Linguistic "Punctuated Equilibrium"

[CarpeMors]

RMW Dixon is notable for having offered a theory of punctuated equilibrium for linguistics, similar in vein to the theory in evolutionary biology offered by Gould. The basic premise is that languages tend to remain constant and relatively unchanged throughout time, and that it is contact with other languages and peoples that drives major language change, which can also be quite rapid language change.

Then of course, there is the whole glottochronology, which as far as I know, can't ever really offer anything meaningful to say that isn't post hoc.

So say I posited this:

*uukibo -> uukshibvo -> uuchivo.

(I'm too lazy to go cut and paste a bunch of IPA symbols right now. The u is obviously meant to be long, the k is obviously in a process of palatalization, ending in a postalveolar affricate, and the b lenites to a bilabial fricative then to a labiodental fricative)

ANYWAY: How long would an evolutionary process like this take? 100 years? 500 years? Is there any sort of rule of thumb? I'm trying to sync up my conworld's languages so that I can apply the evolution and spread of writing amongst them. But part of that is figuring out rates of change. There is language family A which spawns languages {a, b, c, d, e, f} Then there's language family B which spawns languages {g,h,i}. There are other's but in this case the people who spoke language a invented writing (ideophonetic script), and it spread pretty much intact to languages b and c. Languages d,e,f, borrowed it, but used it syllabically, almost like Akkadian.
E.g., there is, say, a glyph for "duke", which languages a, b,c use to mean just that, regardless of what they actually say. In language d the word for duke is "chak". And they write it syllabically, as "ta-ya-ka". (When they originally borrowed the glyphs to use syllabically the word for duke was "tyak". Language speakers d,e,f almost use the syllables like an abiguda, but not quite. By the time the syllabic-ish glyphs of d,e,f spread to languages g,h,i., they used it exclusively syllabically.

I always endeavor to be realistic, so not knowing how to approach this, is a nagging little bother in my mind. The history needs to sync up perfectly. The development and spread of the writing systems can only make sense if the languages are developed carefully.

So, I'd appreciate any thoughts or criticisms or witticisms that any of ya'll might have.

CheerS.

[eldin raigmore]

I don't see why it has to take longer than a single lifetime; or more than three (two? one?) generations.
It might not be very likely, but I don't think there's any reason to think it should be impossible.

[basilius]

CarpeMors: your question is framed in such a way that Wait, I can put it shorter: Dixon's theory is shit.

I don't see why it has to take longer than a single lifetime; or more than three (two? one?) generations.
It might not be very likely, but I don't think there's any reason to think it should be impossible.

The somewhat informal but practical rule I try to observe with SC's I design is: the difference between the (most innovative versions of the) phonetic systems used by three successive generations should not exceed what one can realistically expect for the variation among same-generation lects in one community (e. g. one settlement).

To assess CarpeMors's SC's, I need a detailed list of the SC's with all the intermediate stages (for example, [k] -> [kʃ] does not look like a single-step change to me), including those not affecting the words in CarpeMors's examples (like, the palatalization which seems to happen there is probably phonemic, so there must have been some rearrangement of vowels in between).

My wild guess would be a minimum of 250 years.

Curiously, plain *uukibo -> uuchivo might take less; like, 125 years (shorter if {ch} is [c] and {v} is [β]).

Slower is always possible.

[CarpeMors]

What is "SC"? O.o

Also, I know there is a lot of criticism of Dixon, hence why I put the bit on glottochronology in there, as to me it seems a bit like glottochronology in some aspects. But I don't think my original query has been addressed adequately. First, while languages change, the rate at which various languages change is not constant, even among dialects of the same language. If there's going to be a "rule of thumb", I'd like it to be a plausible one. (*Plausible*=I've made some arguments, or demonstrated the relative worth of something)

I didn't consider direct palatalization of the stop K. I was thinking it affricatized early, and slowly meandered centrally. I may have to adjust my thought process on that.

So, does no one want to take a crack at this question, seriously? Disappointing...

[basilius]

What is "SC"? O.o

Sound change. Local jargon. Sorry.

First, while languages change, the rate at which various languages change is not constant, even among dialects of the same language.

Yes. That's why I wrote:

My wild guess would be a minimum of 250 years.

And:

Slower is always possible.

The latter may need a correction; I don't think more than 100 years without a single SC is OK. But no SC's affecting two specific words, is.

If there's going to be a "rule of thumb", I'd like it to be a plausible one. (*Plausible*=I've made some arguments, or demonstrated the relative worth of something)

I don't think there's anything like a consensus on that among profies (including, on how to count SC's, not to mention measuring their pace).

My reply was about conlanging. Its logic was the following. In normal language transmission people rarely notice systemic differences (except when there's a dialect shift) within three generations, i. e. variation is mostly the same type as found within one generation in the same community.

It can be claimed that sometimes changes go faster, but I feel that I am on the safer side observing my rule. Also, in my experience, the rule does not make it more difficult to model the pace of changes comparable to that from Old English to Early Modern English, or from Latin to Old French, i. e. changes which are usually considered fast.

Also, yes, the criterion I use is intuitive and impressionistic; perhaps it can be restated in a more formal way, but before discussing positional mergers, phoneme frequencies and allophone regroupings, I'd like to see some feedback on some simpler matters.

BTW, if you confess (for example) that you don't understand how phonemic palatalization is related to a rearrangement of contrasts in vowels, no one will sue you (or respond in any other way you may fear). Just ask.