Note: This document accompanied a paper delivered at the thirty-first International Congress on Medieval Studies, Kalamazoo, May 1996. For further information on digital projects in Middle English at Michigan, see the Middle English Compendium page at the Humanities Text Initiative.
Paul Schaffner, 31 Aug 1996.

Tagging the Silences:
Retrospective Encoding of the
Middle English Dictionary

  1. Potential of a fully-encoded e-MED.
    • An example of a bare string search.
    • Additional access points. Search:
      • for a given entry.
      • within a given entry or set of entries.
      • for a word or phrase, wherever it appears among the quotations, and however it is spelled--at least in its own entry, and among cited phrases and cross-references in other entries.
      • for a character string anywhere among the quotations.
      • for a word or phrase in the definition, notes, or etymology.
      • for a spelling listed among the spelling forms.
      • for phonological information
      • for dialect or other labels attached to forms.
      • for a date or date range, whether of manuscript or of composition
      • for a particular manuscript.
      • for a particular work
      • for etyma
      • for the language of the etyma
      • for any field label (law, surg, naut, physiol, theol. hawk)
      • for any semantic label or grammatical, whether italicized as a label propoer ("fig." "refl") or identified by conventional vocabulary (as term of abuse, in conventional comparisons, "oneself")
      • for any cited collocation,
      • for any French or Latin word glossed by the ME
      • for any particular object or subject ("of God" "(animals)"
      • for any Linnaean plant name.
      • for any work cited as a source text from which the ME was translated.
  2. Obstacles to translation from MED-code to TEI-code
    • Granularity

      • Example:
        Text:
        [OE raet; also cp. OF rat, rate, & ML ratus; cp. MDu. ratte, rotte.]

        Tagged:
        <etym>
        <lang>OE</lang>
        <mentioned lang=OE>raet</mentioned>; also cp.
        <lang>OF</lang>
        <mentioned lang=OF>rat
        <distinct type="ov">rot
        </distinct></mentioned>&
        <lang>ML</lang><mentioned lang=ML>ratus
        </mentioned>; cp.
        <lang>MDu.</lang><foreign lang=MDu>ratte
        <distinct type="ov">rotte</distinct></foreign>
        </etym>


    • Misaligned categories
      • Example:
        Text:
        Tho def.art and adj.(2)

        Tagged: <entry type=homograph n=da-adj2>
        <form type=hd>tho</form>

      • Example:
        Text: a1456 York MGame (Add)

        Tagged:
        <date value="1456" calendar="ms" certainty="a">
        1456</date>
        <abbr type="stencil"><name>
        York</name>
        <title>MGame</title>
        </abbr>
        <abbr type="source">
        Add</abbr>

    • Compressed information
      • Example:
        Text: in rape, in (on) a rape

        Tagged:
        <re>
        <form>in rape</form>
        <form>in a rape</form>
        <form>on a rape</form>
        <def>quickly, hurriedly</def>
        </re>

    • Layered information
      Example:
      • Text:
        rib(be

        Tagged:
        <form type=hd>
        <orth>ribbe</orth>
        <pron>ribbe</pron>
        <orth>rib</orth>
        <pron>rib</pron>
        </form>

    • Reticence

      Implied information:
      • Some dates of manuscript and composition.
      • The identity of the manuscript quoted for every quotation, whether preferred or non-preferred, whether specified in abbreviated form or left unspecified.
      • The edition cited.
      • The quality and quantity of the vowels listed in the spelling list (by implication the same as those of the head form, though in this case silence may imply only a healthy agnosticism).
      • The Linnaean character of names presented in genus-and-species form.
      • The relevance of the lemmata or glosses supplied as context when contemporary glossaries are quoted.
      • The form of the etymon when only the language is given, or the language when only the form is given.
      • The relation of a cited phrase to the definitions given in the rest of a sense.
      • The identity of the source text when it is cited simply as "F" or "L."
      • The word to which the elements of a cited phrase belong.
      • The case of a word cited in a phrase.
      • The source of a cited etymon form and variants
      • The identity of the word belonging to the entry in the quotations belonging to the entry.

      Etymological Examples

      • TEXT:
        rank n. [OF renc, ranc]

        TAGGED:
        <etym>
        <lang>OF</lang>
        <mentioned lang=OF>renc <distinct type="ov">ranc</distinct></mentioned></etym>

      • TEXT:
        regence n. [OF]

        TAGGED:
        <etym>
        <lang>OF</lang>
        <mentioned lang=OF>regence</mentioned>
        </etym>

      • TEXT:
        regendren v. [OF regendrer]

        TAGGED:
        <etym>
        <lang>OF</lang>
        <mentioned lang=OF>regendrer</mentioned>
        </etym>

      • TEXT:
        regeneratif n. [From regeneratif adj.]

        TAGGED:
        <etym>From
        <mentioned lang=ME>regeneratif</mentioned>
        </etym>

      • TEXT:
        rapeli adv. [ON: cp. OI hrapalligra]

        TAGGED:
        <etym>
        <lang>ON</lang>
        <mentioned lang=ON></mentioned>: cp.
        <lang>OI</lang>
        <foreign lang=OI>hrappaligra</foreign>
        </etym>


      PHRASAL EXAMPLE

      TEXT:
      yeven ille

      TAGGED:
      <re>
      <form><oRef>
      <ref target=ille.adv>ille</ref></form>
      </re>

      BOTANICAL EXAMPLE

      TEXT:
      Hare's-ear mustard (Conringia orientalis)

      TAGGED:
      <def>Hare's-ear mustard
      <foreign lang=bot>Conringia orientalis</foreign></def>


  3. Assessment of TEI
    Paul Schaffner :: 9 May 1996