3.1.8 Editorial Interventions and Indications
Transcribing text from original 17th and 18th century prints poses some special problems, in particular with regard to general quality or legibility of the source text and transcribability of certain features. Some of the old texts show the signs of age very clearly and are not in a good state, for example having edges torn off, worm holes, stains of whatever nature etc. Additionally, some features of the texts are very hard to capture electronically. The following mark-up elements have been included to deal with problems of this nature.
<add> contains words, phrases or punctuation marks added to the original printed text. Its attributes are:
- resp: indicates who is responsible for the addition. Possibilities are "errata" (which we put directly into their proper places with this method, instead of transcribing the original errata passage at the beginning or end of the whole text) and "annotator", i.e. some unknown previous reader of the pamphlets (one can of course imagine it being the original collector, Bowdler himself). If there is no resp attribute the addition was made by the compilers of the corpus; we have tried to keep these to a minimum.
- place: if the addition is hand-written onto the printed page, this attribute indicates its position, e.g. supralinear, margin.
Examples are found one of the titlepages mentioned above, i.e. EcB1696, where place remains unspecified:
<BYLINE>―<LB>By <GO>W. C.</GO> Esq; <ADD RESP="annotator">Culliford</ADD> ―</BYLINE>
An instance with all attributes is the following from SciB1666:
... or other stars being not in the Ecliptick, are said to have Latitude so many degrees as they <ADD
PLACE="supralinear" RESP="annotator">be</ADD> distant from the Ecliptick toward either pole thereof.</P>
<del> contains words, phrases or punctuation marks deleted from the source text. It includes the attributes:
- resp: "errata", "annotator", or nothing. Cf. the explanation given under <add>.
- type: indicates the method of deletion, e.g. overstrike.
<add> and <del> elements can occur in tandem, if necessary, e.g. in EcB1681:
... and advancing the <IT>Indian</IT> <DEL TYPE="overstrike" RESP="annotator">Commanders</DEL> <ADD PLACE="supralinear" RESP="annotator">Comodities</ADD> to such an odious excess, ...
<corr> contains the correct form of an apparently erroneous part of the original text. It has the following attributes:
- sic: supplies the form(s) as used originally in the source text.
- resp: "errata", "annotator", or missing if the correction was carried out by us. As with <add> and <del> above, we have tried to keep our interventions to a minimum. If the spelling found in the original was a possible, even if somewhat "exotic", basis for the adequate pronunciation of the word in question (17th/18th century pronunciation, that is), we have left it standing. Instances which are listed as spelling variants in the OED were also left unchanged. As far as possible we have thus only corrected apparently nonsensical spellings and obvious misprints, e.g. words printed twice in a row, and cases like abuudant (abundant), bua (but), or Almamacks (Almanacks). As can be seen in the last example, our corrected version is not necessarily the modern regular version (almanacs), because we have only changed the obvious error in the word.
One of our own corrections is found in the next example taken from PolA1646:
for it is a maxime no lesse true then common, <CORR SIC="thar">that</CORR> <IT>Contraries have contrary consequents</IT>.
whereas SciA1644 provides an errata example:
The Souldier will devoure and consume their stately houses and principall Palaces, fire, warre, and the Canon may <CORR RESP="errata" SIC="espoliate">exspoliate</CORR>, &c.
<sic> marks forms in the source text which appear to be erroneous in some way, but which might have been correct after all then, and those which cannot be corrected easily or without a major intervention. For example, "In the year 169, ..." (MscB1700) we have marked "In the year <sic>169</sic>, ..." as clearly one particular year in the 1690s is intended, but we do not know which.
<unclear> contains words or phrases which could not be transcribed with absolute certainty because they are partly illegible through holes or stains in the original print. No attributes are given in this element.
<gap> marks either the omission of some material originally present in the source text, or a gap already present in the original text. Most commonly it is used for ornaments and other pictorial elements which cannot be rendered electronically. Furthermore, it is employed for text parts omitted for sampling reasons (e.g. length, foreign language). It also indicates the place of the original errata passage. Completely illegible passages (because of overstrike or damage) are encoded with this element as well. Lastly, gaps already present in the original text, i.e. blank spaces or missing elements (e.g. notes), fall under this description. Possible attributes are:
- desc: provides a description of the missing element (e.g. "ornament") or gap found in the running text (e.g. "empty space").
- extent: specifies the length of the gap, in (estimated) characters, lines or pages.
- reason: if necessary it indicates why there is a gap.
The following example is from text MscB1670:
and the one of them in <IT>Trine</IT> to the <IT>Mid&rehy;heaven</IT>, and the other casts the same Aspect to the <GAP REASON="illegibility" EXTENT="c. 9 chars.">, or <IT>eleventh Angle</IT>; and Royal <PB N="28"><IT>fixed Stars</IT> <ADD RESP="errata">on the</ADD> prime Cusps of the <IT>Scheam</IT>.
?? Another example where <gap> stands for a nice ornament/picture: find in original texts
<figure> marks one particular kind of gap, namely one caused by the non-transcription of a diagram-like element. <figdesc> is used in these cases to describe the missing diagram.
