[MEI-L] Fwd: looking for advice/suggestions for encoding alternative realisations of duration

Karen Desmond kdesmond at brandeis.edu
Sun Jul 10 01:41:22 CEST 2016 

Dear all,
I am looking for advice about encoding specific durations for notes where
the original notation is not specific, and there may be more than one
interpretation of the durations. This is in reference to encoding mensural
notation, but I am posting here rather than on mensural IG list because I
imagine there might be other similar situations in CWMN (such as the
realisation of notes inégales or swung notes in jazz).

In most standard late thirteenth-century compositions, it is understood
that a breve divides into three equal semibreves and two unequal (the
default is that a semibreve is 1/3 breve), so this is equal to a breve
duration:
<note="semibrevis">
<note="semibrevis">
<note="semibrevis">
And if two semibreves, the second semibreve is 2/3 of a breve:
<note="semibrevis">
<note="semibrevis" num="1" numbase="2">

If 4,5,6,7 semibreves were placed for one breve value, their relative
durations are ambiguous and one solution is to encode them simply as evenly
divided, like for this group of 7:


<note dur="semibrevis" num="7" numbase="1" />

<note dur="semibrevis" num="7" numbase="1" />

<note dur="semibrevis" num="7" numbase="1" />

<note dur="semibrevis" num="7" numbase="1" />

<note dur="semibrevis" num="7" numbase="1"/>

<note dur="semibrevis" num="7" numbase="1" />

<note dur="semibrevis" num="7" numbase="1"/>


This is pretty straightforward (although an editor might want to offer some
suggestions as to how the durations of a group of 7 semibreves might be
interpreted since they were most likely not sung as 7 equal semibreves).

In any case, my question has to do with early fourteenth-century notation
(like in the Fauvel manuscript), when it becomes unclear whether the breve
divides into three equal parts, or into two equal parts (i.e., to use a
slightly later terminology we don't always know whether the tempus level
divisions are perfect, where a breve contains three equal semibreves, or
imperfect, where a breve contains two equal semibreves).

The most straightforward transcription could simply encode the semibreves
as evenly (as above) no matter how many occur in the time of one breve,
whether, 2, 3, 4, or 5 etc.:

<note dur="semibrevis" num="2" numbase="1"/>

<note dur="semibrevis" num="2" numbase="1"/>


OR

<note dur="semibrevis" num="3" numbase="1"/>

<note dur="semibrevis" num="3" numbase="1"/>

<note dur="semibrevis" num="3" numbase="1"/>


OR


<note dur="semibrevis" num="4" numbase="1" />

<note dur="semibrevis" num="4" numbase="1" />

<note dur="semibrevis" num="4" numbase="1" />

<note dur="semibrevis" num="4" numbase="1" />

ETC.

However, what I would like to do (especially to make consonances line up
properly between polyphonic parts) is to offer alternative durational
interpretations for both tempus perfectum and tempus imperfectum. One
suggestion was to use <app>. Something like this, where I would offer a
suggested reading for three semibreves according to tempus perfectum, and
tempus imperfectum, and then also show how they are found in the
manuscript. For example, this is the first group of four semibreves in the
first measure of the triplum:


<app>

<rdg type="p"> <!-- tempus perfectum -->

     <note dur="semibrevis" num="2" numbase="1" oct="4" pname="a" />

     <note dur="semibrevis" num="2" numbase="1" oct="4" pname="g" />

     <note dur="semibrevis"  oct="4" pname="a" />

     <note dur="semibrevis"  oct="4" pname="b" >

</rdg>

<rdg type="i"> <!-- tempus imperfectum -->

<note dur="semibrevis" num="2" numbase="1" oct="4" pname="a"/>

    <note dur="semibrevis" num="2" numbase="1" oct="4" pname="g" />

    <note dur="semibrevis" num="2" numbase="1" oct="4" pname="a" />

    <note dur="semibrevis" num="2" numbase="1" oct="4" pname="b" >

</rdg>


<lem>

<note dur="semibrevis" oct="4" pname="a" />

     <note dur="semibrevis" oct="4" pname="g" />

     <note dur="semibrevis"  oct="4" pname="a" />

     <note dur="semibrevis"  oct="4" pname="b" >

</lem>


</app>


AND this is the next group of three semibreves found in measure 2 of the
triplum:


<app>

<rdg type="p"> <!-- tempus perfectum -->

   <note dur="semibrevis"  oct="4" pname="g" />
   <note dur="semibrevis"  oct="4" pname="f" />
   <note dur="semibrevis"  oct="4" pname="g" />

</rdg>

<rdg type="i"> <!-- tempus imperfectum -->

<note dur="semibrevis" num="2" numbase="1"  oct="4" pname="g"/>

   <note dur="semibrevis" num="2" numbase="1"  oct="4" pname="f" />

   <note dur="semibrevis"  oct="4" pname="g" />

</rdg>


<lem>

<note dur="semibrevis"  oct="4" pname="g" />
   <note dur="semibrevis"  oct="4" pname="f" />
   <note dur="semibrevis"  oct="4" pname="g" />

</lem>

</app>


I would also have to wrap the staffDef element, where the mensuration
attributes is encoded to offer a choice between tempus perfectum and tempus
imperfectum with app and rdg elements.


My concern is that it seems like an awful lot of extra code to add in for
all the semibreves that occur in these pieces. In most of these pieces, the
entire top voice is composed of strings of these groups of semibreves, and
the interpretation really is strictly rule-based - i.e., in tempus
perfectum a group of four semibreves is always interpreted in a particular
way, similarly in tempus imperfectum a group of four semibreves is always
interpreted in a particular way. Is the <app> solution the best way to do
this, or does anyone else have any other ideas? My other concern about app
is that these are not readings occuring in different sources (the more
usual use for the critical apparatus) but rather they are editorial
interpretations. Perhaps something like <orig> and <reg> would work better,
but can you have two different <reg> elements?


The attached examples show the beginning of one piece as it would be
encoded in tempus perfectum, and then in tempus imperfectum. The images
show the two renderings in Verovio. What I want to be able to do is somehow
encode both choices of interpretation within the one file.


Tempus perfectum:

[image: Inline image 1]


Tempus imperfectum:

[image: Inline image 2]


All my best,

Karen

-- 
Karen Desmond, Assistant Professor of Music
Department of Music, Brandeis University
http://www.arsmusicae.org/wordpress




-- 
Karen Desmond, Assistant Professor of Music
Department of Music, Brandeis University
http://www.arsmusicae.org/wordpress
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.uni-paderborn.de/pipermail/mei-l/attachments/20160709/feee581c/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 162254 bytes
Desc: not available
URL: <http://lists.uni-paderborn.de/pipermail/mei-l/attachments/20160709/feee581c/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 160422 bytes
Desc: not available
URL: <http://lists.uni-paderborn.de/pipermail/mei-l/attachments/20160709/feee581c/attachment-0001.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fauvel_nous_MEI_mensural_excerpt_tempus_perfectum.mei
Type: application/octet-stream
Size: 14747 bytes
Desc: not available
URL: <http://lists.uni-paderborn.de/pipermail/mei-l/attachments/20160709/feee581c/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fauvel_nous_MEI_mensural_excerpt_tempus_imperfectum.mei
Type: application/octet-stream
Size: 14853 bytes
Desc: not available
URL: <http://lists.uni-paderborn.de/pipermail/mei-l/attachments/20160709/feee581c/attachment-0001.obj>

More information about the mei-l mailing list