Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!gatech!csulb.edu!hammer.uoregon.edu!arclight.uoregon.edu!worldnet.att.net!howland.erols.net!vixen.cso.uiuc.edu!uchinews!not-for-mail
From: deb5@midway.uchicago.edu (Daniel von Brighoff)
Subject: Creolisation of English [was: Re: Language relatedness a red 
X-Nntp-Posting-Host: ellis-nfs.uchicago.edu
Message-ID: <E4uB3F.GyD@midway.uchicago.edu>
Sender: news@midway.uchicago.edu (News Administrator)
Organization: The University of Chicago
References: <5cnthe$nff@omnifest.uwm.edu> <32eff989.16197851@news.exodus.net> <E4sxDK.1wF@midway.uchicago.edu> <32f01ab4.24688830@news.exodus.net>
Date: Thu, 30 Jan 1997 21:16:26 GMT
Lines: 228

In article <32f01ab4.24688830@news.exodus.net>,
Bill Vaughan <bill@osisoft.com> wrote:
>deb5@midway.uchicago.edu (Daniel von Brighoff) wrote:
>
>>Bill Vaughan <bill@osisoft.com> wrote (in response to 
>>mark@omnifest.uwm.edu (Mark Hopkins)):
>>
>>>English ... may have acquired some of its grammar from
>>>Celtic (the periphrastic tenses are uncannily close to those in Welsh,
>>>in particular the "progressive" tense, which has no counterpart in
>>>German nor in French). 
>>
>>But does in Castilian and Catalan, which clearly must have acquired it
>>from--well, um, Celtiberian?  I've never before encountred the theory that
>>the Celtic verbal system influenced the English one.  In fact, according
>>to my fragmentary recollection of Welsh linguistic history, the
>>"progressive" shows up first in English.  May I ask for references?
>
>Sorry, no references available. When I was studying Welsh a number of
>years ago, I said to myself "Oh, that's where it must have come from."
>The hypothesis sounded a little weird, so I did look for progressives
>in related Germanic languages and in old French, but I had no
>references on old Welsh available. Does the progressive really show up
>first in English? Is it not present in old Welsh? It would be a little
>odd (but not unheard of) for both languages to have developed a
>progressive tense independently. Perhaps some Celtic scholar could
>tell us whether Breton has a progressive tense?

Good questions, and when I have time to do some research on Old Welsh, I
may be able to answer them.  From my study of Middle Welsh, I recall an
overwhelming preference for the synthetic tenses, but to what extent this
reflects a literary affectation (influenced by study of Latin), I cannot
say.

>As for Catalan and Castilian, I have no idea where they acqured their
>progressive tenses from. Maybe Celtiberian, as you say. Maybe an
>independent development.

I was being facetious.  I see no reason to view the emergence of the
progressive in Iberia as anything but an independent development.

Que hi penses tu, Miguel?

>>>On top of this, there was surely a period of
>>>creolization with Norman French, followed by reintegration of the
>>>creole into English, evidenced by the loss of inflections and
>>>regularization of irregulars which is typical of creoles.
>
>>Such as in the English verbal system, where there are approximately 205
>>irregular verbs versus Modern German's...215.  Hmm. 
>
>You said "verbs," I didn't. There are plenty of other irregularities
>and inflections to lose. And I wasn't talking about Modern German, but
>Old English.

Well, if you want to compare Modern English to Old English, the our number
of irregular verbs has actually *increased*, due to analogical extensions
like "dive-dove-dived" (non-American and earlier American
"dive-dived-dived").  And the reason I pick on verbs is that they're one
of the major areas one would expect to see simplifications.  I'm not aware
of a *single* irregular verbs in any of the creoles of my acquaintance.

>>So what are some
>>"irregularities" that post-creole English has jettisoned that its non-
>>creolised relatives, like Dutch, have retained?
>
>Now that's a tougher question. Gender of the definite article, for
>instance. Retained by French, Dutch, 

Retained in Standard Dutch, not in all dialects.  One can make a case for
creolisation in the history of Afrikaans, but not in Hollands.  Pretty
slim evidence.

>and German. Welsh hasn't got it, FWIW. Probably other stuff, my
references are home and I'm at work.
>
>Certainly English is not "Dutch with a French and Latin vocabulary."
>English grammar differs substantially from Dutch. Influence of a
>creole is only an attempt to explain why.

An unnecessary attempt, IMHO.

>> And how many of these
>>changes (like the widespread loss of inflection) are clearly due to 
>>"creolisation" under the Norman French and not trends toward simplifi-
>>cation that began under the Danelaw.  (BTW, any idea how the percentage of
>>Nordic settlers in England in the ninth century compares to the percentage
>>of Normans in the 11th?)
>
>Well, maybe there was creolisation with Norse as well, though I doubt
>it. I'm under the impression that creoles don't arise when the
>languages involved are similar or mutually comprehensible, and I've
>read that Old English and Old Norse were mutually comprehensible to a
>great extent.  It's true that creoles don't arise everywhere two language
>groups abut, though _linguae francae_ often do. So perhaps there was no
>need for an English/Norman creole. 

Creoles arise under very specific circumstances.  (There's a reason why
they're virtually confined to areas of former European colonisation.)
Pidgins (and lingua francas or--if you prefer--linguas francas) arise
wherever two peoples speaking unrelated languages and without recourse to
formal language education have need to communicate.  They only take root
as creoles in situations where the community is so heterogenous as to have
no other common language--such as among populations of slaves taken from
linguistically diverse parts of Africa.

There was no need for a Norman-English creole.  The Norman upper crust
was very small and the vast majority of the population spoke mutually- 
intelligible dialects of Anglo-Saxon.  Now, maybe if you had had a
situation where Danes, Welsh, Saxons, Bretons, Gaels, and Normans were
tossed into close contact in roughly equal numbers and could only
communicate with each other by means of broken French, you might have had
a situation where a creole could have formed.  But I doubt it.

In short:  not every mixed language is a creole, or post-creole.

>OTOH, the upper crust did speak French
>as late as Chaucer's day, though by then it was evidently a learned
>language.  The Norman lords and their Saxon underlings, eg village
>notables, minor Saxon nobility, etc. must have conversed in something.
>Probably a pidgin of some sort.  ("If you speak Norman LOUDLY and
>DISTINCTLY they can understand it just fine." :-) )

Why?  Many Normans had English-speaking nannies or even mothers.  Why
wouldn't they learn English fluently?  Saxons in regular contact with the
upper crust must have been a very small percentage of the population.
And why wouldn't these have learned good French?  We're not talking about
occasional contacts, such as trading partners who see each other twice a
year, we're talking about domestic staffs and such--how long does it take
Mexican busboys in the United States to learn English, even if they live
in a Spanish-speaking community?--and gentry who could afford schooling.

>>I ask these questions because, so far, everyone I've talked to who claims
>>English went through a period of "creolisation" has been using the term in
>>a very different way than do the creolists I've read.  Certainly, the
>>changes English went through in the transition from Anglo-Saxon to Middle
>>English cannot compare to the complete restructuring that characterises
>>the formation of Jamaican creole or Tok Pisin.
>
>I'm not an expert on creoles. But even to me it is clear that creoles
>span a rich range, from Gullah>Ebonics which is mutually
>comprehensible with English, through Jamaican, to Hawaiian Pidgin and
>Tok Pisin which must be learned as a second language by English
>speakers. Furthermore, I'm not wedded to the notion that there was an
>English/Norman creole; it just seems to explain a few things. A
>hypothetical English/Norman creole would have had essentially an
>English syntax, though simplified, with a mostly Norman vocabulary. On
>being reabsorbed into English, some of Old English's irregularities,
>and much of its core vocabulary, would have come back.

Irregularities often get worn away over time without any special mechanism
needing to be invoked.  Creolisation involves catastrophic change, a
complete reworking of the verbal system, for instance.  Where is the
evidence for that in English?

>I don't actually think English per se went through a period of
>creolisation, ie that the line of descent is OE > creole > ME. Rather,
>I think the pattern of descent may be:
>
>Old English >-----+--> Middle English
>                 /
>Norman > creole /
>
>(leaving out any possible Brittano-Roman, Nordic, Celtic and Pictish
>(whatever that may have been)  influences).
>
>>>There is nothing wrong with talking about linguistic evolution. We
>>>know languages change -- we have seen it happen. Of course we must
>>>recognize that simple tree-shaped inheritance diagrams are at best
>>>misleading at best and at worst utterly invalid.
>>
>>The tree-model, the wave-model, and the lattice-model all have their
>>strong and weak points.  Before someone tells me to discard the
>>tree-model
>
>I'm not telling you to discard trees, I just believe trees can never
>show the whole picture. 

*None* of these can show the "whole picture".  They're models, after all.

>Single inheritance can't account for a lot of
>the mechanisms we already know about. Areal influence, for instance --
>how does one model the Balkan sprachbund with trees?  

A perfect case for employing the wave-model.

>(I must admit I
>don't know how areal influences work myself, but they do work. When I
>moved to Trulben (kreis Pirmasens) in 1966, I was struck by the
>dialect they spoke there. It diverged from the "Pfaelzer" dialect
>quite a lot, particularly in dropping final consonants and softening
>G's. "(Guten) Morgen" there was pronounced roughly "moZe."  Funny
>thing is, the G's were softened and the final consonants were dropped
>in just the same places that French does the same thing. Coincidence
>or areal influence? I don't know.)
>
>>I'd like to see a lattice model that can be easily reproduced
>>and understood in two dimensions.
>
>Why? What's so special about two dimensions? Who came down from heaven
>and said all graphs have to be planar? Fortunately there is a theorem
>that says all graphs can be done in no more than three dimensions, so
>I guess if we can't draw them we can sculpt them  :)

What's so special about two dimensions?  Hmm.  Just that every bit of
information I have on linguistics is reproduced in them (ignoring the
thickness of ink on a page or electrons on a screen.)

>Seriously, if a graph isn't planar it can still be drawn on the plane,
>with crossovers, otherwise electronic schematics wouldn't work. So I
>think lattice-models can be reproduced. Whether they are right is a
>different issue.

Have you seen how hairy these things get?  There's a lot one can do with
three-dimensional modelling, but not meany people have the technology
right now to really take advantage of it.  One day, Miguel and Richard
will be able to illustrate their excursus on PIE here with robust lattice
models that I can rotate a the drop of a syllable and my _dtv Atlas zur
deutschen Sprache_ will have unambiguous forty-colour modelling of
German-Romance areal features, but not now, not today.  You may consider
tree-models crude, but they contain a lot of useful information in a
format that is easy to reproduce and digest.


-- 
	 Daniel "Da" von Brighoff    /\          Dilettanten
	(deb5@midway.uchicago.edu)  /__\         erhebt Euch
				   /____\      gegen die Kunst!
