University of Virginia Library


When copying from A the scribe of S made at least 85 errors in a passage which has 2398 (institutional) words in the corresponding Aversion.[16] That means that on average S has one error for every 28.2 words of A. By comparison, in A, there are 14 errors (12 of them corrected by Capgrave),[17] that is one error for every 171 words. So if all errors are considered (arbitrarily) as carrying equal weight S is 6.1 times as inaccurate as A. Evidently, purely on grounds of the frequency with which errors occur, Capgrave had much to be discontented with in S. As will be seen some of the errors had a considerable adverse effect on the sense of the text, often making nonsense of what Capgrave wrote.

For convenience I have grouped the errors in S under three main categories, with further subdivisions. These categories, Errors resulting from Miscopying (i.e. writing something other than what was in the exemplar), Errors of Addition, and Errors of Omission, have a purely descriptive basis, and the arrangement of the material here under these categories is not intended to give rise to a mechanistic approach. Errors may have the same cause whatever descriptive category they fall into. The commonest cause of the errors in S seems to be assimilation, either to what goes before (retentive assimilation), e.g. in clennesse in chastite for in clennesse of chastite (1.1.17 below), or to what comes after (anticipatory


Page 75
assimilation), e.g. Of grete hors of brasse for A grete hors of brasse (1.1.25 below). Some other errors may have been induced by the line-divisions in A. Ultimately many errors seem to be due to the intervention of the scribe's mind, which worked haphazardly and often stupidly.

1. Errors resulting from Miscopying

1.1 Alterations in Wording

On 27 occasions another word occurs in S for that used in A. Of these instances some 9 are probably due to retentive (1.1.17-22) or anticipatory assimilation (1.1.23-25) and 2 (1.1.26-27) may have been induced by the line-divisions in A.

1.1.1  mad  (364r/34; 27/6)  and  (223/9) 
1.1.2  was  (364v/8; 27/20)  shuld'  (223/31) 
1.1.3  ysider 'Isidore'  (365r/9; 29/2)  vsed  (ir/26) 
1.1.4  ston  (365r/23; 29/14)  stonde[18]   (iv/16) 
1.1.5-6  þis  (365r/41; 30/13 (1st), and 366r/11; 32/1)  his  (118r/8 and 119r/21) 
1.1.7  schul  (365v/11; 30/24)  shall'  (118r/26) 
1.1.8  clad  (365v/17; 30/29)  gl|ad  (118v/2-3) 
1.1.9  an  (365v/26; 31/6)  | and  (118v/15) 
1.1.10  þou  (365v/31; 31/10)  yow[19]   (118v/20) 
1.1.11  an 'and'  (365v/38; 31/17)  | on  (118v/31) 
1.1.12  sitter  (366r/3; 31/24)  after [20]   (119r/9) 
1.1.13  þis kyng  (366r/18; 32/6)  þese | kinges   (119r/29-30) 
1.1.14  mene 'means'  (366r/19; 32/7)  meyne 'army'  (119r/31) 
1.1.15  cam  (366r/38; 32/24)  come  (119v/25) 
1.1.16  sum   (366r/43; 32/28)  þine  (119v/32) 
1.1.17  of  (364v/27; 28/7)  in  (224/24) 
1.1.18  se  (365v/5; 30/19)  of  (118r/17) 
1.1.19  neuyr  (365v/6; 30/20)  euer  (118r/19) 
1.1.20  sche  (365v/42; 31/20)  þe  (119r/4) 
1.1.21  ʒe  (366r/26; 32/14)  he  (119v/9) 
1.1.22  on  (366r/37; 32/24)  in  (119v/24) 
1.1.23  to  (365r/25; 29/16 (2nd))  and  (iv/18) 
1.1.24  men  (365v/21; 31/2)  more  (118v/8) 
1.1.25  (366r/7; 31/28)  Of  (119r/16) 
1.1.26  conse|crate  (364v/25-6; 28/6)  contrite  (224/22) 
1.1.27  sekir | merkis  (365v/16-17; 30/29)  sekernes  (118v/2) 
Half of these instances make nonsense of the text and at least seven others substantially alter the sense of what Capgrave wrote.


Page 76

1.2 Changes within Words

In 9 instances letters within words have been altered. One of these is probably due to anticipatory assimilation (1.2.8), and one (1.2.9) was probably induced by the line-division in A.

1.2.1  rose 'Rose'  (364v/29; 28/9)  rese  (224/26) 
1.2.2  exameron 'Exameron'  (365r/11; 29/3)  epaneron'  (ir/28) 
1.2.3  guynosopistis 'Gymnosophists'  (365v/3; 30/17)  guynesopistes   (118r/15) 
1.2.4  fame  (365v/4; 30/18)  same  (118r/17) 
1.2.5  manslauth  (365v/12; 30/25)  mon [in monslaught][21]   (118r/27) 
1.2.6  naked  (365v/21; 31/2)  nakod  (118v/8) 
1.2.7  philisophr'  (365v/26; 31/6)  philiphoser'  (118v/15) 
1.2.8  senate  (366r/27; 32/15)  senade |  (119v/10) 
1.2.9  a|naximenes 'Anaximenes'  (365v/26-27; 31/7)  a maximenes  (118v/15) 

At least three of these instances make nonsense of the text.

There are also 2 instances where contractions in A have been expanded in S incorrectly.

1.2.10  prophecied  (364v/14; 27/26)  prephicied  (224/7) 
1.2.11  a mongis   (366r/25; 32/13)  a mon|ge  (119v/8-9) 

1.3 Wrong Word Order

Just as letters could occasionally be transposed, as in philiphoser' (1.2.7), so too could words. All three instances were probably the result of anticipation combined with attempted correction.

1.3.1  in þe heith a boue was a temple (364v/25; 28/5-6)  in þe heyth was a boue a | temple (224/21-22) 
1.3.2  and whi he was sette þere (366r/7; 31/27)  and why was | he sette þere (119r/15-16) 
1.3.3  þis schal be mad of brasse (366r/31; 32/18)  þis made shall' be of bras (119v/16) 

In the first instance the sense is altered.

2. Errors of Addition

Since S is held to be a copy of A there ought not to be many examples that fall in this category and those that there are have been carefully checked to make sure they do not constitute contradictory evidence. That the S-scribe was prepared to supplement what was in his exemplar is shown by the following three instances. All of them, however, are easily explained as copying errors.

2.1 Addition of Words

2.1.1  on to bettir vse  (364r/30; 27/3)  onto þe better vse  (223/5) 


Page 77

Here one idiom has simply been substituted for another.

2.1.2 The Latin word capitulum is added to a chapter heading before the number xii (A 365r/19 (29/10); S iv/8). What is implied in A is supplied in S.

2.1.3  At mydnyth loke e be redy alle in dikys and. . .  (366r/33; 32/20)  at mydnyght | loke ye þer be redy all' in dykis and . . .  (119v/18-19) 

Here the construction has been changed so that some subject other than A's e is required in S but none is forthcoming, so the sense is seriously impaired by the alteration.

There is one instance of straightforward dittography over a line division in S, possibly induced by the line division in A.

2.1.4  lich as I wil ride | Alle þis (366r/30-31; 32/18)  liche as I will' ride all' | all' þis (119v/15-16) 

Here may also be included a similar instance where only the first letter of a word is repeated over the line division in S.

2.1.5  þerfor owt  (364v/41; 28/20)  þerfore o | oute  (ir/11-12) 

In three instances words have been added through retentive assimilation.

2.1.6  Too grete hor|ses þere be and too naked men standyng be hem  (365r/20-21; 29/12-13)  To gre|te horsis þer be and to naked men | and standing' be hem  (iv/11-13) 
2.1.7  Tho hith he hem grete ricchesse and had hem in ful grete reue|rens  (365r/37-38; 30/9-10)  Tho hight he | hem grete richesse and had hem in full' grete riche|sse and reuerence  (118r/2-4) 
2.1.8  distroye a cite man & woman wal and | hous  (365v/24-25; 31/4-5)  di|stroy a cyte a man and woman wall' and hous  (118v/11-12) 

In all of these instances the sense is affected adversely.

There is one clear instance of a word being added through anticipatory assimilation.

2.1.9  hath bore a | child  (364v/22-23; 28/3)  ha | the a bore a child'  (224/17-18) 

In another instance only the first letter of a word is anticipated.

2.1.10  mor' bounde to do  (365r/11; 29/4)  more bounde d to do  (ir/29) 

Both of these last two errors are easily spotted and the sense thus recovered.

2.2 Addition to Words

Dittography of groups of letters occurs in two instances, the second possibly induced by the line division in A.

2.2.1  euene a noon  (364r/39; 27/10)  euenen anonone  (223/15) 
2.2.2  go|uernaun (365v/20-21; 31/1)  goueruernauns  (118v/7) 


Page 78

Dittography of a single letter occurs in a further instance over a line division in S.

2.2.3  philisophr'  (365v/31; 31/11)  philosop | phir'  (118v/21-22) 

Dittography of letter-strokes (minims) also occurs.

2.2.4  coupled  (364v/17; 27/28)  compled  (224/11) 
2.2.5  sumtyme  (366r/8; 31/28)  sumityme  (119r/17) 

With the possible exception of 2.2.4, which as it stands makes nonsense of the text, all these errors are easily spotted and the sense thus restored.

There is one instance of a word being added through retentive assimilation.

2.2.6  In þe tyme | of tiberius þe emperour'  (365r/24-25; 29/15)  In þe tyme of tyberus oþe emperoure  (iv/17-18) 

Here there may be an element of syntactic confusion, the scribe perhaps thinking of a reading such as * þe tyme of þe emperoure tyberus*. As it stands the text is clearly rendered nonsensical in S but again the error is easily corrected.

3. Errors of Omission

3.1 Omission of Words

In 7 instances words are omitted in S apparently through sheer carelessness.

3.1.1  was a | temple  (364v/33-34; 28/13)  was temple  (ir/1) 
3.1.2  had not a gander'  (365r/6; 28/27)  had not gandir  (ir/22) 
3.1.3  þe . . . puple . . . ded make  (365r/14-15; 29/7)  þe . . . peple . . . make  (iv/2-3) 
3.1.4  as to a god  (365r/16; 29/8)  as to god  (iv/4) 
3.1.5  euyr is þe maistiris wit a boue his disciple  (365v/34; 31/13)  euer is þe meystiris aboue his disciple  (118v/25) 
3.1.6  Thus was þe cite saued and þe kyngis ire softed  (365v/35; 31/14)  ths | was þe cyte and þe kinges ire softed  (118v/25-26) 
3.1.7  in þe eld cronicles  (366r/1; 31/21-22)  in þe cro|nicles  (119r/5-6) 

In 3.1.3, 3.1.5 and 3.1.6 the construction gives a clue to the omission but not in 3.1.7. While the omission of the indefinite article in 3.1.1 and 3.1.2 affects the grammar rather than the sense its omission in 3.1.4 alters the sense radically.

There is one instance of the second of two words ending in the same letter being omitted.

3.1.8  supposed not þat he had be of rome  (366r/43; 32/28)  supposed not he had be of rome[22]   (119v/31) 

There are 4 instances where the scribe has omitted a word through anticipation, his eye presumably moving on to the second of two words beginning with the same letter(s).


Page 79

3.1.9  ferd nevir weel ne neuyr stood (364v/38; 28/17)  ferde neuyr wele neuer stode (ir/7) 
3.1.10  þei knewe summe straunge þingis (365r/28; 30/1-2)  þay knewe straunge þinges (iv/22) 
3.1.11  þat þing whech he had not him\self/ne not myth haue (365v/10; 30/23-24)  þat þinge | whiche he had not hym selue not myght haue (118r/24-25) 
3.1.12  swieres þat wer' keperis for þe body folowed fro ferr' (366r/41; 32/26-27)  swyers þat were kepers for þe body fro ferre (119v/29) 

In a further instance of anticipation the scribe of S has omitted a word through moving on from an s at the end of a word to another s at the beginning of the next word but one.

3.1.13  as I suppose  (364v/20; 28/1)  a suppose  (224/15) 

Except for 3.1.10 all of these omissions through anticipation have a serious adverse effect on the sense.

3.2 Omission from Words

On 3 occasions letters have been omitted from the end of verb forms so as to render them morphologically inappropriate.

3.2.1  spekith  (364v/13; 27/24)  sp|eke  (224/4-5) 
3.2.2  seid p.t.   (364v/13; 27/25)  sey  (224/5) 
3.2.3  named pp.   (365v/33; 31/12)  name  (118v/23) 

On a fourth occasion the inflectional ending is omitted over a page division in S.

3.2.4  be tokneth  (365v/40; 31/18)  be tok ∥  (118v/32) 

The word these is deprived of its last two letters 3 times; on one of these occasions (3.2.7) the error was probably induced by the line division in A (see Part I above, under (2)).

3.2.5  These  (364v/17; 27/18)  The  (224/10) 
3.2.6  þese  (365v/43; 31/21)  þe  (119r/4) 
3.2.7  þe|se  (364v/35-36; 28/15)  þe  (ir/4) 

There is only one instance of omission of a letter from the beginning of a word.

3.2.8  jor|nay  (366r/28-29; 32/16)  ornay  (119v/13) 

In 7 instances letters which occur medially are omitted in S apparently through sheer carelessness.

3.2.9  sanguinis  (364v/19; 27/30)  sanguins (or (?) -gunis)  (224/13) 
3.2.10  tiberius   (365r/25; 29/15)  tyberus  (iv/17) 
3.2.11  satisfie  (365r/40; 30/12)  satyfie  (118r/7) 
3.2.12  likynesse  (365v/14; 30/27)  likenes  (118r/30) 
3.2.13-14  laterane  (366r/2, 3; 31/23, 24)  laterne[23]   (119r/7, 8) 


Page 80

On one occasion omission occurs medially through anticipation.

3.2.15  knythod  (364r/43; 27/14)  knyghode  (223/20) 

Presumably the S-scribe intended to write knyghthode but when he came to the first h he continued as if it were the second one.

Except for 3.2.12, where the effect is to substitute another word, most of the errors in this section would probably have merely checked a reader before he recognized them for what they are.[24]

From the frequency of the errors and their seriousness (at least 35 (41%) of them have an adverse effect on the sense of the text) we are clearly dealing with a good example of scribal 'negligence and rape'. If this example is anything to go by medieval authors evidently had good reason to complain about their scribes. Certainly Capgrave had ample reason to be dissatisfied with S and to reject it. Even when allowance is made for the possibility that there was a mutual understanding between author and scribe that the former would correct the latter's work—as indeed in a sense he did, by rejecting it—if the scribe of S had not been working in Capgrave's scriptorium under the eagle eye of a meticulous author[25] the copyist's work might well have gained currency. Unless an author was supremely vigilant the distortions and corruptions could set in under his own roof.

The present instance of scribal inaccuracy may provide some pointers as to the nature of primary errors and their relative frequencies. Allowance should be made, however, for the possible idiosyncracies of the particular individual scribe. Perhaps the most notable feature of the errors is the high proportion due to retentive or anticipatory assimilation, some 16 (18.8%)—1.1.20-28, 1.2.8, 2.1.6-10, 2.2.6. By contrast there are only 5 errors (5.9%) due to anticipation, where the copyist's eye has moved forward inadvertently to the second of two words beginning with the same letter(s) (or to the second of two instances of the same letter).[26] A second notable cause of errors is line divisions: some 6 errors (7%) may be due to line divisions in the exemplar (1.1.26-27, 1.2.9, 2.1.4, 2.2.2, 3.2.7), and a further 3 errors (3.5%) occur over line divisions (or a page


Page 81
division) in the copy itself (2.1.5, 2.2.3 and (over a page division) 3.2.4). Taken together these two causes account for over one third (35.3%) of all errors. Thirdly, errors often occur in proper names; there are 7 different instances, one of them occurring twice, making 8 altogether (9.4%)—1.1.3, 1.2.1-3, 1.2.8, 3.2.10, 3.2.13-14. A partial cause of at least some of these errors is presumably ignorance on the part of the copyist. Ultimately several of them, and many others as well, must be the result of the intervention of the scribe's mind, albeit working haphazardly and often stupidly. It is hard to assess what proportion of the total number of errors are attributable to this cause but the figure could be quite high, perhaps 25-30%. Although 17 errors are classified under Errors of Addition, only one of these instances seems to be the result of a conscious attempt to alter the text so as to 'improve' it (2.1.3). This feature of the copyist's work suggests that while even a copyist working in an author's scriptorium scarcely had a modern scholar's respect for the authorial text, additions due to a conscious attempt to 'improve' the text probably constitute a very small proportion of primary errors.