Talk:Niger–Congo languages

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Africa (Rated Start-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Africa, a collaborative effort to improve the coverage of Africa on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject Languages (Rated Start-class)
WikiProject icon This article is within the scope of WikiProject Languages, a collaborative effort to improve the coverage of standardized, informative and easy-to-use resources about languages on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Start-Class article Start  This article has been rated as Start-Class on the project's quality scale.

Confusing claims[edit]

This page at first claims that Greenberg's Niger-Kordofanian languages (NK) are the same as Bendor-Samuel's Niger-Congo languages (NC) and therefore include the Kordofanian languages (K). But later on it says that NK = NC + K. What is going on? Perhaps Bendor-Samuel included K in NC, but later linguists took K out?
Jorge Stolfi 23:06, 14 May 2004 (UTC)

Nobody's corrected this, so I will. I'm going by memory, however, so please correct me if need be!
I believe Greenberg also used the term NK, but perhaps only after Bendor-Samuel invented it. Also, some linguists do not consider Kordofanian to be the earliest branch, but rather that it diverged strongly under non-NK language influence; it's possible some of them might use the term NC for the whole family. —kwami 01:55, 18 Apr 2005 (UTC)
That was confusing indeed. I amended your correction a bit, since it was Greenberg himself who introduced Niger-Kordofanian in 1963; Bendor-Samuel (1989) re-introduced Niger-Congo. Niger-Congo is currently the most commonly used term among linguists. — mark 07:32, 18 Apr 2005 (UTC)

Which family is the biggest?[edit]

The Niger-Congo languages are the largest group of the world in terms of different languages.

Any data supporting this claim ?

Try this one for some fairly exact figures on a whole raft of language families:

It shows the Niger-Congo family way in front of everything but Austronesian, and ahead of that too. -- Paul Drye

Difference between language and dialect is quite arbitrary and the difference in numbers isn't that big anyway, so I added "probably". -- Taw


I think the links to the countries in which the example languages are spoken make that part of the article disorderly. Would it be OK to remove them, or is there another possibility to reduce the clutter? Strangeloop (talk) 18:14, 17 Aug 2004 (UTC)


Niger-Congo languages

I've added a navbox to see if we can come up with a nice template to navigate along the major subgroups of Niger-Congo. I've not added it to other articles yet because it needs to be polished a bit first. What do others think? — mark 14:36, 19 August 2005 (UTC)

The box might be helpful in a later stage. This moment is just adds a third outline which even is not displayed properly in my FireFox browser on a wide screen. Currently I think adding more meat to the arcticle itself is of higher priority. For example which features a common to Niger-Congo languages (noun classes, serial verb constructions etc). Hirzel 08:48, 20 August 2005 (UTC)

Fair point. I've parked it here for now. — mark 07:25, 23 August 2005 (UTC)

I'm slowly expanding the article, having added a classification history and a bit on common features recently. Sections on tone, noun classes, and syntax will follow, and maybe a more detailed map. — mark 14:34, 29 September 2005 (UTC)

The navbox has been superseded by the {{Infobox Language family}}. — mark 08:46, 16 February 2006 (UTC)

Time since proto-language?[edit]

Does anyone have any idea how long ago the proto-Niger-Congo language is believed to have been spoken? This would be an important addition to the article. --Saforrest 17:27, 12 May 2006 (UTC)

Is this a valid language family?[edit]

Looking at the lanuage families of the world, I find it strang that the Americas, for example, have a lot of small families where Africa, where humans have supposedly lived for much longer, has so few. It should be the other way arround. Geneticists have found that genetic diversity in Africa is far higher than that of the rest of the world. With respect to languages, it should be the same. What is the reason for this? Have the Niger-Congo people replaced everybody else who might have been there bevfore, e.g. because they had agriculture or bronze age or iron age technologies the other peoples did not? Or is it just that the views of Greenberg dominate in African lingusitics and Grennberg is maybe sombody who has a tendency to lump languages together, while in oother parts of the world, the dominating linguists are applying stricter standards (just a question, I am an outsider to the field)? so is the discrepancy an Artefact due to different traditions in different branches of Linguistics? Is Niger Congo or Niger Cordofanian a language family hypothesis comparable in status to Amerindian or [[Nostratic}} or has a protolanguage or important aspects thereof actually been reconstructed and it has been shown that and how the different subfamilies have developed out of this. If anybody has access to material showing evidence for the validity of this family, please include it in the article. If there is a discussion about the validity of this family or of certain subfamilies being attributet to it in the literature, please include information about it in the article. Nannus 00:32, 9 July 2006 (UTC)

Even though you're right that the field of African linguistics has known more lumpers than splitters historically, the validity of Niger-Congo isn't really disputed. A starting point if you're looking for evidence is Greenberg's The Languages of Africa, though that is not a reconstruction. Reconstructions for Niger-Congo (or actually only parts of it) have been carried out by Mukarovsky and by Stewart. Stewart has worked mostly on reconstructing Potou-Akanic (Ghana, Ivory Coast) and Bantu, and in a recent article (2002) he argues that his Potou-Akanic-Bantu reconstruction is a good starting-point for a reconstruction of Proto-Niger-Congo, comparable in fact (so he argues) to the 'Proto-Germanic-Latin-Greek-Sanskrit' of the poineers of Indo-European reconstruction (Stewart 2002:198-200).
  • Mukarovsky, Hands (1976077). A Study of Western Nigritic. 2 vols. Vienna: Veröffentlichungen des Institus für Ägyptologie und Afrikanstik der Universität Wien.
  • Stewart, John. 2002. 'The potential of Proto-Potou-Akanic-Bantu as a pilot Proto-Niger-Congo', in Journal of African languages and linguistics, 197-224.
mark 16:18, 18 November 2006 (UTC)
It may be that workers in the field generally accept the validity of Niger-Congo, but their standards are fairly generous, to put it charitably. No reconstruction or anything approaching it has been proposed for Proto-Niger-Congo as far as I'm aware – certainly not anything as authoritative as in the case of other, more well-known, families –, not even for most subbranches, even low ones. My impression that the field is still largely in its infancy: scores of languages and worse, even whole language groups, are barely documented or even undocumented, let alone researched on the purely synchronic level even, and the status of numerous groups is unclear. There is no real consensus classification, either, as a consequence, just a laundry list of uncontroversial low-level units, uncontroversial only because they are transparent and obvious, especially in the case of dialect continua, of course.
That's not necessarily a problem, because, for example, in Austronesian there is a similar classification problem (although more points are clear or at least pretty unanimous or uncontroversial), but – apart from various individual languages whose classification as either heavily Austronesian-influenced "Papuan" or heavily "Papuan"-influenced Austronesian languages is actively disputed –, the general extent of Austronesian is largely clear, and there are criteria what languages should have in order to be classified as Austronesian, and there is a reconstruction (though what exactly this reconstruction encompasses has recently been questioned by Ross – apparently it does not account for all of the Formosan languages), even though there is, admittedly, still the issue of Tai-Kadai and in some ways, Austronesianists have been exceptionally fortunate. In the field of Niger-Congo, however, not only individual languages, but entire subbranches (large ones!) have been linked to the unit without sufficient evidence, or even on extra-linguistic grounds (especially when documentation is lacking or missing entirely).
There are a hell of a lot of holes in Niger-Congo, apart from a relatively solid appearing core, and a hell of a lot of work still to do (even the subclassification of Bantu has barely progressed from conventional geographical groupings, though intra-Bantu borrowing has been accounted for in the meanwhile, apparently). Similarly, there are a lot of central open questions about Nilo-Saharan, as well, and in fact Afro-Asiatic doesn't look much better. That Khoisan is spurious and a mere typological-areal grouping has been made clear in the meanwhile, at least.
At this stage, I find it hard to take African historical linguistics (with its Greenbergian superfamily heritage, although his macrophyla are not even his own achievement, but much older hypotheses) very seriously, and have no reason to assume that Africa is so much more coherent phylogenetically than the Americas; in fact, quite the opposite, considering the dazzling, but not unexpected, typological diversity within the African macro-areal.
I do realise the source of the problem – a small handful of experts drowning in data, and overwhelmed with the task of filling the white spots in the documentation (more properly, white or grayish zones in between islands of solid documentation) before they can even begin with proper reconstruction – but it would be better to acknowledge the true, sorely lacking, state of knowledge and revert to an agnostic state akin to American linguistics. Splitting is sounder and much more reasonable and productive methodologically than lumping. Especially considering the central importance of Africa for anthropology and human prehistory, it strikes me as rather irresponsible to deceive the nonspecialist audience like that. It seems to me that among the few experts in African linguistics (there is even barely a specialisation "African historical linguistics" in fact, instead simply experts – who happen to be interested and competent in historical linguistics, either through prior training/exposure or autodidactic approaches – with specific areal interests and expertise within Africa who engage in historical work on the side, as an outgrowth of their studies, not as a central focus; hell, most of them aren't even pure linguists in any way, but multi-experts!), there is a gentlemen's consensus to gloss over the difficulties and the few experts are afraid of making a cut and exposing the truth, namely that the Greenbergian superfamilies do not nearly tell the whole story and Africa is not really that undiverse, and not really that different from the Americas (if at all! It might well be more diverse, even) – the motivation I suspect being reputation and anxiety for desperately needed research money. Admitting that their field is largely castles in the air could reflect badly on Africanists, or badly taken up on the side of nonspecialists (especially nonlinguists and complete laypeople), and could make them look cranky, so keeping the neat appearance (which is possible, after all) is seen as the much easier and preferrable solution. I'm far from wanting to bash Africanists with this – quite the opposite, I have to admire their achievements deeply, in the face of adversity and pitfalls such as politics, I find their work completely humbling in fact, and I empathise with them completely – but I wish they were more honest, frank and simply upfront about the field, instead of insisting in their personal, but insufficiently substantiated beliefs about the interrelationships of African languages.
People need to understand that working with African languages is several orders of magnitude more difficult than working with Indo-European languages. Indo-European is a small, manageable family in comparison, and reconstructing Proto-Indo-European is more like reconstructing the parent language of a subgroup of a subgroup of Niger-Congo, as it basically involves the comparison of a handful of clearly distinct, but closely related (ancient) languages: the effective time-depth to bridge is more on the order of 3000 than 6000 years, while Niger-Congo is at least three times deeper. It's like having not a dozen, but a hundred families like Indo-European at your hands. --Florian Blaschke (talk) 18:10, 13 November 2011 (UTC)

The truth is that this Niger Congo A and Niger Congo B grouping is totally false. The non-Afroasiatic, non-Bantoid, non-Bantu, non-Nilo Saharan languages spoken in Nigeria, Ghana, Benin (formerly Dahomey), and several other places in West Africa are absolutely nothing like the "Bantu" languages that span much of the rest of the continent. The entire theory was a lazy hypothesis from a lazy scholar who is overpraised by others who are too lazy to bother studying the languages for themselves to see how extremely different they are from the "Bantu" languages. This Niger Congo edifice will someday crumble under the weight of more serious and rigorous scholarship. — Preceding unsigned comment added by (talk) 21:21, 1 July 2012 (UTC)

the yoruba language is listed under bantu[edit]

the yoruba language is incorrectly listed as a bantu language it should be listed under benue congo which is a part of the larger volta congo family —The preceding unsigned comment was added by (talk) 11:22, 12 January 2007 (UTC).

Actually, it is listed under Benue-Congo and I couldn't find anyplace in the article suggesting that it is a Bantu language. — mark 12:42, 12 January 2007 (UTC)


Mel, is it really your argument that I have to source the statement that the Bantu languages include most of the languages of central, eastern, and southern Africa, or you will remove it? This is trivially verifiable. john k 20:26, 10 March 2007 (UTC)

So why exactly do you not provide a source? Is your claim that editors don't have to provide sources for what they say (or even explanations of their edits in edit summaries), but that it's up to others to do it for them? --Mel Etitis (Talk) 22:50, 10 March 2007 (UTC)
BUt u dont remove it if it isnt a disputed fact, add a fact tag, but it could b considered a WP:POINT if someone said the sky is blue and u said prove it. The map of Africa shows the distribution of Bantu languages, i dont c the problem. also watch the language thing, if an editor feels someone is being unreasonable just explain it on the talk page. the bad languages weakens your case. I think the cursing editors changes are allowable, esp due to the fact the many things have no [citation needed]--Halqحَلَقَة הלכהሐላቃህ 23:45, 10 March 2007 (UTC)
I apologize for my inappropriate language earlier. As to the rest, like Halaqah says - not everything has to be cited, and the basic claim in my statement is about as close to "common knowledge" as one can get. Columbia says, for instance, that the Bantu "inhabit most of the continent S of the Congo River except the extreme southwest." The map in this article shows more or less my point. The revert was clearly a WP:POINT violation because I was arguing with Mel on another talk page. Would Mel revert me if I made an unsourced edit that George W. Bush is President of the United States, or that the Earth is the third planet from the sun, too? john k 05:21, 11 March 2007 (UTC)
I will be 100% honest, when i saw that edit dispute i knew it had personal roots. It happens to me, where people dont like you disrupt all your edits. It is sad, but what the revert was about was a WP:POINT and yes we should cite references, dont get me wrong but just add fact tags and allows the user to reply, esp if the addition isnt crazy stuff. We need to edit wiki not attack each other.--Halqh حَلَقَة הלכהሐላቃህ 11:23, 11 March 2007 (UTC)

written forms?[edit]

This page lacks any discussion of written forms for the languages under examination. It may be obvious to experts that no written forms exist, but non-experts like myself may find themselves here.

If there are written forms, they should be discussed here.

If there aren't written forms, that fact should be clearly stated. 02:08, 2 June 2007 (UTC)

Well it's not that obvious that no written forms would exist — Swahili for example has been written for more than a millenium, and in the last few hundred years lots of Niger-Congo languages have been 'reduced to writing', as the experts call it. I do agree that this is an important topic, but perhaps this particular article is not the right place for it. Niger-Congo is a huge language family, and no simple generalizations can be made about written forms of the languages that belong to it. Compare the article on Indo-European language, which like this article focuses mainly on linguistic and historical issues and consequently doesn't mention the issue you raise at all. — mark 21:26, 2 June 2007 (UTC)

Niger-Congo A? B?[edit]

The map shows 2 separate shaded areas: Niger-Congo A and Niger-Congo B (Bantu), but I cannot find anything in the article which even mentions, let alone explains, this division. Can anyone help me here? Shanoman 13:54, 7 June 2007 (UTC)

Yeah, I was wondering the same thing. I find it peculiar that these language families can be so arbitrarily grouped together when many of the peoples in question have little to do with one another. This may be the reason why the "language family" is so very broad. I am sure Greenberg and Westermann is not the only person to study these languages, why is there not more sources? Is he your preferred source??

I happen to study Cultural Anthropology and I know that I would not base my conclusions on the testimony of a single author or researcher. Cross-referencing and then cross-referencing again until you have hodge-podge of several names sources from different Universities and backgrounds who generally agree on the matter is the only way to satisfy objectivity. Otherwise, it (this) is simply one man's opinion.

The section on classification history makes sufficiently clear that the hypothesis is not simply the idea of one man. Also, I believe ample sources are cited, at least in that section. See also previous discussions on this talk page, e.g. "is this a valid family?" above. — mark 18:30, 29 August 2007 (UTC)

No history is included as to this language![edit]

I find it quite odd, and in marked contrast to the other language family articles that there is nothing in this article describing the history of the Niger-Congo languages nor is there even a discussion of proposed urheimat locations!Chattanoogan (talk) 02:43, 26 November 2007 (UTC)

  • Perhaps you know, that linguists are quite sure, that the origin of all Bantu was in Cameroon. About the relations between Bantoid and the other languages, they made up their mind quite late, that these groups are one family at all.
  • Another problem ist, that the most similar languages are the Bantu languages, whose speakers moved the longest distances. In the region of Nigeria and farther west, the differences between the subgroups are much bigger.
  • Remember the discussions about the origin of the Indoeuropean family, the ancient languages of which are much better documented (the Tocharic language is only known from inscriptions and has no present successor). There ar equite different current theories. The most convincing one bases on phnonological shifts and seeks the origin in present Ukraina.
  • If the geographically nearest groups are the most different ones, as in western Africa, it is difficult to reconstruct a tree of migrations and splittings.

--Ulamm (talk) 00:21, 27 November 2007 (UTC)

Actual Number of Luyia And Ganda Speakers[edit]

The number of Luyia and Ganda speakers on this page is certainly incorrect: a quick look at Luyia shows the former should be at over 5 million speakers, while the latter should be at least six million. I propose to change the language map to reflect this shortly. Wanyonyi (talk) 08:33, 19 November 2008 (UTC)

Some of the figures date from about 1980, some are newer. I'd prefer to enter a remark about the sources, such as "data base 1980 - 2000"). Otherwise I would have to adapt not only the figures but also their graphic visualization every few months.--Ulamm (talk) 12:10, 26 February 2009 (UTC)

No vocabulary[edit]

Under the heading "Common Features", no common vocabulary is mentioned. —Preceding unsigned comment added by (talk) 13:29, 17 May 2010 (UTC) Most or all of the common features also occur outside Niger-Congo. Tones occur in Chinese, for example. —Preceding unsigned comment added by (talk) 13:33, 17 May 2010 (UTC) Greenberg spoke on this point but his sound laws are very vague. So are his semantics. —Preceding unsigned comment added by (talk) 13:51, 17 May 2010 (UTC) Refusing to include Greenberg's list of cognates may be deliberate, done to hide their vagueness. —Preceding unsigned comment added by (talk) 15:49, 17 May 2010 (UTC)

Good points, but hell no, you can be quite sure that no-one here has deliberately left out proposed cognates to hide the problems with them, or with Niger-Congo. — mark 16:39, 18 May 2010 (UTC)
Everyone is welcome to include Greenberg's cognates in all four articles.
If this is not done soon, I may well include them myself. —Preceding unsigned comment added by (talk) 09:26, 26 May 2010 (UTC)