Each RET text has several potential hierarchies, both bibliographical
and generic. For example, bibliographically, lines can be perceived
as nesting in a column (even if a page has only one column), columns
in a page, pages in a gathering, and gatherings in a book. Critics
see many other structures in texts, based on genre-related assumptions,
and these are perhaps subject to more disagreement than bibliographical
organization, which arises from a factual study of the physical
nature of a book or manuscript. However, even the top-down hierarchies
implicit in RET encoding can be reasonably questioned. From the
viewpoint of the reader, the line being currently read might be
regarded as the top-most unit (<bkdv1 t="ln">
rather than the bottom unit (<bkdv4 tp="ln">).
Under some circumstances, a reader may well believe that the entire
book is contained in the line being read.
Any hierarchy is an interpreted thing. A play, for example, may
be structured (1) as speeches in a scene, and scenes within a
play; or (2) as speech prefix and utterance within a speech, speeches
within a scene, scenes within an act, and acts within the play.
There are certainly too many questions about how any playwright
handles these units to propose one general hierarchy for plays,
as TEI does. Are act divisions as relevant to a play performance
(where classical-scene breaks -- at the entrance or exit of any
character -- provide much more obvious dramatic pointing) as they
are to those scheduling times during which the audience can get
up from their seats and go the bar? To what are stage directions,
including descriptions of settings, properly subordinated? Exits
and entrances and some other stage directions would fit well as
a sub-unit of a speech, but not directions like <em><stage>Thunder.</stage></em>.
DTDs were devised originally for utilitarian documents as a way
of storing necessary information about how their authors and publishers
intended them to be interpreted. Such command files belong to
technical writing, in which precision and lack of ambiguity are
desirable and feasible. However, any reader, critic, or editor
have to step into the mind of that author to write such a DTD.
Under some circumstances, this appropriation of an author's rights
makes sense. For instance, DTDs may legitimately be used by critics
to implement -- that is, to activate -- a critical interpretation
of a text. An SGML editor can take the tag- and character-set-
definitions in any DTD and structure any reader's view and use
of the tagged text. The reader then sees only those relationships
that the critic placed in the DTD. DTDs, of course, have some
limitations. They can only recognize a maximum of two parallel
structures in a text and can only use one of those structures
at a time.
Most readers of a RET text, however, will want to have nothing
to do with devising a DTD, even if they use SGML editors or browsers.
For this large majority, these guidelines offer two things. First,
RET texts have been given a patina of HTML, over the regular SGML
encoding, so that they may be used online with World Wide Web
browsers. These will ignore any tag not defined within HTML, which
of course uses a DTD purposely built for online publishing, not
for text analysis. The NCSA's Beginner's Guide to HTML
gives easy-to-understand instructions on how to encode a document
with these and other HTML tags.
Second, this appendix supplies a simple DTD, modeled on the so-called
`Waterloo DTD' used by Frank Tompa and his colleagues at the Centre
for the New OED at the University of Waterloo and on a version
of this DTD devised for me by Michael Sperberg-McQueen, who presented
it at the CCH Conference on " Early Dictionary Databases"
held at the University of Toronto in October 1993. In his words,
"it enforces *no* constraints on the interrelation of elements,
with the exception that some elements are defined as empty and
thus cannot contain any others."
<!-- Simple DDT for RET 3 -->
<!-- -->
<!-- August 1995 -->
<!-- -->
<!-- Ian Lancashire -->
<!-- -->
<!-- A `Waterloo DTD', modelled on the so-called -->
<!-- `Waterloo DTD' used by Frank Tompa and his -->
<!-- colleagues at the Centre for the New OED -->
<!-- at the University of Waterloo and on a version -->
<!-- of this DTD devised by Michael Sperberg-McQueen,-->
<!-- who presented at the CCH Conference on -->
<!-- "Early Dictionary Databases" held -->
<!-- at the University of Toronto in October 1993. -->
<!-- In his words, `it enforces *no* constraints -->
<!-- on the interrelation of elements, with the -->
<!-- exception that some elements are defined as -->
<!-- empty and thus cannot contain any others.' -->
<!-- Each RET and HTML tag appears here with -->
<!-- a little interpretation. Normally, elements -->
<!-- must have an explicit closing tag. The only -->
<!-- entities defined below belong to HTML. -->
<!-- Note that none of the tags in the TEI header -->
<!-- appear below. -->
<!-- BASIC TAGS -->
<!-- The whole text (TEI P3:1189) -->
<!ELEMENT text - - EMPTY -->
<!-- Preliminaries of book (TEI P3: 988) -->
<!ELEMENT front - - EMPTY -->
<!-- Collection of texts in a book (TEI P3:1006) -->
<!ELEMENT group - - EMPTY -->
<!-- Text in RET series -->
<!ELEMENT retBook - - EMPTY -->
<!-- Composition date of work -->
<!ELEMENT docDatcp - - EMPTY -->
<!-- CITATIONS -->
<!-- Author's name in text (TEI P3: 870) -->
<!ELEMENT author - - EMPTY -->
<!-- Bibliographical reference (TEI P3: 873) -->
<!ELEMENT bibl - - EMPTY -->
<!-- Citation of work (TEI P3: 899) -->
<!ELEMENT cit - - EMPTY -->
<!ATTLIST cit
author CDATA #IMPLIED
work CDATA #IMPLIED-->
<!-- Edition of book (TEI P3: 949) -->
<!ELEMENT edition - - EMPTY -->
<!-- Editor of book (TEI P3: 950) -->
<!ELEMENT editor - - EMPTY -->
<!-- Imprint of book (TEI P3:1019) -->
<!ELEMENT imprint - - EMPTY -->
<!-- Name of printer of cited work -->
<!ELEMENT printer - - EMPTY -->
<!-- Name of publisher of cited work (TEI P3:1111) -->
<!ELEMENT publish - - EMPTY -->
<!-- Name of publ. place of cited work (TEI P3:1112) -->
<!ELEMENT pubPlace - - EMPTY -->
<!-- Quotation from work (HTML and TEI P3:1116) -->
<!ELEMENT quot - - EMPTY -->
<!ATTLIST quot
author CDATA #IMPLIED
work CDATA #IMPLIED-->
<!-- Bibliographical title in citation (TEI P3: 1197)-->
<!ELEMENT bibtitle - - EMPTY -->
<!ATTLIST bibtitle
level CDATA #IMPLIED-->
<!-- CROSS REFERENCES AND HYPERTEXT -->
<!-- Cross-reference locator -->
<!ELEMENT anchor - - EMPTY -->
<!ATTLIST anchor
id CDATA #IMPLIED
CDATA #IMPLIED
<!-- Apparatus for variants (TEI P3: 863) -->
<!ELEMENT app - - EMPTY -->
<!ATTLIST app
from CDATA #IMPLIED
to CDATA #IMPLIED
type CDATA #IMPLIED-->
<!-- Standard edition reference (TEI P3:1059) -->
<!ELEMENT mileston - - EMPTY -->
<!ATTLIST mileston
ed CDATA #IMPLIED
unit CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- Note (TEI P3:1072)-->
<!ELEMENT note - - EMPTY -->
<!ATTLIST note
place CDATA #IMPLIED
target CDATA #IMPLIED-->
<!-- Marker in text to anchor note -->
<!ELEMENT notesymb - - EMPTY -->
<!ATTLIST notesymb
id CDATA #IMPLIED-->
<!-- DICTIONARIES -->
<!-- Example of dictionary headword -->
<!ELEMENT egt - - EMPTY -->
<!ATTLIST egt
n CDATA #IMPLIED-->
<!-- Translation of headword example -->
<!ELEMENT egttr - - EMPTY -->
<!ATTLIST egttr
n CDATA #IMPLIED-->
<!-- Gender of headword -->
<!ELEMENT gender - - EMPTY -->
<!-- Etymology in dictionary entry -->
<!ELEMENT etym - - EMPTY -->
<!-- Explanation in dictionary entry -->
<!ELEMENT explan - - EMPTY -->
<!-- Headword in dictionary entry (TEI P3: 984) -->
<!ELEMENT form - - EMPTY -->
<!ATTLIST form
type CDATA #IMPLIED
category CDATA #IMPLIED
gender CDATA #IMPLIED
inflect CDATA #IMPLIED
modernsp CDATA #IMPLIED
originsp CDATA #IMPLIED
pos CDATA #IMPLIED
syntact CDATA #IMPLIED-->
<!-- Usage field in dictionary -->
<!ELEMENT usg - - EMPTY -->
<!ATTLIST usg
type CDATA #IMPLIED-->
<!-- Cross-reference in dictionary -->
<!ELEMENT xref - - EMPTY -->
<!ATTLIST usg
type CDATA #IMPLIED
lemma CDATA #IMPLIED-->
<!-- HEADINGS AND CLOSINGS -->
<!-- Person addressed in epistle -->
<!ELEMENT addresse - - EMPTY -->
<!-- End material in book (TEI P3: 872) -->
<!ELEMENT back - - EMPTY -->
<!-- Main part of book (HTML and TEI P3: 879 -->
<!ELEMENT body - - EMPTY -->
<!-- Phrase marking work's closing -->
<!ELEMENT closing - - EMPTY -->
<!-- Heading for start of work or section -->
<!ELEMENT heading - - EMPTY -->
<!-- Heading-number for section of work -->
<!ELEMENT headingno - - EMPTY -->
<!-- Signature of writer (TEI P3:1154) -->
<!ELEMENT signed - - EMPTY -->
<!-- Subheading -->
<!ELEMENT subhead - - EMPTY -->
<!-- HTML TAGS -->
<!-- HTML document (HTML tag) -->
<!ELEMENT html - - EMPTY -->
<!-- Document title for WWW browser (HTML tag) -->
<!ELEMENT title - - EMPTY -->
<!-- Heading material for e-text (HTML tag) -->
<!ELEMENT head - - EMPTY -->
<!-- Level-1 heading (HTML tag) -->
<!ELEMENT h1 - - EMPTY -->
<!-- Level-2 heading (HTML tag) -->
<!ELEMENT h2 - - EMPTY -->
<!-- Level-3 heading (HTML tag) -->
<!ELEMENT h3 - - EMPTY -->
<!-- Level-4 heading (HTML tag) -->
<!ELEMENT h4 - - EMPTY -->
<!-- Unnumbered paragraph (TEI P3:1090) -->
<!ELEMENT p - 0 EMPTY -->
<!-- Numbered list (HTML tag) -->
<!ELEMENT ol - - EMPTY -->
<!-- Definition list (HTML tag) -->
<!ELEMENT dl - - EMPTY -->
<!-- Term to be defined (HTML tag) -->
<!ELEMENT dt - - EMPTY -->
<!-- Definition (HTML tag) -->
<!ELEMENT dd - - EMPTY -->
<!-- Unnumbered list (HTML tag) -->
<!ELEMENT ul - - EMPTY -->
<!-- List item (HTML tag) -->
<!ELEMENT li - - EMPTY -->
<!-- Italics (for words) (HTML tag) -->
<!ELEMENT i - - EMPTY -->
<!-- Preformatted text (HTML tag) -->
<!ELEMENT pre - - EMPTY -->
<!-- Block quotation (HTML tag) -->
<!ELEMENT blockquote - - EMPTY -->
<!-- Horizontal rule (HTML tag) -->
<!ELEMENT hr - 0 EMPTY -->
<!-- Forced line-break (HTML tag) -->
<!ELEMENT br - 0 EMPTY -->
<!-- Address of e-text editor (HTML tag) -->
<!ELEMENT address - - EMPTY -->
<!-- Emphasis (for words) (HTML tag) -->
<!ELEMENT em - - EMPTY -->
<!-- Anchor tag for hypertext (HTML; TEI P3: 861) -->
<!ELEMENT a - - EMPTY -->
<!ATTLIST a
name CDATA #IMPLIED
href CDATA #IMPLIED-->
<!-- Image or graphical unit -->
<!ELEMENT img - - EMPTY -->
<!ATTLIST img
id CDATA #IMPLIED
src CDATA #IMPLIED-->
<!-- HTML ENTITIES -->
<!ENTITY amp CDATA '&' -->
<!-- ampersand -->
<!ENTITY gt CDATA '>' -->
<!-- closing angle bracket -->
<!ENTITY lt CDATA '<' -->
<!-- opening angle bracket -->
<!ENTITY quot CDATA '"' -->
<!-- double quotation mark -->
<!ENTITY 160 CDATA '&160;' -->
<!-- non-breaking blank space-->
<!-- MANUSCRIPTS -->
<!-- Addition to manuscript (TEI P3: 850) -->
<!ELEMENT add - - EMPTY -->
<!ATTLIST add
hand CDATA #IMPLIED
place CDATA #IMPLIED-->
<!-- Deletion to text (TEI P3: 922) -->
<!ELEMENT del - - EMPTY -->
<!ATTLIST del
hand CDATA #IMPLIED
type CDATA #IMPLIED-->
<!-- Type of handwriting in manuscript (TEI P3:1008) -->
<!ELEMENT handShft - - EMPTY -->
<!ATTLIST handShft
new CDATA #IMPLIED
old CDATA #IMPLIED-->
<!-- Cancellation of deletion (TEI P3:1134)-->
<!ELEMENT restore - - EMPTY -->
<!ATTLIST restore
type CDATA #IMPLIED
hand CDATA #IMPLIED-->
<!-- Script used by scribe -->
<!ELEMENT script - - EMPTY -->
<!ATTLIST script
type CDATA #IMPLIED-->
<!-- PLAYS -->
<!-- Actor's name in cast list (TEI P3: 850) -->
<!ELEMENT actor - - EMPTY -->
<!-- Classical scene -->
<!ELEMENT classcen - - EMPTY -->
<!ATTLIST classcen
n CDATA #IMPLIED-->
<!-- Performance information on play's title-page -->
<!ELEMENT perform - - EMPTY -->
<!-- Unnumbered division of play -->
<!ELEMENT plydv - - EMPTY -->
<!ATTLIST plydv
type CDATA #IMPLIED-->
<!-- Largest division of play -->
<!ELEMENT plydv1 - - EMPTY -->
<!ATTLIST plydv1
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 2nd largest division of play -->
<!ELEMENT plydv2 - - EMPTY -->
<!ATTLIST plydv2
type CDATA #IMPLIED
n CDATA #IMPLIED
classcen CDATA #IMPLIED-->
<!-- 3rd largest division of play -->
<!ELEMENT plydv3 - - EMPTY -->
<!ATTLIST plydv3
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 4th largest division of play -->
<!ELEMENT plydv4 - - EMPTY -->
<!ATTLIST plydv4
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- Role name in cast list (TEI P3:1136)-->
<!ELEMENT role - - EMPTY -->
<!-- Scene heading -->
<!ELEMENT scene - - EMPTY -->
<!-- Speech in play (TEI P3:1159)-->
<!ELEMENT sp - - EMPTY -->
<!ATTLIST sp
who CDATA #IMPLIED
type CDATA #IMPLIED
direct CDATA #IMPLIED-->
<!-- Speech prefix (TEI P3:1162)-->
<!ELEMENT speaker - - EMPTY -->
<!-- Stage direction (TEI P3:1163)-->
<!ELEMENT stage - - EMPTY -->
<!ATTLIST stage
type CDATA #IMPLIED-->
<!-- POEMS -->
<!-- Shift into verse or prose -->
<!ELEMENT mode - - EMPTY -->
<!-- Unnumbered division of poem -->
<!ELEMENT pmdv - - EMPTY -->
<!ATTLIST pmdv
type CDATA #IMPLIED-->
<!-- Largest division of poem -->
<!ELEMENT pmdv1 - - EMPTY -->
<!ATTLIST pmdv1
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 2nd largest division of poem -->
<!ELEMENT pmdv2 - - EMPTY -->
<!ATTLIST pmdv2
type CDATA #IMPLIED
n CDATA #IMPLIED
rhyme CDATA #IMPLIED-->
<!-- 3rd largest division of poem -->
<!ELEMENT pmdv3 - - EMPTY -->
<!ATTLIST pmdv3
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 4th largest division of poem -->
<!ELEMENT pmdv4 - - EMPTY -->
<!ATTLIST pmdv4
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- PRINTER'S BUSINESS -->
<!-- Subdivision of book (unnumbered) -->
<!ELEMENT bkdv - - EMPTY -->
<!ATTLIST bkdv
type CDATA #IMPLIED-->
<!-- Largest unit of book -->
<!ELEMENT bkdv1 - - EMPTY -->
<!ATTLIST bkdv1
type CDATA #IMPLIED
n CDATA #IMPLIED
format CDATA #IMPLIED
in CDATA #IMPLIED-->
<!-- 2nd largest unit of book -->
<!ELEMENT bkdv2 - - EMPTY -->
<!ATTLIST bkdv2
type CDATA #IMPLIED
n CDATA #IMPLIED
sig CDATA #IMPLIED
side CDATA #IMPLIED
forme CDATA #IMPLIED-->
<!-- 3rd largest unit of book -->
<!ELEMENT bkdv3 - - EMPTY -->
<!ATTLIST bkdv3
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 4th largest unit of book -->
<!ELEMENT bkdv4 - - EMPTY -->
<!ATTLIST bkdv4
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- Name of bookseller (TEI P3: ) -->
<!ELEMENT bookseller - - EMPTY -->
<!-- Change in Compositor -->
<!ELEMENT compshift - - EMPTY -->
<!ATTLIST compshift
name CDATA #IMPLIED-->
<!-- Column (TEI P3: 906) -->
<!ELEMENT col - - EMPTY -->
<!ATTLIST col
n CDATA #IMPLIED-->
<!-- Error in errata list -->
<!ELEMENT error - - EMPTY -->
<!ATTLIST error
from CDATA #IMPLIED
to CDATA #IMPLIED-->
<!-- Correction in errata list -->
<!ELEMENT correct - - EMPTY -->
<!-- Author's name on title page (TEI P3: 945) -->
<!ELEMENT docAuthor - - EMPTY -->
<!-- Publication date on title page (TEI P3: 945) -->
<!ELEMENT docDate - - EMPTY -->
<!-- Edition on title page (TEI P3: 946) -->
<!ELEMENT docEdition - - EMPTY -->
<!-- Editor on title page -->
<!ELEMENT docEditr - - EMPTY -->
<!-- Imprint on title page (TEI P3: 947) -->
<!ELEMENT docImpr - - EMPTY -->
<!-- Title on title page (TEI P3: 948) -->
<!ELEMENT docTitle - - EMPTY -->
<!-- Volume number on title page -->
<!ELEMENT docVolno - - EMPTY -->
<!-- Forme-work (printer's business) (TEI P3: 995) -->
<!ELEMENT fw - - EMPTY -->
<!ATTLIST fw
type CDATA #IMPLIED-->
<!-- Authority to publish statement (TEI P3:1018)-->
<!ELEMENT imprimat - - EMPTY -->
<!-- Cost of book -->
<!ELEMENT price - - EMPTY -->
<!-- Location for erratum item in text -->
<!ELEMENT ref - - EMPTY -->
<!ATTLIST ref
target CDATA #IMPLIED
resp CDATA #IMPLIED
type CDATA #IMPLIED-->
<!-- PROSE WORKS -->
<!-- Unnumbered division of prose work -->
<!ELEMENT ttdv - - EMPTY -->
<!ATTLIST ttdv
type CDATA #IMPLIED-->
<!-- Largest division of prose work -->
<!ELEMENT ttdv1 - - EMPTY -->
<!ATTLIST ttdv1
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 2nd largest division of prose work -->
<!ELEMENT ttdv2 - - EMPTY -->
<!ATTLIST ttdv2
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 3rd largest division of prose work -->
<!ELEMENT ttdv3 - - EMPTY -->
<!ATTLIST ttdv3
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- 4th largest division of prose work -->
<!ELEMENT ttdv4 - - EMPTY -->
<!ATTLIST ttdv4
type CDATA #IMPLIED
n CDATA #IMPLIED-->
<!-- VARIANTS -->
<!-- Textual variant (TEI P3:1119)-->
<!ELEMENT rdg - - EMPTY -->
<!ATTLIST rdg
resp CDATA #IMPLIED
status CDATA #IMPLIED-->
<!-- Correction of typographical error (TEI P3:1152)-->
<!ELEMENT sic - - EMPTY -->
<!ATTLIST sic
corr CDATA #IMPLIED
resp CDATA #IMPLIED-->
<!-- Emendation or addition by editor (TEI P3:1171)-->
<!ELEMENT supplied - - EMPTY -->
<!ATTLIST supplied
reason CDATA #IMPLIED-->
<!-- WORD-LEVEL TAGS -->
<!-- Word cited as object (TEI P3:1186)-->
<!ELEMENT term - - EMPTY -->
<!ATTLIST term
type CDATA #IMPLIED
expansion CDATA #IMPLIED-->
<!-- Character -->
<!ELEMENT c - - EMPTY -->
<!ATTLIST c
type CDATA #IMPLIED-->
<!-- Damage to text (TEI P3: 912) -->
<!ELEMENT damage - - EMPTY -->
<!ATTLIST damage
extent CDATA #IMPLIED
type CDATA #IMPLIED-->
<!-- Font -->
<!ELEMENT f - - EMPTY -->
<!ATTLIST f
type CDATA #IMPLIED-->
<!-- Gap in text (TEI P3: 996) -->
<!ELEMENT gap - - EMPTY -->
<!ATTLIST gap
extent CDATA #IMPLIED-->
<!-- Gloss for words (TEI P3:1001) -->
<!ELEMENT gloss - - EMPTY -->
<!ATTLIST gloss
target CDATA #IMPLIED-->
<!-- Inflection of word -->
<!ELEMENT i - - EMPTY -->
<!ATTLIST i
type CDATA #IMPLIED-->
<!-- Shift into non-English language -->
<!ELEMENT lang - - EMPTY -->
<!ATTLIST lang
type CDATA #IMPLIED-->
<!-- Name (TEI P3:1065) -->
<!ELEMENT name - - EMPTY -->
<!ATTLIST name
type CDATA #IMPLIED-->
<!-- Number -->
<!ELEMENT num - - EMPTY -->
<!ATTLIST num
type CDATA #IMPLIED-->