[tei-council] 13 Names, Dates, People, an d Places

[James Cummings]

Lou's Laptop wrote:
> Brett Zamir wrote:
>> 1) For the line, "Characteristics</soCalled> or 
>> <soCalled>traits</soCalled> are typically independent of an 
>> individual's volition or action and can be either physical, such as 
>> sex or hair and eye colour, or cultural, such as ethnicity, caste, or 
>> faith." I really disagree with including "faith" as something which is 
>> independent of an individual's volition or action.
> But many people are born into a faith, are they not? they cannot be said 
> to choose that any more than they choose their parents? Of course they 
> may go on to choose a different faith, but even when they do, they will 
> often say they are an "ex-muslim" or "ex-jew" or whatever.

It all depends how one defines faith.  That the element's desc says "faith, 
religion, or belief set of a person" highlights this, I think.  You can say a 
person is born into a religion because the family is that religion, as an 
extension to that you can say they are born into a faith, but then the silliness 
of this is exposed if you say someone is born into a particular belief set.  It 
seems completely obvious that no child on birth, and at least for a few years 
after, has any particularly well-defined belief set that corresponds to a 
human-based organised religion.  So while Lou is right that they are 'born into' 
this faith without their volition, it depends whether you view the faith element 
as indicating societal-perceived belief sets or actual belief sets. I won't go 
onto a tangent about how bizarre it is to claim that children have made up their 
minds about their belief sets long before we consider them rationally capable of 
making up their minds on much less serious issues.  That is not the point.  I 
think the viewing of this as a state instead of a trait is something which is an 
editorial decision for an individual project.  I feel the Guidelines make it 
more than clear that none of these 'traits' are necessarily immutable.  It could 
point out that faith is one of those that is particularly disputable to 
highlight the entire fluidity of traits vs states, perhaps.

The interesting thing about these 'traits' though, including the sex element, 
are all members of att.datable which means they not only get a @from and @to 
attribute but a @when attribute and the rest.  I think this is sufficiently 
loose in leaving it to any individual project to decide how they feel about 
this.  I'm only thinking about this because a project I'm involved with is 
recording person elements for some people who quite demonstrably and obviously 
declared changes in faith.

>> 1) Again on some of the "changeable" characteristics, I find it 
>> inconsistent that <education> is a state, but <langKnown> is a 
>> characteristic.
> Yes, I see that this does look a bit odd, and maybe langKnowledge is a 
> boundary case. But education is definitely a state, surely?

I think it is slightly more complicated with both of these.  When 'education' 
(or langKnown was used in corpora it tended to be a single historical fact, 
expressing the highest level of education (or languages known) that person had 
achieved at the point of recording.  So for the context of that document it 
could be seen as a trait of the person, and I think many linguists use the 
concept of educational level in a similar way to faith, socecStatus, etc. 'Show 
me all the answers by secondary school students', etc.   But are treated as 
fixed traits of the person, at the time of the creation of a spoken text.  The 
problem comes when we take that concept of a person and use the same method to 
talk about an abstraction of that person over time rather than at a fixed point 
in time.  In the abstracted general person description, their education is a 
state that changes over time, in the specific instance of any individual 
document, the person's education is a trait of the person at the point in time 
the document was created.  Obviously the date of the document should be compared 
against any dates in any of these important states whenever making any such 
deductions and interpretations.  (Though I worry some people might fall into 
this as a methodological trap! I wouldn't want a corpus of all the works I'd 
ever written to treat the essays I wrote in highschool equally to the articles 
I've written more recently!)  I'm not suggesting the Guidelines change here 
though, just that we should be aware that what we view as a state of a person, 
others will be viewing as a trait of a person with respect to a particular 
document in time.

-James

-- 
Dr James Cummings, Oxford Text Archive, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk