Talk:West Trans–New Guinea languages

Misleading passages

This article currently states,

"In 1957 HKL Cowan linked the Timor languages to the West Papuan family. However, when Stephen Wurm expanded Trans–New Guinea in 1975, he decided Timor-Alor-Pantar belonged there, and he linked it to the South Bird's Head languages in a South Bird's Head–Timor-Alor-Pantar branch of that phylum. Wurm noted similarities with West Papuan, a different family, but suggested this was due to substratum influence."

There are a few serious problems with this. 1) Cowan's West Papua group included South Bird's Head and West Bomberai, as well as the languages currently thought to form the West Papua phylum. Cowan divided these into two subgroups, based upon pronouns, with North Halmahera and West/North/Central Bird's Head in one group, and Timor, South Bird's Head and West Bomberai in the other - exactly the current arrangement. Bear in mind that Cowan did not have a classification of New Guinea languages generally, and this West Papua group was only presented as one relative to his North Papua group ("Een tweede grote Papoea-taalgroepering in Nederlands-Nieuw-Guinea".) This subgrouping is probably wrong, but it's fair to say that the field has yet to develop a coherent opinion on classification at a time depth which encompasses both the current West Papuan Phylum and Trans New Guinea as well these northern families. 2) It was not Wurm who placed the second subgroup with Trans New Guinea, but Voorhoeve, based upon exactly the same pronouns that Cowan had highlighted earlier. Cowan had no knowledge of the more easterly families which formed the basis of McElhanon and Voorhoeve's Trans-New Guinea proposal, and thus had no way of arriving at this insight.24.22.142.28 (talk) 06:16, 11 April 2010 (UTC)

Definitely needs to be corrected. Since you have the sources at hand, why don't you do it? kwami (talk) 06:00, 12 April 2010 (UTC)

I'm banned!

Anyhow, I'd like to see you familiarize yourself with the literature (and I finally got Healey 1964, btw.) You understanding all this would in the long run probably mean more than Wikipedia being corrected.

I'll also observe that Greenberg (1971) placed the Abui language of Alor with the Papuan languages of Timor. Unsure if that's the first such proposal, but it predates the Watuseke paper.24.22.142.28 (talk) 11:34, 12 April 2010 (UTC)

I don't have access to a library, so unless you can email me copies, that's not a very likely direction. Why don't you just rewrite the paragraphs here, and I'll paste them in? (I've got harddrive corruption probs and am trying to prepare for a conference, so this can't be a priority for me right now.) kwami (talk) 16:18, 12 April 2010 (UTC)

"The West Trans–New Guinea languages are a suggested linguistic linkage of Papuan languages, not well established as a group…"

Hmm, what do we mean by this? We're not really sure, and neither is Ross: it's a hazy term for an incoherent concept.

"Ross suspects they are an old dialect continuum, because they share numerous features which have not been traced to a single ancestor using comparative historical linguistics."

This is basically nonsense.

In any event, Ross has no business appealing to "comparative historical linguistics," as he hasn't bothered even attempting to reconstruct any of these families.24.22.142.28 (talk) 09:47, 9 May 2010 (UTC)

No, that means that he has tried a reconstruction, and that neither this nor TAP came out as a coherent group. Or at least he tried what little one can do with just pronouns. I agree that there doesn't seem to be any way to determine whether a group of languages is a "linkage" or not, though, other than having apparent cognates with no reconstuctable ancestor. — kwami (talk) 15:40, 9 May 2010 (UTC)

Redirect from Timor-Alor-Pantar languages

Why is Timor-Alor-Pantar redirected here?

http://en.wikipedia.org/w/index.php?title=Timor-Alor-Pantar_languages&action=history

See WP:NPOV: Ross' opinions aren't supposed to be presented as fact, to the point that we're actually structuring the whole 'pedia around them. Sometimes he's right, but here he is spectacularly wrong. You can call that my original research, but what isn't is that Cowan, Greenberg, Voorhoeve, Wurm and others concur that there is a valid Timor(-Alor-Pantar) family; it is only Ross who disagrees.24.22.142.28 (talk) 10:16, 27 April 2010 (UTC)

Greenberg's opinion isn't worth much when it comes to cladistics; the others (and Greenberg too for that matter) are 35+ years older than Ross. "Only Ross who disagrees" sounds like there was a contemporary debate. Foley would be contemporary, but I haven't seen anything from him apart from what's already in there article. There's also never been a TAP article; I created many of these based on Ross as the most recent classification where there had been nothing before. Of course, you're welcome to create a TAP article if you like. — kwami (talk) 15:24, 27 April 2010 (UTC)

"Greenberg's opinion isn't worth much when it comes to cladistics."

In my experience, it's worth about as much as Voorhoeve's or Wurm's. Here, all these previous researchers happen to be correct. Ross' study is based almost wholly upon pronouns, and, though more recent, cannot be said to be more thorough, or even better informed, as the pronouns were already accurately known decades ago. This field moves very slowly, and it's not at all unreasonable to look at a consensus of the last forty years. We are not going to see active contemporary debates about too much of New Guinean classification, because hardly anyone is doing this.24.22.142.28 (talk) 07:53, 3 May 2010 (UTC)

Granted that the field moves slowly, but Greenberg was a typologist. What did he contribute to cladistics? He popularized some good existing work in Africa, but the only family he found himself was Nilo-Saharan, which, if it turns out to be valid, was perhaps merely luck, since it consisted of all the non-click languages not already classified. He tried the same kind of thing in the Americas, and bombed. In New Guinea, not even his protege Ruhlen follows him.

Voorhoeve worked on limited data, and Wurm said he didn't expect his classification to stand up well over time. Ross only looked at pronouns. I doubt any of them are very reliable, which is a pity. As I said, if you wish to write the articles you want, be my guest.

As for Ross being presented as fact, the article makes it pretty clear that it's tentative. Alor-Pantar languages even has question marks after each superior node in the taxonomy. — kwami (talk) 08:22, 3 May 2010 (UTC)

"…Greenberg was a typologist. What did he contribute to cladistics?"

While also a typologist, Greenberg's taxonomic proposals in all three regions, whatever their merit or lack thereof (and on this we might agree,) were based almost wholly upon "cladistic" claims, and explicitly, even stridently, rejected typology as a valid basis for classification. Of those we're discussing, only Wurm was guided by typology.

In fact, Greenberg's (1971: 810-815) treatment of Timor-Alor (that is, Buna', Makasae, Oirata and Abui) is on the whole quite reasonable, and significantly more extensive than either Cowan's or Voorhoeve's. Perhaps even half of the forms compared in his 92 proposed etymologies - I'd have to count them to be sure - are in fact cognate, and comply with the correspondences visible in the modern dataset, which (as you know) is hardly a given for Greenberg's work. On pages 811-812, you will find a treatment of the pronouns which is indistinguishable from Ross' in method, and no worse in execution.

As an addendum, I doubt that Ruhlen's use of the Wurm et al. classification signified any skepticism at all towards Greenberg's approach (if only!) so much as that the Wurm iteration included more languages under their then-current names - merely correlating the two is quite a task - and a catalog of all languages known at the time was one of the core purposes of the book.24.22.142.28 (talk) 05:25, 6 May 2010 (UTC)

[edit conflict] I know his classifications were cladistic, not typological. What I meant was that his theoretical contributions were in typology. He moved the field forward there. In cladistics, he merely borrowed from others for the most part, and his own ideas have not been confirmed, and often shown to be wrong. But in any case, removing him from your list does not substantially change your argument, so he's a bit beside the point.

Anyway, where would you prefer it redirect? TNG? I sent it here because this article lists the languages included in TAP. — kwami (talk) 05:42, 6 May 2010 (UTC)

I don't see that families which are widely, though not universally, accepted by those active in this small corner of the field should be redirects to groups supported by only one scholar, as if he had entirely superseded the sum of the existing literature. Nor do I have too much trouble with an article for Ross' proposed family, and though I personally find Ross' contribution in this particular region to have been obfuscatory and counterproductive (in some regions he rightly challenged widespread but baseless assumptions,) it would inform readers of what Ross is proposing, which is useful.24.22.142.28 (talk) 06:29, 6 May 2010 (UTC)

So again, what would you prefer? A redirect to TNG? Since you know what you're talking about here, it might be worthwhile for you to just write a TAP article. If brief, it would probably would take less time than our discussion here! — kwami (talk) 06:34, 6 May 2010 (UTC)

Okay, I'll create a stub at some point over the redirect. In this stub we can mention that Ross doesn't agree, and link to this page.

I wonder if this article wouldn't better be titled, "West Trans-New Guinea," or even "West Trans-New Guinea language family" - the "languages" title would be great for an ambitious paper, but is overly presumptuous on Wikipedia - it sounds like an objective method of cataloging them according to a widely-accepted scheme - and the nature of hypercardlike organization, from which wiki sprang, is such that there is no need for a single authoritative hierarchy according to which language articles are arranged.24.22.142.28 (talk) 06:56, 6 May 2010 (UTC)