Lexical Unit Boundary

The logical boundary of a lexical unit (word). It is used to mark the end of a word regardless of its physical layout on the support, allowing the system to reconstruct units split by fractures or line breaks.

Syntax

Any number of Unicode characters enclosed within ampersands &…&. Optionally, linguistic annotations can follow after a semicolon ;.

If the user wishes to easily identify a series of lexical units, the entire sequence can be enclosed within ampersands &…& with individual units separated by a forward slash /.

Overview

• 𝑆 = any sign (Unicode character) ∉ STOP_SET (cf. Technical Catalogue of Symbols) otherwise preceded by the escape character \.
• 𝑁 = number of characters (represented by char or digits).
• λₛ = label, which may also be in abbreviated form, for linguistic annotation.
• λᵥ = value of the linguistic annotation (plausibility scale).
• λₖ = canonical key for the linguistic annotation derived from the corresponding label.

Examples

&ab-c&

&*ab* [c]&

&
abc defg  hijk l
&

&abc/def/hijk&

EpiDoc Outputs

&ab-c&

<ab>
    <w>ab<lb break="no"/>c</w>
</ab>

&*ab* [c]&

<ab>
    <w>
        <unclear cert="medium">ab</unclear> 
        <supplied reason="lost" cert="medium">c</supplied>
    </w>
</ab>

&
abc defg  hijk l
&

<ab>
    <w>abc defg hijk l</w>
</ab>

&abc/def/hijk&

<ab>
    <w>abcdefhijk</w>
</ab>

Structural Analysis (Conceptual AST)

&ab-c&

LexicalUnit
  Plain("ab")
  WordBreak
  Plain("c")

&*ab* [c]&

LexicalUnit
  Unclear[medium, legibility: normal]
    Plain("ab")
  WS
  Supplied[medium]
    Plain("c")

&
abc defg  hijk l
&

LexicalUnit
  Plain("abc")
  WS
  Plain("defg")
  WS
  Plain("hijk")
  WS
  Plain("l")

&abc/def/hijk&

LexicalUnit
  Plain("abc")
  LineBreak
  Plain("def")
  LineBreak
  Plain("hijk")