Wikipedia:Articles for deletion/Language families (Ethnologue)

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was keep/merge; further merge discussion can continue at the talk page. Ƙɽɨɱρᶓȶ 06:48, 30 June 2007 (UTC)

[edit] Language families (Ethnologue)

Language families (Ethnologue) (edit|talk|history|links|watch|logs) – (View log)

Is it really a good idea to have an article whose sole purpose is to uncritically duplicate a single source? Ethnologue does not necessarily represent the current consensus. Also, the "superfamilies" in the article are not only controversial, they aren't even mentioned in the Ethnologue itself. Ptcamn 22:03, 23 June 2007 (UTC)

• Note: This debate has been included in the list of Language-related deletions. -- John Vandenberg 05:19, 24 June 2007 (UTC)
• I share the nom's concern that the article's assumptions about language are based on Ethnologue, which is one group's opinion (albeit an influential opinion). There are other perspectives from which to cover the concept of language families. This suggests the possibility of a merge or a better contextualization. YechielMan 10:24, 24 June 2007 (UTC)

Well, I'm no expert in this area at all. All I can say is that the article neither claims to be the theory of language families, nor a survey of theories. Note the title — Language families (Ethnologue).

Now, is the criticism of the article that it dupicates without critism, or that it fails to reproduce accurately? Please decide which, because they are inconsistant. Either would be good ground for editing the article — addition of criticism, or correction of errors — not deletion of an article.

Both. I don't see how they're inconsistent. Citing criticism against something which is inaccurately reported would be bad, as would accurately reporting something without comparing it to any other points of view. --Ptcamn 00:31, 25 June 2007 (UTC)

I'm not sure why you want to exclude presentation of the Ethnologue material. Ethnologue is not sufficiently notable or reliable for Wiki?

Articles are written by comparing multiple sources. It would be fine if Ethnologue was cited in a general article about language families, alongside other sources. But we shouldn't have articles dedicated to single sources. It could be construed as POV fork. --Ptcamn 00:31, 25 June 2007 (UTC)

Sorry, I'm not an expert, so I sense ruffled feathers, perhaps Ethnologue is not in favour in certain circles, and for all I know they are just plain wrong. However, it'd still be of historical interest. So write it up! I'd love to know! If it is not as simple as right/wrong, what we need is an expert who is not opposed to Ethnologue, but who is opposed to the article, for an impartial judgement. Otherwise it just looks like silencing alternative views. In the mean time, perhaps you could improve the article by adding criticism of Ethnologue (I just don't know where to look) and by correcting any errors.

I have no idea what you're talking about. If people want to know what's in the Ethnologue, they can look at it online. --Ptcamn 00:31, 25 June 2007 (UTC)

In any case, what has been duplicated uncritically, must be retained by the golden principle of Wiki — sited sources are not to be removed without a solid case. Alastair Haines 11:27, 24 June 2007 (UTC)

There is a world of difference between citing a source in an article about some topic, and having an entire article dedicated to copying a single source's data. --Ptcamn 00:31, 25 June 2007 (UTC)

• keep nominator has not given a valid reason for deletion, the problems noted should be dealt with by editing the article  ⇒ bsnowball  17:45, 24 June 2007 (UTC)

  The reason is that the article itself is inappropriate. You can't fix that by editing. --Ptcamn 00:31, 25 June 2007 (UTC)

• keep there are various classification for languages, and obviously they are considerably debated. An article about one particularly frequently used source is appropriate. We have an article on the LC classification, and antoher one on Dewey. DGG 23:41, 24 June 2007 (UTC)

  That's a false analogy. Language classification is nothing like book classification. Book classifications can each have their own merits, based on how useful they are for different purposes. Language classification isn't a question of usefulness, it's a simple factual question of whether and how languages related or not. It's something that can be either right or wrong.

  I would also note that Ethnologue is a tertiary source. They did not invent this classification. This isn't "the Ethnologue classification", it's "the classification that Ethnologue happened to use in its 15th edition". --Ptcamn 00:31, 25 June 2007 (UTC)

  Ptcamn, could you expand on this a bit. As I see it, DGG is demonstrating that multiple notable sources of classification each deserve an article, if the classification has been in use at some time. Has the "Ethnologue" classification never been in wide use? Has the "Ethnologue 15th edition" classification never been in wide use? John Vandenberg 02:04, 26 June 2007 (UTC)

  What sort of "use" do you mean? --Ptcamn 02:29, 26 June 2007 (UTC)

  I am not an expert, so I am open to any definition of use that seems appropriate. Ethnologue "Language families" -wikipedia -Britannica returns a bucket load of hits, so I expect a damn good definition of use in order to discount those ghits. Compare that with the 24 for Encarta, and 300 odd for Britannica. John Vandenberg 02:49, 26 June 2007 (UTC)

  My google results are:

  "language families" 284,000

  "language families" -ethnologue 264,000

  "language families" -wikipedia 252,000

  "language families" -Britannica 259,000

  "language families" -ethnologue -wikipedia -Britannica 217,000

  so of the 284k places using the text "language families" 264k (93%) of them don't mention Ethnologue; 252k (89%) don't mention WP, 259k (91%) don't mention Britannica, and 217k (76%) don't mention any of them. So as for whether any of these is a source that is "used" is open to debate, but it seems as though Wikipedia and Britannica are used at least as much and perhaps more than Ethnologue by websites using the phrase. So much for quantitative measures. Qualitatively no amount of ghits can mean anything. The up-shot is that if we keep this, there is no principaled reason to not have the others (and whatever mysterious sources that the 76% may be relying upon). A final note: many things are in dispute, but rarely do we admit of separate articles of each various source. Take the population of a major city (in the U.S.), we have the US census bureau that gives one number, the state, county and city goverments probably have their own numbers, the chamber of commerce gives yet a different one, advocates for various groups (e.g., immigrants, undocumented immigrants, homeless, various ethnic or cultural groups that have historically been undercounted in formal censuses) give yet a slew of different numbers, so are we fated to have Population of Philadelphia (Census Bureau), Population of Philadelphia (Chamber of Commerce), Population of Philadelphia (Coalition on the Homeless), Population of Philadephia (La Raza) etc. all deserving of article status? Carlossuarez46 23:31, 27 June 2007 (UTC)

Carlos! There are plenty of opposing points of view that can easily be catered for in individual articles (Drink and Drive/Catch a Cab), but there are others that cannot (Capitalism/Communism). How do we determine what is appropriate in this case? The main reason this article is here is because I found the material easy to obtain, but didn't dare put it in the main article, because I felt it was only one "incomplete" treatment of the subject. I think merging it into the main would lead to frustration for some experts, but maybe it would stimulate contributions of sourced criticism and alternative approaches. But then again, it's just a list! Alastair Haines 14:27, 28 June 2007 (UTC)

Google Stats are fun. Of all mentions of language families ( 283,000 ), 215,000 dont mention any of the reference material raised here, so it is easiest to see things by removing them from the results. 1,000 mention only Encarta; not worth mentioning further. 14,300 only mention Ethnologue, 22,000 only mention Brittanica, 23,100 only mention Wikipedia. As an aside, 2,310 mention only Brittanica and Ethnologue, 1,050 mention only Brittanica and Wikipedia, and 865 mention only Wikipedia and Ethnologue. Your point is well made that Brittanica is more referenced for this topic than Ethnologue, but I think there is something to be said for Ethnologue being in the same ball park, given it is not a general encyclopedia. The importance of Ethnologue as a classification of languages hasnt been discounted in my opinion. This Wikipedia article will become more critical over time. Ideally it would outline the changes between the earlier editions and the 15th and then future edition.

wrt to the Population of Philadelphia, language families are not population stats of a single region. This article is providing a taxonomy as provided by Ethnologue. A more accurate comparision is that we keep records of both U.S. Combined Statistical Areas and United States Census data. Both are slightly different ways of breaking up the geography of the U.S., and both are useful and authoritative (i.e. as a series rather than a specific edition). John Vandenberg 16:05, 28 June 2007 (UTC)

Google stats are fun. The question in my mind boils down to whether in the article Language family, we ought discuss the differences among (major) sources, or whether we have separate articles for the topic each according to its own source. Moreover, the articles by source will inevitably be just the list without any meaningful discussion why the tertiary source came down the way it did - which choices were controversial and why did they take that view of the controversy or the evidence? I am not a listophobe (if that's a word); indeed, good lists are good for WP. I can agree that the contents merit inclusion at WP in some form, but just not as a context-free stand alone article. I was trying to come up with an example of how we treat wholly divergent sources of material, and couldn't so I came up with a controversy that so far hasn't hit WP. But with further thought I have come up with the proper metaphor: different English translations of the Bible name and organize the material slightly differently - the ancient Greek, Latin, Aramaic, Hebrew, etc., didn't always come with chapter and verse markers, or even sometimes punctuation in a modern sense. So while we properly cover the Douay Bible and King James Version of the Bible and Revised Standard Version and New Revised Standard Version and others and we cover contents common to each, the Ten Commandments and Psalms, e.g., - we don't have and ought not have articles such as Ten Commandments (Douay), Ten Commandments (KJV) etc., even though they group and translate the commandments (and Psalms) differently - those differences are handled either at the articles about the translation in question (it is with Douay and the names of certain books that differ from the KJV, and in the numbering of the Psalms) or in the article about the passage (like Ten Commandments or Psalms). The grouping of data, the naming of things, is quite similar pedagogically to language classification efforts. Again we shouldn't have various "versions" of each language family by source. The article's contents either ought to be merged with Language family or with Ethnologue. Another indication of why that is so is that were the article expand to the next level of detail - which is also encyclopedic - how Ethnologue organizes each language family it recognizes will lead to further Indo-European language family (Ethnologue), etc. articles, when the contents with various sources' contrasts and bases for analysis should be consolidated at the "family" in question's article, with a general survey at the source's article (particularly if the source is the source of more than just language classification - such as Britannica - or we'll have a multiplicity of articles of each language family, subfamily, etc., by source with no context or critique of what evidence supports or militates against the view taken by the source. Sorry my $0.02 was so long, but there you have it. Carlossuarez46 21:37, 28 June 2007 (UTC)

• Deleteas bad precedent; language classification is the subject of much dispute but we don't need to have an article on each tertiary source's take on it unless we want to have Language families (Britannica), Language families (Encarta), Language families (Wikipedia), etc. Carlossuarez46 19:42, 25 June 2007 (UTC)
• Merge and then delete per the discussion above of this timestamp. Carlossuarez46 21:37, 28 June 2007 (UTC)
• keep. The official site of the ISO 639-3 Registration Authority comments:

It would appear that SIL International as an organization is trusted by the International Standards Organization, and Ethnologue (15th ed) is trusted as data. Perhaps this additional information needs to be added to the article. Of course that doesn't preclude addition of criticism of ISO, SIL or Ethnologue, which should be sourced and cited and added to enhance the article. :) Alastair Haines 01:41, 26 June 2007 (UTC)

This has nothing to do with whether Ethnologue is a reliable source or not (as Carlossuarez46 pointed out, articles like this are inappropriate no matter what source you use), but nevertheless I will point out that ISO 639-3 has since been revised where the Ethnologue data has turned out to be inaccurate (and is still being revised). --Ptcamn 02:04, 26 June 2007 (UTC)

As I understand Carlos' argument, he is saying separate entries for tertiary sources are inappropriate. I agree with that as a general principle, but not as an absolute rule. Many large subject areas have similarly large tertiary literature, with clearly defined groupings — for example, Realism and Idealism in philosophy. These are so notable in themselves, they need no parenthetical descriptor.

However, the argument is specious anyway, given that Ethnologue 15 was specified as the basis for ISO 639-3, it is a primary source wrt ISO 639-3. Primary sources contain errors, in this case it is being peer-reviewed. Ethnologue 16 should end up being a consensus document, but it might not document minority opinions. We certainly can and should do so.

I really appreciate your concern, I suppose people are indeed inclined to accept things uncritically, and to miss the important point that Ethnologue is only now going through a truly global peer-review process, and even then this process is moderated by the organization that publishes Ethnologue! There are valid criticisms of what the ISO have authorized, and surely of whatever ends up being produced. If we have an article on it, we can document criticism, if we don't, we look to be making an editorial judgement for censoring what the ISO have determined. I'd find that hard to defend. Alastair Haines 02:35, 26 June 2007 (UTC)

That ISO relies on Ethnologue for the existence of languages, their spellings, their best names, says nothing about ISO's reliance on Ethnologue for the organization of those languages (correctly named and spelled) into larger groups families. Similarly, ISO relies on [1] for extinct languages, and that site (apparently maintained by a couple of universities, so probably a reliable source as WP goes) also ties its list into families. Once again, there is no indication that ISO buys into the categorization that is the basis of this article. Carlossuarez46 21:02, 27 June 2007 (UTC)

Sounds fair to me Carlos. From a Wiki point of view, though, it's not our job to prefer one source over another, but to report each. If one source seems to get unmerited attention, surely we should add material about alternative sources, not delete the reporting we already have. If the source has questionable methodology or speculative results, there will be criticism we can also report. This is a great discussion because it shows us work that needs to be done, not work that needs to be undone. At least that's how I see it, but I'm an idealist. Alastair Haines 14:13, 28 June 2007 (UTC)

This discussion is long for new people to read (not a criticism). I'll attempt a summary, tear it apart as biased and that will still help summarise everything. ;) Seems the ideas are:

• proposal -- delete -- one controversial source reported without criticism, others don't have similar attention (as yet) Alastair Haines 23:02, 28 June 2007 (UTC)

-- additional argument, we may not even want this level of detail for all alternative classifications

• alternative -- merge -- into Language families or Ethnologue, material here is sourced and relevant elsewhere, why stand alone?
• status quo -- keep -- Ethnologue 15 has special status wrt ISO and this list can be used or not elsewhere as required w/o cluttering another article.

Does anyone know how to make a Wikitable both sortable and collapsible? Whenever I try, the hide/show goes in the wrong place, or the sort buttons go to the wrong line. Sortability helps this list a lot. Collapsibility would make it less intrusive if moved (now or later). Cheers. Alastair Haines 23:02, 28 June 2007 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.