Hi all,

If I can just play devil's advocate for a moment and note that the conversation here has focussed primarily on active digital editing for the creation of new digital editions whereas the TEI Guidelines for text critical apparatus are intended to cover a wider range of representation of critical editions in digital form. The entire reason for the existence of some of the recommendations in that chapter (as obviously in need of revision as it is) is not just for the creation of new digital editions where the <app>s are being created by a loving and caring editor, but also for the retrospective conversion of existing print editions (and related materials) or legacy data migration in all sorts of forms and formats. 

I've been thinking a lot about the use of modal verbs in the TEI Guidelines because I've been getting around to issue #1175 and going through and checking these. (I've mostly been focussing on when we say 'should' and double checking that this is not something that is in fact required by the TEI and should say 'must'.) The TEI Guidelines have said that you 'may' do this which you'll note in the Style Guide means that it is entirely optional whether you do it or not.


This word, or the terms "REQUIRED" or "SHALL", mean that this is an absolute requirement of the TEI Guidelines for production of a TEI conformant file.

This phrase, or the phrase "SHALL NOT", mean that this is an absolute prohibition of the TEI Guidelines for production of a TEI conformant file.

This word, or the adjective "RECOMMENDED", mean that there may exist valid reasons in particular circumstances to ignore a particular recommendation, but the full implications must be understood and carefully weighed before choosing a different course.

This phrase, or the phrase "NOT RECOMMENDED" mean that there may exist valid reasons in particular circumstances when the particular behavior is acceptable or even useful, but the full implications should be understood and the case carefully weighed before implementing any behavior so described.

This word, or the adjective "OPTIONAL", mean that a recommendation is truly optional. One user may choose to follow the recommendation because a particular project requires it or feels that it enhances their work while another project may choose to not follow this recommendation.

Changing this to 'should' or 'recommended' means that while a project may ignore this they are doing so in full knowledge that we recommend against it. My central worry is that I fear projects have used the 'empty <rdg> means this witness is silent at this point' while others that have used the 'not having an <rdg> for that witness means it does not have this bit of text'. Changing this to a 'should' kick those projects out of the fold of doing things 'properly'.  Moreover, suggesting that the way to do this is to have an empty element rules out other methods that may have resulted from retrospective conversion or legacy data migration. For example, those where <rdg> elements whose only content was a <note> element detailing some information about this witness having omitted this bit of text or something similar.

For the record, I agree that in creating a new scholarly digital edition that I personally would use an empty <rdg/> (and would also not use a <lem>) but before making this change I'd spare a thought for those who have taken the TEI at its word that this behaviour is entirely optional, or those doing some form of conversion in the future who may not be able to determine whether a witness is missing or not from an <app> because of the nature of their source media or format.

Your friendly neighbourhood devil's advocate,



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 Elisa <[log in to unmask]>
Sent: 19 August 2018 20:51:42
To: [log in to unmask]
Subject: Re: empty <rdg> elements vs omission of <rdg> elements
Thanks, Marjorie—and your chapter is really helpful for concisely summarizing the distinct approaches we have been discussing, and explaining that neutral position with respect to potentially equivalent witnesses. 

I, for, one agree with Hugh that the proposed change to recommend empty rdg elements is an uncontroversial tiny edit for the Guidelines, but likely connected with more edits to come. I just opened a ticket now  and maybe that will help us connect some related tasks:

Elisa Beshero-Bondar, PhD 
Director, Center for the Digital Text
Associate Professor of English 
University of Pittsburgh at Greensburg
150 Finoli Drive, Greensburg, PA 15601 USA
E-mail: [log in to unmask] | Development site:

Typeset by hand on my iPad

On Aug 19, 2018, at 1:27 PM, Marjorie Burghart <[log in to unmask]> wrote:

Hi Elisa!

Yes, that is what I mean by "neutral recording". Without a lemma, you do not choose one witness or branch of the tradition over the other(s), that is what I call "neutral", but you still record the variance of the text. And I do use empty lem elements when there is an addition in one or more witnesses - with the same logic as for omissions. I've elaborated at greater length on that in chapter 4 of this online textbook: (and also in the online course). Feedback welcome :)

All this indeed depends on the textual scholarship tradition from which you depend. Some people will never use the notion of lemma, other will consider that an edition without a lemma simply cannot be called "critical". I think it is best for the TEI to cater for the needs of each school (as it does), rather than choose one and recommend practices in this department.


----- Mail original -----
De: "Elisa Beshero-Bondar" <[log in to unmask]>
À: "Marjorie Burghart" <[log in to unmask]>
Cc: [log in to unmask]
Envoyé: Dimanche 19 Août 2018 14:20:33
Objet: Re: empty <rdg> elements vs omission of <rdg> elements

Hi Marjorie,
Just to be sure I understand “neutral”, do you mean that when there is no use of lem, the absence of a rdg for a specific witness means something more like, “at this moment this rdg has no material while the others do”, as opposed to, “at this moment, the rdg omits the material in the lem”?  

I work without a base text in most of my projects, because there isn’t now a single agreed-on standard witness and wasn’t at the time of writing. But even so, I see witnesses actively striking out passages (active omission) in copies of previous rdgs. And I also simply see absence of the other witnesses at a point where one late witness has diverged because it has added, say, a whole paragraph of prose. That doesn’t mean the other earlier witnesses actively omitted this material, so I think that might be an example of the more neutral interpretation of an empty rdg. Am I right in that thinking, or are there other cases not to do with “material absent at this point” that might account for an empty rdg element?

Could we imagine an empty lem element? I think we could, depending on why the witness was chosen to be the lem—if it is, for example, a standard canonical text that is a revision of an earlier draft, or a more complete version where others are fragmentary. I don’t work much with lem, though—and I wonder if people tend to use it for the earliest / original witness. I suspect here we may find variations in thinking about what constitutes a lem?


Elisa Beshero-Bondar, PhD
Director, Center for the Digital Text
Associate Professor of English
University of Pittsburgh at Greensburg
150 Finoli Drive, Greensburg, PA 15601 USA
E-mail: [log in to unmask] | Development site:

Typeset by hand on my iPad

On Aug 19, 2018, at 7:15 AM, Marjorie Burghart <[log in to unmask]> wrote:

Hi Franz!

Good point! I woukd say that the exact meaning of an empty rdg is determined by the structure of its app container:
- if the app contains a lem, then it means the empty rdg is an omission;
- if there are only rdgs in the app, then it is not an omission but a neutral revording of the textual tradition.
Best, Marjorie

----- Mail d'origine -----
De: Franz Fischer <[log in to unmask]>
À: Marjorie Burghart <[log in to unmask]>, [log in to unmask]
Envoyé: Sun, 19 Aug 2018 11:32:44 +0200 (CEST)
Objet: Re: empty <rdg> elements vs omission of <rdg> elements

I generally agree, too.

I should just like to point out that om. (omisit/omiserunt) means a
reading has been omitted that previously (in some original version) used
to be there or is supposed to be there.

An empty reading element just says a reading is not present. It might
turn out that this reading is actually a later addition in other
witness(es). In order to convey the full semantics of "om." one might
add @cause, @type or @ana="omission".


Am 18.08.2018 um 19:29 schrieb Marjorie Burghart:
I agree, it ought to say "should", and needs fixing.
Best, Marjorie

----- Mail d'origine -----
De: Hugh Cayless <[log in to unmask]>
À: [log in to unmask]
Envoyé: Sat, 18 Aug 2018 19:16:38 +0200 (CEST)
Objet: Re: empty <rdg> elements vs omission of <rdg> elements

Dear David,

I would say it ought to say *should* (and I may just go ahead and fix it unless there are objections). That chapter is in dire need of a makeover precisely with a view to elaborating best practices that have emerged over the last few years. I would say that <rdg wit=“A”/> means the same thing as “lemma] om. A” in a print apparatus. As you rightly point out, there might be other reasons for leaving out a witness reading, including simply that the editor doesn’t think it’s worth recording. For cases where a witness doesn’t apply because it’s missing part of the work, I might fool around with <witStart/> and <witEnd/>, though I’m not sure how useful they are as currently defined.


Sent from my phone.

On Aug 18, 2018, at 07:03, Birnbaum, David J <[log in to unmask]> wrote:

Dear TEI-L,

The Guidelines, at, say that:

An omission in one witness may be encoded using an empty rdg,

The use of “may” leaves me uncertain about whether an empty <rdg> is consisted Best Practice, and I don’t find any discussion of whether an empty <rdg> elements is preferable to omitting a <rdg> element for a particular witness at a particular location, or about whether the two treatments are supposed to indicate different semantics.

In the absence of explicit semantics or Best-Practice guidelines, it’s tempting to use both methods and invent and impose a difference. For example, in some editions a witness may be used only in certain places, but not throughout the entire collation. Perhaps, for example, Witness B is not usually included because it is secondary to Witness A, but Witness A is missing pages, and Witness B is used as a surrogate for Witness A at those locations. In that case we might use an empty <rdg> in places where Witness A has been included in our collation within a specific <app> and it omits a reading there, while the absence of a <rdg> that points to Witness A would indicate that Witness A has not been part of the collation in  that <app>. That treatment would be specific to a particular edition, and could be documented in the header. I don’t see anything in the Guidelines that either recommends or discourages this approach.

The temptation to include an empty <rdg>, instead of omitting a <rdg> for a particular witness, looks as if it may have been motivated by tabular output. Specifically, in a table where rows are witnesses and columns are locations in the continuous tradition, every witness intersects every moment, and if the witness is silent in one place, the table nonetheless must have a cell. (Whether the cell should be empty or should say something like “omitted” is a separate question.) But TEI parallel segmentation isn’t a table; for example, it combines shared readings in a single <rdg> element, and the combinations may be different inside different <app> wrappers.

My reservation about local semantics and the absence of explicit Best-Practice recommendations arises because I’m looking at the moment at the automated export of TEI output from a collation tool. The collation tool knows which witnesses have readings at which locations, but it will not have access to information about possible different reasons for the absence of a reading suggested in the hypothetical local practice described above. When I transform the collation information into TEI parallel segmentation, then, either I can omit a <rdg> where a witness otherwise in the collation is not attested within a particular <app> or I can include an empty <rdg>. The Guidelines, at the location cited above, say that the empty <rdg> is permitted (“may”), but not that it is recommended, and I see nothing that prohibits, or even discourages, the alternative of omitting the <rdg> for that witness entirely. Users who have implemented one or the other behavior can transform either output into the other automatically, so from an engineering perspective the choice in this context may be arbitrary. But before I flip a coin, I would be grateful for guidance from the Community about whether one or the other representation should be considered Best Practice.