Ethnologue 16/17/18th editions: a comprehensive review
Yüklə 151,3 Kb. Pdf görüntüsü
|
According to E16/E17, ‘this edition contains nearly 60,000 updates and corrections from the previous one’ (curiously, the claimed number of updates between the 15th and 16th editions turned out to be the same as that between the 16th and 17th editions). The meaning of this number is mysterious since it gives an average of eight updates per entry, or, on average, more than one update per field for every entry. But the updates are not evenly distributed, and whatever counts as an update is something very lightweight. Some quick computational comparisons of the 16th edition with the 15th gives the fol- lowing. About 2,500 entries have not been changed at all in the name, dialects, popula- tion, and comment fields (whether explicitly indicated or not). At most, 1,350 entries have been updated and indicated as such (as evidenced by the occurrence of the tokens ‘2005’, ‘2006’, ‘2007’, or ‘2008’). It would have been more informative if E16/E17 re- ported the number of updated entries or the number of updated fields, rather than the obviously diluted number of ‘updates’ (characters?). E18 has improved on exactly this point, reporting on the number of updated entries and the size of the update (at least one field).
6 was the ability to register and thus be able to provide feedback to the Ethnologue editors directly from a specific page. Making it easier to provide feedback is certainly a step in the right direction. 2.6. The language inventory according to E16, E17, and E18. The 7,412 (E16), 7,561 (E17), and 7,532 (E18) entries are categorized as per Table 1. REVIEW ARTICLE 729
6 See http://www.ethnologue.com/ethnoblog/mpl/check-out-new-ethnologue. E16 living
extinct no est
total Macrolanguages 55 —
55 Canonical spoken languages 6,682 373
155 7,210
Deaf sign languages 57 1 71 129
Artificial/constructed languages 0 0 1 1 Pidgin languages 4 3 10 17 total
7,412 E17
Macrolanguages 60 — — 60 Canonical spoken languages 6,857 408
81 7,346
Deaf sign languages 71 3 63 137
Artificial/constructed languages 0 0 1 1 Pidgin languages 12 4 1 17 total
7,561 E18
Macrolanguages 60 — — 60 Canonical spoken languages 6,954 363
— 7,317
Deaf sign languages 137
1 — 138 Artificial/constructed languages 1 0 — 1 Pidgin languages 13 3 — 16 total
7,532 Table
1. The language inventory in numbers, as of E16, E17, and E18. The column ‘living’ counts the number of entries for which E16/E17 lists a speaker number greater than zero in the population field. The column ‘extinct’ counts the num- ber of entries in E16/E17 for which the population field lists zero speakers (or a phrase to this effect). The column ‘no est.’ counts the number of entries where ‘no estimate
available’ or an equivalent phrase occupies the population field. Impressionistically, most of the entries in E16/E17 with ‘no estimate’ are living languages for which no population estimate is given, rather than languages whose living/extinct status cannot be inferred. For E18, there is a language status field, and the columns ‘living’/‘extinct’ then simply count the cases marked as extinct or not. Esperanto [epo] is the sole language included as an artificial (E16)/constructed (E17/E18) language, presumably because it is the only(?) such language known to have native speakers (Bartlett 2006).Afew nonnatively spoken languages—for example, Call- awaya [caw] (Muysken 2009), Gail [gic] (Cage 2003), Leti (Cameroon) [leo] (Dieu & Re- naud 1983), and La’bi [lbi] (Moñino 1977)—are included, but most such known lan- guages, for example, Urban Youth languages (Kießling & Mous 2004), are not included. It is clear that a large number of attested pidgin languages are missing. Due to the transient nature of pidgins, however, information as to the existence of a pidgin is typi- cally more ambiguous than the corresponding information about a language with native speakers. I therefore refrain from discussing the E16/E17/E18 pidgin entries in detail, and refer to the comprehensive listing of pidgins by Bakker and Parkvall (2010). The Bakker & Parkvall 2010 listing differentiates different levels of evidence for the exis- tence of a pidgin, rather than a strictly binary decision of existence or not. I am not qualified to judge the sign language entries, so they are left unreviewed here. The remainder of this review is restricted to languages spoken as a first language. 3. Spurious and missing languages. A number of extant languages are missing from E16/E17/E18, and a number of entries in E16/E17/E18 are spurious, that is, do not exist as languages or duplicate other existing entries. In order to systematically enumer- ate missing and spurious languages from E16/E17/E18, the following method was pur- sued. First, a very large collection of bibliographical references 7 to descriptive work on the languages of the world was annotated as to the language(s) described, causing, for example, any reference to a language missing from E16/E17/E18 to become apparent. Second, the classification according to the research literature was reviewed for every E16/E17/E18 language, causing, for example, duplicate entries to become apparent by competing for the same slot in the classification. Third, a survey of one specific gram- matical characteristic was carried out across the research literature for every E16/E17/ E18 language, causing, for example, duplicate entries to become apparent by being grounded in the same source. 3.1. Missing languages. The languages missing in E16/E17/E18 are listed in Ap- pendix A. To be more precise, a language is listed there as missing if: • extant published literature can make a convincing case that the language exists (or existed; see below), and, • extant published literature can make a convincing case that the language is not in- telligible to any language already listed in E16/E17/E18. An important note is that I do not list languages that are missing solely by virtue of the interpretation of a dialect situation correctly understood (but interpreted differently) in E16/E17/E18. This matter is separately treated in §4. 8 For example, if an E16/E17/E18 entry subsumes a number of varieties with borderline intelligibility, and the facts are cor- rectly indicated (e.g. the names of the varieties given as dialects, and the comments about intelligibility), such cases are not listed here, even if there are good reasons to interpret 730
LANGUAGE, VOLUME 91, NUMBER 3 (2015) 7 See http://www.glottolog.org (accessed 20 January 2012) for more information. 8 A large number of other such cases are taken up in the list of scheduled updates to the ISO 639-3 inven- tory, traceable via http://www.sil.org/iso639-3/changes.asp.
Yüklə 151,3 Kb. Dostları ilə paylaş:
|