On 16 October 2012 22:26, George Corley <[log in to unmask]> wrote:

> People do make actual errors,

Well, is that so? When people pronounce the "t" in "often", is it an error,
or is it just an indication of language change? After all, we know that
spelling pronunciations have driven language change in the past, and those
spelling pronunciations are now considered standard.
The problem of calling such things "errors" is that this presuppose a
standard that is "correct", and any deviation from it is "erroneous". But
we know that language is ever-changing, and faster than most people
realise, which means that whatever "standard" you want to hold on to, it is
probably already out of date by the time it's promulgated.

Now, people do make slips of the tongue, or think one word and say another,
etc. If those are felt as wrong, they will be corrected instantly, usually
by the speaker themself. If not, then who's to say that those are really
mistakes, rather than just the starting point of a language change?

> and a linguist can determine things about
> writing style that can potentially lead to better understanding of the
> text.

I disagree. What you are talking about is text interpretation, which is a
very different exercise from linguistic analysis. Payne has a good
discussion at the end of "Describing Morphosyntax" on why those two are
very different and why the first one isn't and cannot be the job of the
linguist. The two disciplines do feed each other, however using linguistic
analysis to help understanding an already written text is very different
from pretending that such linguistic analysis can help define the writing
style of future texts. I don't believe the linguist has much to say there,
except maybe trying to speak against the worst cases of prescriptivism.

>  Your analogy is broken: astronomers *cannot* affect the orbits of
> the planets short of using a massive rocket or colliding an asteroid (of
> course, if you want to move a planet, you would probably want to consult
> not an astronomer, but an astrophysicist).  However, social science can and
> does have a direct impact on human behavior.  If economists and
> sociologists regularly give advice on how to construct laws and structure
> societies, and psychologists regularly devise treatments for mental
> disorders (and even make value judgements as to what is a disorder --
> eep!), then why can't linguists give advice on writing, based on empirical
> study of what writing conventions make the reader understand the text
> better, or any other quantifiable objective?  Linguists can and do, after
> all, give advice as to how to teach languages, how to treat language
> disorders, etc.

Things that are light-years away from the issue of writing style. I don't
believe my analogy is flawed (yes, except for the fact that I should have
been more precise and should have talked about astrophysicists). As for
those economists and sociologists you mention, I believe most of them don't
pass muster as empirical scientists, putting philosophy and personal
opinions before fact analysis. As for psychologists, that's the first time
I've ever heard that they devise treatments for mental disorders! I've
always known that it was *psychiatrists* that were doing that!
(psychologists are to psychiatrists what biologists are to doctors: they
describe and try to understand. But it's the second one of the pair who
uses the knowledge gathered by the first to eradicate diseases, or even to
decide what is a disease or not!)

Yes, the knowledge linguists gather can be used by education specialists to
devise better language teaching methods, or by speech therapists to treat
language disorders. But it's not up to the linguist to decide on those
issues! They can give data for others to work with, and *that's all they
can, and should, do*. In the same way, even if linguistic data could be
useful for the definition of a writing style (which I'm not sure about), it
would still not be the job of the linguist to define that writing style!
And it would certainly not be the job of the linguist to have opinions on
the source of their data! As any empiricist, linguists are supposed to be
unbiased towards their data. It's not an assumption, or a hypothesis that
needs to be justified. It's a very tenet of empirical science. As soon as
linguists forget that, they cease to be scientists.
Christophe Grandsire-Koevoets.