Base Tag Set for Drama

This base tag set is intended for use when encoding printed dramatic texts, screen plays or radio scripts, and written transcriptions of any form of performance.

Section 10.1 Front and Back Matter discusses elements, such as cast lists, which can appear only in the front or back matter of printed dramatic texts. Section 10.2 The Body of a Performance Text discusses the structural components of performance texts: these include major structural divisions such as acts and scenes (section 10.2.1 Major Structural Divisions); individual speeches (section 10.2.2 Speeches and Speakers); stage directions (section 10.2.3 Stage Directions); and the elements making up individual speeches (section 10.2.4 Speech Contents). Section 10.2.5 Embedded Structures discusses ways of encoding units which cross the simple hierarchic structure so far defined, such as embedded songs or masques. Finally, section 10.3 Other Types of Performance Text discusses a small number of additional elements characteristic of screen plays and radio or television scripts, as well as some elements for representing technical stage directions such as lighting or blocking.

To enable the base tag set for performance texts, the parameter entity TEI.drama must be declared within the document type subset with the value INCLUDE, as further described in section 3.3 Invocation of the TEI DTD. A document using the base tag set for drama and no additional tag sets will thus begin as follows: <!DOCTYPE TEI.2 PUBLIC "-//TEI P4//DTD Main Document Type//EN" "tei2.dtd" [ <!ENTITY % TEI.XML 'INCLUDE' > <!ENTITY % TEI.drama 'INCLUDE' > ]> This declaration makes available all of the elements described in this chapter, in addition to the core elements described in chapter 6 Elements Available in All TEI Documents. The default structure for dramatic texts is similar to that defined by chapter 7 Default Text Structure, as further discussed in section 10.2.1 Major Structural Divisions.

Three additional element classes are used by this base tag set. The dramafront class contains elements which can appear only in the front or back matter of performance texts. The stageDirection class contains a set of elements for specialized stage directions, which can occur between or within speeches. These two element classes were defined in 3.7 Element Classes above.

A default declaration for the comp.drama class is also defined in that section, but a different declaration is needed when the drama base is selected, in order to add all elements which may appear as components of performance texts, in addition to those defined in the core. This is achieved by the following declarations:

<!-- 10.: Class declarations for Performance Texts-->
<!ENTITY % x.comp.drama "" >
<!ENTITY % m.comp.drama "%x.comp.drama; %n.castList; | 
%m.stageDirection;">  
<!ENTITY % mix.drama '| %m.comp.drama;' >
<!-- end of 10.-->

The remainder of the DTD fragment defining the base tag set for drama has the following overall shape:

<!-- 10.: Base tag set for Performance texts-->
[declarations from 10.1: Specialized front and back matter for performance 
texts inserted here ] 

[declarations from 10.2.3: Stage directions inserted here ]

[declarations from 10.3.1: Screenplays and other technical matters 
inserted here ] 
<!--The base tag set for drama uses the standard default
text-structure elements, which are embedded here:-->
<![%TEI.singleBase;[
<!ENTITY % TEI.structure.dtd PUBLIC '-//TEI P4//ELEMENTS Default Text 
Structure//EN' 'teistr2.dtd' > 
%TEI.structure.dtd;
]]>
<!-- end of 10.-->

10.1 Front and Back Matter

In dramatic texts, as in all TEI-conformant documents, the header element is followed by a <text> element, which contains optional front and back matter, and either a <body> or else a <group> of nested <text> elements. For more information on these, see chapter 7 Default Text Structure.

The <front> and <back> elements are most likely to be of use when encoding preliminary materials in published performance texts. These often contain specific textual elements not generally found in other forms of text. These include:

Elements for encoding each of these specific kinds of front matter are discussed in the remainder of this section, in the order given above. In addition, the front matter of dramatic texts may include the same elements as that of any other kind of text, notably title pages and various kinds of text division, as discussed in section 7.4 Front Matter. The encoder may choose to ignore the specialized elements discussed in this section and instead use constructions of the type <div type="performance"> or <div1 type="set">.

Most other material in the front matter of a performance text will be marked with the default text structure elements described in chapter 7 Default Text Structure. For example, the title page, dedication, other commendatory material, preface, etc., in a printed text should be encoded using <div> or <div1> elements, containing headings, paragraphs, and other core tags.

The specialized elements for front and back matter of performance texts are defined as follows:

<!-- 10.1: Specialized front and back matter for performance texts-->
[declarations from 10.1.1: The set element inserted here ]

[declarations from 10.1.2: The prologue and epilogue elements inserted 
here ] 

[declarations from 10.1.3: The performance element inserted here ] 

[declarations from 10.1.4: The castList element inserted here ]
<!-- end of 10.1-->

10.1.1 The Set Element

A special form of note describing the setting of a dramatic text (that is, the time and place of its action) is sometimes found in the front matter.

Descriptions of the setting may also appear as initial stage directions in the body of the play, but such descriptions should be marked as stage directions, not <set>. The <set> element should be used only where the description forms part of the front matter, as in the following examples:

<front>
  <castList>
    <castItem> ... </castItem>
  </castList>
  <set><p>The action of the play is set in Chicago's
       Southside, sometime between World War II and the
       present.</p></set>
</front>

<front>
  <titlePage> <!-- ... --> </titlePage>
  <div type='copyright page'> <!-- ... --> </div>
  <div type='Contents'> <!-- ... --> </div>
  <div type='Introduction'> <!-- ... --> </div>
  <div type='note'>
    <head>Note on the Translation</head> 
    <p> ... </p>
  </div>
  <titlePage type='half-title'>
    <docTitle><titlePart>Peer Gynt</titlePart></docTitle>
  </titlePage>
  <div type='Dramatis Personae'>
    <head>Characters</head> 
    <castList> <!-- ... --> </castList>
  </div>
  <set><p>The action, which opens in the early years of the last
      century and closes about fifty years later, takes place partly
      in the Gudbrand Valley in Norway and on the mountains around it,
      partly on the Moroccan coast, partly in the Sahara Desert, the
      asylum in Cairo, at sea, etc.</p></set>
  <performance> <!-- ... --> </performance>
</front>

The <set> element is formally defined as follows:

<!-- 10.1.1: The set element-->
<!ELEMENT set %om.RR; ((%m.Incl;)*, head?, %component.seq;)> 
<!ATTLIST set
      %a.global;
      TEIform CDATA 'set'  >
<!-- end of 10.1.1-->

10.1.2 Prologues and Epilogues

Many plays in the Western tradition include in their front matter a prologue, spoken by an actor, generally not in character. Similar speeches often also occur at the end of the play, as epilogues. The elements <prologue> and <epilogue> are provided for the encoding of such features within the front or back matter, where appropriate.

A prologue may be encoded just like a distinct poem, as in the following example:

<front>
  <prologue>
    <head>Prologue, spoken by <name>Mr. Hart</name></head>
    <l>Poets like Cudgel'd Bullys, never do</l>
    <l>At first, or second blow, submit to you;</l>
    <l>But will provoke you still, and ne're have done,</l>
    <l>Till you are weary first, with laying on:</l>
    <!-- ... -->
    <l>We patiently you see, give up to you,</l>
    <l>Our Poets, Virgins, nay our Matrons too.</l>
  </prologue>
  <castList>
    <head>The Persons</head>
    <castItem> ... </castItem>
  </castList>
  <set><head>The SCENE</head>
    <p>London</p>
  </set>
</front>

A prologue or epilogue may also be encoded as a speech, using the <sp> element described in section 6.11.2 Core Tags for Drama. This is particularly appropriate where stage directions, etc., are involved, as in the following example:

<epilogue>
   <head>Written by <name>Colley Cibber, Esq</name>
		and spoken by <name>Mrs. Cibber</name></head>
   <sp>
      <lg type="stanza">
         <l>Since Fate has robb'd me of the hapless Youth,</l>
         <l>For whom my heart had hoarded up its truth;</l>
         <l>By all the Laws of Love and Honour, now,</l>
         <l>I'm free again to chuse, &mdash; and one of you</l>
      </lg>
  	  <!-- ... -->
      <lg type="stanza">
         <l>Suppose I search the sober Gallery; &mdash; No,</l>
         <l>There's none but Prentices &mdash; &amp; Cuckolds all a row:</l>
         <l>And these, I doubt, are those that make 'em so.</l>
      </lg>
      <stage>Pointing to the Boxes.</stage>
      <lg type="stanza">
         <l>'Tis very well, enjoy the jest:</l>
		 <!-- ... -->
      </lg>
   </sp>
</epilogue>

In cases where the prologue or epilogue is clearly a significant part of the dramatic action, it may be preferable to include it in the body of a text, rather than in the front or back matter. In such cases, the encoder (and theatrical tradition) will determine whether or not to regard it as a new scene or division, or simply the final speech in the play. In the First Folio version of Shakespeare's Tempest, for example, Prospero's final speech is clearly marked off as a distinct textual unit by the headings and layout of the page, and might therefore be encoded as back matter:

<text>
  <body>
    <div1 type='scene'>
      <!-- ... -->
      <sp who='PR'>
        <l part='Y'>I'le deliver all,</l>
        <l>And promise you calme Seas, auspicious gales,</l>
        <!-- ... -->
        <l>Be free and fare thou well: please you, draw neere.</l>
        <stage>Exeunt omnes.</stage>
      </sp>
    </div1>
  </body>
  <back>
    <epilogue>
      <head>Epilogue, spoken by Prospero.</head>
      <sp who="PR">
        <l>Now my Charmes are all ore-throwne,</l>
        <l>And what strength I have's mine owne</l>
        <!-- ... -->
        <l>As you from crimes would pardon'd be,</l>
        <l>Let your Indulgence set me free.</l></sp>
      <stage>Exit</stage>
    </epilogue>
    <set>
	  <p>The Scene, an un-inhabited Island.</p>
	</set>
    <castList>
      <head>Names of the Actors.</head>
      <castItem>Alonso, K. of Naples</castItem>
      <castItem>Sebastian, his Brother.</castItem>
      <castItem>Prospero, the right Duke of Millaine.</castItem>
      <!-- ... -->
    </castList>
    <trailer>FINIS</trailer>
  </back>
</text>

In many modern editions, the editors have chosen to regard Prospero's speech as a part of the preceding scene:

<sp who="pr">
  <speaker>Prospero</speaker>
  <l part="Y">I'll deliver all,</l>
  <l>And promise you calm seas, auspicious gales,</l>
  <!-- ... -->
  <l>Be free and fare thou well. <stage type="exit">Exit Ariel</stage>
	Please you, draw near. <stage type="exit">Exeunt all but Prospero</stage>
    <note place="margin">Epilogue</note> </l>
  <l>Now my charms are all o'erthrown,</l>
  <l>And what strength I have's mine own</l>
  <!-- ... -->
  <l>As you from crimes would pardoned be,</l>
  <l>Let your indulgence set me free.</l>
</sp>
<stage type="mix">He awaits applause, then exit.</stage>

Prologues and epilogues are formally defined as follows:

<!-- 10.1.2: The prologue and epilogue elements-->
<!ELEMENT prologue %om.RR; ((%m.divtop; | %m.Incl;)*,
  ((%component;), (%m.Incl;)*)+, ((%m.divbot;), (%m.Incl;)*)*)> 
<!ATTLIST prologue
      %a.global;
      TEIform CDATA 'prologue'  >
<!ELEMENT epilogue %om.RR; ((%m.divtop; | %m.Incl;)*,
                            ((%component;), (%m.Incl;)*)+,
                            ((%m.divbot;), (%m.Incl;)*)*)> 
<!ATTLIST epilogue
      %a.global;
      TEIform CDATA 'epilogue'  >
<!-- end of 10.1.2-->

10.1.3 Records of Performances

Performance texts are not only printed in books to be read, they are also performed. It is common practice therefore to include within the front matter of a printed dramatic text some brief account of particular performances, using the following element:

The <performance> element may be used to group any and all information relating to the actual performance of a play or screenplay, whether it specifies how the play should be performed in general or how it was performed in practice on some occasion.

Performance information may include complex structures such as cast lists, or paragraphs describing the date and location of a performance, details about the setting portrayed in the performance and so forth. (See the discussion of these specialized structures in section 10.1 Front and Back Matter above.) If information for more than one performance is being recorded, then more than one <performance> element should be used, wherever possible.

Names of persons, places, and dates of particular significance within the performance record may be explicitly marked using the general purpose <name>, <rs type="place"> and <date> elements described in section 6.4.4 Dates and Times. No particular elements for such features as stagehouses, directors, etc., are proposed at this time.

For example:

<performance>
   <head>Death of a Salesman</head>
   <p>A New Play by Arthur Miller</p>
   <p>Staged by Elia Kazan</p>
   <castList>
      <head>Cast</head>
      <note rend="small type flush left" place="inline">(in order of appearance)</note>
      <castItem>
         <role>Willy Loman</role> 
         <actor>Lee J. Cobb</actor>
      </castItem>
      <castItem>
         <role>Linda</role>       
         <actor>Mildred Dunnock</actor>
      </castItem>
      <castItem>
         <role>Biff</role>        
         <actor>Arthur Kennedy</actor>
      </castItem>
      <castItem>
         <role>Happy</role>       
         <actor>Cameron Mitchell</actor>
      </castItem>
      <!-- ... -->
   </castList>
   <p>The setting and lighting were designed by 
      <name>Jo Mielziner</name>.</p>
   <p>The incidental music was composed by <name>Alex North</name>.</p>
   <p>The costumes were designed by <name>Julia Sze</name>.</p>
   <p>Presented by <name rend="unmarked">Kermit Bloomgarden</name>
    and <name rend="unmarked">Walter Fried</name> at the 
    <rs type="place">Morosco Theatre in New York</rs> on 
    <date value="1949-02-10">February 10, 1949</date>.</p>
</performance>

Or:

performance>
   <p>La Machine Infernale a &eacute;t&eacute;
    repr&eacute;sent&eacute;e pour la premi&egrave;re fois au
    <rs type="place (theatre)">th&eacute;&acirc;tre Louis-Jouvet</rs>
    <rs type="place (theatre)">(Com&eacute;die des
	Champs-&Eacute;lys&eacute;es)</rs> <date>le 10 avril 1934</date>,
    avec les d&eacute;cors et les costumes de
    <name>Christian B&eacute;rard.</name> ... </p>
</performance>

The <performance> element is formally defined as follows:

<!-- 10.1.3: The performance element-->
<!ELEMENT performance %om.RR; ((%m.divtop; | %m.Incl;)*,
  ((%component;), (%m.Incl;)*)+, ((%m.divbot;), (%m.Incl;)*)*)> 
<!ATTLIST performance
      %a.global;
      TEIform CDATA 'performance'  >
<!-- end of 10.1.3-->

10.1.4 Cast Lists

A cast list is a specialized form of list, conventionally found at the start or end of a play, usually listing all the speaking and non-speaking roles in the play, often with additional description (‘Cataplasma, a maker of Periwigges and Attires’) or the name of an actor or actress (‘Old Lady Squeamish. Mrs Rutter’). Cast lists may be encoded with the general purpose <list> element described in section 6.7 Lists, but for more detailed work the following specialized elements are provided:

Cast lists often have an internal structure of their own; it is quite usual to find, for example, nobility and commoners, or male and female roles, presented in different groups or sublists. Roles are also often grouped together by their function, for example:

Sons of Cato:
- Portius
- Marcus

A cast list relating to a specific performance may be accompanied by notes about the time or place of that performance, indicating (for example) the name of the theatre where the play was first presented, the name of the producer or director, and so forth. When the cast list relates to a specific performance, it should be embedded within a <performance> element (see section 10.1.3 Records of Performances), as in the following example:

<performance>
  <p>The first performance in Great Britain of <title>Waiting for
	  Godot</title> was given at the Arts Theatre, London, on <date
	  value="1955-08-03">3rd August 1955</date>. It was directed by
	  <name>Peter Hall</name>, and the d&eacute;cor was by <name>Peter
	  Snow</name>. The cast was as follows:</p>
  <castList>
	<castItem>Estragon: Peter Woodthorpe</castItem>
	<castItem>Vladimir: Paul Daneman</castItem>
	<castItem> ... </castItem>
	<!-- ... -->
  </castList>
</performance>

In this example, the <castItem> elements have no substructure. If desired, however, their components may be more finely distinguished using the elements <role>, <roleDesc>, and <actor>. For example, the second cast item above might be encoded as follows:

<castItem>     
  <role id="vlad">Vladimir</role>:
  <actor>Paul Daneman</actor>
</castItem>

The global id attribute may be used to specify a unique identifier for the <role> element, where it is desired to link speeches within the text explicitly to the role, using the who attribute, as further discussed in section 10.2.2 Speeches and Speakers below.

The occasionally lengthy descriptions of a role sometimes found in written play scripts may be marked using the <roleDesc> element, as in the following example:

<castItem> 
   <role>Tom Thumb the Great</role>
   <roleDesc>a little hero with a great soul, something violent in his
    temper, which is a little abated by his love for Huncamunca</roleDesc>
   <actor>Young Verhuyk</actor>
</castItem>

For non-speaking or un-named roles, a <castItem> may contain a <roleDesc> without an accompanying <role>, for example

<castItem>  
   <roleDesc>Costermonger</roleDesc>
</castItem>

When a list of such minor roles is given together, the type attribute of the <castItem> should indicate that it contains more than one role. The encoder may or may not elect to encode each separate constituent within such a composite <castItem>. Thus, either of the following is acceptable:

<castItem type="list">Constables, Drawer, Turnkey, etc.</castItem>

<castItem type="list">     
   <roleDesc>Constables,</roleDesc>
   <roleDesc>Drawer,</roleDesc>
   <roleDesc>Turnkey,</roleDesc>
  etc.
</castItem>

A group of cast items forming a distinct subdivision of a cast list may be marked as such by using the special purpose <castGroup> element. The rend attribute may be used to indicate whether this grouping is indicated in the text by layout alone (i.e. the use of whitespace), by long braces or by some other means. A <castGroup> consists of an optional heading (represented as usual by a <head> element) followed by a series of <castItem> elements.

<castGroup rend="braced">
   <head>friends of Mathias</head>
   <castItem>        
      <role>Walter</role>
      <actor>Mr Frank Hall</actor>
   </castItem>
   <castItem>       
      <role>Hans</role>
      <actor>Mr F.W. Irish</actor>
   </castItem>
</castGroup>

The following example demonstrates the use of the <castGroup> element to structure the whole of a <castList>, reflecting the way it is presented on the page:

<castList>
   <castGroup>
      <head rend="braced">Mendicants</head>
      <castItem> <role>Aafaa</role>    <actor>Femi Johnson</actor> </castItem>
      <castItem> <role>Blindman</role> <actor>Femi Osofisan</actor> </castItem>
      <castItem> <role>Goyi</role>     <actor>Wale Ogunyemi</actor> </castItem>
      <castItem> <role>Cripple</role>  <actor>Tunji Oyelana</actor> </castItem>
   </castGroup>
   <castItem> <role>Si Bero</role>
              <roleDesc>Sister to Dr Bero</roleDesc>
              <actor>Deolo Adedoyin</actor> </castItem>
   <castGroup>
      <head rend="braced">Two old women</head>
      <castItem> <role>Iya Agba</role> <actor>Nguba Agolia</actor> </castItem>
      <castItem> <role>Iya Mate</role> <actor>Bopo George</actor> </castItem>
   </castGroup>
   <castItem> <role>Dr Bero</role>
              <roleDesc>Specialist</roleDesc>
              <actor>Nat Okoro</actor> </castItem>
   <castItem> <role>Priest</role>      <actor>Gbenga Sonuga</actor> </castItem>
   <castItem> <role>The old man</role>
              <roleDesc>Bero's father</roleDesc>
              <actor>Dapo Adelugba</actor> </castItem>
</castList>

The <castList> element and its components have the following formal definitions:

<!-- 10.1.4: The castList element-->
<!ELEMENT castList %om.RR; ((%m.divtop; | %m.Incl;)*,
                                ((%component;), (%m.Incl;)*)*,
                                ((castItem | castGroup), (%m.Incl;)*)+, 
                                ((%component;), (%m.Incl;)*)*)> 
<!ATTLIST castList
      %a.global;
      TEIform CDATA 'castList'  >
<!ELEMENT castGroup %om.RR; ((%m.Incl;)*, (head, (%m.Incl;)*)?,
                   ((castItem | castGroup), (%m.Incl;)*)+,
                   (trailer, (%m.Incl;)*)?)> 
<!ATTLIST castGroup
      %a.global;
      TEIform CDATA 'castGroup'  >
<!ELEMENT castItem %om.RO; (#PCDATA | role | roleDesc | 
                   actor | %m.phrase; | %m.Incl; )*> 
<!ATTLIST castItem
      %a.global;
      type (role | list) "role"
      TEIform CDATA 'castItem'  >
<!ELEMENT role %om.RO; %phrase.seq;> 
<!ATTLIST role
      %a.global;
      TEIform CDATA 'role'  >
<!ELEMENT roleDesc %om.RR; %phrase.seq;> 
<!ATTLIST roleDesc
      %a.global;
      TEIform CDATA 'roleDesc'  >
<!ELEMENT actor %om.RO; %phrase.seq;> 
<!ATTLIST actor
      %a.global;
      TEIform CDATA 'actor'  >
<!-- end of 10.1.4-->

10.2 The Body of a Performance Text

The body of a performance text may be divided into structural units, variously called acts, scenes, stasima, entr'actes, etc. All such formal divisions should be encoded using an appropriate text-division element (<div>, <div1>, <div2>, etc.), as further discussed in section 10.2.1 Major Structural Divisions. Whether divided up into such units or not, all performance texts consist of sequences of speeches (see 10.2.2 Speeches and Speakers) and stage directions (see 10.2.3 Stage Directions). Speeches will generally consist of a sequence of chunk-level items: paragraphs, verse lines, stanzas, or (in case of uncertainty as to whether something is verse or prose) <seg> elements (see 10.2.4 Speech Contents).

The boundaries of formal units such as verse lines or paragraphs do not always coincide with speech boundaries. Units such as songs may be discontinuous or shared among several speakers. As described below in section 10.2.5 Embedded Structures, such fragmentation may be encoded in a relatively simple fashion using the linkage and aggregation mechanisms defined in chapter 14 Linking, Segmentation, and Alignment.

10.2.1 Major Structural Divisions

Large divisions in drama such as acts, scenes, stasima, or entr'actes are indicated by numbered or unnumbered <div> elements, as described in section 7.1 Divisions of the Body. The type and n attributes may be used to define the type of division being marked, and to provide a name or number for it, as in the following example:

<body>
   <div1 type='scene' n='1'>
      <head>Night&mdash;Faust's Study (i)</head>
      <!-- ... -->
   </div1> 
   <div1 type='scene' n='2'>
      <head>Outside the City Gate</head>
      <!-- ... -->
   </div1>
</body>

Where the largest divisions of a performance text are themselves subdivided, most obviously in the case of plays traditionally divided into acts and scenes, further nested text-division elements may be used, as in this example:

<body>
  <div1 type='act' n='1'>
	<head>Act One</head>
	<div2 type='scene' n='1'>
	  <stage>Pa Ubu, Ma Ubu</stage>
	  <sp><speaker>Pa Ubu</speaker> <p>Pschitt!</p> </sp>
	  <!-- ... -->
	</div2>
	<div2 type='scene' n='2'>
	  <stage>A room in Pa Ubu's house, where a magnificent
		collation is set out</stage>
	  <!-- ... -->
	</div2>
	<!-- ... -->
  </div1>
  <div1 type='act' n='2'>
	<head>Act Two</head>
	<div2 type='scene' n='1'>
	  <head>Scene One</head>
	  <!-- ... -->
	</div2>
	<div2 type='scene' n='2'>
	  <head>Scene Two</head>
	  <!-- ... -->
	</div2>
  </div1>
  <!-- ... -->
</body>

In the example above, the <div2> element has been used to represent the `French scene' convention, (where the entrance of each new set of characters is marked as a distinct unit in the text) and the <div1> element to represent the acts into which the play is divided. The elements chosen are determined only by the hierarchic position of these units in the text as a whole. If the text had no acts, but only scenes, then the scenes might be represented by <div1> elements. Equally, if a play is divided only into ‘acts’, with no smaller subdivisions, then the <div1> element might be used to represent acts. The type should be used, as above, to make explicit the name associated with a particular category of subdivision.

As an alternative to the use of numbered divisions, the encoder may represent all subdivisions with the same element, the unnumbered <div>. The second act in the above example would then be represented as follows:

<div type='act' n='2'>
   <head>Act Two</head>
   <div type='scene' n='1'>
      <head>Scene One</head>
      <!-- ... -->
   </div>
   <div type='scene' n='2'>
      <head>Scene Two</head>
      <!-- ... -->
   </div>
</div>

For further discussion of the use of numbered and unnumbered divisions, see section 7.1 Divisions of the Body.

10.2.2 Speeches and Speakers

The following elements are used to identify speeches and speakers in a performance text:

As noted above, the structure of many performance texts may be analysed as multiply hierarchic: a scene of a verse play, for example, may be divided into speeches and, at the same time, into verse lines. The end of a line may or may not coincide with the end of a speech, and vice versa. Other structures, such as songs, may be discontinuous or split up over several speeches. For some purposes it will be appropriate to regard the verse-structure as the fundamental organizing principle of the text, and for others the speech structure; in some cases, the choice between the two may be arbitrary. The discussion in the remainder of this chapter assumes that it is the speech-based hierarchy which most prominently determines the structure of performance texts, but the same mechanisms could be employed to encode a view of a performance text in which individual speeches were entirely subordinate to the formal units of prose and verse. For more detailed discussion and examples of various treatments of this fundamental issue, refer to chapter 31 Multiple Hierarchies.

The who attribute and the <speaker> element are both used to indicate the speaker or speakers of a speech, but in rather different ways. The <speaker> element is used to encode the word or phrase actually used within the source text to indicate the speaker: it may contain any string or prefix, and may be thought of as a highly specialized form of stage direction. The value of the who attribute however is a unique code, probably made up by the transcriber, which will unambiguously identify the character to whom the speech is assigned. To enforce this uniqueness, the base tag set for drama defines the value of this attribute as IDREFS. This means that the codes included in it must correspond with codes which are specified elsewhere in the document as identifiers for particular elements, typically the <role> element in the cast list where the character is named or described, as discussed in 10.1 Front and Back Matter above.

<!-- in the front matter ... -->
<castList>
   <!-- ... -->
   <castItem> <role id="m2">Menaechmus</role> </castItem>
   <castItem> <role id="pen">Peniculus</role> </castItem>
   <!-- ... -->
</castList>
<!-- ... -->

<!-- in the text ... -->
<sp who="m2"> <speaker>Menaechmus</speaker>
   <l>Responde, adulescens, quaeso, quid nomen tibist?</l> </sp>
<sp who="pen"> <speaker>Peniculus</speaker>
   <l>Etiam derides, quasi nomen non noveris?</l> </sp>
<sp who="m2"> <speaker>Menaechmus</speaker>
   <l>Non edepol ego te, quot sciam, umquam ante hunc diem</l>
   <l>Vidi neque novi; ...</l> </sp>

If present, a <speaker> element may only appear as the first part of an <sp> element. The distinction between the <speaker> element and the who attribute makes it possible to encode uniformly characters whose names are not indicated in a uniform fashion throughout the play, or characters who appear in disguise, as in the following examples:

<castList>
   <!-- ... -->
   <castItem><role id="hh">Henry Higgins</role></castItem>
   <!-- ... -->
</castList>
<!-- ... -->

<!-- in the text ... -->
<sp who="hh">
   <speaker>The Notetaker</speaker>
   <p> ... </p>
</sp>

If the speaker attributions are completely regular (and may thus be reconstructed mechanically from the values given for the who attribute), or are of no interest for the encoder of the text (as might be the case with editorially supplied attributions in older texts), then the <speaker> element need not be used; the former example above then might look like this:

<!-- in the front matter ... -->
<castList>
   <castItem><role id="m2">Menaechmus</role></castItem>
   <castItem><role id="pen">Peniculus</role></castItem>
   <!-- ... -->
   </castList>
<!-- ... -->

<!-- in the text ... -->
<sp who="m2" ><l>Responde, adulescens, quaeso, quid nomen tibist?</l></sp>
<sp who="pen"><l>Etiam derides, quasi nomen non noveris?</l></sp>
<sp who="m2" ><l>Non edepol ego te, quot sciam, umquam ante hunc diem</l>
              <l>Vidi neque novi; ...</l></sp>

More than one identifier may be listed as value for the who attribute if the speech is spoken by more than one person, as in the following example:

<stage>Nano and Castrone sing</stage>
<sp who="nan cas">
   <l>Fools, they are the only nation</l>
   <l>Worth men's envy or admiration</l>
   <!-- ... -->
</sp>

The <sp> and <speaker> elements are both declared within the core tag set (see section 6.11 Passages of Verse or Drama).

10.2.3 Stage Directions

Both between and within the speeches of a written performance text, it is normal practice to include a wide variety of descriptive directions to indicate non-verbal action. The following elements are provided to represent these:

A satisfactory typology of stage directions is difficult to define. Certain basic types such as ‘entrance’, ‘exit’, ‘setting’, ‘delivery’, are easily identified. But the list is not a closed one, and it is not uncommon to mix types within a single direction. No closed set of values for the type attribute is therefore proposed at the present time, though some suggested values are indicated in the list below, which also indicates the range of possibilities.

<stage type="setting">The throne descends.</stage>
<stage type="setting">Music</stage>
<stage type="entrance">Enter Husband as being thrown off his horse.</stage>
<stage type="exit">Exit pursued by a bear.</stage>
<stage type="business">He quickly takes the stone out.</stage>
<stage type="delivery">To Lussurioso.</stage>
<stage type="delivery">Aside.</stage>
<stage type="delivery">Not knowing what to say.</stage>
<stage type="costume">Disguised as Ansaldo.</stage>
<stage type="location">At a window.</stage>
<stage type="novelistic">Having had enough, and embarrassed
     for the family.</stage>

Where possible, the values used for the type attribute on <stage> elements should be defined within the <tagUsage> element of the TEI header (described in section 5.3.4 The Tagging Declaration). For example:

<tagUsage gi="stage">This element is used for all stage directions,
  editorial or authorial. The type= attribute on this element takes
  one or more of the following values:
  <list type="gloss">
	<label>setting</label>
	<item>describes the set</item>
	<label>blocking</label>
	<item>describes movement across stage, position, etc.</item>
	<label>business</label>
	<item>describes movement other than blocking</item>
	<label>delivery</label>
	<item>describes how the line is said</item>
	<label>motivation</label>
	<item>describes character's emotional state or through line</item>
  </list>
</tagUsage>

The <stage> element may appear both between and within <sp> elements. It may contain a mixture of phrase level elements, possibly combined into paragraphs, as in the following example:

<div1 n="1" type="act">
<stage type="setting">
<p>Scene. &mdash; A room furnished comfortably and
tastefully but not extravagantly ...
The floor is carpeted and a fire burns in the stove.
It is winter.</p>
<p>A bell rings in the hall; shortly afterwards the
door is heard to open. Enter NORA humming a tune ...</p>
</stage>
<sp><speaker>Nora</speaker>
<p>Hide the Christmas Tree carefully, Helen. Be sure the
children do not see it till this evening, when it is
dressed. <stage type="delivery">To the PORTER taking
out her purse</stage> How much?</p>
</sp>
<!-- ... -->
</div1>

The <stage> element may also be used in non-theatrical texts, to mark sound effects or musical effects, etc., as further discussed in section 10.3 Other Types of Performance Text.

The <move> element is intended to help overcome the fact that the stage directions of a printed text may often not provide full information about either the intended or the actual movement of actors etc. on stage. It may be used to keep track of entrances and exits in detail, so as to know which characters are on stage at which time. Its attributes permit a relatively formal specification for movements of characters, using user-defined codes to identify the characters involved (the who attribute), the direction of the movement (type attribute), and optionally which part of the stage is involved (where attribute). For stage-historical purposes, a perf attribute is also provided; this allows the recording of different <move> elements as taken in different performances of the same text.

The <move> element should be located at the position in the text where the move is presumed to take place. This will often coincide with a stage direction, as in the following simple example:

<stage type="entrance">  
  <move who="b" type="enter"/>
  Enter Bellafront mad.</stage>

The <move> element can however appear independently of a stage direction, as in the following example:

<sp>
   <speaker>Gent.</speaker>
   <p>Neither to you, nor any one; having no witness
to confirm my speech. <move who="lm" type="enter" where="C"/>
Lo you! here she comes. This is her very guise; and,
upon my life, fast asleep.</p>
</sp>

The <stage> element is defined by the core TEI tag set (see section 6.11 Passages of Verse or Drama). The <move> element is defined as follows:

<!-- 10.2.3: Stage directions-->
<!--Stage is defined as part of the core.-->
<!ELEMENT move %om.RO; EMPTY> 
<!ATTLIST move
      %a.global;
      who IDREFS #REQUIRED
      type CDATA #IMPLIED
      where CDATA #IMPLIED
      perf IDREFS #IMPLIED
      TEIform CDATA 'move'  >
<!-- end of 10.2.3-->

10.2.4 Speech Contents

The actual speeches of a dramatic text may be composed of running text, which must be formally organized into paragraphs, in the case of prose (see section 6.1 Paragraphs), verse lines or line groups in that of verse (see section 6.11 Passages of Verse or Drama), or <seg> elements, in case of doubt as to whether the material should be treated as verse or prose. The following elements, all of which are defined in the core, are available for marking units of prose or verse within speeches:

As members of the class metrical, the elements <l> and <lg> share the following attributes:

`met`	contains a user-specified encoding for the conventional metrical structure of the element.
`rhyme`	specifies the rhyme scheme applicable to a group of verse lines.

As a member of the class divn, the element <lg> also bears the following attributes:

`part`	specifies whether or not the division is fragmented by some other structural element, for example a speech which is divided between two or more verse stanzas.
`type`	specifies a name conventionally used for this level of subdivision, e.g. act, volume, book, section, canto, etc.

In many texts, prose and verse may be inextricably mingled; particularly in earlier printed texts, prose may be printed as verse or verse as prose, or it may be impossible to distinguish the two. In cases of doubt, an encoder may prefer to tag the dubious material consistently as verse, to tag it all as prose, to follow the typography of the source text, or to use the neutral <seg> element to contain the speech itself. When this question arises, the <tagUsage> element in the <encodingDesc> element of the header should be used to record explicitly what policy has been adopted.

The part attribute of the <l> and <lg> elements provides one simple way of indicating where the boundaries of a speech and of a verse line or line group do not coincide. The encoder may simply indicate that a line or line group is metrically incomplete by specifying the value Y or N, as in the following example:

<sp who="Face"><l part="Y">Believ't, I will.</l></sp>
<sp who="Subtle"><l part="Y">Thy worst. I fart at thee.</l></sp>
<sp who="Doll"><l>Ha' you your wits? Why gentlemen! For love &mdash;</l></sp>
<sp who="Face"><l part="Y">Sirrah, I'll strip you&mdash;.</l></sp>
<sp who="Subtle"><l part="Y">What to do? Lick figs</l>
  <l part="Y">Out at my&mdash;</l></sp>
<sp who="Face"><l part="Y">Rogue, rogue, out of all your sleights.</l></sp>

Alternatively, where the fragments of the line or line group are consecutive in the text (though possibly interrupted by stage directions), the values I (initial), M (medial), and F (final) may be used to indicate how metrical lines should be reconstituted:

<sp who="Face"><l part="I">Believ't, I will.</l></sp>
<sp who="Subtle"><l part="F">Thy worst. I fart at thee.</l></sp>
<sp who="Doll"><l>Ha' you your wits? Why gentlemen! For love &mdash;</l></sp>
<sp who="Face"><l part="I">Sirrah, I'll strip you&mdash;.</l></sp>
<sp who="Subtle"><l part="M">What to do? Lick figs</l>
  <l part="I">Out at my&mdash;</l></sp>
<sp who="Face"><l part="F">Rogue, rogue, out of all your sleights.</l></sp>

In dramatic texts, the <lg> or line group element is most often of use for the encoding of songs and other stanzaic material, as further discussed in the next section. Line groups may be fragmented across speakers in the same way as individual lines, and the same set of attributes is available to record this fact. In the following example, an <lg> element is used to represent one verse of a song, which is divided between several voices:

<stage type="head">Song &mdash; Sir Joseph</stage>
<sp who="jopo">
  <lg type="song" part="I">
    <l>I am the monarch of the sea,</l>
    <l>The ruler of the Queen's Navee.</l>
    <l>Whose praise Great Britain loudly chants.</l>
  </lg> </sp>
<sp who="he">
  <speaker>Cousin Hebe</speaker>
  <lg type="song" part="M">
    <l>And we are his sisters and his cousins and his aunts!</l>
  </lg> </sp>
<sp who="rel">
  <speaker>Rel.</speaker>
  <lg type="song" part="F">
    <l>And we are his sisters and his cousins and his aunts!</l>
  </lg> </sp>

These elements are all defined in the core, and are thus available to every TEI document without formality. A more detailed discussion of the encoding of verse is provided in chapter 9 Base Tag Set for Verse.

10.2.5 Embedded Structures

Although primarily composed of speeches, performance texts often contain other structural units such as songs or strophes which are shared among different speakers. More generally, complex nested structures of plays within plays, interpolated masques, or interludes are far from uncommon. In more modern material, comparably complex structural devices such as flashback or nested playback are equally frequent. In all kinds of performance material, it may be necessary to indicate several actions which are happening simultaneously.

A number of different devices are available within the TEI scheme to support these complexities in the general case. Texts may be composite or self-nesting (see section 7.3 Groups of Texts) and multiple hierarchies may be defined (see chapter 31 Multiple Hierarchies). The TEI encoding scheme provides a variety of linking mechanisms, which may be used to indicate temporal alignment and aggregation of fragmented structures. In this section we provide a few specific examples of the application of these techniques to performance texts:

the use of embedded <text> elements
the use of the part attribute on fragmentary <lg> elements
the use of the next and prev attributes on fragments of embedded structures to join them into a larger whole
the use of the <join> element to define a `virtual element' composed of the fragments indicated

Full information and descriptions are provided in other chapters of this document, as indicated in the individual discussions.

When a song appears in its entirety within a single speech, it may be treated as an extended quotation or as an embedded <text> element, or both, according to the preference of the encoder. In the following example, an embedded song is treated as a self-standing text:

<sp><speaker>Kelly</speaker>
   <stage>(calmly).</stage>
   <p>Aha, so you've bad minds along with th' love of gain.
      You thry to pin on others th' dirty decorations that
      may be hangin' on your own coats.
   <stage>(He points, one after the other at Conroy, Bull,
          and Flagonson. Lilting):</stage>
      <q><text><body>
         <l>Who were you with last night?</l>
         <l>Who were you with last night?</l>
         <l>Will you tell your missus when you go home</l>
         <l>Who you were with last night?</l>
      </body></text></q></p></sp>
<sp><speaker>Flagonson</speaker>
   <stage>(in anguished indignation).</stage>
   <p>This is more than a hurt to us: this hits at the
      decency of the whole nation!</p>
</sp>

It might, however, also be treated simply as a quotation:

<sp><speaker>Kelly</speaker>
   <stage>(calmly).</stage>
   <p>Aha, so you've bad minds along with ...
   <stage>(He points, one after the other at Conroy, Bull,
          and Flagonson.  Lilting):</stage>
      <q>
         <l>Who were you with last night?</l>
         <l>Who were you with last night?</l>
         <l>Will you tell your missus when you go home</l>
         <l>Who you were with last night?</l>
      </q></p></sp>
<sp><speaker>Flagonson</speaker>
   <!-- ... -->
</sp>

When an embedded structure extends across more than one <sp> element, each of its constituent parts must be regarded as a distinct fragment; the problem then facing the encoder is to reconstitute the interrupted whole in some way.

As already noted above, the part attribute may be used to indicate that an <l> element contains a partial, not a complete, verse line. The same attribute may be used on the <lg> element, to indicate that the line group is partial rather than complete, thus:

<sp><speaker>Kelly</speaker>
   <stage>(wheeling quietly in his semi-dance,
          as he goes out):</stage>
   <lg type="stanza" part="I">
      <l>Goodbye to holy souls left here,</l>
      <l>Goodbye to man an' fairy;</l>
   </lg>
</sp>
<sp><speaker>Widda Machree</speaker>
   <stage>(wheeling quietly in her semi-dance,
          as she goes out):</stage>
   <lg type="stanza" part="F">
      <l>Goodbye to all of Leicester Square,</l>
      <l>An' the long way to Tipperary.</l>
   </lg>
</sp>

When the fragments of a song are separated by other intervening dialogue, or even when not, they may be linked together with the next and prev attributes defined in section 14.7 Aggregation. For example, the line groups making up Ophelia's song might be encoded as follows:

<div1 n="4" type="act">
  <!-- ... -->
  <div2 n="5" type="scene">
    <stage>Elsinore. A room in the Castle.</stage>
    <!-- ... -->
    <stage type="setting">Enter Ophelia, distracted.</stage>
    <sp who="oph"> <speaker>Ophelia</speaker>
      <p>Where is the beauteous Majesty of Denmark?</p>
    </sp>
    <sp who="qu"> <speaker>Queen</speaker>
      <p>How now, Ophelia?</p>
    </sp>
    <sp who="oph"> <speaker>Ophelia</speaker>
      <stage>Singing</stage>
      <lg next="tl2" id="tl1" type="song" part="Y">
        <l>How should I your true-love know</l>
        <l>From another one?</l>
        <l>By his cockle hat and staff</l>
        <l>And his sandal shoon.</l>
      </lg>
    </sp>
    <sp who="qu"> <speaker>Queen</speaker>
      <p>Alas, sweet lady, what imports this song?</p>
    </sp>
    <sp who="oph"> <speaker>Ophelia</speaker>
      <p>Say you? Nay, pray you mark.</p>
      <stage>Sings</stage>
      <lg prev="tl1" id="tl2" type="song" part="Y">
        <l>He is dead and gone, lady,</l>
        <l>He is dead and gone;</l>
        <l>At his head a grass-green turf,</l>
        <l>At his heels a stone.</l>
      </lg>
      <p>O, ho!</p>
    </sp>
    <!-- ... -->
  </div2>
</div1>

The next and prev attributes are discussed in section 14.7 Aggregation: they form part of the additional tag set for alignment and linking, and are therefore not automatically available to dramatic texts. To enable this tag set as well as the base tag set for drama, the document type declaration might take the following form:

<!DOCTYPE TEI.2 PUBLIC "-//TEI P4//DTD Main Document Type//EN"
                       "tei2.dtd" [
  <!ENTITY % TEI.XML     'INCLUDE'>
  <!ENTITY % TEI.drama   'INCLUDE'>
  <!ENTITY % TEI.linking 'INCLUDE'>
]>

See chapter 3 Structure of the TEI Document Type Definition for general discussion of the way in which TEI tag sets are enabled.

The fragments of Ophelia's song might also be linked together using the <join> mechanism described in section 14.7 Aggregation. The <join> element is specifically intended to encode the fact that several discontiguous elements of the text together form one `virtual' element. Using this mechanism, the example might be encoded as follows:

<text>
  <body>
    <div1 n="4" type="act">
      <!-- ... -->
      <div2 n="5" type="scene">
        <stage type="setting">Elsinore. A room in the Castle.</stage>
        <!-- ... -->
        <sp who="qu"> <speaker>Queen</speaker>
          <p>How now, Ophelia?</p>
        </sp>
        <sp who="oph"> <speaker>Ophelia</speaker>
          <stage type="delivery">Singing</stage>
          <lg id="tl1" type="song" part="Y">
            <l>How should I your true-love know</l>
            <l>From another one?</l>
            <l>By his cockle hat and staff</l>
            <l>And his sandal shoon.</l>
          </lg>
        </sp>
        <sp who="qu"> <speaker>Queen</speaker>
          <p>Alas, sweet lady, what imports this song?</p>
        </sp>
        <sp who="oph"> <speaker>Ophelia</speaker>
          <p>Say you? Nay, pray you mark.</p>
          <stage type="delivery">Sings</stage>
          <lg id="tl2" type="song" part="Y">
            <l>He is dead and gone, lady,</l>
            <l>He is dead and gone;</l>
            <l>At his head a grass-green turf,</l>
            <l>At his heels a stone.</l>
          </lg>
          <p>O, ho!</p>
          <join type="lg" targets="tl1 tl2"/>
        </sp>
        <!-- ... -->
      </div2>
    </div1>
  </body>
</text>

The location of the <join> element is not significant; here it has been placed shortly after the conclusion of the song, in order to have it close to the fragments it unifies.

Like the next and prev attributes, the <join> element requires the additional tag set for linking, which is selected as shown above.

10.2.6 Simultaneous Action

In printed or written versions of performance texts, a variety of techniques may be used to indicate the temporal alignment of speeches or actions. Speeches may be printed vertically aligned on the page, or braced together; stage directions (e.g. ‘Speaking at the same time’) are also often used. In operatic or musical works in particular, the need to indicate timing and alignment of individual parts of a song may lead to very complex layout.

One simple method of indicating the temporal alignment of speeches or actions is to use the corresp attribute discussed in section 14.4 Correspondence and Alignment, as in the following example:

<sp who="m"> <speaker>Mangan</speaker>
  <stage type="delivery">wildly</stage>
  <p>Look here: I'm going to take off all my clothes.</p>
  <stage type="action">he begins tearing off his coat.</stage>
</sp>
<sp id="s1" who="lu"> <speaker>Lady Utterword</speaker>
  <p>Mr Mangan!</p>
</sp>
<sp id="s2" who="cs"> <speaker>Captain Shotover</speaker>
  <p>Whats that?</p>
</sp>
<sp id="s3" who="h"> <speaker>Hector</speaker>
  <p>Ha! ha! Do. Do.</p>
</sp>
<sp id="s4" who="e"> <speaker>Ellie</speaker>
  <p>Please dont.</p>
</sp>
<stage corresp="s1 s2 s3 s4" id="d1" type="delivery">in consternation</stage>
<sp who="mh"> <speaker>Mrs. Hushabye</speaker>
  <stage type="action">catching his arm and stopping him</stage>
  <p>Alfred: for shame! Are you mad?</p>
</sp>

In the original, the stage direction ‘in consternation’ is printed opposite a brace grouping all four speeches, indicating that all four characters speak at once, and that the stage direction applies to all of them. In the example, the <stage> element has been moved to an arbitrary place, and the four speeches with which it is to be associated are specified by identifier as the value of the corresp attribute. This attribute, which is enabled by the linking tag set, provides the simplest way of indicating the temporal alignment of speeches or actions in a play.

More powerful and more precise mechanisms for temporal alignment are defined in chapter 11 Transcriptions of Speech. These would be appropriate for encodings the focus of which is on the actual performance of a text rather than its structure or formal properties. The tag set described in that chapter includes a large number of other detailed proposals for the encoding of such features as voice quality, prosody, etc., which might be relevant to such a treatment of performance texts.

10.3 Other Types of Performance Text

Most of the elements and structures identified thus far are derived from traditional theatrical texts. Although other performance texts, such as screenplays or radio scripts, have not been discussed specifically, they can be encoded using the elements and structures listed above. Encoders may however find it convenient to use, as well, the additional specialized elements discussed in this section. For scripts containing very detailed technical information, the <tech> element discussed in section 10.3.1 Technical Information may also be useful.

Like other texts, screenplays and television or radio scripts may be divided into text divisions marked with <div> or <div1>, etc. Within units corresponding with the traditional ‘act’ and ‘scene’, further subdivisions or sequences may be identified, composed of individual ‘shots’, each associated with a single camera angle and setting. Shots and sequences should be encoded using an appropriate text-division element (i.e., a <div3> element if numbered division elements are in use and the next largest unit is a <div2>, or a <div> element if un-numbered divisions are in use) specifying sequence or shot as the value of the type attribute, as appropriate.

It is normal practice in screenplays and radio scripts to distinguish directions concerning camera angles, sound effects, etc., from other forms of stage direction. Such texts also generally include far more detailed specifications of what the audience actually sees: descriptions of actions and background, etc. Scripts derived from cinema and television productions may also include texts displayed as captions superimposed on the action. All of these may be encoded using the general purpose <stage> element discussed in section 10.2.3 Stage Directions, and distinguished by means of its type attribute. Alternatively, or in addition, the following more specific elements may be used, where clear distinctions can be made:

Some examples of the use of these elements follow:

<camera>Angle on Olivia.</camera>
<view>Ryan's wife, standing nervously alone on the sidelines,
biting her lip. She's scared and she shows it.</view>

Where particular words or phrases within a direction are emphasized (by change of typeface or use of capital letters), an appropriate phrase-level element may be used to indicate the fact, as in the following examples, where certain words in the original are given in small capitals:

<view>George glances at the window--and freezes.
<camera>New angle--shock cut</camera> Out the window
the body of a dead man suddenly slams into
<hi>frame</hi>. He dangles grotesquely,
held up by his coat caught on a protruding bolt.
George gasps. The train <hi>whistle</hi> screams.</view>

<view>Ext. TV control van&mdash;Early morning.
The <name>T.V. announcer</name> from the Ryan interview
stands near the Control Van, the lake in b.g.</view>
<sp who="announcer"> <speaker>T.V. Announcer</speaker>
<p>Several years ago, Jack Ryan was a highly
successful hydroplane racer ...</p>
</sp>

All of these elements, like other stage directions, can appear both within and between speeches.

<sp>
   <speaker>TV Announcer VO</speaker>
   <p>Working with Ryan are his two coworkers&mdash;
Strut Bowman, the mechanical engineer&mdash;
<view><camera>Angle on Strut</camera>
standing in the tow boat, walkie-talkie in hand,
watching Ryan carefully.</view>
&mdash;and Roger Dalton, a rocket
systems analyst, and one of the scientists
from the Jet Propulsion Lab ...</p>
</sp>

<sp> <speaker>Benjy</speaker>
   <p>Now to business.</p>
</sp>
<sp> <speaker>Ford and Zaphod</speaker>
   <p>To business.</p> </sp>
<sound>Glasses clink.</sound>
<sp> <speaker>Benjy</speaker>
   <p>I beg your pardon?</p> </sp>
<sp> <speaker>Ford</speaker>
   <p>I'm sorry, I thought you were proposing a toast.</p> </sp>

<camera>Zoom in to overlay showing some stock film
  of hansom cabs galloping past.</camera>
<caption>London, 1895.</caption>
<caption>The residence of Mr Oscar Wilde.</caption>
<sound>Suitably classy music starts.</sound>
<view>Mix through to Wilde's drawing room. A crowd of suitably
  dressed folk are engaged in typically brilliant conversation,
  laughing affectedly and drinking champagne.</view>
<sp who="tj">
   <speaker>Prince of Wales</speaker>
   <p>My congratulations, Wilde. Your latest play is a great success.</p>
</sp>

10.3.1 Technical Information

Traditional stage scripts may contain additional technical information about such production-related factors as lighting, `blocking' (that is, detailed notes on actors' movements), or props required at particular points. More technical information about intended production effects may also appear in published versions of screenplays or movie scripts. Where these are presented simply as marginal notes, they may be encoded using the general-purpose <note> element defined in section 6.8 Notes, Annotation, and Indexing. Alternatively, they may be formally distinguished from other stage directions by using the specialized <tech> element:

Like stage directions, <tech> elements can appear anywhere within a speech or between speeches.

The elements discussed in the section are formally defined as follows:

<!-- 10.3.1: Screenplays and other technical matters-->
<!ELEMENT view %om.RO; %specialPara;> 
<!ATTLIST view
      %a.global;
      TEIform CDATA 'view'  >
<!ELEMENT camera %om.RR; %paraContent;> 
<!ATTLIST camera
      %a.global;
      type CDATA #IMPLIED
      TEIform CDATA 'camera'  >
<!ELEMENT sound %om.RO; %paraContent;> 
<!ATTLIST sound
      %a.global;
      type CDATA #IMPLIED
      discrete ( y | n | u ) "u"
      TEIform CDATA 'sound'  >
<!ELEMENT caption %om.RO; %paraContent;> 
<!ATTLIST caption
      %a.global;
      TEIform CDATA 'caption'  >
<!ELEMENT tech %om.RO; %paraContent;> 
<!ATTLIST tech
      %a.global;
      type ( light | sound | prop | block ) #IMPLIED
      perf IDREFS #IMPLIED
      TEIform CDATA 'tech'  >
<!-- end of 10.3.1-->

Up: Contents Previous: 9 Base Tag Set for Verse Next: 11 Transcriptions of Speech

Text Encoding Initiative

The XML Version of the TEI Guidelines

10 Base Tag Set for Drama