Digital Renaissance Editions

Become a FriendSign in

Toolbox

4.6. Collation

4.6. Collation

The collation will be contained in a separate file. You will be provided with a template in XML format (see below, 4.6.8-9, for examples). The format will allow you to create collations that are less runic and more accessible as an independent document than the cramped notations made necessary by the printed page. The format of the Textual Companion to the Oxford Shakespeare Complete Works can serve as a partial guide to describing editorial process and decision-making. Brief or longer notes on individual readings are encouraged, as is disagreement with previous editors emendations (where an editor rejects a previous editor's emendation, he/she should consider writing a rebuttal); where textual variants lead to equally-likely alternatives in interpretation, some collation materials may be repeated in the commentary. Because the Collation will be a separate, self-contained document, each lemma will be identified by TLN. From the one collation file, readers will be able to generate a number of different views of textual variants.

4.6.1. Departures from the copy text

The collation should record all editorial departures from the copy text, as far as substantive readings are concerned. This applies to stage directions, as well as text (see "stage directions" above). Minor stylistic corrections to the stage directions in square brackets, such as "Draw[s] his sword," "Enter [the] Doctor," etc., need not be collated; the editorially added matter will be clearly marked by the brackets.

4.6.2. Where there are two early texts of independent authority

The collation must record all cases in which the text not chosen as copy text departs, in substantive matters, from the copy text.

4.6.3. Changes in punctuation

Collate changes in punctuation only when they bear on a textual argument; when modernization entails a choice between two senses possible in the original; or when punctuation is altered to correct the sense of the copy text.

4.6.4. Changes in spelling

Collate changes in spelling only when modernization entails a choice between two meanings possible in the original, or when the spelling bears on a textual choice or argument. In some instances, it may be possible to animate such variants.

4.6.5. Collation of subsequent editions

Collate subsequent editions of importance, particularly all twentieth-century editions whenever a reading is offered that you deem worthy of serious consideration. Collation is of course mandatory when a previous editor's reading or conjecture is discussed in the commentary. As a general rule, all substantive emendations by previous editors should be considered, if only to be rejected (and discussed in the collation).

4.6.6. Historical collation

It is not necessary to give the full textual history of a reading; only the earliest source of a reading should be cited (e.g. where a text went into several quartos, and the later quartos are not of independent authority, you do not need to record these later readings, nor the readings of subsequent editors when these readings are mistaken or misconceived).

4.6.7. When in doubt

When in doubt whether to include a collation, you should query the General Textual Editor assigned to your edition.

4.6.8. Format of collations

The template you will receive provides a framework for the collations. You can copy and paste it into your working document as often as you need.

<coll>
  <tln n="n" />
  <lem resp="Q1"></lem>
  <rdg resp="xxx"></rdg>
  <rdg resp="xxx"></rdg>
  <rdg resp="xxx"></rdg>
</coll>

a) Each collation is surrounded by the tags <coll> . . . </coll>.

b) The TLN can be a single number, a number with a decimal addition, or a range of numbers. The decimal addition is needed for lines where other witnesses have no equivalent (in plays where one particular quarto is the copy text, for example). Express a range of numbers using the full number in each case: <tln n="1003-1009"/>, for example. Note: where you wish to collate a number of repeated instances on different lines, create a note (<note></note>) to list them, or create separate, individual collations for each. Do not include them in the TLN tag.

c) The line number is the line of the current scene in standard MLA format (1.3.143). Again, for a range of numbers use the full number at the start and end of the range.

d) A lemma can be a single word, a phrase, or a longer section indicated by ellipsis (three periods separated by spaces). The lemma is not only the word or phrase for which there are variants, but also the "anchor" which the reader will click on to see the full collation. For this reason, it is important that you design the lemma with three points in mind:

  1. A single-word lemma must be long enough to permit clear visibility of the underline that indicates a variant, and to permit easy clicking. Thus, for example, a single "I" or "a" should be avoided; add a second word to the lemma and to each reading that you list.
  2. Since the computer will find the first example of the lemma in a given TLN range, the lemma must be unambiguous. If a word is repeated in a line, and you are referring to the second instance, you should again include an additional word before or after.
  3. When you indicate a longer passage using ellipsis, the collation will be linked from the first part of the passage, before the ellipsis. Thus this initial section of the lemma must be of sufficient length and be unambiguous.

Note: it is very important that the lemma is exactly the same as it is recorded in your edition including final punctuation. Otherwise the computer will not be able to match it, and there will be an error. The safest way to ensure accuracy is to copy and paste from your edition, making sure to update any collations should you later revise your modern text. If the lemma includes a TLN tag (as may happen especially in prose passages), delete the tag but make a note for the Coordinating Editor, in a separate file.

e) Both for the lemma and for each reading, the attribute "resp=" refers to the edition or editor responsible for the reading. In the template above, the default is "Q1" (DRE style requires a number after each folio, quarto, or octavo mentioned); where the copy text is a quarto the template will provide a default of Q1. You will, of course, change this as necessary.

f) The template provides three spaces for readings. You should delete lines where these are not needed, and add further lines by copying and pasting if you are recording more than three readings.

g) Where a given reading is shared by more than one edition, either cite only the first instance, or create a second <rdg> tag for it. Where it is important to indicate that more than one edition shares a given reading, separate the sigla by a space: <rdg resp="Q1 Q2"></rdg>.

h) List readings in chronological order.

4.6.8.1 Advanced display of complex variants

See this section for the tagging required to display complex variants and ambiguities in the text: relineation between different versions, added passages, variants significant enough that the reader should be alerted to them even when collations are not being displayed, and animation when a single word is irreducibly ambiguous.

4.6.9. Adding notes to collations

As an extensible document, the DRE collation offers editors the opportunity to explain their editorial decision-making to a larger extent than that offered in print editions. Because space to explain or debate editorial choices is unlimited, editors are encouraged to justify not only their unique emendations, but also their adoption or rejection of earlier editorial emendations or editorial conjectures in their collation notes, to make the collation click worthwhile (see 1.2.2, above). Notes may be as simple as a single word or phrase ("dittography"; "turned letter"; "foul case"), but may also articulate more detailed rationales. In each case, enclose the note in the tags <note> </note>, as in the example above.

Sometimes you will want to add a brief note to a lemma or reading: such common information as "and so throughout." In each case, enclose the note in the tags <note> </note>, as in the example in the next section. You can add a note to a specific reading by including it within the <lem></lem> or <rdg></rdg> tags thus:

<lem resp="Dodsley 1">This would . . . and all.<note>shared line</note></lem>

If you want the note to refer to the collation as a whole, add it just before the final tag closing the collation, using the tag <generalNote>:

<generalNote>Editors have variously punctuated this passage to clarify the implied parentheses.</generalNote></coll>

Some frequently used notes should be indicated by adding the appropriate attribute to the <rdg> tag. "Substantively," "Conjectured," and "Not in [text]" can be consistently rendered when the tagging reads thus:

<lem resp="Dodsley 1" subst="yes">Haply</lem>

<rdg resp="Dyce" conj="yes">you--</rdg>

<rdg resp="Malone" omit="yes"></rdg>

4.6.10. Collating press variants

Collations can record press variants in specific copies (witnesses) of an edition. The collations can record the variant readings of witnesses, and can indicate whether they represent the initial state or the proofread state. While it is true that it is often difficult to be sure of the direction in which a change was made, the editor will need to make a judgement as to which reading represents the first or second (or even third) state of the text.

Each collation will include, as a header, a list of witnesses which record the short form, the long form, and the location of the copy:

<listWit>
  <witness id="Q1a" longid="Huth">British Library, Huth 46</witness>
  <witness id="Q1b" longid="Devonshire">Devonshire [full description]</witness>
  <witness id="Q1c" longid="Petworth">Petworth [full description]</witness>
  <witness id="Q1d" longid="Capell">Capell [full description]</witness>
</listWit>

In recording individual variants, you can identify specific witnesses using the short form, and in addition indicate whether the reading records the first or subsequent state. Here is an example of a collation of multiple witnesses in two states:

<coll>
 <tln n="259" />
 <lem resp="Q1b Q1c" state="2">alas, may I</lem>
 <rdg resp="Q1a Q1d" state="1">may I</rdg>
 <rdg resp="F1">(alas may I)</rdg>
</coll>

Note 1: where more than one witness is recorded, the witnesses are separated by a space.

Note 2: where a reading is shared by all extant copies of an edition the generic short form is used (Q1 etc).

For display, the short form will appear when collations are shown colour-coded, the long form in the window that appears when a collation is clicked on.

4.6.11. Sample collations

Please ensure that your collation file includes an initial HTML comment that records the name of the author, the text being collated, the date of creation, and the most recent revision.

Example:

<!--Collations to Fair Em by Brett D. Hirsch. Created 2011-06-01, most recently modified 2013-04-07.-->

It is a good idea to include the date of most recent revision in the file name so that you can be sure not to work on an older file.

Example:

FairEm_M_annotations 2013-04-07.txt

A basic collation:

<coll>
 <tln n="110" />
 <lem resp="Q3">she</lem>
 <rdg resp="Q1">hee</rdg>
</coll>

A collation where the reading originates in the copy text, but where the copy text's spelling has been modernized:

<coll>
 <tln n="57" />
 <lem resp="Q1">villain</lem>
 <rdg resp="Q1">villaine</rdg>
 <rdg resp="Dodsley 1">villein</rdg>
</coll>

A collation with inserted notes:

<coll>
 <tln n="226" />
 <lem resp="Dodsley 1" subst="yes"><SP>Viola</SP> <note>substantively</note></lem>
 <rdg resp="Q1">Vio. <note>and so occasionally throughout play, as at TLN 1213, 2341, and 2345</note></rdg>

A collation where the copy text omits a word or passage:

<coll>
<tln n="159" />
<lem resp="Q2"><SP>Viola</SP></lem>
<rdg resp="Q2">Vio.</rdg>
<rdg resp="Q1" omit="yes"><note>not in Q1</note></rdg>
</coll>

A collation that covers a longer passage:

<coll>
 <tln n="313-314"/>
 <ln>1.2.60</ln>
 <lem resp="Dyce">Come on . . . forwardness</lem>
 <rdg resp="Q1">Duke. Come on . . . intreated / His . . . forwardnesse. <note>as verse</note></rdg>
</coll>

4.6.12. Omissions and changes in font

Use the form <rdg resp="Q1" omit="yes"></rdg> for omissions. In the display on the site this will appear as "Not in Q1." If the omission requires more information, use a standard <note> format. Use "<note>Not italic in Q1.</note>" to indicate an editorial change of type font, where that change bears on a textual argument.

4.6.13. Conjectures and readings of earlier editors

Indicate former editors' conjectures, as distinct from emendations, by "<note>conjectured</note>." Use this tagging to indicate that an emendation accepted into the text, or recorded in the collation, is substantially that offered by a previous editor:

<lem resp="Capell" subst="yes">[Exit attendant.]</lem>

4.6.14. Line breaks

Indicate line breaks in a quotation, by a forward slash, thus: / (with a space before and after).

Previous | Table of Contents | Next