Our language can be seen as an ancient city: a maze
of little streets and squares, of old and new houses, and of houses with
additions from various periods; and this surrounded by a multitude of new
boroughs with straight regular streets and uniform houses. (Ludwig
Wittgenstein, 1958: 18e)
On
Cultural evolution and its linguistic repercussions
Written
language
Concluding
remarks
By “cultural evolution of language” I mean the
changes in language that have accompanied and resulted from cultural, as opposed
to biological, evolution. I’m referring to the evolution that has resulted in
languages that are different (significantly so, I believe) from the earliest
human language--an evolution that has occurred since the biological evolution
that produced the “language faculty” was completed.
The last Note (Grace
2002) ended with a discussion of two different conceptions of what a language
is. In one conception, a language is something that is spontaneously acquired by
any normal human child even in the absence of any guidance from adults and even
when the available evidence is "very restricted and degenerate”. We
referred to language according to this conception as “spontaneously acquired
(SA) language”. Understanding the biological basis for this acquisition (which
only humans are capable of) would, of course, be of considerable interest to
cognitive science.
According to what we’re calling the
“Post-Chomskyan paradigm”, the key to the whole acquisition is the
construction of a grammar for the language in question. The biological basis for
the acquisition--referred to as the “language acquisition device” (LAD)--has
as its main component a universal grammar (UG) which guides the child in
constructing the grammar.
In the second conception, a language is seen as an
instrument whose prototypical function is to encode propositions. We referred to
language according to this conception as “autonomous text (AT) language”.
This conception of language is the product of a long prescriptive tradition in
Western civilization. But this prescriptive tradition would have been
unthinkable without the enormous cultural evolution that had occurred since the
time of earliest language--an evolution that introduced very important
innovations in language use.
However, linguists have tended to ignore the effects
of this cultural evolution. Ignoring it offers them the convenience of
continuing to assume that all of the fundamental syntactic properties of natural
languages have a biological rather than a cultural explanation, while at the
same time freely employing easily accessible languages such as modern standard
English as examples of “natural languages”.
In this Note, I want to point out two major cultural
innovations that have been very significant for language: first, written
language, and subsequently, what I’ll call the “Modern Standard Language”
(MSL). MSLs are characterized by
what has been described as a kind of intellectualization--a kind of
intellectualized superstructure. They originated in
On
Cultural evolution and its linguistic repercussions
One linguist-anthropologist who has called attention
to the significance of past cultural evolution for language is Paul Kay. In an
article entitled “Language evolution and speech style” (Kay
1977), he describes an evolution of languages from less to more autonomous
systems (which is to say, toward what we’ve been calling the “AT
language”). He singles out increasing specialization and division of labor as
the sociocultural changes most responsible for this evolution. He describes
autonomous speech as what is “called for when an addressor needs to
communicate to an addressee with whom he shares a minimum of common experience
precise information on an unfamiliar topic of an affectively neutral kind”
(1977: 28). [His example (1977: 22) of “a prototypic situation for the use of
autonomous speech” is “when one person had to explain to another how to
disarm a complex bomb by one-way radio”].
The mechanism driving this evolution of languages
would have been the more frequent use (as circumstances increasingly called for
it) of a relatively autonomous style of speaking, accompanied by the continual
selection of structural means that facilitated more autonomous speech over those
that did not.
Surely, the single most revolutionary step in the
direction of autonomy must have been the development of written language.
This required a mutual adaptation between writing and spoken language
(since the earliest writing systems had nothing to do with speech, while
earliest language even more clearly had nothing to do with writing).
Much(1) has
written about how people’s conception of language must have been affected by
seeing it written. In particular, it’s pointed out that it was writing that
made possible the objectification of the artifacts of speech by converting them
into texts and thus giving them a permanent form in which they could be
analyzed, discussed, evaluated, and revised.
Walter Ong (1982)
has proposed the designation “primary orality” for the orality of cultures
in which there is not only no literacy, but no knowledge whatsoever of even the
possibility of language being written. Of course, primary orality is the
condition in which the human species has lived for most of its history. More to
the point, it’s the condition in which all
humans lived at the time in which the evolution of the language faculty
was completed.(2)
It seems evident that a primary-orality-like
environment would be required for genuine SA language acquisition to take place.
Although even a culture in the state of primary orality could not be expected to
be completely free of ideology that would influence the language acquisition
process or its results, the influences would surely be very much less elaborated
and distorting than those in literate communities. Being able to study the
acquisition, knowledge, and use of language in such circumstances would be an
enormous opportunity for anyone truly interested in the innate human capacity
for language.
Unfortunately however, it seems highly unlikely that
any culture in this state exists anywhere in the world today. Illiterate
individuals in the contemporary world might come closer than the cultural
“mainstream” (as it’s sometimes been labeled) to representing the
properties of SA language, but contemporary illiteracy should never be thought
of as an adequate substitute for primarily orality, since evidence of written
language is so pervasive in the contemporary environment.
In any case, it does seem that the speakers of
earliest human language--if they were available--would be the ones to look to
for the grammatical properties of what we’ve been calling “SA language”.
They would presumably reflect the effects of the biological evolution that
produced the capacity for language (that is, the so-called “language
acquisition device”) with minimal distortion from subsequent cultural
evolution. Accordingly, to investigate the properties of SA language, we need
either to find language acquired in a cultural environment that approximates
theirs in as many respects as possible or to find some other way to isolate and
subtract the effects of cultural environments on the acquisition process.
But how do we isolate these effects? In fact, the
single language that has most often served as a source of data in Post-Chomskyan
linguistics has been English, and, among what count for the Post-Chomskyan
paradigm as “natural languages”, English is one of those that most nearly
approximate the ideal AT language. It is a prime example of what I'm calling the
"Modern Standard Language", and these MSLs are precisely the languages
that come closest to conforming to the Post-Chomskyan paradigm’s
underlying picture of language(3). The question, then, is how
much and in what ways these appear to differ from earliest language.
It seems, then, that the closest existing
approximation to the AT-language ideal is represented by
what John Earl Joseph (1987) simply refers to as “standard
languages”. Standard English and other standard languages of
For while 'language standards' exist in every
linguistic community, 'standard languages' represent a specifically European
concept, whose defining criteria are based on the attributes of European
languages and on European cultural values. (1987: x).
To avoid confusion, I’m referring to languages of
this kind as “Modern Standard Languages” (MSLs), rather than Joseph’s
simple “standard languages”.
A lot has been written in recent years about language
standardization and standard languages. The label “standard” applied to a
language might seem to imply that its most important characteristic is
invariance. The focus of attention might seem to be on the elimination of
variant forms--whether by the imposition of regulations or by mutual agreement.
However, it has now become clear that the MSL involves much more than
invariance.
It seems to have been developed primarily for
commercial and political ends. However, it has also played a very critical role
in the development of science, especially since the 17th century. (The
subsequent interest by philosophers and logicians in the design of scientific
language seems to have had a particularly important influence in the current
assumptions of linguistics).
Of course, the skills employed in speaking an MSL
don’t account for all of a normal speaker’s knowledge of language. These MSL
skills, skills that come into play for composing and decoding text, are
essentially superimposed on our basic linguistic competence. However, the
standard descriptions of language structures focus particularly on the
structural features that require the MSL skills. This fact tends to make the
other skills appear to be of secondary importance.
Although the MSL developed from European culture (and
its derivatives), its spread to much of the rest of the world has been an
important part of global “development”. It seems necessary if a state is to
belong to the “developed world” that it have a national language that has
been redesigned to function as an MSL. In the immediate post World War II years,
this redesigning was often called “language development” or “language
modernization”. Some
An article by Charles Ferguson,
entitled “Language development” (
The modernization of a language may be thought of as
the process of its becoming the equal of other developed languages as a medium
of communication; it is in a sense the process of joining the world community of
increasingly intertranslatable languages recognized as appropriate vehicles of
modern forms of discourse. (1968:32)
He
goes on:
The process of modernization thus has two aspects:
(a) the expansion of the lexicon of
the language by new words and expressions and (b) the development of new styles and forms of discourse. (ibid)
[italics original].
It’s a mistake to dismiss the changes involved as superficial. At the very least, that’s not how they’re experienced by the speakers of the languages themselves. Japanese students have told me that they find Japanese texts written before the Meiji Restoration virtually impossible to read, and I know speakers of other Asian languages see their own languages as having at least undergone very important changes. However, the only mention of this loss of intelligibility that I’m able to recall having seen in writing (although I’m sure there must be many more) is the following quotation from Charles Gallagher (1969: 65) about Turkish:
The thoroughgoing nature of the Turkish reforms is
shown by the fact that, even putting aside the romanization of the script,
Ottoman Turkish texts of around 1900 are essentially unrecognizable foreign
documents to the modern Turk.
How
are Modern Standard Languages different from earlier language?
One
difference that Kay suggests (on the basis, particularly, of work by Brent
Berlin) is an expansion of the nominal lexicon by the addition of both more
abstract and more concrete nouns.
World and classical languages thus provide richer
resources for communicative subtlety than do local languages in that the former
frequently offer a larger variety of names for a given thing, depending on the
level of abstraction at which the speaker wishes to place his description and
the features of the denotatum or denotata to which he wishes to draw attention.
The larger implication of this finding, should it be corroborated generally, is
that with respect to their nominal systems, world languages provide the means
for more precise and explicit (i.e., more autonomous) communication at whatever
level of abstraction is desired by the addressor. (Kay 1977:
24)
However,
linguists today are still not accustomed to thinking of the lexicon of a
language being systematized in any significant way. Languages have been
represented as having a grammar as their core with the accompanying dictionary
being a list of “irregularities”. For
the post-Chomskyan paradigm in particular the essence of a language resides in a
system of syntactic rules. Changes
in vocabulary are generally regarded as being unsystematic and superficial. Are
there, then, any syntactic differences that distinguish MSLs?
What
Kay, following Ken Hale, suggests is that some languages cannot embed relative
clauses (thus limiting the number of NPs in a sentence that can be modified by a
relative to one). However, Kay (and Hale) are not alone in arguing that the
elaborate syntactic structures of MSLs are exceptional.
According
to Geoffrey Sampson (1997: 74):
It was a cliché of late nineteenth-century
linguistics that the early stages of languages with a long recorded history
showed a development from what was called parataxis to hypotaxis:
from a state in which simple expressions followed one another like beads
on a string, to a state in which logical relationships were made explicit via
subordination of simple expressions into grammatically complex structures. Karl
Brugmann, for instance, took it as indisputable that “Originally people spoke
in sentences having the form of main clauses”.
It
is interesting to note here that Andrew Pawley and Frances Syder propose that
even today’s speakers cannot on the fly and from scratch design sentences
longer than this. In their words (Pawley and Syder 2000:
163),
In a
single planning act (or focus of consciousness) it is not possible for a speaker
to encode novel lexical combinations across independent clause boundaries.
To
quote Sampson further:
This was not a matter of a priori speculation; the
claim was that Homeric Greek and Biblical Hebrew are tangible examples of nearly
purely paratactic languages, regularly stringing together elements whose logical
relationships would be reflected by subordination in any modern European
language. (ibid.)
Walter
J. Ong sees the contrast as being primarily between an oral style (which he
characterizes as “additive”) and a written one (which he calls “subordinative”).
He cites two translations of the first verses of Genesis as illustrative
of the contrast. He writes (1982: 37):
The Douay
version (1610), produced in a culture with a massive oral residue, keeps close
in many ways to the additive Hebrew original
He
contrasts the
The
Some
years ago I identified three syntax-related innovations that seemed to recur
repeatedly in cases of “language modernization”. They surely aren’t the
only such innovations, and probably aren’t even the most prominent. But for
whatever they may be worth, I’ll list them here. They are:
(1)
A considerably more elaborate system of conjunctions. Of course, this isn’t
surprising--it would naturally be expected as part of the mechanism needed for
hypotactic constructions.
(2)
More extensive use of nominalizations. This might come about either through the
introduction of new nominalization devices or extended use of existing ones or
both. (This is reminiscent of the above-quoted suggestion of Paul Kay and Brent
Berlin that expansion of nominal systems was one significant feature of language
modernization.)
David
Olson particularly singles out the role of nominalization, of which he says
(Olson 1994: 118):
It is the device by means of which
"applaud" becomes "applause" and "infer" becomes
"inference." Written texts are peculiar in the way they turn such
actions into entities. Moreover, the device permits the formation of complex
technical discourse.
(3)
More extensive use of passives. This might come about either through the
introduction of new passivization devices or extended use of existing ones or
both. This has been much talked about in the “modernization” of Asian
languages. (It has been dealt with in doctoral dissertations by two of my
students--see Prasithrathsint 1985 and Earns
1993).
I
gave the following quotation from Ludwig Wittgenstein at the head of this piece:
Our language can be seen as an ancient city: a maze
of little streets and squares, of old and new houses, and of houses with
additions from various periods; and this surrounded by a multitude of new
boroughs with straight regular streets and uniform houses.
(Wittgenstein 1958:18e)
This
metaphor is very apt. And it is precisely in such new boroughs, built out from
the base provided by our inborn language faculty, that most of the elaborate
hypotactic syntax of our Modern Standard Languages is to be found.
Something
that Noam Chomsky said offers an interesting example. He says (Chomsky
2000: 146) of his four year old granddaughter:
Does she speak English? What we say in ordinary
discourse is that she has a partial knowledge of the language that she will
ultimately attain if events follow the expected course, though what she now
speaks is not a language at all. But if all adults were to die, and children her
age were miraculously to survive, what they speak would be perfectly normal
human languages, ones not found today.
What I understand him to be saying is that the only reason that what she already speaks would not be counted as any language at all is that we already have our idea of what a representative speaker of English (and each other recognized language) should speak like. We have an idea, for example, of what kind of speaker would make a good informant to provide data for a linguistic description. And in the case of English (or any other MSL), most of us would no doubt be reluctant to accept any speaker as representing the language adequately unless s/he had a good command of its hypotactic superstructure.
In
short, I understand him to be saying that on strictly linguistic criteria, what
she speaks would qualify as a language. Her LAD has done its job; she’s
already acquired what we’ve been calling SA language.
There
is a story that some of the gentle-folk of Victorian England--when informed of
the process of evolution that Darwin had revealed--concluded that the production
of precisely their own kind must have been the intended object of any such
process. To us in our enlightened time, of course, that sounds comically
ethnocentric (or perhaps more accurately, “culturocentric” to use John E.
Joseph’s [1987] apt term). But how different from that
would be the idea that the evolution of the capacity for language had people
capable of producing and understanding autonomous text as its intended object? Still, isn’t that what we’re effectively
assuming when we assume that any explanation of how the language acquisition
process works must account for AT language competence?
Surely
the day will come when linguists will be asked to explain how we have come to
assume that the linguistic skills of an educated elite are the natural product
of the biological evolution of the human species.
NOTES
1.
I think, for example, of a number of works by Walter J. Ong
(e.g., 1978, 1982), by Jack Goody
(e.g., 1977), by David R. Olson (e.g., 1989,
1994), by Roy Harris (e.g., 1980,
1983, 1996, 2000), but
these could hardly be claimed to constitute even a representative sample.
(Back up)
2.
I’m assuming that the biological evolution of the capacity for language had
been completed in all essentials by the time that earliest human language
emerged. If we were not to make this assumption, if we were to assume that the
biological basis of language might have evolved significantly since earliest
language, we would raise the possibility--in fact the probability--that speakers
of different languages might have evolved differently. This would imply the
likely existence of at least some racial differences in language capacity.
However, all of the indications so far have been that, if there are any such
differences at all, they are of no significance whatever.
(Back up)
3.
I’ve attempted to describe the essentials of this underlying picture in Grace 2002.
The most basic points were that (1) the prototypical function of language is to
communicate factual information, and (2) that the prototypical procedure for
this communication of information involves the encoding and decoding of
propositions.
Much
of the collateral damage that I’m attributing to linguistics will continue as
long as these assumptions continue to be accepted as self-evident.
(Back up)
REFERENCES
Chomsky,
Noam. 2000. New Horizons in the study of language and mind.
Earns,
Fumiko Fukuta. 1993. Language adaptation: European language influence on
Japanese syntax.
Gallagher,
Charles F. 1969. Language rationalization and scientific progress. In The Social
reality of Scientific Myth, edited by Kalman H. Silvert.
Goody,
Jack.. 1977. The domestication of
the savage mind.
Grace,
George W. 2002. Collateral damage from linguistics? 1. The post-Chomskyan
paradigm and its underlying assumptions. Ethnolinguistic
Notes, Series 4, Number 21. Internet World Web page at
http://www2.hawaii.edu/~grace/elniv21.html.
(Back up)
Havránek,
Bohuslav. 1964 [orig. 1932]. The functional differentiation of the standard
language. In Paul L. Garvin (ed. & translator), A Prague School Reader on
Esthetics, Literary Structure, and Style.
Harris,
Harris,
Harris,
Harris,
Joseph,
John Earl. 1987. Eloquence and power: The rise of language standards and
standard languages.
Kay,
Paul. 1977. Language evolution and speech style. In Ben G. Blount and Mary
Sanches (eds.). Sociocultural dimensions of language change. Academic Press, pp.
21-33. (Back up)
Olson
1989. Literate thought. In
Olson
1994. The world on paper: The conceptual and cognitive implications of writing
and reading.
Ong,
Walter J. 1978. Literacy and orality in our times. ADE [sc. Association of Departments of English] Bulletin, no. 58, pp.
1-7. (Back up)
Ong,
Walter J. 1982. Orality and literacy: The technologizing of the word.
Pawley,
Andrew, and Frances Hodgetts Syder. 2000. The one-clause-at-a-time hypothesis.
In Heidi Riggenbach (ed). Perspectives on fluency.
Prasithrathsint,
Amara. 1985. Change in the passive constructions in written Thai during the
Sampson,
Geoffrey. 1997. Educating Eve: The 'language instinct' debate.
Wittgenstein,
Ludwig. 1958. Philosophical investigations. Translated by G. E. M. Anscombe. 2nd
Ed. Oxford: Basil Blackwell. (Back
up)
| Home Page | The Ethnolinguistic Notes | The Ethnolinguistic Notes, Series 1 and 2 | Ethnolinguistic Notes, Series 3 | The Ethnolinguistic Notes, Series 4 | Reflections: Language Evolution |
| Reflections: Knowledge of Language | Personal Page | The Human Predicament | Why Write Unpublishable Things? | Modest Proposals | Odds and Ends | Pictures |
