previous next


The Text of Plautus—its special value for the study of Textual Emendation

There is no Latin author the study of whose text has at once such interest and such value for students of textual emendation as Plautus. For the text of Plautus is on the one hand not nearly so certain as the text of Virgil, of which we have some half-dozen complete or fragmentary MSS. dating from the third to the sixth century, nor on the other so hopelessly uncertain as the text of Propertius, of which no MS. exists that is older than the thirteenth or fourteenth century. It is still full of difficulties, in spite of the labours of a large number of scholars for a large number of years, though each month—I might almost say each week—sees a difficulty removed; and now that we have at last a full collation1 of all the important MSS., we may hope to attain before long to a completely satisfactory text.2 The study of the text of Plautus has thus all the fascination of a problem which has not yet been solved, but which evidently can, and sooner or later must, be solved. Even an untrained student may at any moment by an ingenious conjecture remove a difficulty, and thereby open the way to the resolution of a score of similar problems.

Valuable minuscule manuscript of Plautus

And the text of Plautus offers peculiarly useful material to the student of textual criticism from the following reason. It is for the larger part, like the text of most Latin authors, dependent on minuscule3 MSS. of the tenth and eleventh centuries. Now one of these minuscule MSS. of Plautus has for a great portion of the text extraordinary value. It has preserved with rare fidelity the actual text of the archetype, leaving corruptions as they stood, with scarcely any attempt to remove them. It thus throws a wonderful amount of light on the course taken by corruptions in minuscule MSS., as an example or two will show. In the Pseudolus v. 1041 for te nunc the other minuscule MSS. of Plautus read lenonem. Ballio, a leno,” is one of the characters in the Pseudolus, and the word leno is of frequent occurrence throughout the play; but it is plain that the sense of this line requires te nunc, and that lenonem must be a corruption of these two words. The change seems a violent one, and it does not at first sight appear how we could justify such an emendation as te nunc where MSS. read lenonem. A glance at the good codex shows us the intermediate step between the two readings. It has lenunc, having faithfully preserved the miswriting of the archetype—a miswriting not uncommon in MSS. (ch. vi. § 1)—of l for t. Correct this single letter, and the line reads smoothly and metrically:

Macédoniensem quí te nunc flentém facit.
The writer of the copy from which the other MSS. are derived, trying to emend the obvious corruption lenunc, succeeded only in effacing all trace of the true reading. Again, in Pseud. 267 the other MSS. offer an impossible ending of a trochaic tetrameter, dextram: lucri quid détur, potius rém divinamdextram.” The true reading is deseram, as we learn from the good codex, which has destram, t having been substituted for e in the archetype—a substitution which probably dates from a time when the text was written in capitals. Other instances from the Stichus are: v. 573 possit for opus sit, where the good codex has opos sit, this being probably a trace of the old spelling4 of Plautus' time; v. 192 nive repleverit for ni vere perierit, where the good codex has ni vere perlerit. Other examples with the same wrong division of words may be seen in the Persa: v. 587 aequo mihi curat for aequom hic orat, where the intermediate stage is aequo mihi corat; v. 546 qui aspexi equidem for quia specie quidem through qui aspeci equidem. These examples, which might be multiplied indefinitely, may suffice to illustrate the part played by this codex, which so faithfully interprets for us the puzzling varieties of reading that are found in its fellow MSS. The text of most Latin authors depends on minuscule MSS. precisely similar to the ordinary minuscule MSS. of Plautus, but an “interpreter” codex is usually lacking; and that is why a study of the Plautine text is so valuable a training for the emendation of other Latin writers. The MSS. of these will offer us hundreds of readings like lenonem for te nunc, dextram for deseram, aequo mihi curat for aequom hic orat, leaving us without the slightest clue to the origin and course of the corruption. And yet it is a palmary rule of textual criticism that until we can indicate how and why a proposed reading was altered to the reading of the MSS., our emendation cannot be satisfactory or convincing.

Ambrosian Palimpsest in capitals

There is still another feature of the Plautine text which makes a study of it peculiarly valuable training for textual emendation. Unlike the texts of most Latin authors, it is not dependent on minuscule MSS. alone, but has for a considerable part of the plays (for almost the whole indeed of four plays, the Persa, Poenulus, Pseudolus, and Stichus) a MS. in capital letters of the fourth century, the famous Palimpsest of the Ambrosian Library at Milan. All minuscule MSS. have, of course, originally come from capital or uncial texts; and a comparison of the minuscule and majuscule texts of these four plays shows us the nature and extent of the corruptions which a text would commonly suffer in its transmission from majuscule to minuscule form. Thus from a variety of reasons the MSS. of Plautus are capable of teaching us more about Latin textual criticism than those of almost any other Latin author.

It was the discovery of this ancient MS. of Plautus, the Ambrosian Palimpsest, which opened the way to the scientific study of the Plautine text, as at present conducted on the lines laid down by Ritschl. Before it appeared on the scene such corruptions as were shared by all the minuscule MSS. had the credit of being genuine readings, because every known codex agreed in exhibiting them. Against a “consensus” of MSS. textual criticism was powerless. By the help of the Palimpsest, however, which offered a new reading in many of these corrupt passages, Ritschl was able to prove that all that this “consensus” of MSS. implied was that, with the exception of the newly-discovered codex, all our MSS. belonged to one and the same “family”—in other words, were derived ultimately from one and the same archetype or original MS. From an examination into their peculiarities this lost original has been assigned to the eighth or ninth century—no very early date. The readings, therefore, which Ritschl's predecessors had not ventured to alter appear to be nothing else than the corrupt readings of a single minuscule MS. of Charlemagne's time or later. The value of a “consensus” of MSS. receives thus a convenient illustration from the MSS. of Plautus; for the agreement of a mere pair—namely, the Ambrosian with any one of the others—is of far more importance than the agreement of all the minuscule codices that we possess.

The manuscripts of Plautus

Here is a list of our MSS.5:—
  • The “Ambrosian Palimpsest,” now in the Ambrosian Library at Milan, formerly, it seems, in the library of the Irish monastery of Bobbio in North Italy. It is written in capitals of the fourth century. Its readings6 in the critical apparatus of editors of Plautus are indicated by the letter A.
  • The “Codex Vetus,” now in the Vatican Library, formerly in the Palatine Library at Heidelberg. It was written in Germany in the tenth century. Its symbol in editors' critical apparatus is B.
  • The “Codex Decurtatus,” in the Palatine collection at Heidelberg, written in Germany in the eleventh century. Its symbol is C.
  • The “Codex Ursinianus,” in the Vatican Library, written in Germany in the eleventh century. It was this MS. whose discovery at the Renaissance caused so much enthusiasm; and all MSS. of the fifteenth century and later which contain the last twelve plays are copied from it. Its symbol is D.

Of these four MSS., which are our leading authorities for the text of Plautus, A contained all the twenty plays and also the Vidularia, but only a fragment of the MS. has been preserved to us; B contains all the twenty plays; C only the last twelve (in this order, the order also of B and DBacchides, Mostellaria, Menacchmi, Miles Gloriosus, Mercator, Pseudolus, Poenulus, Persa, Rudens, Stichus, Trinummus, Truculentus); D the first three (Amphitruo, Asinaria, Aulularia), with vv. 1-503 of the fourth, the Captivi, followed by the last twelve.

To these we may add three twelfth-century MSS. which contain only the first eight plays in the same order as B (Amphitruo, Asinaria, Aulularia, Captivi, Curculio, Casina, Cistellaria, Epidicus), viz.:—

  • A MS. at Milan (E).
  • A MS. at Leyden (V).
  • A MS. in the British Museum (J).
  • Also a fragment of a MS. in the Vatican Library, containing about 150 lines of the Captivi (vv. 400-555). This fragment belongs to the eleventh century, and is designated by the letter O. Of a lost MS., designated by T, a MS. used by the scholar Turnebus, we have only isolated readings preserved. They show that T was very like B, and therefore of great value.

We have seen that all these minuscule MSS. are ultimately copies of a single lost minuscule codex, assigned to the eighth or ninth century. The readings of this lost archetype, which we can infer from the readings of BCDEVJ etc., are, for convenience, designated by the symbol P; so that, for example, in Stich. 595, where the minuscule MSS. have all of them the corruption una instead of vasa of the Ambrosian Palimpsest, we may indicate7 the two readings in this way: una P, vasa A. The relationship of the several members of this family, the “Palatine” family, of MSS. seems to be as follows8:—

The first eight plays in B and the first three and a half in D were copied from the same original, a ninth- or early tenth-century MS., now lost. But this part of B was corrected from a much better MS., perhaps the archetype itself. EVJO were all copied immediately or ultimately from one original—a MS. possibly of the tenth or early eleventh century, which was itself a copy of the original of BD, so that their text is not of much value. The second part of B, containing the last twelve plays, is probably copied directly from the archetype. In the latter portion of the MS. the task of copying was divided between a number of scribes, to each of whom only a short piece of the original was assigned, with the result that the last seven plays (Pseud., Poen., Pers., Rud., Stich., Trin., Truc.) have been copied with extraordinary fidelity (see above, p. 2). C and D (last twelve plays) are copies, made apparently in the same scriptorium, of a lost (ninth- or tenth-century?) MS., which was probably, like B, a direct copy of the archetype.

If we put these results in the form of a stemma codicum, we have—

It is probable that P, the archetype, was directly copied from a MS. in capitals (i.e. not later than the fifth century); so that our authorities for the text of Plautus resolve themselves into two ancient MSS. in capitals—one preserved, though in a fragmentary condition, viz. the Ambrosian Palimpsest; the other lost, viz. the original from which P was copied.

Using the manuscripts to establish the text

Where we have the evidence of A, the Ambrosian Palimpsest (in capital script), as well as of P, the archetype of all the other MSS. (in minuscule script), we are seldom at a loss to find the true reading. The difficulty is to elicit the genuine text from the evidence of P alone; and, owing to the fragmentary state of the Palimpsest, that is what we have to do in the greater number of plays. The Persa, Poenulus, Pseudolus, and Stichus are, as before remarked, the favourable exceptions. The method, then, to be followed by the textual critic of Plautus is obvious. He must observe the corruptions which the evidence of A enables him to detect in the P-text of these four plays and others, and infer, by deduction from the known to the unknown, that the same corruptions will be present in other parts of the work where the evidence of A is wanting. An example will make this clear. In Trin. 556 the minuscule MSS. (BCD) agree in reading “Me tibi dixisse hoc. Dixisti tu arcano satis,
” “- w -- - --u --- u-
” which will not scan. The Ambrosian Palimpsest, however, offers dixti instead of dixisti, a change which makes the line perfectly metrical: “Me tíbi dixisse hoc. Díxti tu arcanó satis.
” “- w -- - -u --- u-

Similarly eleven lines below, in v. 567, the minuscule MSS. have the unmetrical line “quid tecum, Stasime? De istoc quod dixisti modo.
” “- -- wu -- - --u u-

Here again A reads dixti, a form which restores the metre of the line: “quid técum, Stasime? istoc quod dixtí modo.
” “- -- wu -- - -u u-

Some lines farther on, in v. 602, we have in the minuscule MSS. a line with an extra syllable: “quomodo tu istuc, Stasime, dixisti nostrum erilem filium?
” “wu -- wu --u -w- -u-

Here we have not the Palimpsest to help us; for the leaf which contained this part of the play has been lost. For all that, no one could for a moment question the propriety of departing from the reading of all the minuscule MSS., dixisti, and of substituting for it dixti:quómodo tu istuc, Stásime, dixti nóstrum erilem fílium?
” “wu -- wu -u -w- -u-

For the instances just quoted show that the scribe of the archetype had a habit of writing dixisti instead of dixti. This example is typical of the method which must be followed in reconstructing the text where we have the testimony of the minuscule MSS. only—in other words, the text of the greater portion of the plays. From the portion where we have the evidence of A to check the evidence of P we must discover what are the besetting sins of the P-scribe; for the probability is that corruptions in the P-text in the remaining portion also of the plays are due to the same mistakes.

Classification of the prevalent errors in manuscripts

These prevalent mistakes of the minuscule MSS. of Plautus we shall find to be the prevalent mistakes of all Latin MSS. They may be roughly classified under seven headings:
  • Emendation
  • Transposition
  • Omission
  • Insertion
  • Substitution
  • Confusion of Letters
  • Confusion of Contractions
Of these only the first class, errors of emendation, are deliberate deviations from the text of the original; the others are all ordinary unintentional mistakes of copying. I propose to devote a chapter to each class, and in each chapter, after explaining with a few examples from various authors the nature and cause of the corruption, to give a list of examples from the MSS. of Plautus, and finally, by way of supplementing theory with practice, to indicate certain passages of Plautus, not yet satisfactorily emended, which seem to me to exhibit the corruption in question.

1 In the large Teubner edition by Ritschl's three pupils, Loewe (now dead), Goetz, and Schoell, the last volume of which appeared in 1894. Some additions and corrections will be found in the critical apparatus of the small Teubner text by Goetz and Schoell (Leipzig 1893-6).

2 The text which modern criticism seeks to discover is that of the first edition or, as an ancient edition is generally called, “recension” of Plautus, which is variously referred to the time of Varro by Ritschl, and to the age of Hadrian by Leo (Plautinische Forschungen chap. i).

3 The earliest Latin MSS. were written in capitals till the fifth century. From the fourth century we find MSS. in uncials or rounded capitals (e.g. V is the capital, U the uncial form). From the eighth century onwards minuscule or small writing became universal—in Italy Lombardic minuscule, in Spain Visigothic minuscule, in France and Germany Caroline or Carolingian minuscule. This last variety, introduced in the reign of Charlemagne, and brought to artistic perfection in Alcuin's School of Calligraphy at Tours, is reproduced in our ordinary printers' type. For at the Renaissance period in Italy there had been a reversion to the early Caroline minuscule, and the first Italian printers copied this, the ordinary book-hand of the time. In Germany, on the other hand, the debased form of Caroline minuscule, known as Gothie or Scholastic Minuscule, was in vogue at the era of the invention of printing; and German printers cut their printing-blocks in imitation of this less pleasing script.

4 We have OPOS on an old inscription (C. I. L. i. 52 C. Pomponi Quir[ina] opos). See my Latin Language p. 236.

5 A full account of them will be found in Ritschl's Prolegomena, and in the Introductions to the several plays in the large Teubner edition by Loewe, Goetz, and Schoell. Facsimiles in photography of their writing are given in Part I of Chatelain's Paléographie des Classiques latins.

6 The fullest account of the text—a text extremely hard to decipher—is to be found in Studemund's Apograph of the Ambrosian Palimpsest (Codicis Rescripti Ambrosiani Apographum), Berlin 1889.

7 P is the initial of Palatinus. The text of this lost archetype is usually called the “Palatine” text, because the MS. which most faithfully reproduces it, the “Codex Vetus” (B), belonged to the Library of the Elector Palatine.

8 See my pamphlet, The Palatine Text of Plautus, Oxford (Parker) 1896.

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 (9 total)
  • Cross-references from this page (9):
    • Plautus, Persa, 4.3
    • Plautus, Persa, 4.4
    • Plautus, Pseudolus, 1.3
    • Plautus, Pseudolus, 4.4
    • Plautus, Stichus, 1.3
    • Plautus, Stichus, 4.1
    • Plautus, Stichus, 4.2
    • Plautus, Trinummus, 2.4
    • Plautus, Trinummus, 3.1
hide Display Preferences
Greek Display:
Arabic Display:
View by Default:
Browse Bar: