N. F. Blake (essay date 1965)
SOURCE: “English Versions of Reynard the Fox in the Fifteenth and Sixteenth Centuries,” in Studies in Philology, Vol. 62, 1965, pp. 63-77.
[In the following essay, Blake surveys several editions of Reynard the Fox, noting a trend toward standardizing the English language.]
All writers on the history of the English language agree that the introduction of the printing press was an important landmark in the development of the language. McKnight, for example, writes: “The printing press introduced by Caxton was one of the most important factors in fixing the English language in permanent form.”1 But although Caxton's language has been investigated,2 few scholars have made any study of the language of the other printers of the late fifteenth and early sixteenth centuries to determine how this trend to conformity developed or how quickly the establishment of English in permanent form was achieved. Yet several books were constantly reprinted in the fifteenth and sixteenth centuries and by investigating the changes in orthography made in the printed versions of one of these minor best-sellers, it should be possible to make a contribution to the study of “the process and progress of the move towards conformity.”3 A study of this sort might help to show how the individual master-printers approached the language in which they were printing, the sort of changes they made and whether they attempted to standardize it.
One of the popular books of this period was William Caxton's Reynard the Fox, which he himself translated and then printed in 1481 (WC). This book was evidently so successful that Caxton reprinted it in 1489 (PL). This version is extant only in one copy now in the Pepys Library, Magdalene College, Cambridge; it lacks a couple of leaves at the end. Another reprint was issued about 1500 by Richard Pynson (RP). This version also survives only in one copy, which forms part of the Douce bequest to the Bodleian Library, Oxford. It likewise lacks several leaves at the end. A further edition, this one by Wynkyn de Worde, appeared about 1515 (WW). Only two leaves of this edition are known to exist and they are now in the University Library, Cambridge. The last edition I shall deal with was printed by Thomas Gaultier in 1550 (TG). This edition survives in several complete copies, one of which is in the Bodleian Library and another in the British Museum.4 I shall confine my attention to the editions so far listed, for all subsequent editions, and there were many of them,5 contain such extensive alterations that it is hardly possible to compare the language of these later versions with that of the earlier ones satisfactorily. Nevertheless between the first and the fifth edition there is a span of eighty years which should be sufficient to show whether there was any trend to conformity. Unfortunately only a small part of WW is extant so I have not always been able to use it in tracing the development of certain written forms because WW does not contain sufficient examples.
Before any discussion of the language can be attempted, it is necessary to elucidate the relationship of the various reprints. It is normally assumed that a reprint would be reprinted from the latest printed version.6 This is not the case with Reynard the Fox, for although all the texts are closely related to one another, there is not a straightforward chronological progression in their printing history. PL is naturally a reprint of WC, for WC was the only English version available when PL was printed. RP is not, however, a reprint of PL, but it also is a reprint of WC. WW is likewise not a reprint of RP, but a reprint of PL. TG, on the other hand, is a reprint of WW. Thus RP might be said to stand outside the main line of descent of Reynard the Fox. This fact may be readily proved because the changes which are made in PL do not appear in RP, though they are found in the two later reprints. Consider, for example, the following passages:
WC he that shoef your crowne PL he that shoere your crowne RP he that shoef your crowne TG he that shore your crowne WC but now he sorowed that PL but now he trowed that RP but nowe he sorowd that TG he trowed his iourney
Examples like these could be multiplied. Unfortunately the corresponding passages from WW are not extant for comparison, for the leaves of WW which survive correspond to a part of RP which is missing. Yet it is possible to show from the surviving leaves of WW that it reproduces a mistake made in PL and that therefore it must be a reprint of PL:
WC I wyl al otherwyse on yow yet / abyde I shal brynge PL I wyl al otherwyse oon you yet byte I shal brynge WW I wyll yet al otherwyse byte you I shal brynge TG I wyll yet all otherwyse by you I shall brynge
From the above examples it can be accepted that there is a straight-forward sequence of printing for four of the versions, viz. WC—PL—WW—TG, and that RP stands outside this sequence and is a reprint of WC.7
Throughout the eighty years covered by the survey there is no noticeable change in the haphazard use of i and y. PL differs considerably from WC in its use of these graphemes, but the changes made are purely random. Thus when the compositor was setting from WC a4r, which includes all chapter 1 and some of chapter 2, he changed i to y eight times and y to i four times. In addition e is once changed to y and i once to e.8 If anything y is used somewhat more frequently than i in PL, especially in such words as wyth etc., but both letters are used indiscriminately. There is certainly no attempt at standardization. The same state of affairs is to be found in all the later versions. Individual words are not necessarily spelt in the same way as in the copy-text, but no version shows a particular preference for one letter or the other. In the endings of the preterite of weak verbs and the plurals of nouns, however, e did become the standard spelling by the end of the period. In WC, PL and RP -yd/-id/-ed and -ys/-is/-es interchange freely, though spellings with e are not common. In WW e spellings are introduced a little more frequently; and in TG they become regular. The -yd/-id and -ys/-is forms do not survive in TG. Similarly in TG in the preterite of weak verbs the ending -ed is extended to words which in the earlier versions formed their preterites by adding -d or -de: sauourd and prayde appear as sauoured and prayed.9 The spellings with -ed are not found with any regularity before TG. Although conformity was established in this case, WC's standardized spellings were not always respected in later editions. In WC -y is always used at the end of a word. There is only one exception to this in the whole text: herbi. In PL, however, this final y is often changed to i, so that words can be spelt ending in i or y. The latest three versions also use either spelling. The development of spelling in this period was not always towards conformity.
It is well known that a final e was added or omitted indiscriminately in early printed books. The five versions of Reynard the Fox are no exception. From WC through TG there is no discernible reason for the omission or addition of the e in most instances, and, as in the case of i and y, no version agrees with its copy-text as to when final e is found or not. On the other hand, WC rarely includes a medial e before the adverbial ending -ly. PL often adds e in this position, especially after dentals and stops, so that WC's sharply, frendly and goostly appear as sharpeli, frendely and goostely. RP likewise frequently adds an e, and WW and TG reflect the orthography of PL.
In WC a and o when followed by a nasal interchange freely. This confusion is retained in PL where even the preposition on and the article an are spelt indifferently with a or o. PL often changes the spellings found in WC: songe (sange), domage (damage), and stande (stonde),10 but not systematically. In RP there is a marked preference for the o spellings: vnderstonde, londe, stondyng, except in the preposition and the article where a is common. There are insufficient examples in WW upon which to base any conclusions, but in TG an a is generally found where the copy-text has an o: stande, lande, any, husbande. Regularity is not achieved in TG, though the a spellings are dominant. The preposition, however, is spelt on. In all texts whether they use a or o, a u is frequently inserted between the a/o and the nasal when it is followed by another consonant. This change is found particularly in words of Romance origin: penaunce, commaunde, condiciouns. In PL and RP reverse spellings when the aun/oun is simplified to an/on do occur, thus PL has danger for WC's daunger. But these examples are few in comparison with those which show the change an/on to aun/oun. In TG there are examples only of a u being added; there are no words which drop a u which was found in the copy-text. The spelling aun/oun is not regular yet in TG in words of Romance origin, but this development is one of the few regular trends towards conformity which is found consistently in all the texts.
In WC words like do and see can be spelt with a final single or double vowel. PL does not differ much from WC, though some changes which are made in PL tend towards simplification of the final double vowel: doo and see appear as do and se; but go becomes goo. In RP, however, there is a very strong tendency to double all final single vowels: thus WC's be, se, go, to, do, so, nothyng appear as bee, see, goo, too, doo, soo and noo thyng respectively. The limited evidence from WW suggests that in that text whereas final oo was simplified, final ee was retained; this is the trend found also in TG. In neither WW nor TG are these spellings carried through consistently.
There is no uniformity in the use of ou or ow and au or aw in WC. In PL one can glimpse the beginnings of a tendency to use au and ou internally, as in hauthorn (hawthorn) and coude, and aw and ow finally, as in yow and now. Nevertheless there are many exceptions. RP still uses the spellings indiscriminately. WW develops the trend found in PL: it uses ou and au internally, except in the word downe, which is almost invariably spelt doun(e) in PL, and ow and aw finally, except in you where the ou spelling is regular. TG follows WW, though regularity is not achieved. But whereas in PL there are times when an internal ou or au in WC is changed to ow or aw respectively, there are no occurrences of this reverse spelling in TG so that one may perhaps suggest that a preference had evolved. The spellings ei, ey, ai, ay vary freely among themselves in all texts, except that in TG a slight preference for ay may be noted. The variation between er and ar in such words as merchant and Reynard is decided finally in favour of ar. In WC and PL either spelling is used; in RP ar is found more commonly in common nouns: marchauntes, but er is used regularly in the names of the animals: Grimberd, Reynerd; and in WW and TG ar is the regular form in all words which had previously exhibited variation. One of the most remarkable features of this study of the five versions of Reynard the Fox is how the grapheme ea appears suddenly and becomes accepted as the standard spelling in some words in such a short time. It is rarely found in the three earliest versions which use e or ee instead: grete, heed etc. In WW ea makes its first regular appearance in the word great, though it is also found sporadically in other words in WW. Otherwise WW prefers to represent this long vowel sound by an internal and a final e: it changes PL's breed, leep and feet to brede, lepe and fete. In TG the spelling ea has become almost regular in such words as teache, head, heade (“heed”), great and beast.
As for the spelling of consonants and consonant groups a tendency to conformity can be noticed in the spelling of such words as enough and through. In WC enough, for example, is spelt as inowh and inough. The beginnings of the spread of spellings in -ough is found in PL, where, although many -wh spellings are retained and isolated examples are changed to -uh: ineuh (inewh), there are frequent occasions when the -wh is altered to -ugh: thaugh, inough etc. It is noteworthy that there are no examples of the reverse spelling -ugh to -wh in PL. RP, however, shows no particular advance over WC. But in the short passage from WW extant there are several examples where PL's -wh has been changed to -ugh, and in TG -ugh has become regular. A similar trend to standardization is apparent in the use of the graphemes -tch and -dg-. These spellings are already found in WC, but they are not as common as ch and g(g): cache, juge etc. Already in PL many of the ch and g(g) forms give way to -tch and -dg- respectively: fetche (feche), pledge (plegge) etc., though reverse spellings also occur so that one cannot assume that -tch and -dg- were yet the dominant forms. RP tampers little with WC's spellings of these consonant groups, but in WW and TG -tch and -dg- become the most frequent forms, although they have not yet become the only ones.
It is not possible to trace such a consistent trend to standardization in the other consonant spellings. For example, both k and ck are used in WC. But many of the examples which have k in WC appear with ck in PL: spack (spak), dranck (drank) and stomack (stomak), whereas in RP many of WC's ck spellings are simplified to k: spak (spack) and cok (cock). In WW there is no sign of consistency: sometimes a ck is changed to k and sometimes a k to a ck: spake (spack) and ducke (doke). This variety is also characteristic of TG so that at the end of the period there is as much freedom in the use of ck and k as there had been at the beginning. Standardization did, however, begin to make itself felt in the question of whether a consonant should be doubled or not, either internally or finally. PL differs considerably from WC in its use of single and double consonants, but it does not reveal a decisive preference one way or the other. Sometimes a double consonant is simplified: vylonye (vyllonye) and april (appryl); and sometimes a single consonant is doubled: ballock (balock) and fell (fel). When a word ends in a double consonant followed by an e in WC, there is a tendency to reduce this group to the single consonant in PL: at (atte), al (alle) and had (hadde); though there are exceptions: ranne (ran). RP, on the other hand, exhibits the opposite tendency. Although internally consonants are not regularly doubled, a single final consonant is, particularly if the word is a monosyllable ending in f,...
(The entire section is 6648 words.)