University of Virginia Library

Search this document 


expand section 
expand section 
collapse section 
expand section2. 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 
expand section 

expand section 


Besides the S-scribe's alterations to the A-text which constitute errors there are a large number of non-erroneous alterations of the kind that are usually referred to as 'purely scribal'. Here we are dealing exclusively with forms resulting from the intervention of the scribe's mind, albeit in the main working mechanically and sometimes haphazardly. We still need to learn more about such linguistic alterations; what is known has had to be deduced from collating and studying different versions of texts that survive in a fairly large number of manuscripts. As A. Hudson says, in the only detailed study of this kind (concerned with non-textual variations) known to me, 'Ideally one needs for a check a manuscript copied from a known exemplar'.[27] The present material provides precisely such a check, one that is all the more ideal because of the circumstances in which the copy was made—from an authorial copy in the author's scriptorium. It is hard to imagine circumstances in which the external pressures on a scribe to conform to his exemplar might be greater.[28] However, by the third quire (from which the


Page 82
surviving part of S's work comes), since 'having got into his stride, a scribe ceases to copy the dialect of his exemplar, but translates the text into his own',[29] the S-scribe may have come to terms to some extent with these pressures, such as they were.

In Hudson's terminology the linguistic alterations made by copyists may be 'dialectal' or 'orthographic' ('Tradition and Innovation', p. 371). The basis of this dichotomy derives from the distinction made by the Latin grammarians (who got it from the Greeks) between two of the three attributes of a letter, its figura or written form and its potestas or phonic significance. (The third attribute of a letter is its nomen or name.) Attention was drawn to the importance of this distinction by D. Abercrombie, but there seems to be no basis for his claim that 'it is not easy to discover the relationship between figura and potestas'.[30] The clearest statement I have found is that by the sixth-century grammarian Audax:

Figura litterae quae est? Qua notatur. Potestas quae? Qua in ratione metrica valet, cum aut producta est aut correpta. Et . . . figuram oculis deprehendimus, potestatem mente cognoscimus.[31]
This statement is more or less the same as those by the earlier, fourth-century grammarians Victorinus and Diomedes,[32] except for the addition of the last sentence. In this sentence the distinction between the physical existence of the written form and the mental knowledge of its phonic value is a remarkable anticipation of Saussure's famous distinction between parole and langue.

The difference between 'dialectal' and 'orthographic' alterations is that whereas a 'dialectal' change involves the substitution in the copy of figura B with potestas B for the exemplar's figura A with potestas A, i.e. there is a change in phonic significance (which would give rise to a


Page 83
different pronunciation) as well as in spelling, an 'orthographic' change involves the substitution in the copy of figura B with potestas A for the exemplar's figura A with potestas A, i.e. the phonic significance remains the same despite the change of spelling. An example of the first process, "dialectal" change, would be whech > which where the figura 'e' with the potestas /ε/ is replaced by the figura 'i' with the potestas /I/. An example of the second process, 'orthographic' change, would be þe > the where the figurae 'þ' and 'th' both have the potestas /ð/. Exactly the same distinction is made by A. McIntosh, except that he would call instances of 'dialectal' change 'spoken-language features (S-features)' and instances of 'orthographic' change 'written-language features (W-features)'.[33] Terminology is a problem. Since many features illustrative of 'orthographic' change can show orderly regional distribution the use of the term 'dialectal' in opposition will inevitably lead to misunderstanding.[34] If 'spoken-language features' is used instead then the Saussurean distinction between a mental knowledge of a figura's phonic value and the articulated sound is blurred. The terms which offer the greatest potential for clarity are the Latin ones, figura and potestas, partly because there is no single English word that will serve as an adequate translation of the latter. I shall therefore refer to the first kind of alteration as FP-changes, ones where both the figura (or form) and the potestas (or phonic significance) are altered, and to the second kind of scribal alteration as F-changes, ones where the figura is altered while the potestas remains the same. With this distinction in mind I propose to examine the evidence.

Nothing is known about the S-scribe except that he worked in Capgrave's scriptorium at Lynn. It is probably a fair presumption that he came from somewhere not a hundred miles from Lynn. But it cannot be assumed automatically that his dialect was Lynn dialect. Indeed, in view of the dearth of external information, the best guide to the S-scribe's dialect may be the linguistic data. When analyzed alongside comparable unpublished data from other sources the dialect-features of


Page 84
S would appear to harmonize with those attributed to the northern part of Cambridgeshire (Isle of Ely) north-west and west of Wisbech,[35] an area about fifteen miles from Lynn. This provisional placement, which is compatible with the external information, is based both on the features retained and on those altered in the process of being transferred from A to S.

The argument for this dialect placement is as follows:

(1) The form erde 'earth', in co-occurrence with -th for the pr. 3 sg. indic., indicates the general area to be East Anglia.

(2) The absence of spellings with q- for the reflex of OE hw-, and of spellings such as xal 'shall', probably indicates an area west of the R. Ouse (but Lynn could hardly be excluded).

(3) Given that the forms of the pr. 3 sg. indic. end in -th, the dominant use of spellings with a in 'they' (þai, þay, etc.; see Appendix, §86), all of them alterations from the exemplar, restricts the area of likely provenance to a small sector of the overall domain of erde, and only the western half of this sector is consistent with (2).

Taken together the first two factors, with a little help from the third, indicate a small area of N. Ely and NW. Norfolk bounded by Wisbech and Downham Market in the south and Lynn in the east. Narrowing the area down still further is, however, more difficult.

(4) Within the small area of N. Ely and NW. Norfolk already delineated the form elde 'old' would indicate either N. Ely or Lynn. Unfortunately elde occurs only once in S, where it could be relict from A.

(5) Given (4), there are a number of regular changes made by the S-scribe which suggest N. Ely rather than Lynn: A but > S bot (Appendix, §5), A -th > S -ght (Appendix, §35), A sey > S say (4x, against retention of sey(e) 8x, Appendix, §73.1-2), A seyd/seide > S saide (5x, against retention of seide 7x, Appendix, §73.3-4), A sch- > S sh- (Appendix, §§75.1, 76), A worchip > S worschip (Appendix, §75.3), A strength > S strengh (Appendix, §79). If this analysis is correct one or two forms may be relict, notably S meche < A mech, which occurs once only.

The second part of this argument, (4)-(5), may well be right but seems to me to be rather less secure than the first part, (1)-(2) plus (3).

Since in the present state of our knowledge of Middle English dialectology it is not always possible to determine whether a change made by the S-scribe is of dialectal significance or not I have listed in the Appendix all the linguistic alterations made by the S-scribe when copying from


Page 85
A. This complete listing is of value only because S is a direct copy of A and because of the circumstances in which the copy was made. In what follows references are to the numbers of the sections in the Appendix.

The list in the Appendix includes over a thousand linguistic alterations made by the scribe of S when copying A, approximately one change to every 2.3 words. Not all of these changes are consistent—for example, S has cronycles for A's cronicles (§40.1.3) and cronicles for A's cronycles (§40.2.3)—and some are not sufficiently frequent to indicate a pattern. A few lead to the partial loss or partial blurring in S of distinctions that are maintained in A. For example, S occasionally has þus for A's þis (§87.3) and sometimes þis for A's þus (§88.2), so that A's distinction between the demonstrative adjective/pronoun and the adverb is effectively lost; similarly S occasionally has hem for A's him/hym (§38) and occasionally hym for A's hem (§85.2) so that A's distinction between the singular and plural of the oblique case of the third person pronoun is blurred; and by sometimes writing hir/hir' for A's her'/here 'their' (§84.2) the scribe of S blurs the distinction in A between hir 'her' and her 'their'. Most of the linguistic alterations, however, are consistent (or nearly so) or indicate predominant trends. It is notable that a scribe whose work was 'rejected' for its textual inadequacies did make a number of linguistic alterations with a fair degree of consistency. These changes fall into three groups:—

  • (1) Changes which are of obvious potential dialectal significance.[36]
  • Here belong FP-changes like A cherch > S chirche (§6), A first > S frist (§34), A puple > S peple (§63), A whech > S whiche (§93), and possibly A but > S bot (§5), and F-changes like A sch- > S sh- (§§75-76), and probably the elimination of spellings without gh in words like MnE might (§35), though this could be an FP-change.
  • (2) Changes which may or may not be of dialectal significance.
  • In this category come a number of alterations, such as writing ai/ay for A's ei/ey [38 examples in §§1.1.3, 1.1.5-6, 86.1-2, 86.3.1, against only 1 of the opposite (ey for ai, §1.1.1) and that in a word, MnE master, which is subject to other variation (§1.3.1)], the elimination of double-vowel spellings for ME /ε:/ and /e:/ (§11) and for ME /&c.nv;:/ and /o:/ (§55), writing -ed for A's -id/-yd (never vice versa, §§17, 19), writing -er for A's -ir/-yr [23 examples in §§26.1.1, 26.2.1, against only 1 of the opposite (-ir for -er', §27.2.2)], a preference for -es for A's -is/-ys [19 examples in §§30.2, 30.3.1, against 5 of the opposite (-is/-ys for -es, §30.1)], writing -ith or -yth for A's -eth (§32.1), writing -ing for A's -yng except after letters composed of minims (§§44-5), writing qu for A's qw (§67), a preference for more frequent use of þ, especially final -þe for A's -th (§81), and writing y- for A's ʒ- (§97). When considered together the alterations made to several individual words may also indicate a pattern,


    Page 86
    one of preference for eliminating the 'unusual': for example, the scribe of S eschews ch in MnE besiege (§9) and ew in MnE endure (§25), prefers to include g in MnE assign (§43) and i in MnE reign (§71), adds final d to A's form of MnE should that lacks it (§76.2.1, but cf. §76.2.2), omits w in MnE not (§54), writes preve, the commoner ME form, rather than prove (§66), and dwelt, the commoner p.t. form, rather than dweld (§18.2), and does not spell MnE with(-) with ʒ (§95.1). The elimination of th-spellings for MnE -ght and in MnE little (§80) may also conform to this pattern. Presumably sum was preferred for MnE some to distinguish it from summe 'sum' (§77.1), words not distinguished in A.[37] Most of the alterations in this section are F-changes.
  • (3) Changes which do not appear to be of dialectal significance.
  • Even if some of the alterations included here turned out to be of dialectal significance it would not affect the general point that there are scribal changes which justify separate treatment in such a section. The dominant theme of these changes is a preference for 'fuller' spellings. This theme is illustrated by four main features:—
  • (a) The expansion of abbreviations and contractions.
  • Examples are A & >S and (§2.1), A con- > S con- (§8), A -er > S -er (§26.3), A -er- > S -er- (§29.1), A -es > S -es/-is (§30.4), A -ing/-yng > S -ing'/-yng' (§44.1.2, 44.2.3, 45.2.2), A -m(-) > S -m(-) (§48.1), A -n(-) > S -n(-) (§50.1), A -nne > S -nne (§52.1.2), A -ous > S -ous (§60), A par- > S par- (§62), A pri- > S pri- (§64), A pro- > S pro- (§65), A special > S special (§78), A þt > S þat (§82.1), A -ui- > S -ui- (§92), altogether 112 examples. The predominant trend is confirmed by the S-scribe's treatment of Latin quotations and names: there are 18 instances of the expansion of contractions in Latin words, e.g., A Sanctus Nicholaus in Carcere > S Sanctus Nicholaus in Carcere. Against this trend are 45 examples of the opposite (an abbreviation or contraction being substituted for a full spelling in A, §§2.2.1, 26.1.2, 26.2.2, 27.3, 29.2, 30.5, 48.2, 50.2, 52.2.2, 69, 77.2.2, 82.2, 87.2, 88.1, 94, 95.1-2) but 23 of these occur at the end of a line and were presumably induced by limitations of space.
  • (b) The addition of final e.
  • There are 215 examples of this feature (§13 and cross-references) against a mere 9 of the opposite (omission of -e, §§15, 77.2.1, 99.2.1) 1 of which occurs at the end of a line and was presumably induced by limitations of space. This feature alone accounts for alterations to 9% of all words in the text.
  • (c) The substitution of -e for a final flourish.
  • There are 38 examples of this feature (§§47.2.1, 47.5, 68.3) against 14 of the opposite (writing a final flourish for -e, §§14, 47.6, 84.2.3) 2 of which occur at the end of a line and were presumably induced by limitations of space.
  • (d) The addition of a final flourish, ? for -e.
  • There are 73 examples of this feature (§§12, 27.1, 47.1.2, 47.4, 51, 58, 68.1.1, 95.2-3) against 19 of the opposite (omission of a final flourish, §§27.2, 68.1.2, 84.1.1, 84.2.1) 1 of which occurs at the end of a line and was presumably induced by limitations of space. Since the flourish could in some cases be an otiose stroke its inclusion here as a 'linguistic' alteration is perhaps questionable, but the feature does seem to be part of a general trend; to treat it as a


    Page 87
    palaeographical (or, as McIntosh ('Scribal Profiles', p. 222) would call it, 'graphetic') feature (or sometimes as a paleographical feature) would be to separate evidence that belongs together.
  • The preference for 'fuller' spellings is also indicated by a few relatively minor features, A betokneth > S betokenyth (§16), A -ion > S -ioun (§46.1), A my > S myne (§49), A whan > S whanne (§52.1.1), 8 examples in all; also a glance at the spellings for MnE should (§76) confirms that although four different forms are employed in S the form substituted is always 'fuller' than that in A. Against these minor features may be set 10 examples of -s for A's -es/-is (§30.6) and 1 of within for A's withinne (§52.2.1); the relatively high incidence of the former (against the predominant trend) would suggest that the alteration of A's -es/-is to -s may have had dialectal significance. A notable feature of these instances that run counter to this predominant trend is the high proportion of instances (27 out of 87, or 31%) that were presumably induced by limitations of space. Some other features are also presumably the result of limited space being available, especially A emperouris > S emperoris (§59), and A þe > S þe (§83.3.1). In view of the preference for avoiding double-vowel spellings (§11) the isolated spelling þee (§83.3.2), though recorded once in W. Norfolk, may well be the result of a desire to fill the space between the end of the word þe and the end of the line. Even as apparently elementary a consideration as the space available at the end of a line can affect the spelling of a word.
  • All the alterations that belong in this section are F-changes.

Hudson concluded that 'the forms which are unaltered [should be accepted] as part of the scribe's written dialect', and that, as a general rule, 'a scribe will only copy the linguistic forms of his exemplar when they form part of his orthographic "vocabulary"' ('Tradition and Innovation', pp. 371-372), i.e. he will make FP-changes, ones where in his view the phonic significance of a form in his exemplar does not accord with his spoken dialect, and also, though not always regularly, F-changes, ones where the forms in his exemplar do not accord with what would have been his regular orthographic practice had he been writing without the 'distraction' of the exemplar; for it is implicit in her conclusions that a scribe was prepared to perpetuate in his copy forms from his exemplar which, though within his linguistic competence, he would not normally have used in linguistic performance. Analysis of the material presently under scrutiny confirms these conclusions. It also yields information which allows more precise statements, statements which may be described as cautionary, about the two kinds of linguistic alteration, though allowance must be made for the partial nature of the evidence. In this case FP-changes form a very small proportion of the total number of linguistic alterations; this finding would seem to be only partly accounted for by the fact that the dialect-area of S is not all that far removed from that of A. Of the remaining F-changes a large proportion (over half the total number of linguistic alterations) do not appear to be


Page 88
of dialectal significance. Recently A. McIntosh has asked 'How far does a scribe, seeking to eliminate alien S-features [i.e. make FP-changes] in a text he is copying, take the further step of eliminating alien W-features [i.e. making F-changes]?' ('Inventory', p. 605, n. 1). From the present evidence it would appear that, even when allowance is made for the dialectal proximity of S to A, the S-scribe's concern was more with F-changes than with FP-changes and that he could not have looked upon the making of F-changes as a 'further step'. Indeed he could hardly even have distinguished between FP-changes and F-changes. In other words, the scribe was more concerned with textual transmission than with dialectology. Obvious as this conclusion may seem it is vital that it is not overlooked.