Course materials for/by
Peter L. Patrick. May contain copyright material used for educational
purposes. Please respect copyright.
Notes for LG102, Intro
Sociolinguistics
Week 19, Winter term
Language Origins
How did language, or
languages, begin? We will not answer this question, but will try to
§ identify some powerful
myths concerning language in stories about its origin,
(Myths are often false, but not necessarily so – they are defined not
by being false, but by being powerful
stories within the culture which holds them.)
§ make analogies to
§ and roughly narrow down the possible
age of human language as we know it today.
Myths of Origin
§ Biblical
account in Genesis:
§ God created the first words,
and first spoke to humans.
§ God gave Adam the power to name the
animals.
§ There
was a universal
language for a long time.
§ Humans tried to become god-like
in power, building the Tower of Babel.
§ To
prevent this, and to humble humans, God created linguistic diversity and
unintelligibility.
o This
story teaches several powerful (sometimes harmful) language myths:
§
§
§ There
was a Golden Age of
linguistic perfection.
§
§
Linguistic
diversity is harmful to unity and progress.
§ Herodotus
(Greek historian, c485-420 BC) tells another story of language origins:
§ An
Egyptian king used language to discover who were the oldest people
§ He
isolated two children from birth, with no access to speech.
§ The
first word they spoke was Phrygian bekos,
meaning “bread”.
§ So
the Phrygians (from what is today central
§ According
to Herodotus, Phrygians were the ancestors of the Armenians. (Herodotus was not
always right!)
o Phrygian
was attested c. 800 BC, extinct since c. 6th century BC.
§ Other
words include brater
‘brother’ < PIE *bhrater
and mater
‘mother’ < PIE *mater,
§ from
which you can see it was an Indo-European
language, quite close to Proto Indo-European, and
§ possibly
spoken in the area in which Armenian is historically spoken.
§ Phrygian
was at any rate a close relative of Armenian, and perhaps also of Greek.
§ Most
of the inscriptions remain untranslated today.
o This
story too teaches several powerful language myths:
§ All
children speak a universal
language at birth.
§
§
§ Layman’s approach: a language is a big bag of words, each of which refers to some thing.
§ Modern
linguistics: The building blocks of language are not words, but more abstract
elements.
§ These
abstract elements do not correspond in a simple way to objects/events/processes
in the world,
§ but
rather have an arbitrary and symbolic relation to each other and to the
world.
o Linguists’
notion of a universal grammar is more abstract than the king’s
‘universal language’:
§ Nobody
can speak UG directly!
o Many
linguists also believe that language sets humans apart from animals (why? Go to slide #2),
but
⇨ What precisely do we mean by language?
⇨
When did we begin to have language?
⇨
How did it develop?
Relevance of language origins
for Sociolinguistics – a Darwinian view
We’ve
seen the importance of linguistic variety (different kinds of languages,
speakers & sociolinguistic settings), and linguistic variation (different
ways of communicating “the same thing” within a language) already this term.
Formal
linguistics, esp. syntax, depends heavily on a notion of Universal Grammar – a
unity (or “fundamental agreement in structure”,
Ø “On my theory, unity of type is explained by unity of descent”,
Ø so UG is to be explained by evolution of languages from a common
ancestor.
On
the other hand, biological variation is largely accounted for by error in descent (genetic mutation). Small
differences between parent and child lead to new forms.
Ø This
is also true of language variation, even when most aspects of language are passed
on efficiently (the normal situation of learning one’s mother tongue).
Ø So,
some cases of linguistic unity of type are accounted for by unity of descent –
Ø via
“normal genetic transmission” (with the linguistic sense of ‘genetic’,
not the biological one).
However,
the other great law according to
o
Put another way, social context (i.e. the “conditions of
existence”) is more important than shared
ancestry (i.e. “unity of descent”) to the evolution of language, and
ultimately explains the nature of UG.
o
(more about Darwin’s theories)
But
unity of descent is even less important for language than for organic beings,
because innovations in language don’t face the life or death test of natural selection – you can almost always speak in
a new way without risking death for it.
Ø Because
of this, we can tolerate variety of language more easily, and other factors –
social context, language contact – should have even more influence on language
change.
It’s
easy to go too far in borrowing biological metaphors for language – as many
linguists do. I agree with Labov (2001) that “language does not show an evolutionary pattern in the
sense of progressive adaptation to communicative needs”.
This is
because – although there is
variability, descent, and unity of type, as noted above - there is “no evidence for natural selection” in language.
Ø There
is no process which weeds out the ‘worst’ linguistic forms and keeps only the
‘fittest’ –
Ø If
there was, people would not feel the need to constantly complain about how
‘awful’ the language of ‘young people today’ is!
Other
sociolinguists (e.g. Chambers 2003) argue against the idea that linguistic
diversity serves biological
functions.
Ø It
certainly serves social functions,
but as there are many ways to do this without facing the life-or-death test of
natural selection, the analogy to evolution fails here.
Other
cases of linguistic unity of type – similarities arising among languages which
do not share descent (=are not
genetically related) - are studied under the heading of linguistic typology.
Ø Such
cases are very relevant to understanding pidgin and creole
languages (final week of class).
Finally,
the social contexts in which language operates today are derived from the social
contexts in which humans emerged from our earlier stages of evolution.
Ø Either
way, language origins seems to me an important topic for sociolinguistics –
Ø too
important to leave to the evolutionary biologists, anthropologists, zoologists,
or formal linguists!
How is human language
different from animal communication?
Animals
have ways of transmitting information -- communication systems, some of them
conscious.
These
often involve multiple channels, as with humans: visual, auditory, tactile, and
olfactory.
Linguists’ definitions of language often narrowly focus on speech and hearing, sound systems –
o Too narrowly, considering what we have learned about signed language in the last few decades.
One
definition of language:
1) a
set of discrete vocal sounds,
2) each
of which is arbitrary (i.e. meaningless by itself),
3) which
are productive (i.e. can combine in many ways to produce larger units),
4) each
of which may have conventional (=agreed-upon) meanings,
5) and
which allows its users to communicate about events that are distant in time
or space.
⇨ Substitute “gestures” for “vocal sounds” in
(1), and it’s adequate for signed languages as well.
o I.e., “2nd-order intentionality” or
reflective meaning: we can monitor our monitorings.
Symbols are arbitrary, conventional &
productive, making language a productive symbolic system of context-free
reference. We know that signs stand for meanings, the signs catch our attention
- like a child first realising that letters can spell
words to mean things.
Primates (e.g. apes,
humans) have been around for 60-70 million years.
Ø For
the first 59-69 million, there was certainly no such “language” as this!
Ø We can explore
what kinds of communication animals have, and
Ø In what ways it
differs from or fall short of language,
Ø Including close
relations like chimpanzees and bonobos with their
skills for learning symbolic systems, but
·
None of them quite have language.
But when we talk about language
evolution, we have to remember that
Ø it is not a biological
organism itself, and did not follow the processes and constraints of evolution
that organisms have followed.
Ø Hence
biological terms applied to language are metaphorical:
o
they can be pushed
too far, and will fail or mislead us when they are.
To talk about language evolution is
to talk about the evolution of humans, then, and it is still largely mysterious
and unsolved.
Why should socio-linguists be interested
in language origins?
I propose that to
understand language evolution, we need sociolinguistics:
Ø
o
It seems likely that early language in prehistory
played a role in the development of sociability and society among humans
o
But, is this a chicken or egg role? Did language
facilitate development of society, or vice versa?
Ø
o
where everyone knew all other members of the group,
and
o
social relationships
were fundamental to survival.
Ø It seems
necessary then that the study of language evolution must be informed by
sociolinguistics as well.
And for
a comprehensive understanding of sociolinguistics, we need to know about the
evolution of language:
Ø The variety and diversity of languages, and the
complexity of their uses and functions, are
the main facts which sociolinguistics tries to understand;
Ø These are
obviously the result of evolution from a smaller number of source languages,
and language functions, elaborated over time.
Ø
Ø To study this
elaboration and refinement, at least – and perhaps also the origin of the
earliest languages/functions – we need to know how the processes of evolution
and natural selection applied to our species.
Ø [This has long
been the view of anthropology, which in
o archaeology,
o culture,
o physical
anthropology (incl. evolution),
o and linguistics.]
Natural Selection and
Explanations of
Ø Without concepts
and intentions to express, there would be no language; and
Ø without articulators
(the mouth, the hands/face), there’d be none, either.
Thus there is
Ø interaction
between language and thought, and
Ø interaction between
language and utterance.
We can all observe that infants
growing into children improve their individual interaction along these lines.
These kinds of interaction can also evolve and be improved
over generations by natural selection (Darwin’s 1859 idea):
Ø Favourable traits which
are heritable (=can be passed on) spread throughout a population.
Ø Individuals
that have favourable traits are more likely to
survive, reproduce and pass them on.
It’s possible too (but
controversial) that natural selection may favour the group or species, rather than individual
–
Ø
Other aspects of the origins of
language are beyond the reach of evolution:
Ø broader
universal laws of physics, of mathematics (e.g. of logic), etc., that constrain
what language can be like;
Ø and the genetic
raw material that happened to be available in early (pre-) humans and
populations.
The latter include what we sometimes
call “path-dependent” properties:
Ø Had things
played out differently, they would have been different –
Ø We might not have
language at all, or perhaps not have become the humans we are –
Ø They are a
matter of (pre-)historical accident.
So, some of the explanation of why
language is the way it is is bound to be
unsatisfactory: it just happened to be that way.
Ø Path-dependent
evolutionary processes can lead to properties that are arbitrary or even mal-adaptive, i.e. do not help an organism
to survive.
o
Physical ex. of the ‘wind-pipe’ and ‘food-pipe’ in
vertebrates
o
Linguistic ex.: fact that you can say pretty much the
same thing in a sentence or a noun-phrase – why have both? (Carstairs-McCarthy
1999, Hurford 2007)
Ø
“We may be on the brink of seeing some convergence in our
understanding of issues of genetic diversity, cultural diversity, and
linguistic diversity. It may be possible,
then, to work toward a unified reconstruction of the history of human
populations. It is much needed, because
certainly we do not have such a unified history at the moment.”
(Colin Renfrew, in Bryan
Sykes ed. The Human Inheritance: Genes,
Ø
“[A]rchaeology, genetics, and
linguistics... are now generating many new data and insights. All of them can
be expected to converge toward a common story, and behind them must lie a single history.
Singly, each approach has many lacunae, but hopefully their synthesis
can help to fill the gaps.”
(Luca L. Cavalli-Sforza Genes, Peoples and
We do have reasonable match-up between what linguistic evidence
tells us, and what genetic evidence shows –
Ø
see e.g. this chart from the last source (language
families on the right,
but that’s not even a full description, much less a set of explanations.
A Calendar of Human
Ø 3-4 billion years of life on earth
Ø 60-70 million years ago:
o Primates emerge (includes great apes, lesser apes, hominids).
Ø 4-5 million years ago:
o Hominids (ancestors of humans) branch off from other primates.
o Australopithecus: walked upright, pint-sized brain. “Lucy”.
Ø 2-3 million years ago:
o Homo habilis: brains half-size of modern humans, adapted for language.
o Group activities, cultural patterns, made crude tools.
Ø 2 million years ago:
o Homo erectus: brain
expanded rapidly (1000cc, quart-sized). “Blending”.
o Discovered fire, hunting large game, sophisticated tools. Prelanguage?
Ø 300,000 years ago:
o Homo sapiens
(Neanderthals): brain expanded rapidly, larynx descended.
o Rituals/afterlife, finished tools. Pronounce i,
u, a. Duality. “Eve”.
Ø 40,000 years ago:
o Homo sapiens sapiens (Cro-Magnon): modern brainsize, world
diffusion.
o Advanced art, culture, full language. Modern adult vocal tract.
Ø 5,000 years ago:
o
Pictographic writing: Sumerians (
Ø 3-4,000 years ago:
o Proto-Indo-European (possibly older?). Complexity > many descendants (tree).
o 3 genders (masculine/feminine/neuter). 3 numbers
(singular/dual/plural). Many cases.
o Chinese writing system: pictograms/ideograms/phonograms. Still in use (ex).
Sources:
Andrew
Carstairs-McCarthy. 1999. The origins of complex language: An inquiry into the
evolutionary origins of sentences, syllables and truth.
Andrew Carstairs-McCarthy. 2007.
Jim Hurford.
2007. The origin of noun phrases: Reference,
truth, and communication. Lingua 117(3): 527-42.
Zdenek Salzmann.
1998.
LG102 Intro
Sociolinguistics page
Return to Peter
L. Patrick's home page