Gest of Robyn Hode, A [Child 117] --- Part 02
DESCRIPTION: Continuation of the notes to "A Gest of Robyn Hode" [Child 117]. Entry continues in "A Gest of Robyn Hode" [Child 117] --- Part 03 (File Number C117B)
Last updated in version 2.6
NOTES [10870 words]: INTRODUCTION
It is a rare man that can make a name for himself that lasts across the years. It is still rarer for a name to make a man. Yet that is what happened with Robin Hood.
Dobson/Taylor, p. ix, sneer a little at the ballad scholars who have worked on this story, lumping them with "local enthusiast[s]" and "writer[s] of children's stories." Pollard, p. ix, notes that in recent years there has been an upsurge in Robin Hood scholarship, but most of it sociological -- a study of popular protest. Pollard wishes "to reclaim some of the ground for the historian." And this note -- exceptionally long as it is -- is an attempt to reclaim a bit of it for the folklorist also.
It appears that by 1250 at the latest, the name "Robin Hood," or some close variant ("Robehod," "Rabunhod") was commonly used as a name for un-apprehended prisoners. Knight/Ohlgren, p. 21, mentions a Robert Hod in 1226 who was a fugitive and whose property was given to St. Peter's of York . Baldwin, p. 51, tells of a Robert Hood of Cirencester who committed murder no later than 1216. Holt2, p. 188, lists William Lefevre of Berkshire, who was active 1261-1262, and who came to be known as "William Robehod." Baldwin, p. 52, probably following Holt2, p. 187, says there was a "distinct concentration" of people with the surname "Robinhood" in southeast England in the late thirteenth century. Child notes many more people with the name during the fourteenth century.
There is no reason to think these Robin Hoods were anything but common criminals, or that their name meant anything. As Pollard says on p. 187, "That there was an oulaw persona, possibly based on a person or persons who had once existed, called Robehod or variations of that name, known fairly widely by the 1260s, is not in doubt. But we do not know when or by whom stories about this persona were created, let alone when and by whom some of them were brought together as a narrative recognizably set in the early fourteenth century." What is certain is that, over the next two centuries, "Robehod" became "Robin Hood," the forest outlaw who defied the law and still managed to remain free for many years.
The legend has taken many twists over the years. Presumably it started with those robbers named Robehod. But it came to stand for more. The legend seems to have been at its best in the period from perhaps 1400 to 1500, when the "Gest" and other early ballads were written. It took a severe turn for the worse when Anthony Munday wrote a series of Robin Hood plays, and in the process converted Robin to a banished nobleman, gave him a wife, and otherwise bastardized what until then had been an excellent piece of folklore.
We cannot hope to find the "real" Robin Hood. Many scholars have tried to find an Original Robin over the years; none of their attempts has gained wide support, and most have convinced no one but the scholar himself. Many would agree with Mortimer's statement (p. 23) that "The Robin Hood of later legend was not a historical figure, but there were plenty of robbers and outlaws who were genuine enough." Yes, there are plenty of things named after Robin -- for instance, Wilson, p. 138, thinks the earliest significant record of Robin is the 1322 mention in the Monkbretton Chartulary of "The stone of Robin Hode," in Skelbroke in the West Riding of Yorkshire, near a site which later boasted a Robin Hood's Well. But the earlier records of outlaws named Robin Hood show that this stone is not a memorial of an early robber; it is a relic of a legend. Or, as Holt1, p. 106, declares, "the Robin Hood place names illustrate the spread of the legend, not the doings of the outlaw."
Holt1 (pp. 53-61) summarizes attempts to locate the original Robin; all have problems. Although all can be made to fit some part of the legend, they require ignoring other parts. Given the vast amount of effort expended, it seems clear that the surviving records are not sufficient to find "the" Robin Hood. Either the records are incomplete (to show how poor our sources are for the pre-Tudor period, consider that we don't even know the names of two of King Edward I's children; Prestwich1, p. 126) or there was no one man behind the legend. The summary in Baldwin, p. 42, is probably best: "It is clearly impractical to regard the ballads as even a semi-fictionalized biography of Robin and his followers."
The one thing that seems possible is that there was some early storyteller who created the first cycles of Robin Hood tales. The "Gest" as we have it can hardly be his work, but since it is composite, it may well incorporate portions of his account. Some of the other early ballads may also be close to this early myth-making. But for this, the "Gest" is the single most important source -- being as it is far longer than any truly traditional British ballad on record (it will probably be evident that, in this case, "gest" means "geste" ("song of deeds"), or perhaps "jest," not "guest").
Robin's situation in some ways resembles that of that other great name in British legend, King Arthur. There seems to have been an historical Arthur, although all we know is that he probably fought a battle against the Saxons at Mount Badon. The Welsh made him into the subject of folktale -- but it was Geoffrey of Monmouth whose largely fictional work created the Arthur legend. (For details on this, see "King Arthur and King Cornwall" [Child 30]).
Most of what follows is, of course, based heavily on the work of others, such as Holt and Keen and Knight/Ohlgren. I have tried to summarize the more important suggestions of these scholars, even when I disagree with them. Nonetheless, there seems to be much that is yet to be mined from the "Gest" and the other Robin Hood ballads.
THE EARLY BALLADS
The "Gest" is considered by Holt (Holt1, pp. 15-34.), following Child and others, to be one of only five fundamental pieces of the Robin Hood corpus, the others being "Robin Hood and Guy of Gisborne" [Child 118], "Robin Hood and the Monk" [Child 119], "Robin Hood and the Potter" [Child 121], and "Robin Hood's Death" [Child 120].
There are only a few variants on this list, mostly involving the "Death" -- my guess would be that this is because the Percy version is a mess and all the other copies are late. Holt1, pp. 27-28, do not even acknowledge any of the recent traditional versions of the "Death," and Knight/Ohlgren look at the 1786 English Archer version (Child's B) only where the Percy text fails (Knight/Ohlgren, p. 599) -- even though there are other traditional texts, including Davis's version, which appears to be a slightly damaged and mixed version of a very good original. Fortunately, since the "Death" overlaps the "Gest," its antiquity is not a major concern.
Keen's list of Robin Hood ballads of "proven early origin" (pp. 116-117) is the "Gest," the "Story of Robin Hood and the Potter," "Robin Hood and Guy of Gisborne," and "Robin Hood and the Monk"; he excludes the "Death" even though its plot is part of the "Gest" and so clearly ancient. (
On page 123, Keen in effect appends "Robin and Gandelyn" [Child 115] to his list (while adding that it is only the skeleton of a ballad; in his view, it is a sort of proto-Robin tale). He also points out the much-mentioned connection of the Robin Hood corpus to "Adam Bell, Clim of the Clough, and William of Cloudesly" [Child 116].
Ohlgren, p. 217, lists only the "Gest," the "Monk," and the "Potter" as early, seemingly based solely on external evidence: These three, and only these three, can be shown to predate 1525. "Robin Hood and the Monk" seems to be the earliest, coming from a manuscript of about 1450 (Percy/Wheatley I, p. 105, calls it "possibly as old as the reign of Edward II," but offers no reason for this incredibly early date. Thomas Wright also suggested this date, but Dobson/Taylor, p. 123n1, are openly contemptuous of this date). The manuscript, while well-written, is much-stained and hard to read (Knight/Ohlgren, p. 31); there may be a few textual uncertainties as a result.
The manuscript of the "Potter" is dated c. 1500 by Child and Ohlgren (and Copland in his late sixteenth century edition of the "Gest" also printed a play which seems to have drawn on the same tradition; Dobson/Taylor, p. 208). In fact there is very strong evidence that it is somewhat older, since it was the property of Richard Calle of Norfolk, who was active in the period 1455-1475 (see the notes to the "Potter"). But it is safe to add "Guy of Gisborne" to the list of early ballads, because, while the ballad itself is from the Percy folio, there is a fragment of a play on the same plot from c. 1475.
The list in Knight/Ohlgren, not surprisingly, is similar to that in Ohlgren; they file under "Early Ballads and Tales" the "Monk," the "Potter," the "Gest," "Guy of Gisborne" -- and tack on "The Tale of Gamelyn," "Robyn and Gandelyn," and "Adam Bell, Clim of the Clough, and William of Cloudesley."
EncycLiterature, p. 957, lists the Gest, the Potter, the Monk, and Guy of Gisborne as the "core" of the legend.
Chambers, pp. 132-134, after a nod to "Robyn and Gandeleyn" (which on p. 131 he calls the earliest tale of Robin Hood, never mentioning that it does not use the name "Robin Hood") lists as early ballads Guy of Gisborne, the Monk, and the Potter, plus perhaps the Gest, but not the Death; instead he offers "Robin Hood and Friar Tuck," i.e. "Robin Hood and the Curtal Friar" [Child 123].
The dating of the "Curtal Friar" is a vexing question. The language of our surviving versions of the ballad is rather modern, but that is not an indication of date of origin. The tale as it stands features absurdly many fighters and dogs, but that may be the result of the inflation common in tradition.
The first apparent linking of the Friar and Robin Hood dates from the fragmentary play of "Robin Hood and the Sheriff" (Dobson/Taylor, p. 203), based on the same story as "Guy of Gisborne"; it has a reference to, and perhaps even a part for, "ffrere Tuke." Even more explicit is the play printed by Copland around 1560, often called "Robin Hood and the Friar," which has three characters: Robin, Little John, and Fryer Tucke (see the versions on pp. 286-290 of Knight/Ohlgren or pp. 210-214 of Dobson/Taylor). Both of these plays predate the earliest version of hte ballad of the Friar (Dobson/Taylor, p. 209).
From about the same time as "Robin Hood and the Sheriff" comes the so-called Tollet Window -- a panel window of the Morris Dances and May games, reproduced in GutchI, p. 349, and RiversideShakespeare, p. 1478, and alluded to on Dobson/Taylor, p. 62. It was thought by GutchI, p. 338, to have been painted in the time of Henry VIII but based on originals from the time of Edward IV.
The window shows in its bottom three panels an unknown man, a lady (presumed to be Maid Marian), and a friar (presumed to be Friar Tuck). There is no overwhelming reason to think the first figure is Robin -- but neither is there any other obvious candidate. However, RiversideShakespeare, p. 1478, believes that Robin is not the man to Marian's left but the hobby-horse above her. Obviously the presence of Robin in this context is debatable -- and, hence, so is this early connection with Friar Tuck. In any case, we note that this is a century after Langland's reference to Robin, and more than half a century after the Staffordshire Friar Tuck.
Logic says that the Friar is not integral to the legend -- if there had been a genuine cleric in Robin's band, for instance, why is he not mentioned when Robin dies? And why do we see Robin going to mass in Nottingham in the Monk?. We do meet Friar Tuck in the play version of "Guy of Gisborne" (Baldwin, pp. 68-69; Cawthorne, p. 188), but this might be the source of, rather than inspired by, the "Curtal Friar."
There isn't even absolute proof that the "Tuck" of later legend is the same as the Curtal Friar of the ballad. We are forced to admit that the data is not sufficient to reach a certain conclusion about Tuck. I personally think him a later addition; in any case, I will not base arguments on the "Curtal Friar." For how Tuck came to be associated with Robin, see the section on "Who made Maid Marion?"
In sifting through these materials, Keen sounds a useful warning:"we must remember that we are not dealing with a host of different stories, but with a host of versions of the same story, and that what is significant is the similarity of tone, the forest setting, the animus against the law and its officers, the callous indifference to bloodshed, and not the differences of detail. At the same time we must remember that we are not dealing with a series of individual characters, but with a type-hero, the outlaw, who, though he may appear under more than one alias, remains essentially the same, and what is significant about him is not his name or his individual acts, but his conventional attitudes" (pp. 126-127). Although, just to show how confusing these things are, Pollard, p. 12, says that "We are not dealing with one Robin Hood character: we are dealing with several."
THE TEXT OF THE GEST
Chances are that we do not have the text of the "Gest" in anything like its original form. The place names it mentions make it almost certain that it was written by a Yorkshireman (see the note on Stanza 3) -- and a Yorkshireman who rarely travelled beyond his home county.
Yet the text as we have it is in fairly generic Middle English, with almost no signs of northern dialect (Brandl, according to Clawson, p. 7-8, detected what he considered "Northern rhymes" in certain sections, but Clawson notes that such rhymes are in fact found throughout, and are in any case found in other parts of the country. There is nothing distinctly northern about the poem). Chaucer could almost have written it; certainly he would have understood it with little difficulty. There are some Robin Hood ballads in northern dialect, such as "Robin Hood and the Bride," a variant of "Robin Hood and Allen a Dale" [Child 138] found in the Forresters manuscript, but the "Gest" in its printed forms is not one of them.
And yet, this is the period when regional dialects of English were at their strongest and most distinct, and because English was only slowly regaining its role as an official language, "authors in the twelfth, thirteenth, and fourteenth centuries generally wrote the English that they spoke -- whether in London, Hereford, Peterborough, or York" (Burrow/Turville-Petre, p. 5). Admittedly the "Gest" is more likely from the fifteenth century. But the expectation would still be that it would contain local linguistic forms.
The fact that it is so free of Northernisms strongly argues that there was a recensional stage when these characteristics were purged. What's more, because the surviving prints are all in essentially the same dialect, all our surviving copies must derive from this de-Northernized copy of the text. This needs to be kept in mind in evaluating our surviving witnesses. Dobson/Taylor, p. 6, suggest that "the next move in the investigation of the Robin Hood legend would seem to lie with linguistic scholars." But this challenge was not taken up until Ohlgren suggested it to Lister M. Matheson, and even Matheson's work is very preliminary.
Matheson, on p. 210 of Ohlgren/Matheson, declares that the printed editions of Pynson, de Worde, Goes, and Notary have all adapted the text to fit their preferred dialects, but adds that "a number of Northern spelling and forms survived this process.... Their appearance suggests strongly that the original author was indeed a Northerner and possibly a Yorkshireman." I must confess that I do not see how his methodology can support such a strong conclusion; his method is to compares the prints against the suggested regional dialects -- but not to compare the prints against each other in a meaningful way. Only by this means could he determine the residual dialect before the various changes.
Matheson does suggest, based on his analysis, that the source for the Pynson and de Worde editions was not a lost print by Caxton, because in that case the spellings would have been more standard. This conclusion is probably strong enough to stand. It does not mean that there was no Caxton print, but that it was not the common source. Pynson or de Worde might have used a Caxton original, but not both.
Like most of the Robin Hood ballads (and, of course, like the romances), we have no field collections of the "Gest" -- it is likely that it never existed in tradition. What we have are printed editions. Child's text is based on seven of these, which he calls a, b, c, d, e, f, and g -- a system usuallybut not always followed by the later scholars. The prints may be briefly described as follows:
a: "A Gest of Robyn Hode," is in the National Library of Scotland. The call number in Advocates Library H.30.a. Often referred to as the "Lettersnijder edition," based on the font used. A photo of the front graphic can be found in the photo section preceding p. 223 of Ohlgren, and a photo of the whole first page is on p. 107 of Ohlgren/Matheson. Isaac's plates 92-93 show the layout of two interior pages. Contains all or parts of Child's stanzas 1-83, 118-208, 314-349 -- just under half the total. It is Dobson and Taylor's A.
b. "A Lytell Geste of Robyne Hode," printed by Wynken de Worde. The surviving copy is in the library of the University of Cambridge, Selden 5.18. Photos of the frontispiece can be found in Ohlgren (again, in the section preceding p. 223), on p. 113 of Ohlgren/Matheson, and in Holt, p. 14. Dobson and Taylor cited it as B.
c. Bodleian, Douce e.12 (called Fragment #16 by Child). Duff-Bibliog #361. Two leaves. Portions of stanzas 26-60 only, said by Duff-Bibiog, p. 100, to have been taken from a binding and to be the central leaves of a quire. A photo is on p. 121 of Ohlgren/Matheson. Dobson/Taylor refer to Child's c and d under the siglum D.
d. Bodleian, Douce f.1 (called Fragment #17 by Child). Portions of stanzas 280-350 only. A photo is on p. 125 of Ohlgren/Matheson. Dobson/Taylor refer to Child's c and d under the siglum D. The pages were placed in binding strips and have been trimmed; this has resulted in the loss of text at the beginning of lines as well as at the top and bottom of pages. Unusually, this edition indents alternate lines, so that some lines are more defective than others.
e. Bodleian, Douce f.51(3) (called Fragment #16 by Child). Portions of stanzas 435-450 only; from stanza 443 on, only the ends of the lines survive. A photo, showing the extent of the damage, is on p. 100 of Ohlgren/Matheson. It is reported to have been extracted from the binding of a book (Oates, p. 3). Dobson/Taylor collectively cite e, p, and q under the symbol P.
f. "A Mery Geste of Robyn Hoode," British Library C.21.c. Printed by William Copland, meaning that it is from 1548 or later although before 1570. Since Copland registered a Robin Hood play in 1560, and Copland's print contains two dramas as well as the "Gest" (Dobson/Taylor, p. 208), it is likely that 1560 is the year of printing -- although Dobson/Taylor suggest that Copland had printed the plays in an earlier separate form, in which case the date must be after 1560. A photo is on p. 129 of Ohlgren/Matheson. Dobson and Taylor made the unfortunate decision to ignore Child's sigla and cite this as C. A single leaf of another Copland edition is Oxford, Cordington Libraray, All Souls college, k.4.19. It has been hypothesized that this is a later edition; I do not know if this has been proved.
g. "A Mery Iest of Robin Hood," Bodleian Library, Z.2.Art.Seld. Printed for Edward White, who was active well into the seventeenth century (e.g. Wikipedia reports that he printed the 1611 third quarto of Shakespeare's "Titus Andronicus." He or a relative was also among the first to license "Greensleeves") He may well have known Anthony Munday, of whom more below. Gutch, p. 141, suggests on the basis of a Stationer's Register entry that this copy was printed in 1594.
Since Child's time, two more small fragments have been discovered. For reasons to be seen, I am labelling them p and q rather than h and i.These were studied in detail by Oates, and the descriptions are from his paper.
p. The "Penrose fragment," formerly owned by Boies Penrose but now in the Folger Shakespeare Library. A full leaf and a portion of a second, recovered from a book binding. Stanzas 227.4-235.2, 243.2-250.4, 312.4-319.3, 327.3-335.1. Dobson/Taylor collectively cite e, p, and q under the symbol P.
q. The University of Cambridge fragment. Found in a book binding and presented to Cambridge University in 1917. Contains 220.1-227.3, 319.4-327.2. Dobson/Taylor collectively cite e, p, and q under the symbol P.
Thus far is fact. Beyond that we must rely on inference. What follows summarizes information we derive from the contents of the prints (typefaces, etc.)
The type of a (Lettersnijder) is Lettersnijder 98 -- that is, 20 lines are 98 millimeters tall, making the type 13.9 point (in the modern usage of 72 points=1 inch.) The orthography is very peculiar. The first page is set entirely as prose -- Oates, p. 9, makes the reasonable suggestion that it was originally intended to be set as poetry, but then it was decided to include the woodcut of the mounted archer at the top, and the text had to be reset and dramatically compressed to make room for it.
Based on the samples in Isaac (plates 92, 93), the spaces between words are very small -- in a lot of cases, there are no spaces at all. The only punctuation marks are points which are placed almost at random (certainly not where we would place periods; some hardly even qualify as comma breakss) and a handful of section marks, some of which indicate line breaks. It also lacks stanza breaks.
The first letters of lines are capitalized, but in Isaac's first sample, almost nothing else (e.g. in lines 50.2-58.1, we find the following: "lancaster," "seynt mari abbey," "criste" (christ), and four instances of "robyn" -- balanced by one instance of "Robyn," as well as "Caluere." If you can see a pattern in that, you're smarter than I am.) In the second sample, proper names are regularly capitalized ("Robyn," "John," "Scarlok," although not "wylluam" or "much"), as is the pronoun "I." This second section also typically spells "The(e)" with a y and a superscripted e -- a usage not found in the first sample.
I rather suspect, based on the usage, that there were two typesetters, one more familiar with English orthography than the other.
Gutch1, pp. 80, 142, contends that Lettersnijder was issued by Myllar and Chepman in 1508, and Holt1, p. 122, also refers to it as among "the Chapman (sic.) and Myllar Prints of 1508." This is understandable but a mistake. Chepman and Myllar were authorized to print mass books and other materials in Scotland in 1507, and published for about twenty years (Isaac, introduction to Myllar and Chepman).
The largest single collection of works from their press is Advocates H.30. This book contains in one binding no fewer than eleven quarto books. The first nine of these are typographically similar, and seven of the nine contain a colophon or other markings associating them with Myllar and Chepman. The three with dates are all from 1508: Porteus of Noblenes, Chaucer's The Maying, and the Knightly Tale of Gologros and Gawaine. (For the full list of contents, see Isaac or p. 144 of Gutch1)
The natural assumption is that the last two items in the volume are also from Myllar and Chepman, especially since item #10, The Twa marrit wemen and the wedo, is attributed to the Scottish poet Dunbar. But it is notable that every one of the properly attributed Myllar and Chepman prints, according to Isaac, is in a Textura face. The Avocates copy of the "Gest" is not in Textura; it is, of course, in Lettersnijder.
The link to Myllar and Chepman appears dubious on other grounds. The small catalog of their known works includes two by Dunbar, one by Henryson, and Blind Harry's Wallace. Their other works, if not as obviously Scots by authorship, are strongly Scottish in style -- Hahn's edition of "Golagras and Gawain," based on the Myllar/Chempan edition, is so broadly Scots that it is not until line 76 that he can go a whole line without a gloss! Whereas at least 80% of the lines in the "Gest" make perfectly good English sense as printed, without need for explanation. And, as Clawson says on p. 2 (cf. Isaac), the incipit to the Advocates text of the "Gest" reads "Here begynneth" (English), not "Here begynnis" (Scots), a reading which would surely have been "Scotticised" even if nothing else had been.
Thus the strong weight of evidence is that Chepman and Myllar did not print the "Gest." There is, indeed, no reason to think that the printer was Scottish.
Beyond that we can say little, because the Lettersnijder font was common around the beginning of the sixteenth century. Most printers who used Lettersnijder were Dutch, and there are a few instances of errors which make sense in Dutch (e.g. "mijn" for "mine"; 200.3), so it is highly probable that it was the product of a Dutch press. Knight/Ohlgren, p. 80, and Ohlgren/Matheson, p. 101, mention an attribution to Jan van Doesborch of Antwerp, but this is speculation; the only real support for the belief is the fact that van Doesborch printed books for the English market. But Isaac, notes to Laurence Andrew,e, mentions a belief that van Doesborch published only books associated with Andrewe, and there is no reason to think the "Gest" should be so associated.
Because we do not know the printer, the the date is uncertain; the period 1510-1520 is often suggested, but it might be a decade or two earlier. Holt1, p. 15, merely suggests that it was published in Antwerp between 1510 and 1515.
It is clear that compositor did not know English very well, he also shows signs of inexperience in his craft. In particular, he seems to have had trouble with inverted letters, such as n/u and, once or twice, m/w. There may also be a few instances of mistaking the letter thorn for a d when it should have been transcribed th. (See the note on Stanza 179. This may indicate that the common ancestor of a and b still used eth and/or thorn. I have not spotted any instances which might arise from confusion caused by a yogh.)
(Incidentally, although a has the most problems with inversions, b also has a few, in 299.1, 305.3, 363.2. This leads me to wonder if there wasn't a printed version which preceded both a and b with many inversions, most but not all of which b corrected.)
Child, p. 40, offered a handful of instances which made him believe a more primitive than b, and this opinion has been repeated many times. I did not consider Child's short list of examples sufficient to be decisive, and Ohlgren/Matheson, p. 101, also admits doubts.
Wynken de Worde's b text is without doubt the earliest of the complete copies. De Worde (the successor of England's first printer Caxton) worked from 1492 to 1534, although the piece has no internal dating. The colophon says that b was "Enprented at London: In fletestrete at the sygne of the sone" (Ohlgren/Matheson, p. 98). And de Worde did not move to Fleet Street until 1500. Thus the earliest possible date is actually in that year.
However, de Worde -- although his typography was always behind the times (Binns, p. 110, says that "most of his printing was of indifferent quality and some of it was thoroughly bad") -- gradually changed his fonts and his collection of clip art (he started using pure Textura-style blackletter but eventually acquired Roman and Italic and even Greek type; as Moran points out on pp. 26-38 -- although the Greek is perhaps the most unreadable font I have ever seen in my life).
Binns, p. 109, suggests that the "Gest" was printed around 1498-1500, when de Worde was busily printing other romances -- "Bevis of Hampton," "Sir Eglamour," and "Guy of Warwick." (E.g. Duff-Hand-List, p. 2, lists as his only four books certainly dated to 1498 the "Description of Britain," the "Morte d'Arthur," the "Canterbury Tales," and the "Legenda Aurea.") This makes excellent sense but suffers from the fact that a date before 1500 is ruled out by the colophon.
Based on the facsimiles, it appears de Worde published the "Gest" using his Textura 95 font (Duff's #8; facsimiles in Isaac, figures 2, 3, 7, 8 and Duff-Bibliog, plate XIV, where it is called #4). The number "95" refers to the size of the type -- it means that 20 lines of type were 95 mm. tall. In other words, 20 lines equalled 270 points, meaning that it was about 13 point type (as we would describe it today).
Isaac, facing figure 1, says that Textura 95 was "the most frequently found of all de Worde's types in the sixteenth century"; he used it for his entire career. Duff-Bibliog, pp. 127-129, lists 103 books believe to have been printed by de Worde before 1500; 82 of these use at least some Textura 95, and 26 appear to use it exclusively. However, it did evolve somewhat; in this periord, there were multiple forms of the letters a, d, h s, v, w, and y (Isaac, figure 1). The heading line of de Worde's edition of the "Gest" uses four of these letters, in states a-1, d-1, h-1, and y-2. The y is datable: de Worde was using y-1 in 1502, but by 1506 had shifted to y-2 (Isaac, notes to plates 2 and 3).
So the date almost has to be after 1503. But on other grounds, the earlier, the better. The illustration at the head of the print, which shows a woman, a man carrying a sword backwards, and a man who appears to be a herald. The artwork has no relevance at all to the "Gest," and de Worde gave up a large portion of his clip art (as well as some fonts of type) when he made the move; much of the material, in fact, ended up in the hands of another printer, Julian Notary (Duff-Printers p. 131). Had de Worde printed the "Gest" before his move, or long after, he could probably have used better art.
Another argument for a not-too-late date is the fact that, in around 1507, de Worde his rival Richard Pynson began a policy of cooperation (Isaac, notes on Pynson). This ended a strong rivalry that had existed between the two. Given that de Worde and Pynson both seem to have produced editions of the "Gest," this is an argument that the de Worde edition was printed before their agreement.
This is strong evidence for Ferguson's date of around 1506, (Oates, p. 7); this date is also found in the Short Title Catalogue of Book Printed in England, Scotland & Ireland, 1475-1640 (Ohlgen/Matheson, p. 112). My own date, based on examination of the facsimilies independent of the above, was c. 1505.
All that being said, someone really needs to examine the actual printed copy, not just facsimiles (which may not be the exact size of the original), checking all the letters; my suspicion is that, using Isaac's data, we could offer a much more exact date.
Of all the copies of the "Gest," de Worde's appears to have been the most used. No fewer than three readers but their names in it (Ohlgren/Matheson, p. 117). One called himself "George Poll" (Powell) and urged readers to kiss his "briche and buttocks." A second simply says "By me John"; this is perhaps John Cony, who signed that name to two other books which were bound with the Gest, "The assemble of goodes" and "The Frere and the Boye" (interestingly, another copy of the latter poem is also bound in the volume containing sole copy of the "Potter").
The third name is entered twice, with different spellings: One claims the book is "Avdary Holman[']s," the other says it is "By me avdery homan of titsey." Audrey Holman also put her name in two of the other books bound with the "Gest." Ohlgren devoted significant effort to trying to locate Audrey Holman, eventually coming up with three candidates (Ohlgren/Matheson, pp. 117-120). His most likely candidate is his #2; we don't have her dates, but her older brother was born in 1571 (meaning that she was probably a few years younger), and she was still alive in 1621. She eventually married William Masters and had two children. Thus she cannot be the original owner. Still, the fact that the book went through at least three and probably four owners before being entered into the Bagford collection shows how popular it was.
It is has been stated that c and d are from the same original -- note, e.g., that Dobson/Taylor cite them under the same siglum, although they do not quite state that they are the same edition. However, even a casual glance at the letter forms shows they are distinct.
Ritson thought c to have been printed by Wynken de Worde -- but dated it 1489 (Child, p. 40). Duff-Bibliog, p. 100, has no doubt that it is by de Words, noting that "though in the earlier type it has the later I, and Caxton's I does not occur. It cannot be earlier than 1500, and quite probably was printed a year or two later." Ritson's date, at least, is impossible, because de Worde was Caxton's assistant until Caxton died in 1491 (Duff-Printers, p. 23); de Worde could not produce a book of his own before 1491, and the evidence is that it took him several years to start publishing large numbers of books (perhaps because he did not have Caxton's skills at compiling and editing). Knight/Ohlgren, p. 87, mention the attribution but not the year. Oates, p. 6, accepts the attribution to de Worde, and allows that it predates b, but does not offer a date.
The type is a good argument for the attribution to de Worde, but because there are so many Texturas floating around, it isn't quite proof. And, if it is from de Worde, why then are there so many differences from b? The differences are rarely substantial, but they are numerous.
Farmer instead suggested John Rastell as a printer (Child, p. 40). Rastell's dates are disputed; Child claims 1517-1536, but Isaac's introduction to Rastell suggests that he was in business from about 1512. (He also has the distinction of being the first English printer to handle music and text in one pass.) However, Rastell is another printer using those ubiquitous Textura types, so I doubt this can be demonstrated with certainty. I will say that, based on the facsimiles in Isaac, it doesn't look like Rastell's style.
Gutch1, pp. 80, 141, follows Ritson in saying that Copland's f print seems to have been derived from b, and Clawson, p. 3, declares it "apparently a reprint of b." This is clearly true; I noticed the matter independently before I saw the (brief and undocumented) claim in Gutch. It is strange to note that Child and other recent editors seem to have paid little attention to this fact -- Child cites the variants in f without saying anything about the ancestry of that print.
It is hardly surprising that William Copland followed the text of de Worde, because it is believed that William Copland was either the younger brother or the son of another printer, Robert Copland -- and Copland actually worked for Wynken de Worde early in his career (Isaac, introduction to Copland; Duff-Printers, p. 146), and apparently was responsible for editing some of de Worde's editions (Duff-Printers, p. 7); he was also mentioned in de Worde's will (Duff-Printers, p. 139).
Thus it is very likely that William Copland would have worked from a copy of de Worde's own earlier printing -- indeed, it is possible that Robert Copland worked on b. (Ohlgren seems to think it more than possible; on p. 114-115 of Ohlgren/Matheson, he suggests that the "rose garland" used in the archery contest of stanza 398 may have been an interpolation by Copland. The obvious difficulty with this is, if Copland had been rewriting the "Gest," why didn't he fill in the several lacunae in the poem? And we find other mentions of rose garlands in the Robin Hood literature; see, e.g, Knight, p. 7).
Ohlgren/Matheson, p. 186, notes the somewhat curious fact that there seem to have been many early editions of the "Gest," but that production then slacked off. Ohlgren thinks there was a 45 year gap between the Notary and Copland editions. Since his date for Notary's print is conjectural, the gap may not have been that long -- but it was probably substantial. Ohlgren's suggestion is that copies ceased to be printed because Henry VIII turned Protestant and Robin Hood was very Catholic. This does not account for the whole gap, because Henry was still quite Catholic, thank you, in 1520 (and even 1530), and never ceased to regard himself as Catholic. But it might explain part of the gap.
White's g text rarely gets much attention, simply because it is so much later than the others. It is instantly clear that the text has been much modernized, although this does not prove whether it is from a good or a bad source. We will cover its affinities below.
From the lineation, it will be evident that the two p and q fragments are from the same edition. It is also generally accepted that e is part of the same print (although not necessarily part of the same copy of that print). It is also clear from the fact that the first verses of q come before the first verses of p, but the last verses of p come before the last verses of q, that the two were not properly bound in a single quire. Oates, pp. 5-6, is convinced that they were mis-collated -- that is, the edition had its pages out of order.
This raises an interesting point. The epq text is widely attributed to Richard Pynson. The suggestion seems to go back to Duff-Bibliog, p. 100, based on a single leaf of q (even though he admits that the "collation [is] not known), and is accepted by Isaac (preface to images 92 and 93 of the "Gest"), and was accepted without question by Oates (p. 4), Dobson/Taylor (pp. 71-72), and Ohlgren (Ohlgren/Matheson, p. 98). In terms of the type, this fits -- epq seems to be in the Textura 95 that Isaac (in the notes preceding plate 13) says was Pynson's standard type.
But almost everyone had a Textura 95: de Words (Isaac, before plate 1), Pynson (Isaac, figures 13, 14, 15, 19 -- indeed, based on figure 19, Pynson's collection of ornaments includes several which appear to me to be exactly the same as those de Worde used in the "Gest.), Hugo Goes (Isaac, before plate35; Goesacquired his Textura from de Worde), Robert Copland (Isaac, before plate 45), John Scolar (Isaac, plate 47; he and his successor Charles Kyrfoth, like Goes, had their Textura from de Worde), John Skot (Isaac, before plate 50), Thomas Berthelet (Isaac, introduction to Berthelet, says that this is another instance where that printer acquired it from de Worde), John Byddell (yet another had worked for de Worde and may have gotten some of his type; Isaac, introduction to Byddell), John Herford (Isaac, introduction to Herford).
Plus Julian Notary had a Textura 92 (Isaac, before plate 26), as did Ursin Mylner (Isaac, before plate 44). There were Textura 93s in the library of John Rastell (Isaac, before plate 36), Henry Pepwell (Isaac, before plate 48), Peter Treveris (Isaac, before plate 53), and Richard Bankes (Isaac, before plate 55). Even Chepman and Myllar, in Scotland, used a Textura 93 similar to de Worde's Textura 95 (Isaac, introduction to Chepman and Myllar).
This list could easily be extended, especially given how freely de Worde spread his favorite font around. And, as Duff-Bibliog points out on p. ix, "it is clear that almost all early English printers well understood what is now called 'leading', that is, producing a greater space between the lines by inserting slips of metal, so that we find the same type often with two, sometimes with three, different measurements." Thus simply measuring the height of the type is not sufficient to determine which font it is.
Ohlgren says on p. 101 of Ohlgren/Matheson that epq uses the forms of w and s found in Pynson's Textura 95. This appears to be correct based on the samples in Isaac, but the sample is too small. The fact that epq seems to be in Pynson's type is not quite proof.
Matheson, on p. 203 of Ohlgren/Matheson, affirms that the orthography of epq matches Pynson's. This too is strong evidence at a time when printers followed very different standards. But it appears from the footnote on p. 249 that Matheson used only a small collection of facsimiles, meaning he didn't have much material to work with.
According to Binns, pp. 110-111, Pynson was a Norman; he perhaps began as a bookseller rather than a printer. he probably learned the printing trade from Guillaume Le Talleur of Rouen, and in 1490 took over the printing business of William de Machlinia of Belgium. He moved to Fleet Street in 1500, began to work on government documents in 1503, became Royal Printer in 1508, introduced Roman type into England in 1509, and retired in 1528, dying two years later. According to Binns, p. 512, his listed output consists of law books, official publications, and missals. Steinberg/Trevitt, p. 48, declare that Pynson "obtained a virtual monopoly of law codes and legal handbooks."
And note the description of Pynson's work. Steinberg/Trevitt, p. 48: "Pynson published some 400 books, technically and typographically the best of the English incunabula." Or Binns, p. 112, "Pynson was without doubt the finest printer of his day. He had a fine range of types and used them well. His press-work was superior to that of his contempraries. He used illustration more sparingly and more effectively than de Worde, and was much more successful with his decorative initials and borders." And yet he decided to print something completely different in the "Gest,' and when he did so, he got the pages in the wrong order?
The matter is trivial; we are less concerned with the printer of epq than its text, but I do think caution is indicated. The one important result of Ohlgren's examination is that, if epq is indeed by Pynson (and I think it likely, just not certain), then it almost certainly dates from 1505 or earlier, when Pynson adopted a different form of w.
Ohlgren manages to assign printers to every edition except a (Ohgren/Matheson, p 98). In addition to Pynson for epq, de Worde for b, Copland for f, and White for g, he argues that c is the work of Hugo Goes of York, while d comes from the press of Julian Notary.
I wouldn't consider either attribution to be very strong. The connection of c with Goes is also found in the Short Title Catalog, but the font (as noted above) proves relatively little. Since Goes, de Worde, John Scolar. and Thomas Berthelet all had copies of de Worde's Textura 95, and Pynson had something quite close, any of them could have been responsible -- indeed, the way the text is printed looks to me a bit more like the sample of Scolar in Isaac than the sample of Goes. The Short Title Catalog suggests 1506-1509 as the date, but with a question mark.
Our knowledge of Goes is very limited; according to Isaac, we have three addresses for him (London, Beverly, and York), but the two former addresses were taken from materials now lost; our only datable book was printed at "York, in the Street called Steengate" in 1509 (Isaac). We have records of only three books by him (Binns, p. 129), and only one -- the Directorium Sacerdotum -- still survives.
We certainly cannot rule out the possibility that Hugo Goes printed the "Gest" -- a work which would likely be popular in Yorkshire. On the other hand, we note that his one known book was in Latin, and the other two also sound like they were intended for clerical use and were in Latin. From such works to the "Gest" is rather a stretch. And while the survival of early books is rather a matter of chance, the fact that we have so many surviving books by de Worde, and so few by Goes, is at least a slight argument against Goes as the printer.
Ohlgren does not absolutely deny the possibility that de Worde published c. On p. 122 of Ohlgren/Matheson, he says that if it is by de Worde, it must be earlier than b -- a statement which he does not justify. But he goes on to mention the point made above, that b and c have significant differences, which he considers strong evidence that c is not by de Worde. This is true but not decisive; I think we must consider the printer of c uncertain. What the differences do prove is that c can be treated as an independent witness.
Ohlgren does point out on p. 123 a suggestion that the Goes edition might have encouraged people to name all sorts of places in Yorkshire after Robin. This is possible but beyond proof.
The attribution of d to Notary is based on the use of Textura 92 (Ohlgren/Matheson, p. 126), but the difference between Textura 92 and Textura 93 (or even Textura 95) is really only a difference in leading. Plus Notary wasn't the only printer using a Textura 92. Ohlgren says he was the only "major London printer" to use Textura 92, but offers no reason to think d came from a London printer. The Short Title Catalog dated it "c. 1515?" -- but this was apparently only a guess.
There is another argument against the attribution to Notary, and that is the list of materials Notary printed. The list on p. 129 of Duff-Bibliog lists seven items. Six are in Latin and appear to be church books. The only exception is a print of Chaucer's "Mars and Venus." Notary seems to have been aiming for a rather highbrow market; the "Gest" hardly fits!
There is agreement that all these prints have a recent common source, possibly a lost printed copy but more probably (given the dates of Pynson and de Worde) a manuscript, and clearly not the original, since all copies share certain defects. Further evidence for a recent source is shown by the fact that all the copies are quite closely similar. I do not think any reasonable scholar would dispute this point.
What, then, is the relationship between these prints?
Dobson/Taylor, p. 8, suggest that a is "apparently a cheap reprint of a previous and now lost edition by Richard Pynson," i.e. of epq. This follows from a comparison made on p. 9 of Oates, who compared the 70 lines for which epq and a both survive. Oates found several significant differences between a and epq, but six times as many cases where the two agree with each other against b. It is clear that they represent a single phase of the text, and it is likely that one is a copy of the other.
Oates is convinced that a is a copy of epq. And his evidence extends beyond the textual. The woodcut at the head of the Lettersnijder edition is a copy of one used by Pynson in his edition of the Canterbury Tales. But (contrary to, e.g., Holt1, p. 122) it is enphatically a copy -- the images can be seen side by side on pp. 104-105 of Ohlgren/Matheson, and the Canterbury version differs in the face, the spurs, the ribbons on the horse, and other details from the Lettersnijder version; in addition, Lettersnijder is cropped more closely. Oates believes -- and I think it almost certain he is correct -- that Pynson used that same illustration in his edition of the "Gest," and the Lettersnijder printer then copied it (and, as mentioned, forgot to leave room for it!).
Matheson seems to confirm this, declaring on pp. 200, 203 of Ohlgren/Matheson that the spelling of a closely matches epq. He does not a few variants in a which are valid English alternatives rather than errors, and suggests that this might mean that a native English speaker was involved in the typesetting of a. It strikes me as at least as possible that the copy of epq used to create a had a few corrections written into it -- but it might also be that these variants are from the typesetter who knew English, as opposed to the one (responsible for the majority of the remaining text) who did not.
There is a secondary point: If the Lettersnijder edition is derived from Pynson, it must be post-1490, when Pynson began printing, and likely post-1495. Probably Lettersnijder is later than that. If Duff is correct in dating Pynson to 1500, then a date after 1510 seems likely for Lettersnijder. (On the other hand, Ohlgren/Matheson, pp. 107-108 suggests a date in the early 1490s for Pynson, which allows Ohlgren on p. 110, to claim a date of c. 1495 for Lettersnijder.)
Looking at the other substantial copies, it is instantly clear that f and g go together -- g in fact looks like a modernized copy of f, perhaps compared with a copy of b; most of the differences between f and g are cases of an archaic form in f being replaced by a more modern form in g. Clawson, p. 3, calls it "very similar" to f.
On this basis, I would be inclined to date g as late as possible -- a Jacobean date would be far better than an Elizabethan, and frankly, I'm inclined to suspect that the attribution to White is deceptive and the piece actually printed in the reign of Charles I. f also has some signs of modernization, although far fewer than g,
It is also clear that f and g go with b. The relationship between b and f is noted on p. 130 of Ohlgren/Mathison, with the observation that f has had its language modernized -- although Ohlgren seems to have missed a few points about the copy of b used to produce f (Ohlgren does not examine g in any detail, merely calling it a "close copy" of f -- which is true of the basic text, but g modernizes f even more than f modernized b). Ohlgren/Matheson, pp. 132-133, suggests that Copland printed the work in part because of its anti-clerical tone.
There are strong indications that the copy of b used by the compositor of f was damaged. A good example is in stanza 305. The text of b has Little John say "No lyfe on me be lefte." All fg can offer is "That after I eate no bread," which is so utterly feeble that the only possible explanation is that the exemplar was damaged. In stanza 400, b has "And bere a buffet on his hede, I-wys right all bare," while fg give us "A good buffet on his head bare, For that shal be his fine," which fails to rhyme and is inept anyway. These readings suffice to prove the kinship of fg. The relationship to b is less instantly obvious but will be evident to anyone who goes over the collation.
Child does seem to have realized that fg were relatives of b, but he does not really describe the situation, if indeed he even thought in terms of a stemma. But it seems clear that we have two basic groups, which we might call Pynson and de Worde. Pynson consists of epq and a, with a having value only where epq is defective (admittedly, more than 80% of our knowledge of the Pynson text comes from a). de Worde consists of bfg -- and, because b is complete, this means not only that g has no value (as was recognized, e.g. by Dobson/Taylor and Ohlgren) but also that f has no value.
Unfortunately, the fragments c and d are all so short that their affinities cannot be firmly established. My feeling is that c and d are closer to the b group than to a, but not as close to b as are. This conflicts with the opionon of Ohlgren, p. 122, who thinks (on the basis of spelling rather than text) that it is another copy of Pynson. But if that is the case, why is it so distinct from a? I don't think Ohlgren's opinion can be sustained. The best guess is that it is independent.
Where the fragments are extant, they can give us some help. But the two combined include less thant a quarter of the "Gest." For the largest part of the poem, we are stuck choosing between a and b -- or, indeed, between b and conjectural emendation.
Although we cannot prove whether epq/a or b is the older text, Child (p. 40), Dobson/Taylor, and Knight/Ohlgren (p. 80) all consider a to be the more primitive -- but Child's evidence is summarized in a single note on p. 40 listing about a dozen variants. The primary evidence, really, is that a was incompetently typeset (note that there is a homoioteleuton error as early as the second stanza), meaning that the typesetter wasn't fiddling with it. Child in particular takes a as his copy text insofar as it is extant; he uses other readings only where it appears badly corrupt. Both Child and Knight/Ohlgren follow their copy text so closely as to alternate between spelling Little John's name "Lytel" where b is the copy text and "Litell" where a is extant -- an obvious absurdity.
As Ohlgren/Matheson states on p. 101, "Since 1899... all of the poem's editors have repeated Child's assertion that the Lettersnijder edition [a]... is the earliest surviving edition... and hence it has been given pride of place in various critical editions, even though it is in an imcomplete state. It has even supplanted the almost-complete Wynkyn de Worde edition [b]." This even though, as Ohlgren continues, "Lettersnijder is not only a decidedly poorer version of the text but also an almost incompetent copy of an earlier version by Richard Pynson, which now must be recognised as the earliest surviving edition of the poem."
Even before reading Ohlgren's comments, I didn't buy Child's argument. Child's collation method seems almost designed to obfuscate (particularly since he was inconsistent in how he recorded variants), but if we convert it to an inline collation, it was easy to see the two groups mentioned above: a on the one hand and bfg on the other.
It is at this point that the fact that the text we have is not northern becomes important. The common ancestor of a and b was not the original -- and if a preserves this edited text better than b, that doesn't really make it much closer to the original.
Hence I think Child's extreme preference for a exaggerated. True, it has older grammatical forms. But recall that it is probably Dutch, typeset by a Dutch compositor. Many of its errors are pure and simple goofs -- e.g. in 6.4, "vnkoutg" for "vnkouth"; 15.4 "mynge" for "mynde." Clearly the compositor of a simply transcribed the original mechanically.
Wynken de Worde, although born in the Low Countries, was thoroughly familiar with English, and his work was designed to make English audiences comfortable -- and, indeed, to standardize the language. His press made a habit of updating grammatical forms (Steinberg/Trevitt, p. 58). His text of the "Gest" has surely been touched up, so if the question is solely one of grammatical form, a is generally to be preferred. But there is no hint that de Worde made substantial revisions. Where the difference is one of fundamental meaning, as opposed to grammatical form, it seems to me that b has as much authority as a, and the poem should be re-edited on that basis.
The fact that Pynson and de Worde and (apparently) three other printers all issued versions of the "Gest" around the beginning of the fifteenth century is obviously a testimony to its popularity. But the fact that Pynson and de Worde have noticeably different texts is also noteworthy. If two printers, who sometimes worked together and were for very long based on the same street, produced substantially different versions, this clearly implies that one is not dependent on the other, although it is likely they are based on a common recent source.
Bottom line: The text of the "Gest" needs to be re-edited eclectically, based on the Pynson and de Worde types, with c and d consulted where extant and conjectural emendation sometimes necessary, especially in the places where Pynson is lost.
Fortunately very few of the differences between the texts are substantial -- the main reason why the texts are considered to go back to a single fairly recent original. But at least one variant, in stanza 53, is potentially significant; see the note on that verse.
If we were to grade the condition of the text, we would probably list it as "fair." There is no real doubt as to the general course of the narrative, meaning that the text of the "Gest" is in better shape than, say, the text of the "Death." But the amount of minor damage is extensive. As a result, I have included a textual commentary following the commentary on the content of the "Gest."
Based on the close similarity between the surviving texts, the archetype of the surviving versions (that is, their most recent common ancestor) probably dates from the reign of Henry VI or Edward IV (i.e. between 1422 and 1483), with the latter reign more likely than the former; this is obviously the latest possible date of composition But it is nearly certain that there were several generations of copies between the poet's autograph manuscript and the last common ancestor of our surviving copies. The various common errors, such as the lost first line of stanza 7, demonstrate this.
THE DATE OF THE GEST
If the "Gest" is not contemporary with the events it describes, when was it in fact written?
The dating of the poem remains a matter of controversy. GutchI, p. 81, claimed a date from the time of Chaucer, or the reign of Richard II (1377-1399) or Henry IV (1399-1413), which is not quite the same thing, but close. Chambers, p. 134, thinks he can detect signs of fourteenth century language in the "Gest." Child rejected this but left room for a date c. 1400. Knight/Ohlgren, p. 81, reject even this -- but their argument that the poem cannot have had a long life in manuscript is not logically sound.
Even if we allow for the possibility of rewrites to modernize the language, the "Gest" is unlikely to be earlier than the fourteenth century, simply because the saga of Robin Hood seems to be exclusively English. Unlike, say, the story of King Arthur, the Robin Hood tradition seems to be solely the possession of the English and English-speaking Scots (Holt1, p. 114). Given that the poem is clearly the work of a professional composer (see the section on the "Gest" as a romance), this requires a date after English was reasserting itself as a language of the middle and upper classes, which can hardly be before 1300.
Clawson, pp. 5-6, goes over Child's text and counts instances of inflexional -e and -es, counting 252 in all, or about one every other stanza. He argues for these as instances of fourteenth century usage (repeating the claim on p. 128), but this is far from decisive. These endings certainly were still used by Chaucer, and were gone by the time of Malory, but there are a few still in Charles of Orleans, and a provincial dialect might have preserved them longer than London did.
Holt, p. 192, mentions Clawson's observation that the poem throughout preserves Middle English inflexional endings (and of a type, it appears, more typical of the regions outside London), but also points out that no study of the language has been made since Clawson's 1909 work -- unfortunate, since knowledge of Middle English dialects has greatly increased since them. And inflexional -e alone can't prove much.
Vocabulary isn't really much help. There are a few strange words in the "Gest," some of which will be mentioned in the notes, but they are no hint to date because we don't know their meaning! Nor are there many words which changed their usage between the fourteenth and fifteenth centuries. We do note that there is no mention of the office of Ranger, an office probably instituted in the early fourteenth century and known to have been in existence in 1341 (Young, p. 163) -- but there is no mention of the older office of forester, either, so that's no help.
Ohlgren, p. 217, argues that the original was made in the reign of Henry V (1413-1422) or the first reign of Henry VI (1422-1461), but advanced no direct evidence.
Ohlgren argues that the poem, although written in Lancastrian times, was set in the reign of Edward III, perhaps on the basis of Laurence Minot [c. 1300-1352? (see note on Stanza 353). That the poet tried to set the poem in the reign of Edward III is certainly not inherently impossible, but it is not compelling. Minot seems to have been a northerner (Kunitz/Haycraft, p. 358), but his poems apparently survive in only a single manuscript, so there is little reason to think he was popular outside court circles. Nor can I detect any other allusions to his work (e.g. he often referred to Edward III as a boar -- Sisam, p. 254 -- and there is no hint of that in the "Gest").
Keep in mind that Edward III, once a hero-king, "outlived his own generation and his own usefulness, and became a considerable liability to the throne during his last years" (Ormrod, p. 35). Also, Edward III relied on parliament far more than earlier kings, and while he was anything but a constitutional monarch, that did mean that he had to redress grievances. And this was remembered. Why would a Robin Hood have arisen in this time? A date in the reign of Edward III is tempting to us now because (as we shall see) Langland's 1377 mention of Robin Hood is the earliest datable reference. But the elements of the poem suggest several different dates. We shall deal with these below.
In this connection we might note that Henry V (reigned 1413-1422) kept very tight reign on criminals, but his son Henry VI did not (1422-1461 plus 1470-1471), and his government was riven by faction (Wolffe, pp. 116-117). There was also much disorder in the reign of Henry IV (1399-1413), as that king tried to hold the throne he had usurped from Richard II. Might the disorder of the times have given rise to an interest in an alternate source of order?
Holt2, p. 10, observes that "Robin... was the product of a society where the threshold which separated lawful behavior from self-help by force of arms was indistinct and easily crossed." This, of course, was true for most of the middle ages. On the other hand, it was probably never more true than in the 1450s, at the beginning of the Wars of the Roses (see, e.g., Wagner, pp. 186-187, regarding the Percy-Neville feud).
Ohlgren, in his later writings, seems to have reconsidered his original dating. On p. 185 of Ohlgren/Matheson, he strongly urges a date toward the end of the Yorkist period, choosing 1483 as a somewhat arbitrary approximation. This, I think, is impossibly late, given that Ohlgren is arguing that Pynson's first printing was from around 1495. Although the primary texts of the "Gest," by de Worde and Pynson, are similar enough to have a recent common ancestor, they are also defective enough that it is hard to believe the original could be only twelve years old at the time Pynson printed it!
I think we are forced to admit that we don't know the date of the final editing of the "Gest," although it is probably fifteenth century; my personal date would be in the second quarter of the century -- but with older components. If it were much older than that, given the northern base of the legends, it would probably be much harder to understand.
Keen, followed by Holt1, pp. 35-36, does note that the three shorter early ballads have very different "feel": The "Potter" is humorous, with little real violence but a lot of tricks. Pollard, p. 12, in fact calls Robin a "trickster" in this tale -- although, in the "Gest," it really appears that Little John, not Robin, is the trickster. Nor is that the only instance -- e.g. in "Robin Hood and Allen a Dale" [Child 138], John is impressed into the role of Bishop, and rather than asking three times whether there are objections to the marriage, he asks seven times.
By contrast, The "Monk" and "Guy," especially the latter, are very bloody; in describing the latter, Pollard (p. 12) calls Robin a "cold-blooded killer." Pollard, p. 96, counts "nine homicides in the early ballads," although on p. 97 he grants that this is far fewer than the hundreds slain in "Adam Bell" and admits that the outlaws rarely inflict injury on the victims they rob. Compare this to Fulk FitzWarin, who kills fourteen of King John's knights on their first meeting (Ohlgren, p. xix), and more thereafter.
Pollard's suggestion, on pp. 98-99, is that Robin is appropriating forms of violence allowed by the rules of chivalry -- although, it should be noted, he has to take several of the ballads collectively to make this argument.
The "Death," if it be granted as ancient, is of course more a tale of treachery than anything else.
If the diverse nature of these ballads tells us anything, it is that the material of the legend is old enough that several different poets worked on it, each taking it in a different direction. We note that the "Gest," although composite, does not use any elements of the "Monk," the "Potter," or "Guy," and merely uses the content, not the lyrics, of the "Death." This implies a very large amount of material, of which the "Gest" takes only a small subset.
I will admit that I have held very different opinions over the date of the "Gest." Any suggestion must be extremely tentative. Right at the moment, however, I would be inclined to a date around the early 1450s, although based on materials from the earlier fifteenth and perhaps even the late fourteenth centuries. And the historical framework, if there was one, probably dates from the early fourteenth (which may, indeed, be the period when the name "Robin Hood" ceased to be that of simply a successful outlaw and became that of a courteous outlaw concerned with justice and propriety). It also seems likely that there was a revision of sorts, cleaning up the northern dialect although not changing the plot. Ohlgren's suggestion that this took place in Yorkist times is plausible, although I would prefer the period prior to 1475 to give more time for divergences to crop up.
Last updated in version 2.6
Go to the Ballad Search form
Go to the Ballad Index Song List
Go to the Ballad Index Instructions
Go to the Ballad Index Bibliography or Discography
The Ballad Index Copyright 2018 by Robert B. Waltz and David G. Engle.