zompist bboard

[Mwanatabu]

Hey Janko,

I emailed you a couple of months ago regarding the upcoming revival of Yen and some changes in the numeric inventory. Did you get my message?

[Janko Gorenc]

It all depends on the reconstruction of PIE (Proto-Indo-European) and whether or not you ascribe sound values to the proposed laryngeals (whatever those values may be) as well as which spelling system you use. Overall, the two reconstructions you have are more-or-less just two ways

Hi,

Thank you for your message.
I keep both versions (ways) in my collection.
Your reconstruction I'll put there where I have an additional variant, different ways of counting in one and the same language.

Janko Gorenc
_________________
My collection numbers from languages and dialects changes incenssantly.
Today 30.6. 2010 my collection includes: I have data for numbers from 14555 (14015 languages, dialects (both: Conlangs and Natlangs), and 536 additional variants.

[Janko Gorenc]

[jatzail]

no. i speak only about the languages i know. i looked your list a year ago and there were a couple of mistakes in natural languages (ancient and modern)you were adviced about some of these mistakes and i don't know if you corrected. i can suppose mistakes also in other languages because you have the tendency to take as true everything you are said without control personally (i know you cannot control 5 thousand languages but do it with the more important one -also dead).

[sangi39]

I think I'm going to have to agree with both Jatzail and Fajrulo on this one. As Fajrulo says, the number database is a very useful resource, especially for gainin inspiration, comparing numbers across languages and language families, etc. and collecting them all is a good thing in this respect. However, following along the lines of Jatzail, there does seem to be a quite "copy and paste" methodology to the whole endeavour.

Perhaps it'd be best to take Zompist's approach in his "numbers from 1 to 10 in 5000 languages" in which languages are arranged according to language family, then subgroup, then language and then dialect, i.e. gradually dividing the languages/dialects into smaller and smaller groups. IMO this does go some way to avoiding the dialect vs. language question since whether or not you say, for example, Norwegian, Danish and Swedish are dialects rather than languages, they'll still more or less end up being grouped in the same way, possibly under "Continental Scandinavian". Similarly Serbian and Croatian, instead of being dialects of one or the other they may both be labelled under "Bosnian/Croatian/Serbian" which would include the equally linguistically vague Bosnian.

A similar approach could likewise be used for conlangs, creoles, pidgins and mixed languages where they are listed in order of descent and relation to their single or multiple parent languages. In other words, I don't feel that a simple division between conlang and natlang then arranged alphabetically is the best layout for such a database.

This approach of course would involve much more work regarding what language family a language belongs to as well as its descent pattern and relation to other languages but realistically this shouldn't be too much of a problem if more contributions were made by the readers of the database in order to aid Janko's efforts.

On a personal note, as well, I always find the conllection of numbers in conlangs quite a strange idea given that, as Viktor said, many are abandoned quite early on. I know I've abandoned around 20 in the 8 years I've been conlanging and only made any real progress in my most recent projects. It seems important to me then that a conlang should really meet a basic criteria before its numbers should be included in the list, say, for example, having not just a phoneme inventory and a list of cases, tenses, moods, etc. and a basic typological overview, e.g. SVO/OVS/VSO and so on, but also a detailed account of allophony and the use of various grammatical features and how various sentences are constructed beyond simply "cat ate mouse" and some form of derivational morphology and a complete swadesh list. At least with that level in the conlang it's taking on the shape of an actual language rather than a set of lists.
_________________
In the name of the Father, the Son and... wait... that can't be right... can it? No. It can't be.

By the flow of the Red River we worship humanity and live our lives on its banks.

[Janko Gorenc]

[Janko Gorenc]