[tei-council] Re: damage - what is it good for?

[Lou Burnard]

Thanks to everyone for their input.  I've now checked in a series of 
changes which not only preserve <damage> but add <damageSpan> and a new 
attribute @group designed to do the sort of grouping of bits of damage 
Gabby describes here. The text already suggested using <join> for this 
purpose, but no-one has ever understood that element, probably because 
of the extreme obscurity of its discussion ...  When I feel a lot 
stronger I might also add an example showing how you might use it for 
this purpose  to SA.

I'm holding back on further changes to PH . As far as I know the 
unresolved issues are only
(a) what <surface> means -- waiting for input from Conal
(b) what to do about hand/handDesc etc. -- I need to digest Arianna's 
comments
(c) some more examples of use of <fw> are needed

Comments on other bits gratefully received!

Lou


Gabriel Bodard wrote:
> We've never used <damage>, I don't think. (Although there was at some 
> time talk of using it to link a <supplied> and <gap> that are really 
> part of the same bit of damage. I didn't really see the point, since 
> the whole problem with linking such things was that they might cross 
> div or word structures. And now it appears that would have been abuse 
> anyway.) As you describe it, however, it is the sort of thing that an 
> epigrapher *would* want to record--deliberate or other damage to the 
> surface of the stone not affecting our ability to read the text. (For 
> deliberate I guess we use <del>.)
>
> What *would* certainly be useful, however, is some way to indicate 
> precisely what you describe: separate bits of damage (whether 
> indicated by <gap>, <supplied>, <damage>, or <unclear>) that are in 
> fact the result of the same gouge out of the front of the stone, for 
> example. Traditionally both epigraphers and papyrologists indicate 
> that a gap and a supplied are part of the same damage by enclosing the 
> two in a single set of square brackets. E.g.:
>
> [abc ...]
>
> (we can restore "abc" but not the rest of whatever was in that bit of 
> damage... as opposed to:
>
> [abc] [...]
>
> = "abc" was lost because of (e.g.) surface wear, something else was 
> lost because of (e.g.) the stone being cut down for re-use, and can't 
> be restored)
>
> So in summary I've never used <damage>, and the only time I'd be 
> likely to want to could arguably be handled by <unclear>--but unclear 
> for us has the very specific meaning that the letters so marked would 
> not be unambiguous outside of their context. Therefore I'd vote for 
> keeping it, but not scream and shout if I'm out-voted. :)
>
> Best,
>
> Gabby
>
> Lou Burnard a écrit :
>> As I proceed through PH wreaking havoc, I have come upon the <damage> 
>> element. This is allegedly used to mark a part of a manuscript within 
>> which there has been some damage to the carrier, e.g. by rubbing or 
>> singeing or spilling marmalade, but not so much as to make the 
>> transcriber unsure of what the writing actually says (if that were 
>> the case, the <unclear> element should be used), nor so extensive as 
>> to make the writing (or the carrier) actually disintegrate or 
>> disappear (for which the <gap> element is available).
>>
>> As defined, <damage> respects textual structures even less than the 
>> other elements. If it is to be kept, it should probably be given a 
>> sister <damageSpan> (analogous to <delSpan>) so that it can point 
>> across div boundaries for example.  Though even then there isn't any 
>> really satisfactory way of dealing with things like circular spots of 
>> damage in the middle of the page, which have to be split up into 
>> numerous <damage> elements.
>>
>> But since it is really about the state of the carrier, not the text, 
>> why would you want to record it anyway? I am sorely tempted to just 
>> remove it and see who protests....
>
>