Dear all,
I would like to model the following scenario:
1. The French National Archives have the registration registers of the
conservatoire from 1822 to 1914.
2. On the physical document classified as "AJ/37/396", we can read that someone
called "Gisèle Grandpierre" entered the conservatoire on october 17, 1908, and
studied the harp.
3. It is written that Gisèle Grandpierre is a male student, but my musicologist
colleagues know that only women studied the harp at that time.
4. We need to model three things:
- The existence of a student studying the harp is deduced from reading
document "AJ/37/396".
- A diplomatic transcription of the gender field: "Male".
- An interpretative transcription of this gender field: "Female" which
leads to a gender attribute assignment.
With core CIDOC-CRM entites, I think I can model some aspects of this scenario:
- "AJ/37/396" is a E18 Physical Thing ---[P128 carries]---> E90 Symbolic Object
(the readable text of the whole registration record, URI: <AJ/37/396_E90>).
- <AJ/37/396_E90> ---[P106 is composed of]---> the field "gender" (URI:
<AJ/37/396_gender_field>) ---[P190 has symbolic content]---> the string "Male
». This is a diplomatic transcription, even if it is not made explicit.
- <AJ/37/396_E90> ---[P67 refers to]---> E21 Person (URI:
<E21_Gisèle_Grandpierre>)
- The E55 Type "female gender" is associated to <E21_Gisèle_Grandpierre>
through a E13. It is an "interpretative act" that differs from what is written
in the archives. E13 seems adequate. This E13 acts like an interpretative
transcription.
But I have two problems:
1) I think this "gender attribute assignment" should be, for sourcing and
motivation purpose, associated:
- to <AJ/37/396_gender_field>, to highlight the "error" in the sources
- to <AJ/37/396_E90> (which express data on the musical course), to
highlight the fact that <E21_Gisèle_Grandpierre> studied the harp.
- to higher-level knowledge in the name of which musicologists can infer
that if someone studied the harp in 1908, then it was a woman…
I don't know how to do it.
2) I don't like the "P67 refers to" predicate between <AJ/37/396_E90> and
<E21_Gisèle_Grandpierre>, because I think that infering the existence of a
CIDOC-CRM entity (here, a E21) is the result of an interpretative act, and
should then be signed by the researcher, motivated and sourced. Do I need
something like E13 but for "Entity Creation" rather than “Attribute Assignment”
resulting from a reading? Or did I miss something in the core CIDOC-CRM
documentation?
I studied CRMtex, and think I can model the diplomatic and interpretative
transcriptions with it, but I still do not know how to connect the
interpretative transcription and the fact that a researcher decides to
postulate the existence of a female student (E21) studying the harp.
Again, associating attributes (here, gender) through interpretative acts to
this E21 should not be a problem thanks to E13.
And I have the feeling that CRMinf is a bit… too much for this scenario ;)
Many thanks for your insight on this issue.
Thomas Bottini
———————————
Thomas Bottini
Institut de Recherche en Musicologie — IReMus UMR 8223, CNRS
———————————
Thomas Bottini
Institut de Recherche en Musicologie — IReMus UMR 8223, CNRS
_______________________________________________
Crm-sig mailing list
[email protected]
http://lists.ics.forth.gr/mailman/listinfo/crm-sig
