Description of the TEI-Encoding for the Psalmcatenae Project


Table of Contents

  1. Header
    1. Defining Glyphs
    2. Defining the Layout
    3. List of Authors
  2. Macrostructure
    1. Divisions
    2. Notes
    3. Anonymous Blocks
    4. Segments
    6. Anchors
    7. Examples
  3. Microstructure
    1. Page Breaks
    2. Line Breaks
    3. Uncertain Readability
    4. Gaps
    5. Ekthesis
    6. Regularization
    7. Abbreviations
  4. Hints

Defining Glyphs

    <glyph xml:id="linking-sign-1">
        <glyphName>LINKING SIGN ONE</glyphName>
            <graphic url="./images/linking-sign-1.png"/>


The xml:id is used to link a glyph in the text to entity in the character declaration:

<g ref="#linking-sign-1"/>
The # sign is used to mark the value as a reference. The graphic element contains the path to the image of the glyph (url).

Defining the Layout

    <layout n="layout of the pages">
        <p xml:id="middle-of-the-page">middle of the page</p>
        <p xml:id="main-text">main text</p>
        <p xml:id="left-margin">left margin</p>
        <p xml:id="top-margin">top margin</p>
        <p xml:id="right-margin">right margin</p>
        <p xml:id="bottom-margin">bottom margin</p>
        <p xml:id="on-the-line">on the line</p>
        <p xml:id="above-the-line">above the line</p>
        <p xml:id="margin-top-right-side">margin top right side</p>
        <p xml:id="margin-top-left-side">margin top left side</p>


The content between the <p> tags can be more verbose. It should provide a description of the layout position.

Important is the xml:id attribute because it is used as reference in other elements in the document body. For example a division in the middle of the page should carry a rendition attribute with a reference to the corresponding layout element:

<div rendition="#middle-of-the-page"> ... </div>
References start with the # sign.

List of Authors

    <person xml:id="person-1" role="author">
        <bibl corresp="#bibliography-title-1">Asterius, fr.</bibl>


At the moment not physical persons are distinguished but authors of specific editions. This seems to be appropriate because it is necessary for example to identify Anonymous as author of so far unknown fragments and Anonymous as author of the fragments published by Dorival.



<div xml:id="vat-gr-1422-1-2" type="bibletext" subtype="commented" rend="red ink" rendition="#middle-of-the-page"> ... </div>


The xml:id is used to identify an entity in the transcription. The value can be freely choosen, however, it should be somehow consistent, e.g. name of the manuscript / manuscript identifier + number of the psalm the entity belongs to + counting upwards. The identifier must be unique within a document.

The values of the type attribute are restricted to the following values:

Divisions are terminated by a page break. Every section on a page like the different colums builds a division on its own.


<note place="right margin" rend="red ink" xml:id="vat-gr-1422-1-6" xml:lang="grc" corresp="#vat-gr-1422-fr-1-1"> ... </note>


The place attribute contains the position of the note on the page. Possible values are:

The corresp attribute gives a reference to the entity the note belongs to. Here the note is associated with a commentaryfragment with the id vat-gr-1422-fr-1-1

Anonymous Blocks

<ab xml:id="vat-gr-1422-1-2" xml:lang="grc"> ... </ab>


The anonymous blocks are only used because TEI requires them


<seg type="commentaryfragment" source="Basilius of Kappadokia" xml:id="vat-gr-1422-fr-1-1" corresp="#vat-gr-1422-ps-1-1a" rendition="#main-text"> ... </seg>


The type attribute can take on of the six values:

The corresp attribute contains a reference to a psalm text (or another text) the segment belongs to. In almost all cases this is by convention an identifier of an anchor.

The source attribute has the name of the author as mentioned in the text of the manuscript. Often the value is Anonymous.

When a commentaryfragment runs over a pagebreak it is necessary to set the prev and next attributes with the xml:ids to indicate that the texts belong together.


<quote type="patristic" corresp="#person-9" xml:id="vat-gr-1422-quot-1-5" source="fons ignotus (ed. Dorival IV 367)"> ... </quote>


The value of the type attribute must be one of:

The value of the corresp attribute is a pointer to a element of a list of authors given in the document header.

When a quote runs over a pagebreak it is necessary to set the prev and next attributes to indicate that the texts belong together.


<anchor xml:id="vat-gr-1422-ps-1-1a" type="psalmtext" n="Ps (LXX) 1,1a" corresp="../psalms-LXX/psalms.xml#Ps-LXX-1-1"/>


The anchor element is only used for the markup of subdivisions in verses of the psalmtext.

The corresp attribute is a pointer to the text of the psalm in another file which contains the LXX text according to Rahlfs.

The value of the n attribute appears in the list of commentaryfragments at the end of the diplomatic html version.



<seg type="commentaryfragment" xml:id="vat-gr-754-fr-9-83" rendition="#right-column" 
        source="(Athanasius)" corresp="#vat-gr-754-ps-9-26c">
   <quote type="patristic" subtype="quotation" xml:id="vat-gr-754-quot-9-86" 
             source="Athanasius, schol. in Ps 9,26c" corresp="#person-2">Ὅμοιον ... νοσσιάν:</quote>

The value of the corresp attribute on the seg element is a pointer to an anchor element of the bibletext.

The subtype attribute on the quote element designates the relation of the text of the manuscript to the text of available editions.

Suggested values are:

Hexaplaric Variants

<seg type="hexaplaric" xml:id="vat-gr-754-hexaplaric-10-13" rendition="#right-column" 
        source="Aquila, Theodotion (Ps 11,1)" corresp="#vat-gr-754-ps-11-1a">A Θ τῷ νικοποιῷ·</seg>

Please note that the seg of hexaplaric variants does not contain a quote.


<seg type="glosse" xml:id="vat-gr-754-glosse-10-4" rendition="#right-column" 
        source="Anonymous" corresp="#vat-gr-754-ps-10-3a">
    <quote type="patristic" subtype="unknown" xml:id="vat-gr-754-quot-10-13" 
	          source="fons ignotus; in Ps 10,3a?" corresp="#person-18">κατέβαλεν·</quote>


Page Breaks

<pb n="25r"/>


Since divisions start and end with pagebreaks the pagebreak element stands below the div with attribute type="transcription".

Pagebreaks mark the beginning of a page.

Line Breaks

<lb n="2" break="no"/>


Linebreaks mark the beginning of a line.

Linebreaks can have an attribute break with the value no which indicates that the linebreak appears in a word.


<hi rend="ekthesis">Μ</hi>ακάριος

Other possible values for the rend attribute on hi are:

Uncertain Readability

... το<unclear reason="damage">ῖς</unclear> ...


... <gap unit="character" quantity="2" reason="erased"/> ...


There are two types of regularization: One for normal regularization without attribute type on reg and one with value correction for little errors.

    <reg type="correction">βουλῇ</reg>


  1. Simple abbreviations for authors of the Septuagint:
    <abbr source="Aquila" type="bibleversion"><hi rend="overline">Α</hi></abbr>
  2. Nomina Sacra
        <abbr type="nomSac">χυ</abbr>
  3. Tachygraphy: The am Tag stands for abbreviation marker
            <abbr type="tachygraphic">γ<am>ῆς</am></abbr>
  4. Suspension
            <abbr type="suspension">ἀλεξαν<hi rend="superscript">δ<am>ρείας</am></hi></abbr>
  5. Common abbreviations
        <abbr type="common"><g ref="#abbreviation-1"/></abbr>
  6. Compression
        <abbr type="compression">Ὡρ<am>ι</am>γ<am>ένους</am></abbr>
  7. Compression and suspension
    	    <abbr type="compression-suspension">Εὐσ<am>ε</am><hi rend="superscript">β<am>ίου</am></hi></abbr>
  8. Compression and tachygraphy
            <abbr type="compression-tachygraphic">κ<am>α</am><am>τὰ</am></abbr>

To sum up, the possible values for the type attribute on abbr are:


For easier understanding and work the values of the xml:id attributes should follow this rules:

The flow of work should go from outside to inside: divisions then anonoymous blocks then segments and quotes with anchors. In the next step the markup of the microlevel elements follows.