For comparison only, around 18 years ago I was involved in a liturgical project which transcribed and edited a number of English medieval benedictine liturgical service books. In those days we modified TEI P4 to provide custom elements and referenced a full repository of the Corpus Antiphonalium Officii with pointers. (Where we came across new antiphons/responds/etc that weren't in CAO we added them manually. 

The structure we used in that CAO textbase was something like: 


<ant previd="c2119" nextid="c2121" id="c2120">
    <usage ms="Alb2" code="01035000">[16r] lauds-a1</usage>
    <usage ms="Bat" code="01035000">[11r] lauds-a1</usage>
    <usage ms="Cdm" code="01035000">[16v] lauds-a1</usage>
    <usage ms="Ely" code="01035000">[12v] lauds-a1</usage>
    <usage ms="Evm" code="01035000">[26v] lauds-a1</usage>
    <usage ms="Evm" code="01036009">[21r] lauds-a1</usage>
    <usage ms="Hyd" code="01035000">[12v] lauds-a1</usage>
    <usage ms="Mch" code="01035000">[14r] lauds-a1</usage>
    <usage ms="Wcb" code="01043000">[50r] lauds-a1</usage>
    <usage ms="Wor" code="01035000">[9v] lauds-a1</usage>
<aBody wit="CAO-C CAO-G CAO-B CAO-E CAO-M CAO-V CAO-H CAO-R CAO-D CAO-F CAO-S CAO-L Hyd Alb2 Evm Mch Cdm Ely Wor Wcb Bat">
De Sion veniet dominus omnipotens
  <rdg wit="CAO-C CAO-G CAO-B CAO-E CAO-V CAO-H CAO-R CAO-F CAO-S CAO-L Hyd Alb2 Evm Mch Cdm Ely Wor Wcb Bat"> ut salvum faciat populum suum</rdg>
   <rdg wit="CAO-M"> ut salvum faciet populum suum</rdg>
   <rdg wit="CAO-D"> qui salvum faciet populum suum</rdg>

This was pointed to from the body of the individual edition with something like: 


<Day num="5" code="01035000">
        <rubric type="dayTitle">Feria quinta &punctus; </rubric>
        <service name="lauds"><rubric type="serviceTitle">In laudibus &punctus;</rubric>
          <antiphon>&A; <xptr href="Antiphons" type="aBody" from="ID(c2120)"/></antiphon>
          <incipit type="psalm">&P; Miserere &punctus; </incipit>
          <antiphon>&A; <xptr href="Antiphons" type="aBody" from="ID(c2121)"/></antiphon>
          <incipit type="psalm">&P; Domine deus salutis &punctus; </incipit>
          <antiphon>&A; <xptr href="Antiphons" type="aBody" from="ID(c1920)"/></antiphon>
          <incipit type="psalm">&P; Domine refugium &punctus; </incipit>
          <antiphon>&A; <xptr href="Antiphons" type="aBody" from="ID(c2503)"/></antiphon>
          <incipit type="psalm">&P; Cantemus &punctus; </incipit>
          <antiphon>&A; <xptr href="Antiphons" type="aBody" from="ID(c2415)"/></antiphon>
          <incipit type="psalm">&P; Laudate &punctus; </incipit>

The benefit of this was there was very little running text per se in the editions themselves. (We didn't care about psalms or content of lectio, the project was more concerned with the order of antiphons and responds in the particular services.) The strength is that distributed pointing which then enabled us to generate information like the <usage> elements (stored only because it was quicker/easier for display than looking them up on the fly.

Today, of course, I'd do this very differently but I would probably still avoid the idea of considering it to be a performance text (with stage directions, etc.). Liturgy isn't really my area of interest otherwise I might have updated the XML after the PI passed away.

Just in case that historical view is helpful.



Dr James Cummings, [log in to unmask]

School of English Literature, Language, and Linguistics, Newcastle University

From: TEI (Text Encoding Initiative) public discussion list <[log in to unmask]> on behalf of Efraim Feinstein <[log in to unmask]>
Sent: 26 September 2018 06:11:16
To: [log in to unmask]
Subject: Re: Marking up liturgical texts

Another potential option, depending on how you think of those instructions (part of the text or an annotation): tei:note/@type=instruction

On Tue, Sep 25, 2018, 2:54 PM Daniel Schwartz <[log in to unmask]> wrote:
Thank you both for the examples.

On Tue, Sep 25, 2018 at 1:43 PM Paul Schaffner <[log in to unmask]> wrote:
True in general for phase 1 texts (thank you), but in this case I
did some quick and imprudent edits before posting the file...
Couldn't help myself :)

On Tue, Sep 25, 2018, at 14:36, Lou Burnard wrote:
> The P5 version of this along with many other liturgical texts is of
> course included in EEBO/TCP,and thus readily available from the Oxford
> Text Archive and elsewhere:
> try
> for example
> On 25/09/18 20:25, Paul Schaffner wrote:
> > Daniel,
> >
> > Since I'm quoted in your query, I'll try to put my text where my
> > mouth is.
> >
> > Here's a crudely tagged (P4-ish) copy of the Book of Common Prayer, 1549
> > edition (STC 16270A):
> > Using mostly drama tags, but a very limited generic tag set designed for
> > the bulk tagging of tens of thousands of diverse books. An advantage of
> > using a limited tag set to begin with is that it forces you to think of
> > the overall structure of the markup before getting lost in the detail...
> >
> >
> > pfs
> >
> > On Tue, Sep 25, 2018, at 12:19, Daniel Schwartz wrote:
> >> I just agreed to help out with TEI encoding of a liturgy project. If list
> >> members could provide links to examples of TEI used for liturgical texts
> >> and/or schemas for the same that would be most appreciated. Thank you.
> >> All best,
> >> Dan
> >>
> >>
> >> On Mon, Apr 3, 2017 at 1:51 PM Paul Schaffner <[log in to unmask]>
> >> wrote:
> >>
> >>> I have raised this question before, back when we were encoding
> >>> lots of variants on the Englsh and Latin prayer books. We never did
> >>> find a wholly satisfactory encoding for liturgical texts, partly because
> >>> they are so various amongst themselves (what works for one will
> >>> not necessarily work for another), and partly because they can be
> >>> so various within a single text: the div-level organization can be
> >>> very confusing, without any clear hierarchy, with one labelled chunk
> >>> succeeding another without evident organization; within these
> >>> chunks, once finds a mixture of (often unmarked and often abbreviated)
> >>> 'rubrics' (i.e, instructions, or narratives that serve in place of
> >>> instructions;
> >>> descriptions of alternative versions, etc.); readings, often in the
> >>> form of quotations or centos of quotations; prayers, sometimes
> >>> in the form of litanies, sometimes not; call-and-response type
> >>> structures
> >>> (antiphons); hymns; etc., all rather jumbled together.
> >>>
> >>> Most of the time, when the format supported it, we opted for
> >>> dramatic tags: <speaker> for  "Priest:"  "People:"; <stage>
> >>> for rubrics (instructions); and <sp> for short direct speeches.
> >>> But this structure could not be maintained, since dramatically
> >>> formatted text would immediately give way to (say)
> >>>
> >>> Lectio IV.
> >>>
> >>> or
> >>>
> >>> PRAYER.
> >>> ...
> >>>
> >>> which might or might not be spoken by the same person who had
> >>> just been given a <sp> tag. Leading one to doubt whether to
> >>> create a new div at that point; or a floating text within the
> >>> currently open <sp>, or something different altogether.  We might
> >>> start tagging rubrics as <stage>, then find a rubric that turned into
> >>> a multi-page commentary, for which <stage> seemed very
> >>> awkward at best.
> >>>
> >>> In a word, though we used the drama tags, and supported
> >>> their semantic appropriateness, the actual structure of the documents
> >>> invariably seemed to make them problematic to apply, or to
> >>> apply consistently or confidently. I think in the end we would have
> >>> done better to have used a much more generic div/p/seg structure
> >>> -- though that has its problems too.
> >>>
> >>> pfs
> >>>
> >>> On Mon, Apr 3, 2017, at 14:21, Lou Burnard wrote:
> >>>> Well, yes, <seg> CAN be used like that, to cover all sorts of edge
> >>>> cases. (Though you may find it needs to be wrapped up in a <p> or
> >>>> something similar).
> >>>> But if you;re interested in marking up the liturgy semantically, why not
> >>>> do the job properly, using <sp> and <p> and <lg> for the various
> >>>> scripted parts, and <stage> for the "directions" ? Alternatively, you
> >>>> could use <note>, perhaps with a @type attribute to indicate that it's
> >>>> an instruction. Either of these options has the advantage that it can
> >>>> appear within or between paragraphs.
> >>>>
> >>>>
> >>>>
> >>>> On 03/04/17 16:57, Hayim Lapin wrote:
> >>>>> Hello all,
> >>>>>
> >>>>> A colleague working on liturgical texts (in fact, seasonably
> >>>>> appropriate: the text for the passover seder) asked how her group
> >>>>> might encode liturgical instructions ("here the cup is raised"; "In X
> >>>>> circumstances the following is recited").
> >>>>>
> >>>>> I suggested <seg> using @type and @subtype attributes as necessary to
> >>>>> specify further.  However, my guess is that there are people on this
> >>>>> list who have far more experience than me with this sort of thing.
> >>>>>
> >>>>> Any suggestions?
> >>>>> Many thanks,
> >>>>> --
> >>>>> Hayim Lapin
> >>>>> Professor of History
> >>>>> Robert H. Smith Professor of Jewish Studies
> >>>>> 2115 Francis Scott Key Hall
> >>>>> College Park, MD 20190
> >>>>> +1 301 405 4296
> >>>>>  |
> >>>>
> >>> --
> >>> Paul Schaffner  Digital Content & Collections
> >>> University of Michigan Libraries
> >>> [log in to unmask] |
> >>>
> >>
> >> --
> >> Dr. Daniel L. Schwartz
> >> Associate Professor of History
> >> Associate Director of the The Center of Digital Humanities Research
> >> <>
> >>
> >> Texas A&M University
> >> 4213 TAMU
> >> College Station, TX 77843-4213
> >>
> >> *Paideia and Cult <>:
> >> Christian Initiation in Theodore of Mopsuestia* (Center for Hellenic
> >> Studies, 2013)
> >> http;//
> >> <>
> >

Paul Schaffner  Digital Content & Collections
University of Michigan Libraries
[log in to unmask] |

Dr. Daniel L. Schwartz
Associate Professor of History
Associate Director of the The Center of Digital Humanities Research

Texas A&M University
4213 TAMU
College Station, TX 77843-4213

Paideia and Cult: Christian Initiation in Theodore of Mopsuestia (Center for Hellenic Studies, 2013)