Change and What Remains

[Dr. John E. McLaughlin]

Larry Trask wrote:

> The English that I speak is, in a reasonable sense, a daughter of the
> English my parents spoke.  And the English the young people are speaking
> back home, which is already noticeably different from my own, can be
> regarded as a daughter of the English I speak.  In that sense, I
> suppose, a language can co-exist with its own daughter.  Is that what
> you mean?

> My 22-year-old niece has acquired some of the vowel changes collectively
> known as the Northern Cities Shift.  As a result, her vowel system is
> conspicuously different from mine, and I occasionally have trouble
> understanding her.  Her parents (my generation) have no trouble
> understanding her, because they're exposed to the youngsters' speech all
> the time, but they don't talk like her.  But I don't get home very
> often, and, when I do get home, I'm startled by the young people's
> English.

> All of the several generations can talk to one another, just as I could
> talk to my grandparents when they were alive.  But my grandparents
> didn't sound exactly like my parents, either.  And my Aunt Catherine,
> who's now pushing 100, sounds remarkably different from the youngsters,
> and even from me.

> As I said, ceaseless change.  The incomprehensible English of King Alfred
> is separated from us by scarcely more than forty generations.
> Chaucer's speech, separated from us by no more than 25 generations,
> would also be utterly incomprehensible to us.  Even Shakespeare, only
> about 16 generations ago, would probably be largely incomprehensible to
> us if we could hear him speak -- and some specialists believe we would
> not be able to understand him at all.  What will our descendants, 20 or
> 25 generations from now, make of the sound recordings we'll be leaving
> them?

> So, apart from the trivial case of me and my niece, how can a language
> co-exist with its own descendant?

Normally, I agree completely with you, Larry, but this time I've got to
disagree, if only in pointing out that your definition of "language" in this
discussion is not standard usage.  In the sense that you've described,
namely, that any change makes a different language, then you and your niece
are speaking different languages, you and your Aunt Catherine are speaking
different languages, indeed, you and I are speaking different languages, as
I imagine you and your wife are both bilingual in each other's non-English
and different tongues.  According to this definition, language is strictly a
one-generation, one-location event, to be changed to the language of the
next generation in the same location or to be changed whenever one crosses
the county line.

In one sense, you are indeed correct--when using the definition that a
language comprises a particular set of generative, transformational, and
phonological rules on a nearly identical lexicon with an identical phonemic
and phonetic inventory.  One generation slightly fronts a vowel over the
previous generation.  The next generation might raise that vowel a touch.
The following generation might switch the alveolar fricative in one word to
a postalveolar affricate to make it rhyme with another word.  In this sense,
using the constant change definition of language, you would have to admit
(as you did) that several parents, grandparents, and great-grandparents can
coexist with multiple daughters, granddaughters, and great-granddaughters.

But if you use the standard definition of "language", that is, two speech
varieties which do not exceed X% of mutual comprehensibility, then your
familial examples don't work anymore.  You must look at rates of change in
different communities and particular linguistic features that develop in one
community that drastically affect mutual intelligibility versus other
changes that do not have as radical an effect.  In this definition of
language, a community that has experienced a slower rate of change, where
the changes are such that mutual comprehensibility is not radically
affected, can be compared to a sister community where the rate of change has
been faster and has included changes that tend to always affect mutual
comprehensibility.  In this case, community A can be said to still speak the
mother language and community B is speaking a daughter language.  Actually,
this scenario with a standard definition of language is simply the same
situation as described using your tighter definition of language--only at a
longer time scale.

Let me reiterate and give more details about Shoshoni and Comanche.  The
phonological, morphological, and syntactic rules of Eastern Shoshoni are
very similar to the rules that must be reconstructed for Common Shoshoni.
The differences between modern Eastern Shoshoni and modern Western Shoshoni
(now completely separated by the states of Utah and Idaho where other
dialects are spoken) are minimal and mostly optional variants.  There is
only a small percentage of different vocabulary (at least 90%
lexicostatistical similarity between the most extreme examples from western
Nevada and central Wyoming).  Speakers from Fallon, Nevada have reported
virtually no problems in communicating with speakers from Wind River,
Wyoming.  About 1720 or so, some groups of Eastern Shoshoni speakers
separated from the main group and headed south into the Texas Panhandle.  A
treaty with the Spanish in 1786 lists the names of the treaty signatories
with Spanish translations.  The list is easily analyzed into Common Shoshoni
(predating certain vocabulary that has changed in both modern Eastern
Shoshoni and Modern Comanche).  Some examples from 1786, 1828, and 1865 will
illustrate some features of the languages (Cm = Comanche; Sh = Shoshoni; in
Modern forms: y = barred i, j = y; NR = not recorded):

                    1786               1828               1865
                    (Spanish source)   (French source)    (Spanish source)
'crane, heron' (Cm) ...condata         NR                 different word
Modern Cm--different word
Modern Sh--koonty
'shoe, foot' (Cm)   ...nampat          naap               napé
Modern Cm--napy
Modern Sh--nampe
'sit on mtn' (Cm)   toyamancare        NR                 NR
Modern Cm--tojamakary
Modern Sh--tojamankary
'red' (Cm)          enca...            eica..., eca...    eca...
Modern Cm--eka
Modern Sh--enka

Notice how the loss of preconsonantal /n/ was completed in Old Comanche
between 1786 and 1828.  Yet this one change was the major breaking point
between mutual intelligibility with Shoshoni and mutual unintelligibility.
Before that, the tribes of west Texas spoke a dialect of Shoshoni (as the
near identity of 1786 forms and Modern Shoshoni forms attests).  After 1828,
the bands spoke Comanche.  Their self-identification didn't change (Shoshoni
nymy, Comanche nymy), but their speech did.  All other changes between
Shoshoni and Comanche are found elsewhere as dialectal variants in Shoshoni,
but losing the preconsonantal /n/ on top of several minor changes that had
already occurred was too much phonetic (and according to my preferred
analysis, phonemic) difference to make up for.

This seems to me to be a clear case of a parent co-existing with a child,
using a standard definition of "language".  Even when using a more
restrictive definition of "language", this shows how quickly a single
radical change can affect an entire community, breaking off communicative
ability with the parent community in about 40 years.

John E. McLaughlin, Ph.D.
Assistant Professor
mclasutt at brigham.net

Program Director
Utah State University On-Line Linguistics
http://english.usu.edu/lingnet

English Department
3200 Old Main Hill
Utah State University
Logan, UT  84322-3200

(435) 797-2738 (voice)
(435) 797-3797 (fax)