A posteriori lexical, syntactical, and grammatical evolution

This forum is for constructed languages, both those invented by UniLang members and those already existing.

Moderators:Ashucky, Dormouse559

do_shahbaz
Posts:68
Joined:2020-01-16, 0:45
Gender:male
A posteriori lexical, syntactical, and grammatical evolution

Postby do_shahbaz » 2020-01-16, 12:02

I am creating an a-posteriori conlang based on a real Indo-European language becoming extinct 1000 years ago. Although I have read and re-read the latest attested texts in the language's corpus, and have managed to do off with phonology* (mostly) and "solid" unevolving words such as numbers, as of now I am utterly stuck with creating new words, because I can't seem to know in which direction to carry on its lexical evolution. I already know of lexical change and its types, borrowing, calquing, etc. however:

1) In Indo-European languages, word for "hand", "dog", "do", "die", the numbers, etc. tend to change more slowly and less drastically, but the words "tail", "girl", "kick", among others get replaced and re-invented more often. Are there trends, universal and IE-specific, that governs which words and expressions are to be more "conservative" and "innovative" then the rest?
2) How much should a language change from 1000 A.D to 2000 A.D, lexically, syntactically, and grammatically?
3) Can expressions such as "please", "hello", etc. hold out for 1000 years (in English the former displaced "pray" [as in "Pray do tell me Mr. Bingley..."])
4) Do wars, sprachraum-expansion, migration, and other social phenomena impact the speed of language evolution, and in what way (accelerating or decelerating)?

I shall post other questions of mine once each of them comes to mind.

N.B. I'm not sure whether to create this thread in this forum or the next one (a.k.a General Languages). The discussion would certainly have implications on real-lang linguistics; I guess I shall move the discussion there if it gets too interesting.

*okay, I admit that am still stumped on some issues with the phonology. Which brings me to my last question :
5) Is it wise to start off with the lexicon of a conlang even when the phonology remains somewhat unresolved?

User avatar
Dormouse559
Language Forum Moderator
Posts:6939
Joined:2010-05-30, 0:06
Real Name:Matthew
Gender:male
Country:USUnited States (United States)

Re: A posteriori lexical, syntactical, and grammatical evolution

Postby Dormouse559 » 2020-01-18, 1:34

do_shahbaz wrote:I am creating an a-posteriori conlang based on a real Indo-European language becoming extinct 1000 years ago.

I'm curious which language you're starting with; if it has any close relatives with more recent attestation, those are worth a look. The location and history of the conlang are also relevant.

1) In Indo-European languages, word for "hand", "dog", "do", "die", the numbers, etc. tend to change more slowly and less drastically, but the words "tail", "girl", "kick", among others get replaced and re-invented more often. Are there trends, universal and IE-specific, that governs which words and expressions are to be more "conservative" and "innovative" then the rest?

Just pulling off the top of my head here; I'm sure there's a more exhaustive answer out there on the interwebz. How common a word is is one of the factors. Less common words are more likely to get replaced or modified. The distinctiveness of a word is also a factor; if words used in similar contexts start to sound alike due to sound change, one or more could be changed (cf. pen, pin -> ink pen, stick pin; in some English dialects with the pin-pen merger).

A quick Wikipedia search got me the Dolgopolsky list, which gives the 15 most stable words from a sample of 140 languages. The Swadesh list and its variants are much better known, but they're not necessarily about measuring the likelihood a word will be replaced. An advantage to the latter is that it's larger, so more practical from a conlanging standpoint.

2) How much should a language change from 1000 A.D to 2000 A.D, lexically, syntactically, and grammatically?

At least some? There isn't a hard-and-fast rule here because all languages change at varying rates and in different ways. Certainly, I'd expect a language spoken now to be quite different from its ancestor a millennium ago, but exactly how much and how can vary widely. A thousand years gets you from Vulgar Latin to highly innovative French, but it also gets you to the conservative Sardinian. It also gets you Romanian, which is innovative in some respects but retained a case system, unlike most other Romance languages.

3) Can expressions such as "please", "hello", etc. hold out for 1000 years (in English the former displaced "pray" [as in "Pray do tell me Mr. Bingley..."])

I don't see why not. It also depends on what "hold out" means for you. "Hello" has antecedents going back further than a thousand years; they just weren't used exactly the same as "hello".

Really, I don't see why any given vocabulary item couldn't last a thousand years. It's more the impression one gets from looking at the vocabulary overall that matters for believability and/or naturalism.

4) Do wars, sprachraum-expansion, migration, and other social phenomena impact the speed of language evolution, and in what way (accelerating or decelerating)?

Possibly, but not all in the same way, and the effect will vary based on the details of the situation.

N.B. I'm not sure whether to create this thread in this forum or the next one (a.k.a General Languages). The discussion would certainly have implications on real-lang linguistics; I guess I shall move the discussion there if it gets too interesting.

The concerns of conlangers and linguists regularly overlap. As one of the conlang forum mods, I find your topic to be a good fit here. If you feel the topic really must be moved, send me a PM.

*okay, I admit that am still stumped on some issues with the phonology. Which brings me to my last question :
5) Is it wise to start off with the lexicon of a conlang even when the phonology remains somewhat unresolved?

Maybe, maybe not. Every part of a language feeds into every other part, so you might as well ask whether it's wise to work on the phonology without having resolved the syntax, or the lexicon without having resolved the morphology. I'd say to work on the language in whatever order makes sense for you.
N'hésite pas à corriger mes erreurs.

User avatar
Ashucky
Posts:1745
Joined:2010-11-09, 18:35
Real Name:Andrej
Gender:male
Location:Ljubljana
Country:SISlovenia (Slovenija)
Contact:

Re: A posteriori lexical, syntactical, and grammatical evolution

Postby Ashucky » 2020-01-18, 12:56

Dormouse559 wrote:
do_shahbaz wrote:4) Do wars, sprachraum-expansion, migration, and other social phenomena impact the speed of language evolution, and in what way (accelerating or decelerating)?

Possibly, but not all in the same way, and the effect will vary based on the details of the situation.

Just wanted to expand on this topic a bit more. Three external factors may affect the speed of language evolution: geography, range, and stratum.

Geography: the more varied the terrain is, the greater the dialectal variety may be. If the speakers of your language live in a mountainous area where travelling from one valley to another is severely hindered by the terrain, the language is more likely to diversify quickly because there is little contact between the isolated pockets (=valleys) of the speakers, and a change in one group of speakers (valley A) has a much lower chance of spreading to another group (valley B).
By contrast, if the language is spoken in a flat area where transport is quick and easy, the language will appear more homogeneous with less variability, simply because any kind of change can quickly spread throughout the entire population.

Range: languages spoken in a small area tend to be more complex than languages spoken across a large expanse. A small language is more likely to be conservative and retain more of the complexity of its parent language (or even increase its complexity over time), while a language that spreads quickly across a vast area tends to lose more of its original features and complexity.

Keep in mind that range and geography tend to go hand in hand.

Stratum: by this I meant the perceived rank or prestige the language has in the society. Is there an substratum (another language of lower prestige) or an adstratum (a language of a higher prestige)? This will generally have an affect on vocabulary (not so much on grammar).

A subcategory of stratum can be standardisation. If there is a particular dialect (the prestige dialect) and a strong central government, then it's more likely that all other dialects will eventually be replaced by the prestige one, thus making the language more homogeneous (and such dialects also tend to be less complex).
Slovenščina (sl)English (en)Italiano (it)漢語 (zh)Español (es)Suomi (fi)Svenska (sv)日本語 (ja)فارسی (fa)Nešili (hit)
The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge.
Največji sovražnik znanja ni nevednost, marveč iluzija znanja.

do_shahbaz
Posts:68
Joined:2020-01-16, 0:45
Gender:male

Re: A posteriori lexical, syntactical, and grammatical evolution

Postby do_shahbaz » 2020-01-22, 5:26

Thank you for all of your suggestions; each of them are very much appreciated.

Dormouse559 wrote:A quick Wikipedia search got me the Dolgopolsky list, which gives the 15 most stable words from a sample of 140 languages.


Is there an extended version of the list? One that ranks concepts according to their cross-linguistical semantical stability?

There isn't a hard-and-fast rule here because all languages change at varying rates and in different ways. Certainly, I'd expect a language spoken now to be quite different from its ancestor a millennium ago, but exactly how much and how can vary widely.


Can the extent of the lexical and grammatical changes be measured? For instance, how much of the lexicon would you expect to change? (loanwords excluding)

I'm curious which language you're starting with; if it has any close relatives with more recent attestation, those are worth a look. The location and history of the conlang are also relevant.


The language I'm trying to "re-construct" is... Khotanese! a language spoken in the southern edge of the Tarim Basin until 1000 years ago. Further information may be found here:

https://cbb.aveneca.com/viewtopic.php?f=6&t=7090

It happens to have some contemporary few not-so-near relatives: Wakhi and (somewhat more distantly) Sarikoli. However, the two languages are spoken in the mountains, while Khotanese is spoken in the plains of an oasis, bordering a river; the latter is therefore bound to be exposed to more Turkic influence. The geography alone (Turkic influence excluding) did impact the language: 2000 AD Wakhi/Sarikoli is shown to be somewhat more conservative than the latest attested Khotanese texts, in, for instance, their less drastic, or lack, of certain consonant lenitions. I'm curious as how to develop Khotanese even further.

Honestly, I am a bit hesitant in sharing the details of my projects, which is why that wasn't the first thing I did in this thread/ But it seems that the peoples of this forum can be "trusted"... (no offense). And to the admins, I say that I have no intentions, by linking to another conlanging forum, of promoting said forum.

User avatar
Dormouse559
Language Forum Moderator
Posts:6939
Joined:2010-05-30, 0:06
Real Name:Matthew
Gender:male
Country:USUnited States (United States)

Re: A posteriori lexical, syntactical, and grammatical evolution

Postby Dormouse559 » 2020-01-23, 8:23

do_shahbaz wrote:Is there an extended version of the list? One that ranks concepts according to their cross-linguistical semantical stability?

I haven't found an extended version of the Dolgopolsky list, specifically. While looking for one, I found out about the Leipzig-Jakarta list, which has about the same concept but includes 100 items. In both cases, they're aggregating the behavior of many different languages, so remember that these aren't law, just expressions of broad tendencies.

Can the extent of the lexical and grammatical changes be measured? For instance, how much of the lexicon would you expect to change? (loanwords excluding)

I don't know how much change I'd expect, nor do I know how I'd quantify it. Particularly, grammatical change seems like it'd be difficult to put a number on. Let's say, generally speaking, that a language spoken now and its ancestor from a millennium ago will not be mutually intelligible, though their genetic relationship will still be evident.

Really, Wakhi and Sarikoli and your historical information are the most useful guides here. While, as you say, the two languages might not be that close to Khotanese, they give you a sense of what related languages have been up to (all the way through the present day!). Here are the bits I noticed:

1) 1000 years ago, certain Khotanese features were more innovative than in modern Wakhi and Sarikoli
2) You've concluded your reconstruction will have stronger Turkic influence.
3) It's spoken on plains.

The first thing suggests an innovative tendency that you could choose to continue. Think about comparing Wakhi, Sarikoli and Khotanese to a common ancestor (Proto-Iranian?) and to each other to get a sense of what changes they share or not.

To get the Turkic influence, you'll need to decide the nature of that influence. Off the top of my head, here are some relevant questions: Which Turkic language or languages were involved? When did the influence start? What was the power balance between Khotanese speakers and Turkic speakers? How did it change over time? (A language with low prestige is more likely to borrow from a language with high prestige than the other way around.) Were Turkic speakers more represented in certain areas of life or social strata, or did they bring new concepts to the Khotanese? (If so, vocabulary related to those areas/concepts has a higher probability of being borrowed from Turkic.) What sprachbund effects are active in the Khotanese range, and does it adopt them, too?

As Ashucky explained, languages spoken in areas with few barriers to communication often simplify morphologically. Khotanese could follow this path, given its location on a plain.

Honestly, I am a bit hesitant in sharing the details of my projects, which is why that wasn't the first thing I did in this thread/ But it seems that the peoples of this forum can be "trusted"... (no offense). And to the admins, I say that I have no intentions, by linking to another conlanging forum, of promoting said forum.

I believe you'll find that I'm on that other forum, so no need to worry. :)
N'hésite pas à corriger mes erreurs.


Return to “Conlangs”

Who is online

Users browsing this forum: No registered users and 13 guests