Tales (Eight-text).
We know, of course, from the scansion and from grammar alike, that
half should be halfe ; and this is the reading of HI. We can then,
out of curiosity, consult Cm. E. (Dd. fails), and we find that they
likewise have the erroneous reading half. This is because they may,
or they may not, pay respect to HI. I think E. got half from Hn.,
which it so often copies in minute details, glosses, side-notes, and all.
Prof. Tatlock does not, in this respect, give HI. its due. Thus in
4686, Hn. has :—
His walet biforn hym in his lappe.
And Pt. Cp. L. Cm. E. [Dd. is deficient] all follow like a flock of
sheep. But the sentence has no verb, and the line scans vilely. HI.
has : —
His walet lay byforn him in his lappe.
Surely this insertion of lay is an emendation as to which there can be
no doubt. Whence did it come, if not from Chaucer ]
In A831, Hn. (followed by the rest) has :—
Lat se now, who shal telle the firste tale.
But HI. has :—
Let se now who schal telle ferst a tale.
If this is not an emendation by Chaucer, it is at any rate worthy of
him ; it is better expressed, and the line is much smoother.
A1376. Hn. (followed by Pt. Cp. L.) has :—
Biforn his celle fantastyk.
Here is no sense, and only eight syllables.
E. Cm. Dd. give us : —
Biforn his owene celle fantastik.
Here (since owene = ow'ne) we get ten syllables, but still no sense.
It is an attempt at emendation, and is wrong.
But HI. has :—
Byforne in his selle fantastyk.
I should prefer Biforen. But, passing over this, we here recover the
missing word in, which is obviously right. It is not due to a revision
by Chaucer, but, as in other cases, to a recovery of Chaucer's text.
A2555-6. Hn.:—
And if so falle / the chief teyn be take
On outher syde / or ellis sleen his make.
The Harleian Manuscript 7334. 39
Here sleen is impossible, as we require the 3rd pers. pres. sing,
subjunctive. Pt. Cp. L. Cm. E. all follow Hn. ; but HI. has the
right form slee, which of course is Chaucer's word. And if we must
not owe an emendation to HI., it may suffice to say that Dd. has sle.
And Dd. is an honourable MS.
I hold it as good as proved that HI. sometimes has the right
reading where the other seven MSS., or most of them, have gone
wrong; and in this sense, at any rate, it is an "inspired" MS., being
in true contact with the author (how, I do not know) where other
good MSS. fail. There is therefore no reason why it should not
sometimes be right in cases of much variation and considerable doubt.
Whence it is obvious that it should always be consulted.
On the other hand, I have no doubt that it is often unsafe and
treacherous. Sometimes the scribe was stupid and wrote nonsense, as
has been shown ; and there is reason to suspect that he did not always
follow his copy, but may have made " emendations " on his own
account. And it may also be the case, that the compiler whose
exemplar was before him did the same thing. It can hardly be always
an honest record ; and hence comes all the trouble. For this reason,
Prof. Tatlock's paper is very much to be commended, as it is very
helpful.
It is, however, not a little strange that the very second example
which he adduces (p. 5) to show that HI. can give a good emendation,
is precisely one where I am of quite a different opinion. I must
therefore quote it ; only we require the context, upon which all turns.
The passage in HI. (B4377-80) runs thus :—
Whan ]?at ]>e moneth in which the world bigan,
That highte March, whan God maked first man,1
Was complet, and passed were also
Syn March bygan, tway monies and dayes tuo, &c.
Let us reckon this up. We start from March 1 ; then we count
onwards to March 31, when the month is complete ; also, two months
more (April and May), which brings us to May 31 ; and 2 days more,
which lands us in June ! This will not do at all ; because the day
meant is May 3. It is all explained in my note to B4045, p. 250 of
vol. v. Not that I claim the credit of explaining it, if any credit
there be ; for it was explained long ago in Thynne's Animadversions
on Speght's Chaucer, p. 62 ; and again by Mr. Brae, who calculated it
1 The other 7 MSS. have first maked or first made.
40 Canterbury Tales (Eight-text).
carefully by astronomy, and found that the hour at which chantecleer
crew was 9 A.M. on May 3. The right reading is, of course : —
Sin March bigan, thritty dayes and tuo.
Sin March bigan indicates the starting-point ; a completed March
brings us to March 31 ; and the thritty dayes can only refer to April,
for the plain reason that March has thirty-one days ! So April is
also complete. Then come two days more, which are May 1 and May
2, also complete ; and then we are safely landed in May 3. The
scribal side-note in Dd., " id est, secundo die Maii," is of no authority
and quite wrong. If the scribe had written " id est, tertio die Maii,"
he would have written sense. I am quite willing to accept a reading
of HI. as against other MSS. in many cases, but certainly not here.
Why Chaucer is here charged with " carelessness and confusion " for
having stated a simple fact in a playfully roundabout manner, I do not
know.
As to the so-called " nine-syllable " lines, I should like to say a
few words. I think it is very likely that I may have needlessly
" amended " some of them by taking readings from the Harleian MS.,
and I am sorry for it. I do not think the history of this matter is
sufficiently well known. I remember the time when the mere suppo
sition that Chaucer ever wrote such lines was a despised heresy ; and
I have good cause to remember it. My first pronouncement on the
.subject,1 which for some time brought down upon me quite a storm of
contradiction, and contempt, appeared forty-three years ago, in Morris's
Chaucer, ed. 1866, vol. i. p. 174. I there dared to assert that the
following lines (all from the Harleian MS.) can be "properly scanned
by making the first syllable stand alone."
May / with al thyn floures and thy greene ; A1510.
Ther / by auenture this Palamoun ; A1516.2
Now / it schyneth, now it reyneth faste ; A1535 ; &c.
James Russell Lowell, whose gracious presence I remember and revere,
in his delightful essay on " Chaucer " in " My Study Windows " told
us that " his [Chaucer's] ear would never have tolerated the verses of
nine syllables, with a strong accent on the first, attributed to him by
Mr. Skeate [sic\ and Mr. Morris [who expressed no opinion on the
subject]. Such verses seem to me simply impossible in the pentameter
1 The same heresy had been broached at least twice previously. But I found
it out without help, all the same.
2 However, the Six-text has: — "There as, by auenture," &c. For other
similar cases, see Prof. Tatlock's article, p. 12, note.
The Harleian Manuscript 7334. 41
iambic as Chaucer wrote it." But those were early days (1871). I
retorted by quoting 13 consecutive lines of this character from
Tennyson's "Vision of Sin," to show that such lines are still in
use : —
" Then / methought I heard a mellow sound ; " &c.
And I sometimes ask myself whether people ever read, with any
attention, the English dramatists about which they write so glibly. I
give some examples.
Win / my love, and I will make thee great.
Greene, James IV, A. i. sc. 1.
Were / I baser born, my mean estate ; id. iii. 3.
Will / them, even as they love their queen ; id. v. 1.
Frank/ly tell me, wilt thou go with me ?
Peele, Battle of Alcazar, A. ii. sc. 4.
Phil/ip, whom some call the Catholic King ; id. iii. 1.
Bar/barous and bloody Tamburlaine.
Marlowe, Tamljurlaine, I. ii. 7.
Con/quer, sack, and utterly consume ; id. II. iv. 2.
Jer/ome's bible, Faustus ; view it well.
Marlowe, Faustus, i. 1.
Holmo,fuge! whither should I fly? id. ii. 1.
Lay / hands on that traitor Mortimer. Edw. II, i. 4.
Those / that hate me will I learn to loathe.
Marlowe, Massacre at Paris, i. 2.
I could quite easily add a score more of examples ; see A Student's
Pastime, p. 376. Our old writers were perfectly familiar with the
device. Abbott's Shakespearian Grammar quotes a considerable
number.
Stay, / the king hath thrown his warder down ; Rich. II, i. 3.
The above explanation will show that, if I have weakly turned a
nine-syllable line into one that is more acceptable to many others a
little too often, it was not because I have any prejudice against them
myself. It is strange to remember a time when I had to sustain this
theory against all comers.1 It encourages me to imagine a time when
1 It is amusing to see Prof. Tatlock (p. 16, note 1) speaking of " the archaistic
language of the 16th-century Court of Love," when I remember the hard things
said of me when I denied that Chaucer wrote it. Let me quote a specimen.
" The criticism which would exclude from the Chaucerian canon such things not
merely as the Court of Love and The Flower and the Leaf, but in whole or in
part [I do not deny a part] the existing English version of the Romance of the
Rose . . cannot be allowed to pass quite unchallenged by those whom it does not
satisfy."— Saintslmry, A Short Hist, of Eng. Literature (1898), p. 119.
42 Canterbury Tales (Eight-text).
my explanation of the " Evolution of the Canterbury Tales " may be
first of all tolerated, and then accepted as being obvious to all.
The best tests will be (1) to examine the history of the Clerk-Merchant
link (E1213-1244) in all the MSS. ; (2) to explain why many MSS.
know nothing about it ; .(3) to find out what MSS. besides HI. put
D1293-4 and T>1307-8 in their right places; (4) to trace the
history of B1163-90, F673-708, and E2419-F8.
I think Prof. Tatlock does me some injustice in saying (p. 2, note
4) that my views on the " Evolution " depend upon " the contents
and arrangement of seven (not all) of the published MSS." Surely
I tried to make it plain that I depended upon the contents and
arrangements of the thirty-six MSS. described in the "Trial-Table"
prefixed to the " Six-text " edition, at the back of p. xxi*. Here is
arithmetical proof that my evidence is five times greater than it is
alleged to be. It makes a difference. Thirty-six is a fair proportion
of the whole, and ought to give good tentative results. To say that
I only consider seven MSS. is very careless ; for I mention fifteen more
on p. 16, and three more on p. 22. My point is mainly this : — that
if you only really understand to what "type " a given MS. belongs,
you can understand the readings a great deal better, and can, not
unfrequently, guess them beforehand. For example, I have now
examined the eighth published MS., Camb. Dd., and find that its
readings usually exhibit the peculiarities of an " edited " text ; the
" type" to which the "Trial-table " shows it to belong. So there may
be some truth in my scheme after all. I proved, for example, that
"H and I form an inseparable group," and "that E and F do" the
same (see Prof. Tatlock's article, p. 21, note 1). He adds — "Dividing
B, as I believe we should, into two, we shall have, therefore, eight
groups instead of nine." All of which I have said already; only
" as I believe we should " ought to be changed into " as we must"
As to the remarks in note 1 on p. 23, I find it said that I " half-
exclude" A252 b, c. It was not my doing. I had to accept the
numbering in the Six-text, or else to throw out all the numbering
of the rest of Group A. Also, that I excluded six genuine lines
after D44, " which are known to me in three MSS." Yes, but they
are not in the Six-text, and I forgot all about them till I came to
write the Notes in vol. v. ; by which time vol. iv. had been printed
off. It will be easily understood that my edition of Chaucer is by no
means final, but was meant to be an improvement upon its predecessors,
and to serve the immediate wants of students ; it had to be completed
The Harleian Manuscript 7334. 43
within a limited time, or not at all. The shortness of human life has
to be considered.
It seems a pity to trouble about the fact that the Manciple's Pro
logue hints at early morning (HI 6), while the Manciple's Tale, when
prefixed to the Parson's, belongs to the afternoon. The solution is
extremely easy, as the Hengwrt MS. shows us. The Manciple's Tale
there precedes the Man of Law ; J and that is why the Man of Law
is so emphatic about the progress of the day, and laments that it is
already ten o'clock. But in his very first arrangement of the Tales,
Chaucer moved the Tale down so as to precede the Parson's, without
troubling to remove the inconsistency, because he could easily do that
when he rearranged the Tales again finally. And that day never
came. There is no difficulty. The poet was always putting off that
final duty ; a most fatal procedure.2
NOT is it worth while to make a difficulty of the reading Ten of
the clokke in 1. 5 of the Parson's Prologue. It is certainly wrong,
but the error was in the MSS. from the beginning] apparently a
stupid error on the part of the scribe, made by confusing 4 p.m. with
the tenth hour of the day. Indeed the error is valuable ; for it proves
that Chaucer never revised his text line by line ; or, at any rate, he
never did so as far as the Manciple. To this day, the French text
shows us that there are gaps (which he never discovered) in the Tale
of Melibeus. The rearrangement of Tales and even the alteration of
links did not involve a line-by-line revision of the whole text. He
was not the man to trouble much about it, beyond banning Adam
Scrivener once for all.
As to this reading (Ten) there really need be no difficulty.
Chaucer says the afternoon was already advanced, and another Tale
had to be told before sundown; indeed, earlier than that, for the
early-rising Pilgrims would surely cease to ride at 6 ; and the sun
set about 7. Hence the readings 6, 7, 8, 9, 10, 11, 12 are all
impossible'^ and as the afternoon was already well advanced, we
cannot read 1 or 2. Hence the only possible readings are three, four,
and fyue ; every one of which is dead against the readings of the
majority of the MSS. Of these three readings, four is the only one
that agrees both with the given length of the shadow and with the
twenty-nine degrees of height ; and even a few MSS. have this
1 As in the ill-arranged Christ Church MS.
2 All that Chaucer did at the time was to erase a name in Hn. in Group I,
line 1, and to tell Adam to write Manciple instead.
44 Canterbury Tales (Eight-text).
reading. It is also, in any case, the most likely hour of the above
three. Let us sometimes rest content with a reasonable possibility.
Even the immaculate editor of the future will have to put up with
making suggestions which many critics will still condemn.
Classification 0f % I
I HAVE already shown how to classify the leading MSS. in my
paper on " The Evolution of the Canterbury Tales " ; but as the
method does not seem to be sufficiently understood, I now try to
make it clearer. The statement in Prof. Tatlock's paper on the
Harleian MS. (p. 2, note 4) — that my evidence " is the contents and
arrangement of seven of the printed MSS." — is misleading, as I con
sulted throughout Dr. Eurnivall's Trial-Tables, which give the arrange
ment of thirty-six MSS. ; and at the same time I carefully examined
the numerous printed specimens of connecting links, as given in the
Preface to the Six-texts, being more than seventy in number.
I have now further examined a considerable number of the
descriptions of MSS. in Miss Hammond's excellent book ; and I find
nothing that I cannot reasonably explain. Her system of classification
really differs from mine very slightly, but she obtains no secure
results. The method of analysis there pursued is insufficient; for
when the latest forms of arrangement have been analysed and shown
to have earlier arrangements behind them, it obviously becomes
necessary to reverse the order by synthesis, and to show chronologically
how one arrangement grew out of the other. Of course many of the
MSS. are really due to contamination of types > but for all that there
are four main types that arose successively from the attempt made by
the author to rearrange his material. This he did tentatively and
from time to time, not always making all the cho.nges that were
rendered necessary} and sometimes even temporarily or entirely sup
pressing a link; but we can usually trace with sufficient clearness
what he was about, and why he made such changes as appear. Of
course a little patience is necessary. For no one will ever understand
this matter until he comes to see that some of the Prologues were at
once readapted to fit new circumstances, while some were left to be
dealt with afterwards. Nothing but arrangement in the chronological
1 It is notorious that the Shipman's Tale was originally meant to have been
told by the Wife of Bath ; see B 1200-9. Of. p. 43 above, lines 3-14.
Classification of the Manuscripts. 45
order of development, beginning with the earliest, can ever explain
the whole matter. And it is this which I believe I can do ; though
it is quite possible that a few points can hereafter be better put. All
depends, as it obviously must, on the history of the links. Of these,
such as are genuine require careful handling; whereas such as are
obviously spurious are worse than worthless. It is well to point out
the latter, but they must always be neglected. "We can tell them,
fortunately, by their exceeding badness.
I use Dr. FurnivalPs numbering of the Tales, with arabic numerals
instead of roman ; thus he calls the Clerk's Tale " IV. 1 " ; for which
I write 4a. His arrangement is that of the Ellesmere MS., which (as
I think we all agree) is the latest of all. Miss Hammond uses for the
Clerk's Tale the symbol E 1, because it is the first tale in Group E ;
but as no MS. whatever presents the Tales in the order of the Groups
A, B, C, &c., such symbols are more likely to puzzle than to help us.
The list of Tales in E is as follows : —
1. Prol., Knight, Miller, Reeve, Cook. 2. Man of Law. 3. Wife,
Friar, Summoner. 4a. Clerk. 4ft. Merchant. 5a. Squire. 5ft.
Franklin. 6. Doctor, Pardoner, la. Shipman; b. Prioress; c. Sir
Thopas ; d. Melibee ; e. Monk ; f. Nun's Priest. 8a. Second Nun.
Sft. Canon's Yeoman. 9a. Manciple. 9ft. Parson.
As I have already explained my views rather fully, I now propose
to make several simplifications. For example, it is not worth while
to consider no. 1 ; it is always at the beginning. Nor even Gamelyn ;
it is sufficient to say that it is not in the Hengwrt MS. ; nor in
those of the Ellesmere type. Neither need we trouble much about
Groups 6, 7, 8, 9 ; though I note a few points. Except in the
Hengwrt MS. and those of the Ellesmere type, the order is 8, 6, 7, 9 ;
all unconnected, except by spurious prologues. Group 7 is, in several
MSS., split up, but the final order of its tales (as all agree) is that
given above. The real trouble is with Groups 2, 3, 4, 5 ; and it is
upon those that everything turns.
We might almost go further, and omit 3 ; since it is always
entirely independent of the rest. But its position (as will appear)
affects the others. Groups 2, 4, and 5 demand the most careful
watching.
Omitting for a moment all connecting links, we find these
arrangements : —
Hengwrt MS.— 1| 2. 5«. 4ft. 5ft. 4a.
40 Canterbury Tales (Eight-text}.
A. "Petworth" type.1— 2. 5a. 4ft. || 4a. 5ft.
B. Lansdowne MS.— 2. 5«. || 4a. 4ft. 5ft.
C. Harleian MS.— 2. || 4a. 4ft. 5a. 5ft.
C*. Ellesmere MS.— 2. |] 4a. 4ft. 5a. 5ft.
Observe particularly the symbol ||. It marks the position of Group
3, and always effectually separates tale from tale, and forbids any
linking. It is extremely instructive.
Some MSS. do not conform to any of these types ; but they are
not (I think) numerous, nor do they agree among themselves ; and I
believe their arrangement to be due to contamination or to casual
causes. They are therefore best neglected.
Miss Hammond's groups (p. 172) are as follows : —
I. Ellesmere. II. Harleian. III. Selden. IV. (a) Harl. 7333,
&c. j (b) Corpus, Lansdowne; (c) Harl. 1758, &c. ; (d) Petworth;
(e) Hatton.
In Group III, she places the Selden MS. by itself; but it is
hardly worth considering in this connexion ; the order of tales is : —
1. 4a. 3. 4ft. 5a. 2. 7. 8. 6. 5ft. 9.2 This gives no help, as the
positions of 2 and 5ft are (practically) absurd. In her Group IV,
I take the sub-groups (a) and (e) to be here of very little value, whilst
(c) and (d) hardly differ ; but (b) has, as shown above, so material a
difference from (c) and (d) that, for our present purpose, it must be
considered separately. It is better, perhaps, to give the names of the
Tales, as follows :
Hengwrt MS. || Man of Law; Squire; Merchant; Franklin;
Clerk.
A. Petworth. Man of Law; Squire; Merchant. || Clerk;
Franklin.
B. Lansdowne. Man of Law; Squire. || Clerk; Merchant;
Franklin.
C. Harleian. Man of Law. || Clerk ; Merchant ; Squire ;
Franklin.
Note how the Clerk was regularly advanced, till he took the
place of the Squire ; and how 3 (denoted by ||) was advanced, at the
1 I call it the " Petworth" type, because a considerable number of MSS.
exhibit a type very like it ; but the Petworth MS. itself has 7 a, b, after
, whic
Gamelyn, which is quite an unessential difference. We want to know about
2, 4, 5.
2 Furnivall gives this order ; a
order given on p. 172 is misprinted.
2, 4, 5.
2 Furnivall gives this order ; and so does Miss Hammond at p. 187. The
Classification of the Manuscripts. 47
same time, before him. Everything depends upon the position of
these three — the Squire, the Merchant, and the Franklin.
It may seem strange that the Man of Law is here included ; but
he is concerned with a most important link. I call this link " m " ;
and its history must be understood. Its form in HI. I call " m,"
It does not occur in the Hengwrt MS. ; but it must have been
written very early ; for it occurs in MS. Camb. Mm., in which the
parts of Group 7 are still separate, and in MS. Camb. Ii., which is a
" Petworth " MS. on which the Hengwrt MS. has exerted a strong
influence. It is found both in Petworth and Lansdowne, always as a
Man of Law-Squire link, and was, in fact, the old Squire's Prologue,
in which capacity it appears in all the old black-letter editions. It is
called " The Squiers Prolog " or " Prologus Armigeri " in at least
fifteen MSS., and is assigned to him in three more ; all being of the
Petworth or Lansdowne type. But in the Harleian MS. it is wrongly
assigned to the "Sompnour," as the scribe seems to have perceived;
for he left it incomplete. The odd thing is that this error has crept
into two other MSS., viz. Eawl. Misc. 1133 and Eoyal 17, where the
reading " Sayde the Sompnour" is merely absurd, because in both
instances the thoughtless scribes have not only written "the Prolog of
the Squier " a few lines above, but actually prefix it to the Squire's
Tale ! In one MS. only, viz. Arch. Selden, it happens to precede
the Shipman's Tale, and hence has been called the " Shipman's
Prologue."
The fact is, that when the Clerk was advanced next to the Man of
Law, this Prologue became absolutely useless, and was meant to have
been cancelled.
My notion is, that instead of deleting the whole Prologue, the
sole precaution taken was to erase the name of squier l in the phrase
" Sayde the squier," and to strike out the headline that described the
Prologue. Hence the scribe of HI. gives us no headline, and (seeing
perhaps the s of the erased squier) wrote sompnour at a venture,
though the following Tale was that of the Wife of Bath ! And
finally, finding that he was on a wrong tack, never completed the
Prologue, but left off in the middle of a sentence, and simply added —
"Here endeth the man of lawe his tale" ; and so cut the tale off from
all that followed. The casual assignment to the Shipman is against
all the evidence ; but I have allowed myself (as Tyrwhitt did) to make
believe that it is right, merely for the purpose of preserving the
1 Just as a came was erased in Hn. in Group I, line 1.
48 Canterbury Tales (Eight-text).
28 lines, which are certainly all genuine. As in the Six-text, they
are numbered Bl 163-1 190.
In the "edited" MSS. of the Ellesmere type, they are very
properly suppressed. Miss Hammond thinks that these MSS. may
have been edited by Chaucer himself, as they are edited with some
judgment. I would rather say that they may perhaps, if edited
in his lifetime, have been edited by friends who had received his
permission to do so. For they exhibit at least two blunders. They
give the component parts of the Monk's Tale in the wrong order; and
they quote the « Host-stanza " at the end of the Clerk's Tale,
which the Lansdowne and Harleian scribes had rightly suppressed.1
As regards the order of the Tales, the Harleian MS. is alone
authentic.
The history of this link " m " can be thus tabulated. Hengwrt : —
2. 5a. Pt:— 2m5a. L :— 2m5a. HI :— 2m || 4o. E:— 2 || 4a.
The advancement of Group 3 to the position between 2 and 4a
most effectively severed the Man of Law from the Clerk.
In the Hengwrt MS., the Squire, Merchant, and Franklin are
all linked together; but the Clerk is unconnected with anything,
though it has the ineffective Host-stanza at the end. We must
consider each link separately. The old Squire-Merchant link I have
called "n." It consists of F673-708, used in a different way from
that in my edition ; i. e. different as regards the Franklin only.
Hence, in Hn., F675, we duly find " Quod the Marchant," which
is quite right and consistent. So also in Pt., which keeps the same
order, viz. Squire-n-Merchant. But in L., Group 3 was advanced to
follow the Squire, which severed the connexion altogether ; and con
sequently "n" disappears from it. But it was not done with; being
revived in HI., in the form which I call "n," as a Squire-Franklin
link. This was extremely easy, because all that was required was
to retain the word Squire, but to alter " Marchant " to " Fraunkeleyn."
There wTas a slight difficulty as to the pronunciation, because the latter
word was properly trisyllabic, as in A331. However, we may read it
in F675 as "Fraunk'leyn," and so get it into the line.
F696 had still to be dealt with ; but luckily, in Hn. and Pt. it was
a "nine-syllable" line, viz. '•' What, / Marchaunt ! pardee sire, wel
thou west." Hence it was easily altered to " What, Frank/eleyn,"
&c. ; and all was complete as regarded F696. There still remained
1 Exactly the same may be said of the Nun's Priest's Epilogue, which is
genuine enough, but quite useless as a link. It appears in but few MSS.
Classification of the Manuscripts. 49
F699; where Hn. Pt. have— " That knowe I wel, sire, quod the
Marchaunt, certeyn." ! Here the alteration to " quod the Fraunke-
leyn " not only kept the rime, but made a better line. This change
from " Squire-n-Merchant " to Squire-rc-Franklin could not be bettered ;
so it remained permanently, and is found in MSS. of the Ellesmere type.
Surely this was Chaucer's own doing. Observe how common
" Squire-n-Merchaiit " is; it occurs in a large number of MSS.
and in the black-letter editions ; so it was not merely casual.
We have next to consider the Merchant-Franklin combination.
They are closely connected in Hn. by a link which I call " x, y." It
is necessary to use a double symbol, because the Six-text places " x "
in one group, and " y " in another, though " x " and " y " are in
contact under all circumstances. The former is now counted as
E2419-2440, and the latter Fl-8 ; and Hn. has the combination
Merchant-xy-Franklin ; to which the readings correspond. Thus
E2425 (Hn.) has '" Marchauntes tale"; and Fl,2 appear in Hn. as
follows : —
" Sire Frankeleyn, com neer, if it your wiile be,
And sey us a tale; for certes ye."
But this is obviously a scribal blunder; the words "com neer"
are in the wrong line, making Fl too long, and F2 too short.2 Of
course what was meant is : —
" Sire Frankelyn, if it your wille be,
Com neer, and sey us a tale ; for certes ye."
Thus understood, the combination Marchaunt-xy-Franklin, as
given in Hn., is quite right.
But in the next arrangement, in Pt., Group 3 was placed in the
position shown by || ; i. e. immediately after the Merchant, which at
once severed him from the Franklin, destroyed the value of the
link "xy," which could only follow the Merchant (see E2420),
and left the Franklin isolated. It was then easy to alter his
position, and to substitute Clerk-Franklin for Franklin-Clerk.
Besides the now useless link " xy," it must be remembered that the
Clerk's Tale had an appendage of its own in the Host-stanza, or " h,"
which linked on to nothing. It is clear that Chaucer's intention, at
this time, was to drop both connexions, at any rate for the moment,
and to suppress both " h " and " xy," leaving the Clerk and Franklin
1 Certayn or certeyn occurs thus as a tag more than 30 times ; see E646, 694,
960, 1417, &c.
2 It looks as if the line had been dictated piecemeal— Sire Frankeleyn— if it
your wille be— com neer— if it your wille be (repeated)— And sey, &c.
EIGHT-TEXT ED. CANT. TALES. E
50 Canterbury Tales (Eight-texf).
unconnected, and trusting to a future day for connecting them. But
the scribes did not understand this, and persisted in using up the old
material in various ways; the chief difficulty being in E2420-5.
Thus the scribe of Mm. dropped "h," and wrote in "xy" unaltered
(perhaps the easiest thing to do), as if the Tale preceding the
Franklin's was the Merchant's, and not the Clerk's. MS. Pt. also
dropped " h," and altered E2425 to-—" By mony ensamples it proueth
wel." This answers the purpose after a poor sort, as if it referred to
the " archewyues " in Chaucer's Envoy ; but the line is short. Had
it been "By mony ensamples olde it proueth wel," it would have
scanned. MS. Harl. 7333 kept "h," but dropped " xy " ; which did
no great harm. MS. Trin. E. 3. 15 did the same, but also dropped
F709-728, which was a mistake. MS. Harl. 1758 kept "h," and
dropped part of " xy," viz. the first eight lines (E241 9-2426). This-
also servecl the purpose, but was inartistic, because it required a
sudden change from the stanza to couplets; a difficulty which the
scribe courageously surmounted by turning E2427-F6 into a couple
of seven-line stanzas, with the daring rime of can with am ! MSS.
Royal 18 and Barlow 20 do precisely the same ! MS. Sloane 1685
and the Lichfield MS. not only drop "h," but also E2419-26, which
leaves a gap. Briefly, they all alike make a sad mess of it. The
attempt to preserve " xy " was beyond their powers ; it was achieved
in quite a different way, by the skilful hand of the author.
We now come to the Lansdowne type. Probably owing to the
difficulty caused by this dropping of " xy," Chaucer seems to have
made rather a bold move, which solved that difficulty altogether. He
advanced Group 3 still further, and reversed the order of the Clerk
and Merchant, so that he now had to deal with the order Clerk,
Merchant, Franklin. He provided no link for the Clerk and Merchant,
but left them in mere contact. What he did with regard to the
Merchant and Franklin, we have no very clear evidence, owing to the
unlucky chance that the three MSS. of this type, viz. the Corpus
MS., the Lansdowne MS., and Sloane 1686, all abruptly terminate
the Merchant's Tale with E2318, and leave a large gap between this
line and F9. Thus the chief features of this type are the preservation
of " m," the Man of Law-Squire link, and the total absence of any
link between either the Clerk and the Merchant or the Merchant and
the Franklin.1
1 It is worth saying that, although B3961-80 (not in Hn., but found in L.)
are not in Cp. and Sloane 1686, they are found in several MSS. of the Petworth
type, and must have been added rather early.
Classification of the Manuscripts. 51
We now come to a consideration of the Harleian MS. Here
Group 3 was again advanced yet another step, forming a barrier
between the Man of Law and the Clerk, which rendered " m,"
the Man of Law-Squire-link useless, so that it ought to have been
cancelled. But it wrongly appears, here, for the last time, in the
inefficient form " m " ; after which we hear no more of it in the latest
or "Ellesmere" scheme, although the 28 lines that compose it are all
genuine. One obvious result of this was to render the position of
the Squire, for the first time, entirely free.1 Not only so, but the
Clerk, Merchant, and Franklin, were all left free in L. (as far as we
know) ; and it was easy to arrange all four, the Squire, the Clerk, the
Merchant, and the Franklin, in any position. But Chaucer had
already tried prefixing the Merchant to the Franklin, as in Hn. ; the
Clerk to the Franklin, as in Pt. ; a second prefixing of the Merchant
to the Franklin, as in L. ; but was still dissatisfied. But now it was
possible to prefix the Squire to the Franklin, which pleased him. The
Man of Law was now isolated, and so remained. The other four
Tales were in the order : — Clerk, Merchant, Squire, Franklin ; and all
were free. There were thus no less than three links to be supplied,
and this was done with such consummate skill that the hand of the
master is obviously at work. Let us see how he did it.
First, as to the Clerk and Merchant. This combination was a new
thing, and he rose to the occasion. Taking advantage of the fact that
his Clerk's Tale ended with the line : — " And let him care and wepe,
and wring, and wayle," he wrote an entirely new Prologue, com
mencing with E1213 — " Wepyng and wailyng, care and other sorwe,"
and continued for 32 lines, to El 244; thus providing the firmest and
most satisfactory link in the whole set. This link I call " z." It is
amusing to see how much younger the Merchant has grown ! In Hn.
and Pt., he had a grown-up son (Six- text, F682-694); but now he
has only been married for two months (El 234). This (as Tyrwhitt
pointed out) brought the unfortunate Thynne to grief ; for, finding in
his MSS. both the old Merchant's Prologue (F673-708) and the new
one (E1213-44), he impartially printed them both; so that the very
man who, on Fol. xxxii, back, tells us that he had a grown-up son,
goes on to explain, on Fol. xxxiii, that he had only been two months
1 In Hn. it is linked to the Merchant, which it precedes. In Pt. it is linked
both to the Man of Law (before it) and the Merchant (behind it). In L. it is free
from the Merchant ; and finally, in HI. it is freed (for the first time) from all old
connexions, and could therefore obtain new ones.
52 Canterbury Tales (Eight-text}.
married ! This is what comes of not understanding how the various
links were from time to time altered.
Chaucer had next to connect the Merchant with the Squire ; and
this he does with great ease. He simply takes the old Merchant-
Franklin link in Hn., viz. x, y (E2419-F8); leaves the allusion to
the Merchant in E2425 ; and puts Squire for Franklin in Fl.
The correct reading of Fl, 2, as shown at p. 49, had once been : —
Sir Fraunkeleyn, if it your wille be,
Com neer, and sey us a tale ; for certes ye.
But this had appeared, incorrectly, in Hn. and Pt., in the extra
ordinary form : —
Sir Fraunkeleyn, com neer, if it your wille be,
And sey vs a tale ; for certes ye.
And sure enough, the scribe of HI. got hold of this, and produced
this extraordinary result : —
Sir Squier, com forth, if it your wille be,
And say vs a tale ; for certes ye.
The former line is all very well, but the second is surely wrong ; and
accordingly, we find in Cm. E. Dd., the well-known and acceptable
lines : —
Squier, com neer, if it your wille be,
And sey somwhat of loue ; for certes ye.
This is yet one more example of what I have said already, viz. that an
error which occurs in Hn. is very apt to persist in many copies.
Lastly, Chaucer had to connect the Squire with the Franklin ;
and this he likewise did with perfect ease. He simply revived the
old Squire-Merchant link, viz. " n," or F673-708, which had been
used in Hn. and Pt. The reference to the Squire in F673 was all
right ; but those to the Merchant required alteration. He had to deal
with F675, 696, and 699. But I have already had occasion to
explain this, in tracing the history of the link "n" at p. 48 ; so that
I need not repeat what I have there said. Perhaps it is proper to
add that the Harleian MS. is here defective ; but it is fair to assume
that it had, at this point, the same readings as are found in Cm., Dd.,
and E., and the other late MSS. in which the Franklin follows the
Squire.
I hope I have now made the matter clear, though it is rather a
difficult maze to follow. The point is, shortly, that there are no less
than four arrangements of the tales in Groups E and F. This is
Classification of the Manuscripts. 53
simple fad, whatever the explanation may he. All that I do is to
place these arrangements in chronological order, and to show what
was done from time to time. The four arrangements are : (1) Squire,
Merchant, Franklin, Clerk; (2) Squire, Merchant, Clerk, Franklin;
(3) Squire, Clerk, Merchant, Franklin; (4) Clerk, Merchant, Squire,
Franklin. And the last of these was final. Those who do not accept
my explanation will still have to deal with the orders (1), (2), and
(3) ; and must explain them how they can. I think they will fail.
Even these four arrangements do not quite exhaust the list, because
(5) the Ellesmere MS. slightly alters the positions of other Tales, by
carrying Group G lower down; though it agrees with HI. in the
arrangement of Groups E and F. If we leave MS. Hn., or order (1),
out of consideration, there still remain (2), (3), (4), and (5).
These last four agree with those which, in my own edition (iv.
xxiii), are called types D, C, B, and A respectively; where the
arrangement A, B, C, D gives the order as reckoned back from the
final one, and so reverses the chronology. All that I now do is to
place Hn. by itself, at any rate as regards the Tales in Groups E
and F, because it gives the order Squire -Merchant -Franklin-
Clerk, which occurs (perhaps) nowhere else. This order is by no
means casual, and involves much more besides ; for the MS. welds
Squire-Merchant-Franklin into one group, by the use of links "n"
and " xy " ; and that is why I take it seriously. The alternative is to
ignore it, and to say that its arrangement is due to contamination of
(possibly) (2) and (3). I will not now contest the point. For the
sake of argument, let it be granted. It still remains to explain (2),
(3), and (4) ; which is what I claim to have done. If my account of
Hn. is wrong, it has only to be disregarded ; and the remainder still
stands. Then, going back to my old arrangement of the MSS. into
classes A, B, C, D, the Tale of the Man of Law and of those in
Groups E and F stand thus, inserting the Units, and denoting Group
D by | 3 |, because it always forms a barrier : —
D. — Petworth. — Man of Law-m-Squire-n-Merchant | 3 | Clerk-
xy-Franklin.
C. — Lansdowne. — Man of Law-m-Squire | 3 | Clerk Merchant
Franklin.
B.— Harleian.— Man of Law-ra | 3 | Clerk-z-Merchant-icy-
Squire-[w]-Franklin.
A.— Ellesmere.— Man of Law | 3 | Clerk -z-Merchant-^-
Squire-w-Franklin.
54 Canterbury Tales (Eight-text}.
Here B and A coincide (in excepted). The link n is the same as n
(F6 7 3-708), but has changes in the names referred to; and xy is
similarly altered from xy (E2419-F8). My belief is that this employ
ment of old links with new functions or relations is Chaucer's own
doing, though some scribes certainly wrote (or preserved) links on their
own account. Even if it is not so, the facts are unaltered.
Briefly, I call attention to the following facts, as appearing from
the evidence of many MSS. (leaving out Hn. for the present).
(a) The same set of lines (m = Bl 163-1 190) appears both as an
old Squire's Prologue and (wrongly) as m (in HI.) as a useless pro
logue. In MS. Arch. Selden it appears as a Shipman's Prologue. It
appears in the Preface to the Six-text as " Man of Law's End-link, Eeal
Shipman's Prologue." I should call it " Man of Law's Endlink or
Eeal Squire's Prologue ; Sham Shipman's Prologue." It is a Squire's
Prologue in MSS. Corpus, Sloane 1686, and Lansdowne (of the
Lansdowne type), and Sloane 1685, Barlow, Eoyal 18, Harl. 1758,
Lichfield, Harl. 7333, Laud 739, Eawl. Misc. 1133, Hatton, Trin.
E. 3. 15, li, Eawl Poet. HI, Eoyal 17, Eawl. Poet. 149, Mm, Helm-
ingham, Trin. E. 3. 3. To say that these twenty MSS. are all wrong
is mere assumption, and explains nothing. In my view they are correct.
(&) The same set of lines (n = F673-708) appears both as an
old Merchant's Prologue and as n (in E.) as Words of the Franklin.
It is called in the Preface to the Six-text the " Squire's Endlink as
Merchant's Prologue." I should call it " Merchant's Prologue," after
wards used as "Words of the Franklin." However, it is always a
Squire's Endlink. It occurs as a Merchant's Prologue in MSS.
Hatton, Barlow, Sloane 1685, Eawl. Misc. 1133, Eawl. Poet. 149,
Laud 739, Harl. 1758, Eoyal 18, Lichfield, Mm., li. [Some MSS.,
as Harl. 7333, Eoyal 17, Trin. E. 3. 15, employ «z" as a Merchant's
Prologue, which is quite inappropriate; for that is the Clerk-
Merchant link, and the Clerk does not precede the Merchant in those
MSS. MS. Eawl. Poet. 141, which is imperfect, does not contain
the Merchant at all] Here are eleven MSS., all belonging to the
same set as those enumerated in (a), employing F673-708 as an old
Merchant's Prologue. That there was once an old Merchant's Pro
logue, as distinguished from the present one (z), is further rendered
probable by the absence of " z," i. e. the standard Merchant's Prologue,
from many of the MSS. that give us " n." I do not call the use of
these lines as a Merchant's Prologue a wrong use, but rather an old
use. I take it to be quite correct.
Classification of the Manuscripts. 55
(c) The same set of lines (xy = E2419-F8) which now con
stitute the Merchant's End-link and the Squire's Head-link (or, to give
the true name, the Squire's Prologue x) were once used as a Franklin's
Prologue. As to this, the MSS. vary, according to the Tale which
precedes the Franklin. When used as a Clerk-Franklin link, as in Pt.
and many other such MSS., their shape is variable and quite unsatis
factory, as has already been said (p. 50). The only MSS. of an old type
which employ them in a wholly satisfactory and perfectly natural
manner are the Hengwrt MS. and its follower Ii.,2 in both of which
the Merchant is followed by the Franklin. And this is one reason
why I think the order of the Tales in MS. Hn. should receive
attention.
ON THE WORD AEIVEE IN THE GENERAL PROLOGUE.
Let me first quote the following admirable summary. " On line
60' of the General Prologue, ariue or armee. The MS. readings are
not yet collected. Caxton ed. I. has aryue, ed. II. arme ; Thynne
has armye ; Speght army ; Urry army ; Morell aryve ; Tyrwhitt
armee. Skeat prints aryve, with condemnation of armee as giving no
good sense, and due probably to the misreading of the spelling ariue\e\.
Skeat is criticised by Fliigel (Jour. Gc. Phil. i. 124-5, Anglia xxiv.
443-6), with citations for armee, in the sense of ' military expedition.'
The New Eng. Diet, takes Tyrwhitt's reading, not Skeat's, for annota
tion." — Hammond, Chaucer, p. 273.
There is more to be said. I think the true reading is arivee, after
all; or it may be spelt aryue. My original reasons were these.
(1) If we are to take the durior lectio, surely that is ariuee. (2) It
scans better, for the word noble should have its e elided before a
vowel, as in 1. 708 : — " He was in chirche a nobl' ecclesiaste." So
here : — " At many a nobl' ariuee hadd' he bee." (3) It is gram
matical ; we can say " he had been at many a noble disembarkation."
And surely that is graphic. Whereas we can hardly say "he had
been" at many a noble army, or military expedition. Surely we should
say with. And this weighs with me most.
And now for more.
I am told by Mr. Aldis Wright that the reading in MS. Hengwrt
(to which I now always look first) is really ariuee, and that armee in
1 In E., before E2149, is the rubric— "The Prologe of the Squieres Tale";
and in Del. — " Incipit prologus Armigeri."
2 See the description of MS. Ii. 3. 26 at p. 61.
56 Canterbury Tales (Eight-texf).
the •" Six-text edition" is due to a mistake.1 The Cambridge MS.
has arytte, which cannot be mistaken. And now that Dd. is printed,
we find that it has lost lines 1-252 ; but Dr. Furnivall fills up the
gap from MS. Egerton 2726, which has : — " At many a noble aryve
hade he be " ; actually with v, not u. All these facts require explana
tion rather than dismissal. If aryvee is not in the N. E. D., it may
Some day be in the Supplement.
For the form is quite legitimate. Cotgrave has : — " Arrivee, f. an
arrivall, accesse, abboard, or comming to." Minsheu's Spanish
Dictionary (1623)2 has: — " Arribada, or Arrivada, f. an arriuall,
comming to shoare, or to harborough, ascending, or mounting vpward."
The chief trouble is that the 0. F. word does not appear in Godefroy's
Old French Dictionary ; in which, unfortunately, the deficiencies are
numerous. Perhaps it was commoner in Anglo-French.
At any rate, I have found an example in print. In the Black
Book of the Admiralty (Eecord Series), vol. i. p. 427, I find:—
" Item, les choses desusditz deyvent estre cryes et publics plusieurs
foitz et sovent par totes niefs [ships] de ceste presente aryve, et en
chescon des escheles [companies], afin qe chescon ait conisaunce et
plenier memoire de yceux." The preceding sentence helps us a
little : — " Item, qil soit crie et defendu, qe nul ne desende [that no
one may disembark] sanz le comandement de maistre." From MS.
Cotton, Claudius E. viii ; fol. 249.
Again, at p. 433 : — " lesditz maistres et quatre compaignons les
puissent prendre a leur arrivement a terre " [leaving the ship]. From
MS. Sloane 2423 (A.D. 1373); fol. 3, back.
The two first " Items " are from Ordinances given to the French
fleet in 1338 ; before Chaucer was born.
It is probably by mere luck that this form has been preserved.
Had it been spelt with i instead of y, the chances are that it would
have been written anne by a scribe. In the same volume I find, at
p. 438 : — " [est] advenu que nostredit admiral a commis soubz luy
en aucuns portz ou es armees ou entreprinses qui se sont faictes. par
ladite mer gens de legere facon." The reference is rather to a naval
than a military enterprise.
1 The librarian of the National Library of Wales (where the MS. now is) tells
me that "the stroke over the i is very faint ; but as seen with a magnifying-glass
I think there is undoubtedly a mark present."
2 I was once reproved for quoting this book by some one who wrote from the
United States to tell me that I ought to have known that there is no such
volume in existence. I quote from the copy in my own possession.
General Prologue, I 60. Allusions to Persius. 57
Again, at p. 439 : — " es prouffitz communs de la guerre, tant a nos
armees comme aultrement " \ where armees may be meant. I do not
pretend to decide.
At p. 441 : — " quant au regard des navires [various reading,
armees] et entreprinses, qui se ferout [feront ?] par ladite mer, nostredit
admiral demourra en icelles armees chef." Was it usual for an
admiral to be chief in armies?
Again, on the same page : — " Et sil advenoit aucun cas, es armees
et entreprinses, ou nostredit admiral ou son lieutenant seroient en
personne."
I desire to add that I am much obliged to Dr. Furnivall for
verifying all these passages at the British Museum. He reports that
they are printed exactly as they are written.
THE TWO ALLUSIONS TO PERSIUS.
It is known that E721 was suggested by 1. 2 of the Prologue to
the Satires of Persius. I now find that Chaucer was indebted to
another passage in the same short Prologue for the remarkable form
Pegasee (for Pegaseus) in F207 ; where a marginal note in the
Ellesmere MS. has equus Pegaseus. I have noted (Chaucer's Works,
v. 376) that the poet was thinking of the adjectival form Pegaseus
rather than of the substantival form Pegasus. This is not quite right,
but very nearly so. For a side-note in the Cambridge MS. Dd. tells
us a little more. It runs thus : id est, "equus Pegaseus, Percius 4to."
Here either "4to" is an error for " 14o," or it is short for "quatuor
decimo," as the allusion is obviously to the 14th line of the same
Prologue, viz. "Cantare credas Pegaseium nectar"; the only allusion
to Pegasus that occurs in Persius, and only twelve lines distant from
the line alluded to above. This shows that Chaucer evolved the form
Pegaseus as a sb. from the adjectival form Pegaseius. — Notes and
Queries, 10 S. xii. 6.
A METHOD OF DESCRIBING A CHAUCER MS.
By making use of my article on " The Evolution of the Tales "
and the Comparative Tables (and Notes) given above, it is possible to
give, in a compendious form, a sufficiently minute account of a Chaucer
MS. that will go a long way towards determining its character. I give
such an account below.
58 Canterbury Tales (Eight-text).
MS. Mm. 2. 5 ix THE CAMBRIDGE UNIVERSITY LIBRARY.
A new symbol is required to denote the Spurious Shipman's
Prologue of 12 lines. I denote it by " s*." And another, to denote
the spurious Canon's Yeoman-Doctor link ; viz. " t*."
My list of Symbols is as follows.
Tales. — 1. Prologue to Cook. 2. Man of Law. 3. Wife to
Sompnour. 4a. Clerk. 45. Merchant. 5a. Squire. 55. Franklin.
6. Doctor, Pardoner, la. Shipman. 75. Prioress. 7c. Sir Thopas.
Id. Melibee. 7e. Monk. 7/. Nun's Priest. Sa. Second Nun. 85.
Canon's Yeoman. 9a. Manciple. 95. Parson.
Special Links.— m. Old Squire's Prologue (B1163-1190).
m. nameless Prologue (the same lines) ; used as a Shipman's
Prologue in MS. Arch. Seldeii only.
n. Old Merchant's Prologue (F6 73-708). n. Words of the
Franklin (the same lines).
h. Host-stanza (at end of Clerk) ; not used as a link. h*. The
same, in a more or less spurious form ; used as a link to the
Franklin.
x, y. Old Franklin's Prologue (E241 9-2440; and Fl-8) ; as in
Hn.
x, y. The same lines, as used in HI. and E.
x*, y*. The same lines, in a spurious or debased form.
z. Clerk-Merchant link (El 2 13-1 244). See HI.
e. Nun's Priest's Epilogue (B4637-4652).
Gam. Tale of Gamelyn. s*. Spurious Shipman's Prologue, t*.
Spurious link between the Canon's Yeoman and the Doctor.
The contents of MS. Mm. 2. 5 may be thus expressed :
1. Gam. s* 7 a, b, e (ending with doude). 2 m 5a n 45. 3. 4a
x, y, 55. 7/ (with complete Prologue). 8 t* 6. 7c, d. 9.
Here the full stops mark where the Groups end.
I add Notes, giving some more minute particulars, gleaned from a
partial examination. Such a note as " 637-8" means that lines 637
and 638 are really found in the MS. ; on the other hand, such a note
as [3155-6] means that the lines within square brackets are omitted
in it.
1. (Group A) Prologue to Cook. The first folio is lost ; [1-90].
[252 5, c]. 637-8; 1250; 2681-2; 2779-82; 3045. [3155-6].
[3721-2]. 4375-6. Gamelyn.
s* 7 a, 5, e. (Group B). Shipman, Prioress, Monk, s* = spurious
MS. Mm. 2. 5 in the Cambridge University Library. 59
Shipman's Prol. of 12 lines ; "begins — No we frendes saide oure oste so
dere ; ends — beganne. Printed in Pref. to Six-text, col. 29. [1220-3].
1461-2 : An hundred fraunkes for a wyke or tweye For certayn
bestes J>at I moste bye (sic). 1467-8 ; like Pt. [1881-3078]. 3197-
3204; 3611; 3764. Monk's Tale en Js— cloude ; 3956.
2. (Group B). Man of Law. 1-1162 ; joined to Squire-Merchant,
m 5a n 46. (Group P>, 1163-1190; F9-672, 673-708; E1245-
2418). Squire-Merchant. B1179 ; Seide )>e Squier, &c. ; see Pref.
to Six-text, col. 21. F671-2. F675 ; Qwod pe Marchante ; see
Pref. to Six-text, col. 43. [F679, 680]. El 305-6 ; And 316: pou take
a wiffe "pat to ]>e is vntrewe Full ofte tyme it scball ]>Q rewe ; cf. Pref.
to Six-text, col. 72. E1645 ; ends— a Hire here (cf. 1647). [E1646-
7]. El 7 7 7-8 ; As ]>at sche bare daunsynge in hire honde So fressh
scho was and J>erto so likande. E1815 ; ends — wise sone. [E1816].
El 8 18* (after 1818) ; So hasted Januarie and moste be done.
[E1927-8]. E2355 ; ends—^i his sight a-^eyne. [E2356-7].
3. (Group D). Wife to Sompnour. 45 follows 44. 222 ; Knyves
and ringes and pwrses wel faire. 382 ; And eke pat he hadde sus-
peciown and lelousnes. [575-584]. [609-612]. [619-626], [717-
720]. 1295-6 ; And eke— tyme. 1307-8 ; wrongly placed after
1294. [1731]. 1732 ; followed by— To kepe }owe from )>e peynes of
fendes blake. 1872-4 ; only one line — Than burell folke in richesse
and wynnynges. [2004 6, c; 2012 6, c; 2048 6, c].
4a x, y, 56. (El-1212, 2419-2440, Fl-8, 709-1624). Clerk-
Franklin. Ell 7; For of wedloke cometh grete emprise (!). E143.
E173. El 195-1 200 follow E1212. E2425 ; By J>is marchaundes (I)1
tale it proueth wele ; see Pref. to Six-text, col. 50. [E2433]. Fl, 2.
Sir fraunkeleyn com nere 2 $if it ^oure wille be And saye us a tale, for
certes 30. [F4]. F9 18 precedes F9 17. [1147-8]. [1191-6]. 1266
precedes 1265. [1433-4]. [1455-6]. [1493-8]. 1530 precedes
1529, and runs — Certes me thynketh it were right grete routhe.
1556-7. 1567-8.
7/. (Group 3957-4636). Nun's Priest. 3961-80. 4112-4.
4136-7. [4233-8]. [4425-4486] ; 62 lines lost. [4637-4652].
8. (Group G). Second Nun and Canon's Yeoman. 73-4.
1 This most important reading shows that the reference to the Merchant in
E2425 persisted even after, in the Petvvorth scheme, E2419-2440 was made to
follow the Clerk. This oversight was corrected, in the Harleian aud Ellesmere
schemes, by changing x y into x y, as I have explained at p. 52.
2 Com nere belongs to F2 ; read Sir Fraimkeleyn, Jtf it jourg wille be, Com
nere, and saye us a tale ; for certes Je.
GO Canterbury Tales (Eight-text}.
156-9. That 36 wille ^owe goue?*ne no-thyng amys He wille
sane and kepe tille 36 be dede For ^oure clene lyffynge and and (sic)
^our goodnes And brynge $owe streight in-to heuene blisse (spurious).
213-4; I leue al pis thynge pan pis I dar wele saye (only one line).
325. [326-337]. [432]. 563-5. 1238-9.
t*6. (Group C). Doctor- Pardoner. Spurious Prologue of 14
lines ; begins — When pat pis ^oman his tale ended hadde : ends —
Nowe goode men quod, he herkeneth euerychon. 287-296. [297-8],
299, 300. 305-6. 356 ; Aboute pe herte or els pe longe (!). 395-
761; on 4 misplaced leaves-, 486-577 precede 395-485; and 670-
761 precede 578-669. 486-490. 603-4.
7 c, d. (Group B : 1881-3078). Sir Thopas, Melibeus. [1995].
[2042-4]. 2105-7. [2108].
9. (Groups H, I). Manciple and Parson. [H47-52].
117; ends with degree (error for decree). 118; Who wille nowe
telle a tale lat see (!) ; cf. Pt. Cp. L.
As for Melibeus and the Parson's Tale, I have not examined
them.
The above description shows clearly that this MS. is of the
" Petworth " type, and agrees with the Petworth MS. in a large
number of cases where readings differ. Its variations from that MS.
are but few. For example, it preserves B3764, which Pt. omits. The
lines in D1872-4 do not agree with Pt., but with Cp. L. In E1777
it has bare (as in Cp. L.), not ladde (as in Pt.).
E2425 has Ipis marchaundes tale (as in Cp. L.) ; not like Pt. ;
though F675 agrees with Pt. in reading the marchante.
But it has one very striking variation from Pt. in the fact that
Group 7, instead of being complete and succeeding 6 (as often) is split
up into no less than three fragments, viz. 7 a, b, e near the beginning,
If (after 5b) and 7 c, d in the position in which we should expect to
find the whole group. We find something of the kind in Hn., which
has 7 e, / near the beginning, and 7 a, b, c, d after 6 ; and there are
other MSS. which split up this Group in other ways. Indeed the
Petworth MS. itself differs from nearly all MSS. of the usual
" Petworth " type by the fact that it places 7 a, b after Gamelyn,
but 7 c, d, e, f in the usual place. This suggests that the Tales in
that Group (as perhaps in others) were, in the first instance, written
separately from each other, and were collected afterwards. And
perhaps we may further conclude that both the Petworth MS. and Mm.
are (as regards type) earlier than the other MSS. of the "Petworth"
MS. Mm. 2. 5 in the Cambridge University Library. 61
type in which this Group is complete ; and perhaps Mm. is (as regards
type) the earlier of the two.
In her wonderfully comprehensive volume on "Chaucer," Miss
Hammond draws attention (p. 192) to the curious fact that this MS.
has some of its Tales numbered, but according to a system which does
not agree with the order in the MS. itself. " These numbers are : —
Reeve iii, Man of Law v, Wife of Bath vii, Summoner ix, Franklin
xii, Prioress xviii, Sir Thopas xix, Parson xxiv " ; and she refers us
to similar markings found in the Corpus MS. (Oxford).
In describing this Corpus MS. (at p. 188) she notes that " the
numbers in this Codex are : Reeve iii, Cook iv, Man of Law v,
Wife of Bath vii, Pardoner xvi, Shipman xvii, Sir Thopas xix,
Melibeus xx, Monk xxi, Nun's Priest xxii, Manciple xxiii,
Parson xxiv."
She notes further that the two systems agree, " although neither
MS. agrees in existing order with the numberings."
In this respect I think there is an oversight. The numberings are
simply due to the order of the Tales in the Corpus MS. itself ; so that
there is no mystery about them.
This can be shown from her own description of that MS. " Order
of Tales :— A Gamelyn Bl Fl D El E2 F2 G C B2 H I " • where
A represents Group A in the Six-text, &c. If we write this order
out, and number the Tales as we proceed, we have this result : —
Knight i, Miller ii, Reeve iii, Cook (including Gamelyn) iv, Man of
Law v, Squire vi, Wife vii, Friar viii, Summoner ix, Clerk x, Mer
chant xi, Franklin xii, Second Nun xiii, Canon's Yeoman xiv, Doctor
xv, Pardoner xvi, Shipman xvii, Prioresse xviii, Sir Thopas xix,
Melibeus xx, Monk xxi, Nun's Priest xxii, Manciple xxiii, Parson
xxiv. All that has happened is that the scribe of Mm. has taken his
numbers from a MS. of the Corpus-Lausdowne type.
MS. Ii. 3. 26 IN THE CAMBRIDGE UNIVERSITY LIBRARY.
Without taking notice of lost lines and peculiarities of reading,
the contents may be expressed by the following formula : —
1. Gam. 2 m 5a n 4& xy 5&. 3 . 4a. 8 . 6 . 7 . 9.
This only differs from the Petworth type by the remarkable position
of " 3. 4a," which follows " xy 5b " instead of preceding it. This is
like the position found in Hu. ; and what it means is easily seen.
62 Canterbury Tales (Eight-text}.
The chief feature of Hn. is that, immediately after "2," it has the
connected sequence " 5a n 46 xy 56." But in forming the Petworth
type, the first step was the writing of " m," the old Squire's Prologue
(B 1163-1190), which joined 2 to 5a, giving "2m 5a." The scribe
of li. has used this "m" to good effect, and has thus made out the
long sequence " 2 m 5a n 46 xy 56," i.e. a Clerk-Squire-Merchant-
Franklin group. The Petworth type cut off " xy 5b " from the end of
this, and placed it after " 3. 4a " • an arrangement which (with con
siderable modifications) persisted ever after. Hence li. 3. 26 presents
a curious variety of the Petworth type, on which the Hengwrt MS.
has exerted a remarkable influence.
POSTSCEIPT.
I HAVE tried to explain that it is quite hopeless to comprehend the
order of the Tales in any MS. unless we carefully study the connecting
links that intervene between them. It is not enough to state whether
they are or are not connected. We must know likewise the nature of
that connexion.
I have shown above, for example, that the link "xy" occurs also
as " xy" the difference being that one of the lines is altered. It would
make the matter still clearer if I had gone further, and denoted the
use of "xy" in the Petworth MS. by the symbol "x* y*"; using
the asterisk, as elsewhere, to show that the lines, as they occur in that
MS., are practically corrupt, inasmuch as they have riot been properly
altered so as to make them suit their new position ; whereas in the
forms "xy" and "xy" they are correctly adapted to their surround
ings. In order to show this in such a way as to make it fairly obvious,
it is necessary to give the actual readings of the variable lines ; which
I now do.
(1) xy. This is the connecting link between the Merchant and
Franklin in Hn.
The complete passage is as follows ; where I denote the invariable
lines by numbers only.
THE MERCHANT'S TALE : E1245-2418. After which— U Here is
ended the Marchantes tale / of lanuarie. -ff Here folwen the Wordes
of the Worthy Hoost to the Frankeleyn.
Postscript. 63
E2419. Ey goddes mercy / seyde oure Hoost tho
Now swich a wyf / I prey god kepe me fro
Lo whiche sleightes / and subtiltees
In wommen ben / for ay as bisy as bees
Ben they / vs sely men for to deceyue
And from a sooth / euere wol they weyue
2425. By this Marchantes tale / it preueth weel
(Here follows E2426-2440 ; and next—}
Fl. Sire Frankeleyn / com neer if it your wille be
And sey vs a tale / for certes ye
(Here follows F3-8.)
11 Explicit. H Here bigynneth the Frankeleyns tale.
THE FRANKLIN'S TALE: F709-1624.
NOTE. This is all consistent. The reference in E2420 is to the
wife of Januarie ; and the preceding Merchant's Tale is correctly
alluded to in E2425. I have explained above the (unessential) mis
reading in Fl, 2; viz. that "com neer" belongs to F2, not to Fl ;
see p. 49.
(2) x* y*. This is the " manipulated " connecting link between the
Clerk and Franklin in Pt. It should have been rewritten, to suit the
preceding Clerk ; but the work of adaptation seems to have been left
to the scribes. The complete passage is as follows —
THE CLERK'S TALE: El-1194: 1201-12: 1195-1200. After
which — II The prologe of the Fraunkeleyn.
E2419. Ey goddis mercy seide oure hoost thoo
Alle euel wyues god kepe vs fro
For mony scleightes and sub[t]ilites
Bene in wommen that bene euere bisy as bees
Vs foly men forto desceyue
For from the soth euere wil thei weive
2425. By mony ensamples it proueth well
(Here follows E2426-2440; and next—)
Fl. Sir Frannkeleyne come ner if it your wille be
And say us a tale for sertys ye
(Here follows F3-8.)
IT Here endeth the prologe of the Fraunkeleyn. And here
bygynneth the Fraunkeleyns tale.
THE FRANKLIN'S TALE: F709-1624.
64 Canterbury Tales (Eight-text).
NOTE. This is consistent, at the end, as regards the Franklin.
But the placing of this Tale next to the Clerk (apparently adopted as
an experiment only, for it was altered afterwards) required an adapta
tion which it did not fully receive ; though we may well suppose
Alle euel wyues to refer to the last stanza in the Clerk's Tale, which
(in Pt.) contained the lines El 195 and El 197 — Ye Archeivi/es, &c.,
And sdendere wyves, &c. Note also the alteration in E2425, which
gets rid of the allusion to the Merchant's Tale, since it no longer
preceded the Franklin.1
(3) xy. This is the connecting link in HI. between the Merchant
and Squire ; skilfully and correctly adapted.
THE CLERK'S TALE: El-1212 ; new Clerk-Merchant link : E1213-
1244; THE MERCHANT'S TALE: E1245-2418. After which— Here
endith the marchauntes tale.
E2419. Ey goddes mercy sayd our hoste tho
Now such a wyf I pray god keep me fro
Lo whiche sleightes and subtilitees
In wommen ben for ay as busy as bees
Ben thay vs seely men for to desceyue
And from a soth . euer wol thay weyue
2425. By this marchaundes tale it proueth wel
(Here follows E2426-2440; and next—)
Fl. Sir Squier com forth if that your wille be
And say vs a tale for certes ye
(Here follows F3-8.)
11 Her endith the prologe. 1T And her bygynneth the Squyeres tale.
THE SQUIRE'S TALE: F9-616 (gap in MS.).
NOTE. This is cleverly done. The author has here the same read
ings as at first with respect to the Merchant ; and simply alters Fl so
as to suit the Squire. But it is characteristic that (if the scribe is
right) he did not at the moment amend F2 ; let us hope the scribe is
wrong ! In any case, it was well mended at last ; see p. 52.
If each of the other links, viz. " m," " n," and " z," be carefully
studied at length in the same way, they will all become intelligible at
last All depends upon the patience of the reader.
1 The readings in Cp. L. cannot be determined, as these MSS. are defective
just at this very point.
PR Chaucer Society, London
1901 cPublications_i
A3
ser. 2
no. 43
CIRCULATE AS MONOGRAPH
PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET
UNIVERSITY OF TORONTO LIBRARY
Live Music Archive
Librivox Free Audio
Metropolitan Museum
Cleveland Museum of Art
Internet Arcade
Console Living Room
Books to Borrow
Open Library
TV News
Understanding 9/11