University of Virginia Library

Introduction to Manilius, 1903, p. xxxii.

The more naive the scribe, the more often will the assumption prove correct; the more sophisticated, the less often. This, no doubt, is why critics of this school tend to reject 'the more correct but the less sincere' manuscript in favour of 'the more corrupt but the less interpolated', as Housman elsewhere observes ('The Application of Thought to Textual Criticism', Proceedings of the Classical Association, 1921, xviii. 75). Still, any reasonable critic will prefer the work of a naive to that of a sophisticated scribe, though he may not regard it as necessarily 'better'.

I believe that an attempt has been made in the case of certain Old and Middle English texts, but how consistently and with what success I cannot judge. In any case I am here concerned chiefly with works of the sixteenth and seventeenth centuries.

It will, no doubt, be objected that punctuation may very seriously 'affect' an author's meaning; still it remains properly a matter of presentation, as spelling does in spite of its use in distinguishing homonyms. The distinction I am trying to draw is practical, not philosophic. It is also true that between substantive readings and spellings there is an intermediate class of word-forms about the assignment of which opinions may differ and which may have to be treated differently in dealing with the work of different scribes.

For the sake of clearness in making the distinction I have above stressed the independence of scribes and compositors in the matter of accidentals: at the same time, when he selects his copy-text, an editor will naturally hope that it retains at least something of the character of the original. Experience, however, shows that while the distribution of substantive variants generally agrees with the genetic relation of the texts, that of accidental variants is comparatively arbitrary.

Some discussion bearing on it will be found in the Prolegomena to my lectures on The Editorial Problem in Shakespeare (1942), 'Note on Accidental Characteristics of the Text' (pp. l-lv), particularly the paragraph on pp. liii-liv, and note 1. But at the time of writing I was still a long way from any consistent theory regarding copy-text.

A 'substantive' edition is McKerrow's term for an edition that is not a reprint of any other. I shall use the term in this sense, since I do not think that there should be any danger of confusion between 'substantive editions' and 'substantive readings'. I have above ignored the practice of some eccentric editors who took as copy-text for a work the latest edition printed in the author's lifetime, on the assumption, presumably, that he revised each edition as it appeared. The textual results were naturally deplorable.

He believed, or at least strongly suspected, that some were due to the printer's desire to save space, and that others were 'the work of some person who had not thoroughly considered the sense of the passage which he was altering' (ii.195).

Nashe, ii.197. The word 'reprint' really begs the question. If all an 'editor' aims at is an exact reprint, then obviously he will choose one early edition, on whatever grounds he considers relevant, and reproduce it as it stands. But McKerrow does emend his copy-text where necessary. It is symptomatic that he did not distinguish between a critical edition and a reprint.

Without the italicized phrase the statement would appear much more plausible (though I should still regard it as fallacious, and so would McKerrow himself have done later on) but it would not justify the procedure adopted.

This may, at any rate, be put forward as a general proposition, leaving possible exceptions to be considered later (pp. 33 ff.).

Again he speaks of a 'reprint' where he evidently had in mind a critical edition on conservative lines.

Boas's text is in fact modernized, so that my theory of copy-text does not strictly apply, but since he definitely accepts the B-text as his authority, the principle is the same.

Or consider the following readings: 1604, 1609 'Consissylogismes', 1611 'subtile sylogismes', 1616 'subtle Sillogismes'. Here 'subtile', an irresponsible guess by the printer of 1611 for a word he did not understand, was taken over in 1616. The correct reading is, of course, 'concise syllogisms'. Boas's refusal to take account of the copy used in 1616 led him here and elsewhere to perpetuate some of its manifest errors. In this particular instance he appears to have been unaware of the reading of 1611.

At another point two lines appear in an unnatural order in the manuscript. The genetic relation of the texts proves the inversion to be an error. But of this relation Simpson seems to have been ignorant. He was again content to rely on the copy-text.

The proviso is inserted to meet the case of the so-called 'bad quartos' of Shakespearian and other Elizabethan plays and of the whole class of 'reported' texts, whose testimony can in general be neglected.

That is, certainly not in the text, and probably not in the general apparatus: they may appropriately form the subject of an appendix.

This is the course I recommended in the Prolegomena to The Editorial Problem in Shakespeare (p. xxix), adding that it 'at least saves the trouble of tossing a coin'. What I actually wrote in 1942 was that in such circumstances an editor 'will naturally retain the reading of the copy-text, this being the text which he has already decided is prima facie the more correct'. This implies that correctness in respect of substantive readings is one of the criteria in the choice of the copy-text; and indeed I followed McKerrow in laying it down that an editor should select as copy-text the one that 'appears likely to have departed least in wording, spelling, and punctuation from the author's manuscript'. There is a good deal in my Prolegomena that I should now express differently, and on this particular point I have definitely changed my opinion. I should now say that the choice of the copy-text depends solely on its formal features (accidentals) and that fidelity as regards substantive readings is irrelevant—though fortunately in nine cases out of ten the choice will be the same whichever rule we adopt.

I mean substantive variation, such as occurs in all but the most faithful reprints.

Simpson's procedure in taking the 1616 folio as copy-text in the case of most of the masques included, although he admits that in their case Jonson cannot be supposed to have supervised the printing, is much more questionable.

Some variation is certainly due to error on the part of the folio printer, and this it is of course the business of an editor to detect and correct so far as he is able.