Re: isolating is equivalent to inflected

On Sun, 2005-12-04 at 21:27 -0800, Gary Shannon wrote:
> I'm sure this is old hat to people with actual
> linguistic training, but it was like a major epiphany
> to me this evening.
>
> A few years back I was playing around with a way to
> create a conlang by a two-step process whereby the
> grammar was worked out entirely using English leixcon,
> and then after all the grammatical issues were
> resolved the whole thing would be re-lexed. I went for
> a pure isolating version of English as my starting
> point and created all kinds of particles and marker
> words. I just happened across my notes from 1999 and I
> noticed that my pure isolating English is, with only
> cosmetic changes, equivalent to pure inflected
> English. For example:
Not really isolating/inflexional, but rather isolating/agglutinative are
closely related. An agglutinative language like Finnish has a collection
of affixes which you just tack on, in a particular order, each with a
single meaning, so you've got a root meaning "book", a pluralising affix
and accusative affix, and you get book-es-him or in the genitive
"book-es-ov"; -es- denotes the plural, whereas -him or -ov denote the
accusative or genitive. That's somewhat similar to the mostly isolating
sentence "John will have been being stupid for ten minutes".

Over time, the agglutinations might become weakened and "es-him" becomes
"sen", whereas "es-ov" becomes "rev" (via "zov"), so you get "book-sen"
and "book-rev" (in contrast to the singular accusative with i-umlaut and
the singular genitive in "ev", thus /byk/ vs "book-ev"); in this stage
you can't really separate out the plural morpheme, nor the accusative,
nor the genitive; you have a separate accusative plural, a separate
accusative singular,

I understand there are various ways to distinguish between
agglutinations, inflexions and separate words (a category which includes
clitics such as English's apostrophe-s possession). Agglutinations and
inflexions I think is mostly on the amount of things a particular
morpheme expresses; -es- above denotes one thing, and in any noun in the
plural you can find the "plural morpheme", wheres in the inflected forms
"rev" and "sen", you can't point to the plural morpheme. I think the
distinction between words and (agglutinations and inflexions) is mostly
on what they attach to, or where they go in a sentence; an (aggl. or
infl.) goes on every relevant word, whereas a word is much more free. I
think, someone who knows grammar better can probably clarify. But
contrast:

        The thief made off with the Queen of England's jewels and the
        Mona Lisa.

which has a more isolating structure with the words "with" and "'s"
attaching to phrases ("England" only has one possession word directly
attached to it---"of England"; the "'s" attaches to the whole phrase
"the Queen of England"), with:

        The thief made_off her-England her-Queen with-jewels and
        with-Mona Lisa.

which has "with-" attached to precisely the things the thief made off
with, rather than the whole phrase.

One interesting point though ... You know how in English we typically
have to use the pronouns, whereas in Spanish the pronouns are optional
(that is, Spanish is "pro-drop" language). Ostensibly this is because in
Spanish you have the inflexions which specify the person and number, so
the pronouns are redundant. I'm told though that Chinese and other
languages that don't have any information about person/number are
usually prodrop, and it's usually the ones like English or French which
have a bit of info on the verb about them, but not much, which aren't
pro-drop. (For instance, English has one inflexion -s which denotes the
third person singular, whereas in Chinese you have nothing.)

(Of course, I notice that in colloquial English it's somewhat common to
not use the pronouns---I'm not sure if this is something completely
different, or if it's pro-drop tendency and part of the language's
conversion from the highly inflexional Common Germanic to a much more
isolating language.)

...

> OK. Tell me I'm all wet on this one, but (to me, at
> least) it's an interesting speculation that raises the
> question: is the difference between an isolating
> language and an inflected language little more than
> how it was first written down?
Not entirely, as an inflexional language can't do something like "with
this and that" == "with that and this", they have to do "with this and
with that" == "with that and with this" (the former egs would be more
like "cat and he" as in "I took his cat and he walked away"). But
obviously whether a language is isolating or agglutinative or
inflexional is not a simple categorisation that the language
automatically tells you about.

In a reply, Paul Bennett brings up a related cycle of language
evolution:

On Mon, 2005-12-05 at 01:40 -0500, Paul Bennett wrote:
> On Mon, 05 Dec 2005 00:27:38 -0500, Gary Shannon <fiziwig@...> wrote:
>
> > I'm sure this is old hat to people with actual
> > linguistic training, but it was like a major epiphany
> > to me this evening.
>
> The process you describe is called grammaticalization. Sounds like a made
> up word, and it very probably is, but it's something about which many
> books have been written, some of them good. I haven't read nearly enough
> on the subject, though it fascinates me.
>
> The cycle seems to most usually be:
>
> 1) isolating
> i) particles connect to words, thus
> 2) agglutinative
> ii) sound changes erode affixes, and analogy levels them, thus
> 3) flexional
> iii) sound changes and analogy effectively erode the inflections
> completely, and new particles are adopted from auxilliaries, thus
> 4) isolating
>
> I don't know whether the cycle runs backwards in natlangs. I can sort of
> see vaguely how it might, but I don't have a full understanding of the
> whys and wherefores.
Things can skip stages ... for instance the evolution of Latin to the
modern Romance languages has a new verb form coming from an
isolating-like postposed inflexion of "habere" becoming things like the
French inflexion -ai (I do believe), without at least the language going
via an agglutinative stage.

Also, the English clitic 's for possession was originally an inflexion
which went backwards, from being an inflexional feature of the language
to being an isolating feature. (It's a backwards step rather than
forwards, because if it were forwards new particles (like "of") would
replace the inflexion; instead, the inflexion has become a particle.)

Such is my understanding...

--
Tristan