Re: Incorrect use of "it's" in package control files -- file mass bug?
Thomas Bushnell BSG <tb@becket.net> writes:
> "is/ea/id" isn't really "that". The normal word for "that" is
> "ille/illa/illud". "is/ea/id" is a special kind of weak demonstrative
> which has to be translated differently in different contexts.
I should be more precise.
<is> is really an anaphoric pronoun. Elementary grammars often list
it as "demonstrative" or "personal" (!). Their reasons are several,
but they are actually incorrect. It's neither demonstrative nor
personal; it's anaphoric.
It gets listed as personal because of the idea that every language
"must" have six personal pronouns, one for each gender/number
combination. But like it or not, Latin simply doesn't have third
person personal pronouns. It once did, in the prehistory of the
language, on the stem "s-", which is cognate to ancient Greek's
vestigial third-person pronoun, and which survives in Latin's <suus>
adjective and the reflexive <se>. But the rest of the <s-> pronoun's
forms were lost.
Still, elementary grammars figure they have to give you *some*
equivalent for English's "he/she/it". So they give you the closest
they can find.
As for why they call it demonstrative, that's because first, English
doesn't *have* anaphoric pronouns, and second, because often the best
translation is "this" or "that". It's actually quite indistinct
between "this" and "that", so some grammars say it's a "weak
demonstrative".
Of course, that's confusing, because the purpose of a demonstrative
adjective (or pronoun) is to lend locative emphasis. So how do you
add locative emphasis, but weakly, in a way which is indistinct about
location? Well, you don't.
An anaphoric adjective has a function very similar to a definite
article. It serves to mark a noun as "the one you already know
about", or "the one the listener is familiar with". What makes it
different from a definite article is that it is optional. In
languages with a definite article, to omit one is to imply that it
would be wrong to put it there. But in Latin, you can omit the
anaphoric adjective without implying that it would be inappropriate,
so it's not a definite article, it's an adjective. (But, tellingly,
it is cognate to the ancient Greek definite article IIRC.)
(There are also some special uses of English "the", as when we ask "is
he related to THE George Washington?" In this case "the" is not
really anaphoric [though it's clear why the anaphoric article is
pressed into use], and Latin for these constructions uses <ille>, not
<is>. Definite articles get used for more things than just their core
anaphoric use.)
So <is> is not a demonstrative at all, and it's wildly incorrect to
pretend that it is the Latin word for "that". The Latin word for
"that" is <ille>. The word <is> is a category of pronoun (or
pronomial adjective) which English simply lacks--though we make up for
it by using the definite article, which Latin lacks.
Translating <is> thus requires sensitivity and awareness of the total
context. Sometimes it is best translated by an English personal
pronoun, or a demonstrative, or something of the sort. But it would
be very wrong to think that it simply is the word for "that", or even
one of them. It's not.
Now, that said, what about <hoc est>? What's going on there? Why not
<id est> (which, after all, would be suitably anaphoric when it means
"that is", wouldn't it)?
Well, here a different factor comes into play. Latin prefers to use
demonstratives (real demonstratives) to refer to fragments of what has
just been said or referred to. For example, in English you might
introduce two noun phrases, and then refer to one as "the former" and
the other as "the latter". But in Latin you use demonstrative
pronouns: <hic> meaning "the latter", and <ille> meaning "the former".
The locative sense here is that "the latter" is closer to the
occurence of <hic>, and so it's "here", while "the former" is farther
away (because earlier), and thus "there": <ille>. Indeed, these are
very common, because (being pronouns) they are much less clunky than
"the former" and "the latter".
If you translate these uses of <hic> and <ille> rigidly as "the
former" and "the latter", you get very awkward English. Such a
translation would be incorrect. A good translation has to think about
the total context, and try to get the idea across in natural English.
Often just using "he" or "it" will be sufficient; sometimes restating
a name is necessary to disambiguate, sometimes there is no good
solution. (In my work with Peter Abelard, he often uses *three*
different pronouns this way: <hic>, <ille>, and <ipse> to keep three
distinct references apart. It's maddening to try and translate.
Classical Latin doesn't use <ipse> this way, but Medieval does.)
So in Latin, one uses <hoc est> in a restatement or paraphrase,
because one is explaining what just preceded, what is "here". To use
<illud est> would be very strange, and as I said, it is quite rare.
Generally there would need to be some distance on the page, and a
contrast with <hoc est>. For example, one might introduce two
concepts, and then go on to paraphrase each, explaining the first with
<illud est>, and the second with <hoc est>. And <id est> is also
uncommon, because the natural phrasing would be <hoc est>. To use <id
est> would simply mean something different. Normally <id est> is
going to say something about an *object* ("the object you already know
about"), and not the preceding concept or phrase in the text.
But then why does English have the symbol "i.e." instead of "h.e."?
Well, here you have to look to Neo-Latin: the Latin of the humanists
and other post-medieval scholars, sixteenth century and later. They,
for whatever reason, began using <id est> in this way. English got
its abbreviation from them. But their usage was distinctly new. I'm
not sure why <hoc est> fell out of favor that late. It may be because
most Latin was no longer written in Latin, but was translated from
European vernaculars, but that's just a guess.
Now do you see why your elementary Latin grammars and dictionaries are
just not up to the task?!
Thomas
Reply to: