previous next

Errors of Insertion

Insertion of a gloss

The sense of the text was in MSS. often elucidated by writing an explanatory word, called a gloss,1 over the word of the text which required explanation, or occasionally in the side margin. When an omission in the text had to be corrected the same course was taken. The supplied word was written over the word before which it had to be inserted. It might also be written in the margin, in which case a couple of dots were generally put at the place in the text where the insertion was to be made. Small wonder then that a copyist was often at a loss to know whether an interlinear or marginal jotting should be treated as an explanation or as a supplement. There was still a third possibility, that it might be a correction designed to take the place of something wrongly written in the text; for although the practice of careful scribes in this case was to place dots2 under the wrongly written word, syllable, or letter, this precaution was often omitted. A good example of the two wrong treatments of a gloss is found in Cas. 517. The line should read: “cúr amem me cástigare, id pónito ad compéndium.” “why do I like punishing myself? Well, we may dispense with the explanation.” In P the words cur amem me were wrongly written curam eme, and this phrase was in the original of BVEJ explained by the suprascript gloss curam exime. The scribe of the original of VEJ mistook the gloss for a supplement, and wrote curam eme curam exime castigare etc.; the scribe of B mistook it for a correction, and wrote curam exime castigare etc.

A gloss may have been wrongly inserted from the margin in

nam míhi haec meretrix quae híc habet, Phronésium,
suum nómen omne ex péctore exmovít meo.

These lines are in the minuscule MSS. followed by an unmetrical line which most editors regard as a marginal gloss of the original: “Phronesium, nam phronesis est sapientia.

Epid. 382 sqq. seems to be a very similar passage: “non óris causa módo homines aequóm fuit
sibi habére speculum, ubi ós contemplarént suom,
sed quí perspicere póssent cordis cópiam,
” where the minuscule MSS. insert between possent and cordis copiam the words cor sapientiae (a corruption of cor sapientia est; see ch. vii. § 5), igitur perspicere ut possint.

Glosses in the form of small explanatory words written between the lines

The insertion of interlinear glosses is especially common. The minuscule MSS. have over and over again inserted in the text small words like ego, tu, ut and the like, which originally were written between the lines, and served the purpose of facilitating the reader's comprehension of the construction of the passage. Mil. 702 stands thus in A: “ istam semel amíseris
líbertatem, haud fácile in eundem rúsum restituís locum;
” but in the minuscule MSS. the pronoun te, originally written between the lines to explain the construction of restitues, has found its way into the text—though it still preserves a trace of its adventitious character in that it stands not before but after in:haut facile in te eundem rusum (rursum) restitues locum.” An example from the original of CD is

bene cónfidenterque ádstitisse intéllego,

where te stands in CD between confidenterque and adstitisse, and spoils the metre of the line. Another is

est etiam hic óstium
aliúd posticum nóstrarum harunc aédium,

where, for aliud of AB, in the original of CD stood aliud autem, again spoiling the metre. The Ambrosian Palimpsest affords a good instance of the insertion of a gloss, in the narrower sense of the word “gloss” (i.e. the explanation of a difficult word), in

noctem ín stramentis pérnoctare pérpetim,

where it ends the line, not with perpetim, as P does, but with perpetim totam, the second word being clearly an explanation of the O. Lat. adjective perpetim (class. Lat. perpetuam).

Mention is made of this kind of corruption as early as Galen (xvii. 1 p. 909), who says that the words οἷον ὀδύνη ὀδύνην παύει are not in the MSS. κατὰ Διοσκορίδην of Hippocrates: φαίνεται μὲν γὰρ ὡς ἐχηγήσει προσγραφὲν ὑπό τινος, αὖθις δὲ εἰς τοὔδαφος ὑπὸ τοῦ βιβλιογράφου μετατεθεῖσθαι.

One of the simplest instances is the insertion of the interjection o, which was commonly written above a vocative as a mere indication of the case. This interjection, usually written with an accent-stroke to indicate the length of the vowel, is the origin of our sign of exclamation (!). It occurs in B at Cist. 727, etc.

Miswriting left uncorrected

Another common cause of insertion is that a word or syllable was wrongly written by a copyist, who then left his miswriting uncorrected (cf. ch. ii. § 2). When a scribe wrote a word wrongly, he would, unless he were an exceptionally conscientious scribe, leave the word as it stood, neither erasing it with a penknife, nor deleting it with strokes through the letters, nor—the usual method of indicating an error in MSS.—putting dots under (sometimes above) the wrongly written letters. He left all that to the “corrector” (p. 41), the senior who revised each MS. after the scribe had finished it, and whose duty it was to make the necessary corrections and supply defective punctuation. A wrongly written word or syllable which escaped the notice of the “corrector” would thus be left to stand side by side with the same word or syllable properly written

There is a curious instance of this in the Nonius MSS. All our MSS. of Nonius have at the end of the paragraph on Mictilis (p. 137 M.) the inexplicable word meri. The next paragraph is headed by the word Maestaret, written in the MSS. mertaret; and the explanation of the intrusion of meri is simply this: — The scribe of the archetype, who read mestaret of his original as mertaret, changed his mind as he was copying the fourth letter and determined to write the word with rt in ligature, since the word in his original was written with a ligature — the ligature st, which in MSS. is often hardly distinguishable from the ligature rt (ch. vi. § 1). Thanks to his negligence in leaving his unfinished word as it stood, with the fourth letter wanting its cross-stroke, this adventitious meri has kept its place in all subsequent copies, to the bewilderment of editors.

In Mil. 203, where A reads rightly “écce avortit: níxus laevo in fémine habet laevám manum,” the minuscule MSS. preserve an error of some early scribe, who first wrote autem for avortit, and then, immediately discovering his mistake, did not expunge autem but merely added avorlit in its proper form. They read: “ecce autem avortit nisus leva in femina habet levam manum.” Before the discovery of the Ambrosian Palimpsest editors corrected the manifest corruption leva in femina to laevo in femine, but did not venture to remove autem, preferring to explain the scansion of the line by an unlikely theory that avortit might be pronounced as a disyllable in Plautus. In Mil. 187 the word written correctly after the miswritten word has unfortunately become assimilated to it, and so we find in all the minuscule MSS.: “ut eum qui hic se vidit verbis vincta vinetane esse viderit,” where A gives the true reading: “út eum, qui se hic vídit, verbis víncat, ne is se víderit.

In Truc. 289 the error is harder to detect, because it is only a syllable which has been wrongly written and allowed to stand. The syllable fo of foras, in the phrase ad foras, has been first written eo (on the confusion of f and e see ch. vi. § 1), so that ad foras (for which A gives, probably rightly, ad foris) appears in the minuscule MSS. as adeo foras.

This is an error whose ravages in the text of Plautus, and indeed of all Latin authors, have not, I think, been sufficiently realised. In Pseud. 1187 the reading of P,quid domino quid socii quid somniatis mea quidem haec habeo omnia,” was taken by Ritschl as proof that two lines had been fused into one by some such error as caused the fusion into one of Men. 165-6 and 278-9, already quoted (p. 52). But it is much more naturally explained as a case of a miswriting, quid socii for quid somniatis, which was not corrected, but was left to stand side by side with the true reading. Unfortunately we cannot decipher the whole line in A, but its beginning is plain: “QVIDDOMINOQVIDSOMNIA,
” and there is little doubt that the true form of the line is: “quíd? domino? quid sómniatis? méa quidem haec habeo ómnia.


A special phase of this error is known as Dittography, where the mistake which has been left uncorrected consisted in the writing of a word twice. The word inde in Capt. 490, for example, is wrongly repeated in OJ, inde inde; the syllable te of advorte (Pseud. 277) in A, which reads advortite (cf. ch. v. § 8). Dittography is, however, not nearly so common an error in MSS. as haplography (ch. iii. § 1). The commonest case is the repetition at the end of a word of the letter that begins the following word, e.g. quissim for qui sim (quisim). A complicated example of dittography in the ancient MS. of Cicero de Republica is SECUTUTUSECUTUS for secutus (ii. 33. 57).

Dittography of a word (or letter) most naturally occurs at the end of one page and the beginning of the next. Two MSS. of the Liber Glossarum reproduce not merely the pagination of their original but also this error, in inteatro for in theatro, where the one in stands in the copies, as it did in the original, at the end of one page, and the other in at the beginning of the next (see Goetz Lib. Gl. p. 223). Dittography of a letter is also a common occurrence in the transition from one line to another, e.g. Corneliaana for Corneliana in the Leyden MS. of Nonius (188 M. 24), where the three last letters stand at the beginning of a new line.

Insertion of any marginal jotting

Finally, it should be noticed that not merely a gloss but any jotting in the margin (or elsewhere) may find its way into the text. Thus the words huc usque, which often appear in the margin of MSS.—sometimes to show how far a “corrector” (or copyist) has gone, sometimes as an indication of the end of a quotation or special passage in the text3 — have been repeatedly inserted in the text of a thirteenth-century MS. of Aulus Gellius (see Hertz's Introduction p. lvii).

Caput (also written Kaput, or merely .c.), a marginal adscript indicating the beginning of a new paragraph, has found its way into the text of our MSS. of Nonius Marcellus at p. 48 M. 27, where a paragraph ends: idem populus caput. The Leyden MS. of Nonius ends a paragraph (p. 108 M. 9) with oppido .c. (see Philologus 1896 p. 167).

Other marginal adscripts of common occurrence in MSS. are: r[equire],4 qu[aere], d[cest], nota (often written no with a stroke above, like the contraction for nomen and occasionally for non, p. 76), nota bene, and expressions of admiration like mire, optime. Index words, showing the contents of paragraphs, were often written in the margin. For example, in the margin of the Laurentian codex of Nonius Marcellus, opposite the paragraph (86 M. 10): “Caecuttiunt. Varro Gerontodidascalo: “utrum oculi mei caecuttiunt? An ego vidi servos in armis contra dominos?”” is written the index-heading Cecuttiunt, lippíunt. In the Harleian copy of this MS. this marginal adscript has been treated as if it formed the first part of the line (beginning with oculi) opposite which it stood in the margin, and the passage is copied in this way: “Caecuttiunt. Varro Gerontodidascalo: “utrum caecuttiunt lippiunt oculi mei caecuttiunt” etc. The same Harleian MS. has sometimes incorporated in the text the marginal adscript quaere (written in contracted form) of its Florentine original. The quae inserted curiously into sentences (e.g. 107. 27 “incideret quae in mortis malum”; 114. 25 “Cicero Tusculanarum quae lib. v”) in the copy is nothing but a misreading of this contraction (cf. p. 96 below).

On the insertion of a syllable in cases of substitution like considero for consido, see ch. v. § 8.

List of Examples

Additional examples of insertion:—

(1) Of gloss:

(2) Miswriting left uncorrected:

(3) Insertion of marginal jotting:

Pseud. 445SIMO. Quis hie lóquitur? meus hic ést quidem servos Pseúdolus.” For quis we find in the minuscule MSS. síquis, the si being taken from the “nota personae” Si(mo). (For other examples see Leo's note on Poen. 474.)

(4) Dittography:

The Laurentian MS. of Nonius supplies us with an instance of dittography like that mentioned on p. 32. The scribe has miscopied the se loco potuerant (107 M. 22) of his Leyden original as se loco se potuerant, with dittography of se.

I suspect that a gloss has been inserted in the following passages:

In the following I see (with Spengel) a miswriting left uncorrected:

1 Glossa (Gk. γλῶσσα), sometimes glosa, properly meant “a difficult word” (cf. glossema), but is often used to mean “the explanation of a difficult word.”

2 Hence our word “expunge.”

3 Here are some instances of the phrase. In a Bodleian tenth-century MS. of Bede (Laud. Misc. 159) the words usque hic requis[itum] est stand on the margin of fol. 71 r; and no corrections or glosses occur on the following pages. Similarly a tenth-century MS. of St. Augustine in the Vatican Library (Pal. Lat. 202) has on fol. 73r, in the top corner of the page, usque hic, and on fol. 175v, at the foot of the page, huc usque relegi. A Monte Cassino MS. (No. 494), containing a life of St. Remigius, has on fol. 57 v usque hic scripsi. We have the other use of huc usque in D in the Pseudolus, where at the first line of the letter of Phoenicium (v. 51) there is in the margin Epistola, and at the last line (v. 73) huc usque.

4 R. for require is often found in the margin opposite a corruption in the text, whether placed there by the scribe himself, by the corrector, or by a subsequent scribe who made a copy of the MS. It has, however, other uses. Thus in a Bodleian MS. of Sidonius (Hatton 98), opposite laudibus imperatoris of the text, we find in the margin (fol. 118r) require hujus imperatoris nomen; opposite Brictanos (sic) of the text stands in the margin (fol. 118v) require de Britannís.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 3.0 United States License.

An XML version of this text is available for download, with the additional restriction that you offer Perseus any modifications you make. Perseus provides credit for all accepted changes, storing new additions in a versioning system.

hide References (26 total)
  • Cross-references from this page (25):
    • Plautus, Mostellaria, 2.2
    • Plautus, Poenulus, 2.1
    • Plautus, Pseudolus, 1.3
    • Plautus, Pseudolus, 1.5
    • Plautus, Pseudolus, 4.3
    • Plautus, Pseudolus, 4.7
    • Plautus, Stichus, 3.1
    • Plautus, Trinummus, 2.2
    • Plautus, Trinummus, 3.2
    • Plautus, Amphitruo, 1.prol
    • Plautus, Amphitruo, 2.2
    • Plautus, Captivi, 1.1
    • Plautus, Captivi, 3.1
    • Plautus, Casina, 3.1
    • Plautus, Epidicus, 3.3
    • Plautus, Menaechmi, 1.2
    • Plautus, Miles Gloriosus, 2.2
    • Plautus, Miles Gloriosus, 3.1
    • Plautus, Truculentus, 1.1
    • Plautus, Truculentus, 2.2
    • Plautus, Truculentus, 2.4
    • Plautus, Truculentus, 2.7
    • Plautus, Truculentus, 3.1
    • Plautus, Truculentus, 4.4
    • Cicero, De Republica, 2.33
  • Cross-references in notes from this page (1):
    • Plautus, Pseudolus, 1.1
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: