Talk:Mayan languages/Map and graphics debate

adding images and restructuring
I added two images: a tree diagram of the mayan language family and a map of their distribution. I pulled out the list of Mayan languages and I will instead make a list at the end of the article with links to the respective Ethnologue ISO-codes and wikilinks to their articles.

This is the material I removed:

Maunus 11:41, 27 August 2006 (UTC)
 * Cholan-Tzeltalan languages
 * Chol language
 * Ch'orti' language
 * Chontal language
 * Tzeltal language
 * Tzotzil language
 * Huastecan languages
 * wastec language
 * Chicomuceltec language
 * Kanjobalan-Chujean languages
 * Chuj language
 * Tojolab'al language
 * Jakaltek language
 * Q'anjob'al language
 * Mocho' language (also called Mototzintleco)
 * Akatek language
 * Quichean-Mamean languages
 * Mam language
 * Tektitek language
 * Ixil language
 * Awakatek language
 * Q'eqchi language
 * Poqomam language [poqomam]
 * Poqomchi’ language
 * Quichean languages
 * K'iche' language
 * Kaqchikel language
 * Tz'utujil language
 * Quiche-Achi languages
 * Sipakapense language
 * Sakapultek language
 * Uspantek language
 * Yucatecan languages
 * Itza' language
 * Lacandon language
 * Mopan language
 * Yukatek language
 * Mayan Sign Languages
 * Yucatec Maya Sign Language


 * I'Maunus 11:46, 28 August 2006 (UTC)ve inserted the material again in the form of an infobox of the mayan language family.


 * The Chuj branch is missing from your infobox. Also I notice that Jakaltek dropped out in your tree diagram, which is probably justifiable linguistically (very close to Q'anjob'al) but contrary to ALMG - I'd support you but it's worth mention. Finally, I like your linguistic features edit, it is clearly less idiosyncratic than my own initial attempt, but I think that a few of the things which you cut (effects on Guatemalan Spanish, Lenkersdorf's Sapir-Worfean thesis about Tojolabal, and to a much lesser extent the Lakoff reference), while not necessarily of strictly linguistic interest, are of interest to a general encyclopedic user. Would you object to my putting them back in, conforming of course to your improved format? --Homunq 17:46, 28 August 2006 (UTC)
 * Nope, go ahead. It was a table syntax error that made the Chujean-Qanjobalan branch fail to show. Poptí is another name for Jakaltek, which is used by Nora C England upon whose book I have based the diagram (and she is a big supporter of ALMG). The lakoff stuff I don't really find relevant because 1. his book is not about the maya and he is not a specialist in mayan languages, and other scientist deal much better with the spatial description stuff in mayan languages (there is a lot of literature references about this in the Mesoamerican Linguistic Area article.) and also to me it is not as much the embodiment part that is interesting but the way in which Mayan languages (and mesoamerican languages in general) use them grammatically: this is the part about relational nouns derived from bodyparts. The lenkersdorf experiment stuff I didn't understand the way it was described, but seemed like a very simple way to connect ergative grammar to the linguistic relativity hypothesis. I think if it should go any where it should be in the article Sapir-Whorf hypothesis in the experimental support part. The effects on Guatemalan spanish I also don't think are documented the very same features are common in rural central mexican spanish and while some linguits argue that here it is due to influence from Nahuatl (which has the sdsame constructin as Maya) it is not really possible to unequivacally attribute it to substrate influence (the same constructions are found in European spanish dialects) If you do chose to put it back in though I will not object, although I may reword it. Maunus 21:04, 28 August 2006 (UTC)
 * Lakoff: OK, just add a bullet for semantic calques. Influence on Spanish: I have definitely heard this construction more in Guatemala and Chiapas than in Milpas Altas, DF (though the latter was only the outer orbit for me when I was in the DF); I think this should go somewhere but will defer to you on whether here or MLA. Lenkersdorf: I was trying to simplify a whole book of argumentation into a quick paragraph, of course it comes across as simple. How about I put it in Sapir-Worf with here "(This feature has been used to advance the Sapir-Worf hypothesis in relation to the Tojolabal language.)" Honestly, Lenkersdorf's thesis doesn't really rise to the level of experimental support - he has deep experience and several arguments but all of them are just suggestive, not conclusive.--Homunq 18:48, 29 August 2006 (UTC)
 * Sounds fine to me.Maunus 20:20, 4 September 2006 (UTC)

Distribution map, genealogy tree and other minor criticisms
The size of the words should correlate to the populations - the worst offender is Itza', a probably extinct language which is writ large.--200.6.247.89 00:50, 2 December 2006 (UTC)
 * I've made a note that letter sizes do not reflect spaker populations. (in fact the small lettes are just used for languages with long names in the highlands were they are spoken s close to eachother)Maunus 10:50, 3 December 2006 (UTC)

Also in the genealogy tree figure, Q'eqchi' is misspelled, and Achi appears in parentheses after Sipakapense and Sakapultek for some reason. If we accept the ALMG position, then all three are separate languages. If we go with the OKMA view, then Achi is a dialect of K'iche', but S and S are still languages. &mdash; MikeG (talk) 04:23, 2 December 2006 (UTC)
 * Fixed the geanalogy tree. Maunus 10:50, 3 December 2006 (UTC)
 * Oh no there's one more mistake. The eastern and western branches are switched - If someone can take care of it I would appreciate it otherwise I'll do it later.Maunus 20:19, 4 December 2006 (UTC)

While we're on the subject of errors, I think both the 4 million speakers and the 5 non-Guatemalan Mexican languages are low. I count 8 Mexican languages (Yukatek Maya, Chol, Tojolabal, Tzotzil, Tzeltal, Wastec, Aguacateco, Chontal de Tabasco). Also, adding up just the Mexican speakers from http://es.wikipedia.org/wiki/Lenguas_de_M%C3%A9xico, I get 2051675; taking 40% of Guatemala (based on http://es.wikipedia.org/wiki/Demograf%C3%ADa_de_Guatemala) gives 5055975; and 10% of Belize (http://es.wikipedia.org/wiki/Belice) gives 28107, totalling 7135758. Those last two numbers are ethnic and not linguistic, but since language is the primary ethnic identifier in the region the estimate is decent. I'm conservatively going up to "over 6 million speakers" and 8 Mexican languages but please check me. --200.6.247.89 13:12, 2 December 2006 (UTC) ps. If we could attach speaker numbers to all the languages, I'd say this article would be ready for GA status or better.

Looking at the "sounds" section: it generally only includes sounds that are in ALL mayan languages, but then it also adds the long vowels. On that basis, we should include other sounds (such as the "sixth vowel" written ë) that are in many-but-not-all the languages. Also: if we want to include a picture for the heiroglyphs section, I could easily upload (from any one of a number of sources) a sample from the Dresden codex, from folio 7 or something, one of the list of figures with 4 blocks and 2 numbers above them... I think that would be a clear fair use but do others agree? Certainly it would make a nice counterpoint to the petroglyphic figures elsewhere. --200.6.247.89 21:45, 2 December 2006 (UTC)
 * I put in the stucco glyphs used on the Maya script article for now. Scans of the dresden codex will be greatly appreciated. There is no copyright on photographic representations of the codices, only on drawings (as far as I know).Maunus 20:18, 4 December 2006 (UTC)


 * Wow, thanks, Maunus, that's great (going from "sounds" to "phonological overview"). The one thing that's been lost is the ALMG orthography for all those sounds (though obviously nonexistent for the three proto-mayan-only sounds) - I can make a good stab at it if you don't but you're obviously the expert so I'll give you a chance to do it. --200.6.247.89 14:44, 4 December 2006 (UTC)
 * I placed the AMLG overview in the final section on mesoamerican writing systems, I am not sure if it reflects the most up to date orthography though and I would be happy if you cheked it.Maunus 16:04, 4 December 2006 (UTC)
 * Things are coming along. Can you please add ALMG "q" and "b" (as opposed to "b'")? Again, I'll take a stab if you don't, but I'm the dilettante here. Also, I know that h and y need footnotes (H also used silently to prevent dipthongs, and Y pronounciation varies, at least in Eastern Kaqchikel, between word-final and word-internal - I think it has two valid pronounciations.) And I've heard that the proper IPA for "glotallized consonants" is actually a "creaky" modifier on the following vowel... should this go in a footnote too? I'll check all your work over later from my sources but I'd rather not innovate on the page, currently I'm just watching you.--200.6.254.170 19:25, 6 December 2006 (UTC)
 * But mayan consonants are ejectives and according to the version of the IPA handbook I am using (1999) they are shown with a ['].I am not an expert on ALMG - the only book I have using that orthography is Englands 1998 book - changes after that I don't know nor do I know all the variants for different languages, so please go ahead and boldly add anything you think is wrong, missing or could be expressed better. You don't seem a dilletante at all. (have you considered registering an account?)Maunus 22:33, 6 December 2006 (UTC)

Mayance in the infobox
Should we have the term "Mayance languages" appearing in the infobox? I had thought that this term was not widely used these days...--cjllw | TALK  02:51, 5 December 2006 (UTC)
 * Check out Indo-European languages which has "Indo-Germanic" in the corresponding place. I think that means that that slot is reserved for less-preferred terms, so this is fine. (ps. Also note that that page gives priority to its "category template" over its "infobox"... given the flow of the text, from general-interest to detailed, I think that may be the better choice.)--200.6.254.170 20:35, 6 December 2006 (UTC)

Tables of soundchanges
I am unsure whether the scematic overview table of soundrules from ProtoMayan to modern languagegroups is too big or not necessary. I thouhgt it would be a good graphical way of showing the soundchanges that are the basis for the classification of he languages. It may however be too linguistic (the soundchange formula forexample) or too obscure or too much a repetition of the preceding prose section. I would like comments.Maunus 19:19, 7 December 2006 (UTC)


 * I think everything after the disappeared consonants should be moved to proto-mayan (which, as I have suggested, should be merged with Classic mayan into "Ancient mayan languages", but that's a separate issue). It's interesting for me, but probably doesn't belong here.
 * As for the table (once moved to proto-mayan), as it is, it doesn't add much... but if you put separated the modern language branches each to their own column, then filled in the unchanged sound in the extra columns, and colored the changed sounds, it would start to have some punch. Up to you whether you feel like spending the effort. --Homunq 20:15, 7 December 2006 (UTC)
 * I do think it is important to show the linguistic basis on which the classification of mayan languages as a language family rests - and that is that they can be explained as having evolved from a proto language through a series of soundchanges. If what you suggest is adding the entire phoneme inventories of all the modern languages with a difference in colour showing the changed versus unchanged phonemes I don't find that a good idea. The idea is to show that the basis for classifying the subgroups of the mayan languages is a number of shared innovations between languages. This entails showing which phonemes changed into what under which conditions and in which language groups. I don't think the way you suggest would do that well. I must admit that I don't understad your idea completely. Maybe you could begin the work here on the talk page to show what you mean more precisely? If other editors feel that the comparative phonological framework doesn't belong on this page but on proto-mayan we'll move it there though, no problem.Maunus 21:01, 7 December 2006 (UTC)
 * I've begun reworking over on Talk:Proto-Mayan. --Homunq 22:07, 7 December 2006 (UTC)
 * I've done some changes here taking your ideas into account. I think it is better shown with actual cognate words because then we don't run the risk of providing false information. The colours can be changed easily to fit with madmans maps.What do you think?Maunus 11:45, 8 December 2006 (UTC)
 * Wow.--Homunq 19:05, 8 December 2006 (UTC)
 * I take it that that means acceptable. Feel free to improve or suggest though.Maunus 21:03, 8 December 2006 (UTC)

Reworking maps
Your notes, Maunus: >>Achi is missing from the Quichean proper branch. <<
 * So, this should be added within the K'iche', Kaqchikel, etc. group of 5, right??

Also graphically I would prefer a serif font like the one I used in the other map, but I'll leave it up to your taste to decide if that's appropriate.
 * I'll try a a serif font.

However the apostrophes used should be straight (') looking like commas.
 * Hmmm, they already are straight up and down. How do they appear to you??

I would also change the yellowish background colour which looks kind of dull white or something crisper would be preferrable in my taste and the letters should be compltely black (and weren't you going to assign colour codes to the branches?).
 * Yeah, I'm changing the colours. The yellowish background is not pretty.

Maybe the fuzzy colour and letters is a problem with the resolution? or maybe it would look more crisp in SVG format instead of PNG.
 * Maybe it is the resolution. I'm going to try to increase the size of the final language names.  I did upload an SVG version here, but the SVG looked bad (in part because the 'translator' changed my Arial Narrow to Arial).

Also Homunq had the idea to add the approximate years of splitting by glottochronology and applying speaker numbers to all the branches - I think this would be cool but a bit more laborous Maunus 09:20, 18 December 2006 (UTC)
 * It would be nice if the X axis (the horizontal axis) would more faithfully represent time (instead of the more regular "family tree" format. As noted, that would be quite a bit more work, but more interesting.  I would also like to show, on this chart, the extinct languages, perhaps in italics.


 * I have corrected the chart, with more to come later, Madman 15:09, 18 December 2006 (UTC)

more on Maps
I have completed the genealogy tree, adding Maunus' recent suggestions, as well as increasing the size of the font of the present language names, changing Chujan to Chujean and other changes.

Yes, the original format is SVG, but the upload in SVG looked just awful. I would be happy to send the original format file.

I have thought about the suggestion to add dates and numbers of speakers to the genealogy tree, but I believe that the tree is almost too large as it is, and adding that amount of data would either increase the size or decrease the legibility beyond usefulness. A table of speakers would be useful, but I am suggesting that it be a separate table.

I will still be updating the language name map, and I'm thinking of coding the languages names to the size of the existing speaker population. In fact, I'm thinking that I should also update the history map at the start of the article, although it does look very pretty (still, I have to wonder about the usefulness of colour-coding the present day countries on that map).

Anyway, 1 down and 1 or 2 or more to go. Madman 20:27, 19 December 2006 (UTC)
 * It looks very good Madman. How about if we put the speaker numbers in a template?Maunus 21:34, 19 December 2006 (UTC)


 * Indeed, Madman has done a great job on the diagram. It might however be too crowded/complicated to convey speaker population info in the map itself, tho'. We could instead create a separate List of Mayan languages article, which could be used to tabulate useful data for each -such as alt. names, family, number of speakers, location- which could be complementary to this.--cjllw | TALK  01:44, 20 December 2006 (UTC)


 * Also noting, the equivalent article on the german wiki de:Maya-Sprachen seems to have already constructed a population table based on SIL International data, if it checks out then it might be useful to adopt rather than reinvent the wheel...--cjllw | TALK  07:07, 20 December 2006 (UTC)
 * I have translated and added the list to List of Mayan languagesMaunus 08:49, 20 December 2006 (UTC)


 * Sheesh, that's some fast work, Maunus! I had in mind a different presentation, but that's something which can be worked on over time...Thanks!--cjllw | TALK  08:57, 20 December 2006 (UTC)


 * Would you all mind if I sorted the List of Mayan languages to mirror the genealogy tree? Madman 13:46, 21 December 2006 (UTC)

I have inserted a new map. The language name colors reflect the language groupings and the size of the name reflects the size of the speaker population. Thoughts?

Regarding the genealogy tree, I can change the background color if you (all) think it is intrusive. I'm thinking it needs something more than plain white, though. Thoughts?? Madman 05:03, 22 December 2006 (UTC)


 * Great map! The colours and population size work really well. Please sort the list of mayan langauges if you want. I am in favour of plain white for the genealogy tree and also I think that it does have room for approximate split dates (they will only go in the junctions and in a small black font)Maunus 10:59, 22 December 2006 (UTC)


 * Thanks for the compliments. I am pleased with the map, and am surprised with the stories that it tells:


 * How Yukatek is really the only surviving representative of the Yucatan branch.
 * How robust the Quichean group is.
 * That Ch'orti' seems out of place (I even checked Ethnologue to ensure I had it in the right location).
 * That Wastek is also way out of place - no wonder it is the most divergent.


 * Yes, I am tickled about this map because it really helps me understand the inter-relationships between the various languages.


 * Regarding the tree, if you furnish me with dates, I can put them in the tree (or I can upload the latest SVG version if you want to do it yourself, Maunus). I can change the background to white at that time, as well.  Madman 14:03, 22 December 2006 (UTC)


 * The reason Chorti is out of place is that the proto-Cholan language was spoken as a prestige language by Mayas from Palenque to Copán and throughout the Mayan highlands - at the same time the commoners spoke their respective languages of the k'ichean, mamean etc branches. Actually the Chorti language springs from proto-Choltian which was the southern varant of proto-Cholan (they only changed the l to r in postcolonial times). Maunus 14:18, 22 December 2006 (UTC)


 * That sounds like noteworthy info to me, I've made a stab at including it, Maunus please revise. --Homunq 06:05, 24 December 2006 (UTC)


 * As for the splitting date I though it would be easy to find reliable dates - but looking throuhg my material it has proven slightly more difficult.Maunus 14:32, 22 December 2006 (UTC)


 * I have also inserted this map into the German article, despite my poor Deutsch. I notice that there are Mayan Languages articles several other Wikipedias, including Dansk.  Would you be interested, Maunus, in putting this map in there??  Madman 14:45, 22 December 2006 (UTC)


 * Yes I might, although I don't generally edit there.Maunus 04:41, 24 December 2006 (UTC)


 * Tojolabal should go under (south) of Tzeltal, right west of the concave corner of Chiapas.--Homunq 17:25, 26 December 2006 (UTC)

New language descent map?
Maunus et al.: Should I re-work the present map showing the descent (or outgrowth) of the various branches of Maya to match the names (and colours) used in the article? I'm thinking "yes" but wanted to discuss it first. Madman 22:34, 3 January 2007 (UTC)
 * which map? the genealogy map? I think the present colour symbolism work quite well even if the shades aren't completely the same. Of course it would be preferrable if the language names were consistent throughout.Maunus 09:05, 4 January 2007 (UTC)
 * I'm sorry for not being clear. I meant the first map captioned: "Approximate migration routes and dates of the proto-Huastec and other Maya-speaking groups".  The language names don't match and it might be good to have this map (as pretty as it is) more similar to the other two.
 * In other news, I changed Yukatek => Yucatec in both my images. I would be happy to make any other changes, as you feel appropriate.  Madman 14:54, 4 January 2007 (UTC)
 * Maybe you missed it above, Tojolab'al is slightly misplaced in the map (should go south of Tzeltal). Also if you felt like reordering the quichean languages to put Tzut and Kaq above Kiche (thus leaving K'iche closer to its nearer relatives), but that's minor. --Homunq 16:16, 4 January 2007 (UTC)
 * I did miss that. I will make these changes, as well as uploading SVG versions of the two images so that the "source code" (so to speak) is available for all. Just in case.  Madman 19:10, 4 January 2007 (UTC)
 * The changes have been made. By the way, I ran across this genealogy tree, in French.  This table attempts to add a chronology to the horizontal dimension, as we discussed.  And the squareness of the chart seems to be cleaner somehow than my attempt at curves.  I also realized that Ch'olti' does not appear on my tree (which I will fix later).  Interesting, at least.  Madman 04:41, 5 January 2007 (UTC)
 * I like that one very much and would enjoy seeing that layout here (een thouhg i like your curves). Maunus 08:20, 5 January 2007 (UTC)
 * I second. I'd like that one, if possible with curves and tildes before the years. One thing I wonder: would it be possible to make a language-agnostic version - using say italics for the branches and bold for the actual languages, thus avoiding any English words and making something (possibly) usable on other wikipedias? I understand that English is atypically liberal in allowing foreign (native) spellings, so this may not actually be so very useful...? --Homunq 16:04, 5 January 2007 (UTC)
 * This differs from the other tree on the status of Sipakapense and Uspantek. Which is right? --Homunq 17:09, 5 January 2007 (UTC)


 * Our genealogical tree follows Campbell (1997) the french one follows Sharer (1998) (not a linguistic book) and the Ethnologue. I trust Campbell more on historical linguists than the Ethnologue which is good at other things. And I think that we shall stick with the classification scheme we have here counting Uspantec as springing from the greater quichean node and sipakapense and sakapultek as languages from the core k'ichean node on the same level as k'iche', achi' and kaqchikel-tzutuhil. Maunus 17:54, 5 January 2007 (UTC)
 * Sorry, I meant Sakapultek, not Uspantek. Good, we keep it as it is here. But I leave it to you to act on whether the text should say these two are "closely" related to K'iche' and Achi - given the debate on the issue, I'd guess not.--Homunq 18:29, 5 January 2007 (UTC)


 * I don't know of any debate on the issue - the ethnologue just bases its groupings largely on mutual intelligibillity something that dosn't necessarily reflect linguistic history- campbells grouping are based on linguistic history. Maunus 18:34, 5 January 2007 (UTC)
 * A lot of debate for one word... but the impression given by the word "closely" in the article is that Sip and Sak are CLOSER to Kich and Ach than are Kaq and Tzut - which, while I understand your distinction between intelligibility and history, would still tend to be refuted by the Sharer grouping and is NOT EVEN SUPPORTED by the Campbell one as presented here. (By "debate" I meant "different geneologies".) No sense more back-and-forth here: since I can't check your sources, I trust that if you don't remove the word, you have some reason to support this.Homunq 19:30, 5 January 2007 (UTC)
 * There's no need to hang on to the wording "closely related to" if that can be interpreted as more closely than for example cakchiquel and tzutuhil. I was only objecting to the genealogy presented in the french version which sees the two languages as springing from the greater k'iechean branch. Maunus 19:45, 5 January 2007 (UTC)