Re: All unrelated "groups" are now called "families"!

On Mar 21, 2:43 am, darkstar...@xxxxxxxx wrote:
benli...@xxxxxxxxxx wrote:
On Mar 20, 9:51 pm, darkstar...@xxxxxxxx wrote:

benli...@xxxxxxxxxx wrote:
On Mar 20, 11:49 am, darkstar...@xxxxxxxx wrote:
benli...@xxxxxxxxxx wrote:
On Mar 20, 12:46 am, darkstar...@xxxxxxxx wrote:
I just can't remain silent. What's the big idea? All the unrelated
"groups" are now being called "families"! I look up "Turkic" in wiki,
and suprisingly I find out it's a "family". Once, I've even seen
"Germanic" being called a family (obviously mistaken but marks the
trend)!!

Do these people have even the slightest idea about taxonomical logic?

You don't believe in Altaistics or glottochronology -- that's okay,
but that doesn't mean that closely related taxona like Turkic or some
of the sub-Papuan groups are "families". Family is a big and deep
taxon, for which we only know distant internal relations, whereas a
group is a closely related taxon, whose internal relations are obvious
even to a casual observer, because they still maintain a sufficient
degree of mutual intelligibility, similarity, cognation, etc. How can
such a simple logical point cannot be understood?

If you take an appearant isolate, like Turkic, and you discard all the
accumulated knowledge about Altaic relationships, then please still
call it a "group", because a "family" means there are few known
relations INSIDE the taxon, NOT OUTSIDE the taxon.

Who, at what historical period, is supposed to have used this
"correct" taxonomy which you are advocating?

Ross Clark

At least, Russians have. I've never seen anything like that in their
work. I also wonder about Germans and other countries.

So you're proposing to reform English terminology on the basis of
Russian?

Ross Clark

[top-posting corrected]

To reform something you need a basis for a reform, but if this
confusion is seen as normal, there's really nothing to rant about,
there's just no classification in linguistics (except a Germanic
family inside a Germanic family).
I do not object to debating as long as people understand logic...
Logic is the basis for discussion...

I have no problem with logic. Your use of the word "now" in the second
line suggested that you were noting a recent deterioration in
standards of terminological usage, and that's why I referred to
"reform".

You seem to be advocating a two-term system in which obvious and
closely related genetic/taxonomic units would be called "groups" and
the less obvious, more distantly related, would be called "families".
The only obvious logical rule here would be  that you couldn't have a
family within a group. But you could have groups within groups, and
families within families, I think? E.g. East Polynesian (G) within
Polynesian (G) within Oceanic (F) within Austronesian (F)?

There are prefixes like sub- and super-(or macro-) well-established in
many taxonomical fields. And even subsubsub-, etc. I could never
understand why comparative linguistics should be so terminologically
primitive and use such vague designations as just "languages".

But some linguists have used just such prefixes, e.g. Wurm and his
colleagues in the Pacific, and Greenberg.

If you don't know how to settle it, at least you can use English as a
standard. Br. vs. Am. can be seen as a typical dialectical difference,
"Germanic" as a typical group, "Indo-European" as a typical family,
"Nostratic" as a typical macrofamily (superfamily). It's all been
settled in biology for centuries, why linguistics is supposed to be so
different in the worst sense of it?

Apparently linguists have not felt it as such an urgent need. Anyhow,
I thought that phylum, class, order etc. were defined by their rank in
the hierarchy, not by any absolute degree of similarity. How is this
determined?


Now in terms of actual English usage among linguists, I have two
questions:

(1 ) Do the people whose terminology you were deploring above actually
commit the above logical error? Or do they just use "family" in a
looser sense than you would like? For me and I think many linguists,
"group" is simply not a technical term, and means no more than "set",
i.e. doesn't have the strict genetic implications that "family" does.

In that case, that's just an Anglophone problem, as I've pointed out
above. The same noun for different objects.

Well, objects which have an essential common feature, though they
differ in other ways. Since when is this a problem?


Indeed, some Russians have been wondering, like, "What? Now they call
Papuan languages 'families'? Does that mean these are like IE?" The
probable answer is that they are not. Even though I know nothing of
Papuan, the likely idea is that they mostly established only rather
closely related groups there, because finding true families is much
more difficult, but called them "families" instead. I'm not sure.

I don't know where your "probable answers" and "likely ideas" are
coming from, but you could try to find out instead of just theorizing.
In fact Papuan includes both "groups" and "families" in your
terminology. "Papuan" itself is not a genetic unit of any kind, just a
label for non-Austronesian languages in the New Guinea area.

Hence, this terminological conflation (?). Which says nothing whether
Papuan languages constitute a family or macrofamily or even a higher-
order cluster. Most likely, the latter, because the time depth is
immense (~40 kys? from genetical studies).

Of course, the matter in question has a distant allusion to
lexicostatistcs, since it can be rather difficult to establish a taxon
depth without precise statistical measurement. But in the obvious
cases, like "very similar like Germanic, Turkic", "not very similar",
the use of generic terms is simply objectionable.

OK, I understand a bit better what bothers you. You may have to get
used to the fact that it doesn't bothe other people so much.

Ross Clark

(2) Can you point to any English-writing linguist (present or past)
who has consistently used the terminology you advocate?

No, I'm judging from Wikipedia. That must reflect a students' trend,
since they edit that stuff most often.

.

Relevant Pages