Religious affiliation keywords for factoids

nathangibson commented 3 years ago

@dlschwartz @lunadine @Torabik Now the question of what keywords we use for religious affiliation factoids (see #892). There are currently some possible inconsistencies with the keywords or else I don't understand the system, since religions (what I would call "macro-affiliations") are in the form of "isms" whereas sub-groups ("micro-affiliations") refer to the people ("-ans", "-ites", etc.).

I would think we'd want to be able to classify affiliation broadly or narrowly, as indicated by the source. So is the mismatch between, say "Christianity" as a religion vs. "Melkites" as people a problem? There are also some issues (inconsistencies?) with their relationships to parent keywords such as "Institutions".

My concern here is not to deal with all potential issues of these keywords, but to nail down a consistent system for what we refer to in religious affiliation factoids, and also to figure out how we should formulate keywords we need to add. I understand the taxonomy is a work in progress and is not yet a full-fledged ontology!

Religion keywords, as children of https://syriaca.org/keyword/religion

Some but not all of these also belong to https://syriaca.org/keyword/institutions. Also, I don't see any Samaritanism.

Subgroups (typically children of https://syriaca.org/keyword/institutions):

Not all of these have been assigned to a parent religion.

Varia

There are several keywords under Institutions that are unclear to me, whether they are meant to refer to people or something more abstract:
There are a number of heresiographical descriptions that are mostly irrelevant to me and are usually pejorative, but technically fall under the same category, e.g. https://syriaca.org/keyword/bardesanites. I haven't collected all of these.
https://syriaca.org/keyword/rabbinic-judaism doesn't match the pattern for subgroups -- it's not "Rabbinic Jews" nor is it labeled as an "Institution".
https://syriaca.org/keyword/sabians are sub-categorized under "Islam" (I think they should be separate.)
https://syriaca.org/keyword/melkite (singular) seems to duplicate https://syriaca.org/keyword/melkites (plural)

One possible approach (perhaps overkill) would be to use adjectival forms for everything, keeping the "isms" and plurals as glosses, e.g. "Christian", "Rabbinic Jewish", "Melkite", "Coptic", etc. But perhaps the grammatical form of the word is not so important?

lunadine commented 3 years ago

@nathangibson I very much like the idea of macro and micro affiliations. If you have not already found it online, you could take a look at this list of tags for inspiration (see: Mass-Tagging). This is a Wiki-talk but nonetheless dealing with keywords for religious affiliations. You will for example find tags like "Abrahamitic Religions" (which could be useful depending on your research questions) or for all kinds of conversions e.g., "Conversion to Islam" or "anti-" tags (I saw you already have some of these in syriaca.org). I guess it could also be interesting to have a keyword like "Refusal to convert". However, I am not sure how often this occurs in Usaybia's History and I have a tendency to get too much into detail with factoids...

nathangibson commented 3 years ago

@lunadine Wow, that is an interesting list. I think a lot of it gets into issues that are beyond the scope of our project, in that I'm not trying to track individual attitudes or decisions, but rather see associations or affiliations with certain groups. Or is that being too hyper-focused?

However the "conversion from" and "conversion to" has me intrigued. Including every possible combination of religions like the Wikipedia category "Converts from Judaism to Islam" would seem endless. But if each keyword for a religious group had a "conversion to" and "conversion from" sub-category ("Conversion from Islam", "Conversion to Islam"), that could solve some problems. Like for Abu al-Barakat, a single factoid could be labeled both "https://syriaca.org/keyword/conversion-from-judaism" and "https://syriaca.org/keyword/conversion-to-islam". Then there would be no need for a second factoid about his previous Jewish affiliation unless that is mentioned in a separate context.

@dlschwartz What do you think?

dlschwartz commented 3 years ago

@lunadine @nathangibson I think this is a really interesting idea. XML can do this very easily. The question is how straightforward this is using the TEI. There doesn't seem to be an easy solution that doesn't include the creation of attributes or something like that. The suggestion you've made above Nathan works pretty well. The main drawback is that it requires a large number of new keywords. Every religion in an already large list would require two additional keywords. Let's keep thinking about this.

dlschwartz commented 3 years ago

@nathangibson @lunadine I've started a Google sheet for religious affiliations that begins to reconcile some things. There's quite a way to go on this though.

nathangibson commented 3 years ago

@dlschwartz Thanks for this! The more I think about it, the more "conversion-from-[religion]" and "conversion-to-[religion]" keywords seems like a clean solution. But I'm not the one who would have to create and maintain the keywords :-)

I will have to implement something soon, due to our project coming to a close, but it is ok if it is not a final solution. (It's understandable if you don't have the bandwidth to resolve it right away.) I'm thinking that if we code with conversion-to and conversion-from (even if those keywords don't go online yet), it wouldn't be that hard to generate additional factoids from those for "previous affiliation state" if we all decide that conversion-to/from is not the right way to go.

Regarding the spreadsheet, I can see why this is complex! It appears to me that most of the sub-groups I listed above are not yet in the spreadsheet -- not sure whether you want us editing stuff there directly. Also as a religion I don't see "hinduism" there yet.

In Taxonomy but not anywhere in spreadsheet (or Gazetteer)

anglicans hinduism ikhwan-al-safa-brethern-of-purity russian-orthodox unia

In Taxonomy with straightforward Gazetteer match but not listed in spreadsheet CBSC column

copts greek-orthodox maronites melkites nestorians protestants roman-catholics syro-catholics (=syrian-catholic?)

In Taxonomy with problematic Gazetteer match but not listed in spreadsheet CBSC column

assyrians (ethnicity or religion?) chaldeans (which?) syro-malankar (which?)

dlschwartz commented 3 years ago

@nathangibson This is great, thanks! I can put these in but feel free to add anything to the spreadsheet. It can serve as a clearing house for this discussion until we get further along.

By the way, I cringe when I see the word Protestant in this list. There must be over a thousand sub-groups. Yikes! Perhaps we would limit it to Protestants who had missionary endeavors in the regions where Syriac Christianity has historically been present. I'm not sure I could make even that list without a lot of work. In fact, I'm not sure off the top of my head that I could get beyond Anglican and Presbyterian.

Thanks Nathan.

dlschwartz commented 3 years ago

I've noted this in the spreadsheet but I'll mention it here too. The lists have both adjective and plural noun forms of these keywords, i.e. "Jacobite" and "Jacobites". In other instances there is just one form "Anglicans" but no "Anglicanism" or "Mandaeism" but no "Mandeans". I think that the taxonomy should standardize this into one grammatical form applied throughout.

srophe / syriaca-data