-
Speaking bodies I: How language happens: the recent science of language
Is non-language necessary to language? Could there be 'meaning'
without a physical world? How can language have evolved? What are the evolutionary
precursors of language in animals? What changes in primate brains have allowed
language to develop? What happens in bodies when we use language? Is language
somehow an encapsulated function, or does it draw on the whole of the brain,
the whole of the body? We know language is social, but how should we understand
the embodiment of social interaction among speakers? This first session
of the Speaking bodies minicourse will set out an integrated way
of understanding language as part of human bodies, which in turn are parts
of the physical earth.
- Speaking bodies general introduction
- 1. An emphasis, orientation,
framework
- 2. What is at stake?
- 3. Basic principle: language as structural influence
- Speaking bodies I.
How language happens: the recent science of language
- 1. Bodies able to speak: evolution of the cortex
- 2. History of the evolution of language
- 2. Imagine the effect of a sentence
Speaking bodies general introduction
It's particularly important to remember
that divisions into embryology, physiology, psychology, sociology, and therapy
are not part of nature, and that, in what is important to us, there is really
only one study - human neurobiology.
Motto for Institute of Psychoanalysis
1952
Embodiment can mean a lot of things, but
for this minicourse it means mainly two related things:
- 1. a theoretical emphasis - philosophical
and scientific - on the fact that we are physical bodies in a physical
world,
- 2. and a practical emphasis on bodily
experience, for instance in writing or therapy.
The purpose of this minicourse, like the
purpose of my earlier workshops, is to demonstrate how these emphases work
themselves out in relation to a particular topic or human capability, in
this case language.
I picked language this time because many
of you have been working on it as a topic, either in linguistics or in writing
or in TLA.
1. An emphasis, orientation, framework
There are many legitimate and useful ways
to think about language - many possible frameworks. The embodiment emphasis
that will be demonstrated in this series of talks is not commonly found
in orthodox language studies. The scientific developments it is based on
are quite recent, and the implications of these developments mostly have
not had time yet to trickle down into the many fields they will eventually
influence. There are however less recent language theorists whose thinking
is compatible with these ideas. You'll find some of them in the bibliography.
Places where these ideas ARE already being
worked out in exciting ways include the Linguistics Department at the University
of California at Berkeley, and the Cognitive Science Department at the University
of California at San Diego.
2. What is at stake: wholism vs dualism
For many centuries humans have thought
of the universe as created by an external maker or makers. When this creationist/transcendent
view of origins first began to clash with scientific understandings of the
natural world, scientists such as Descartes (1600s) evaded religious persecution
by saying humans are made of two parts: a physical part and a non-physical
part. Animal and human bodies - the physical parts - are machines that can
be studied by scientific means, but the human mind, soul or reasoning faculty
(animals were considered not to have these) would be non-physical parts
separately created by the deity and inserted into the physical mechanism
that is the body.
More recently we have begun to be able
to imagine a universe that creates itself, and within this vision we have
begun to be able to imagine the 'higher' cognitive abilities of humans as
evolved within increasingly complex and marvelous bodies fully embedded
within a richly self-creating surrounding world.
The fear that accompanies this new way
of thinking has been a fear of discrediting human gifts, reducing them to
something mechanical. This fear can be relieved by realizing that if we
do not artificially separate soul or mind from body, we do not have to think
of bodies as machines. We can instead think of them as physical organisms
whose full resources we have not yet discovered. Instead of reducing
cognition or soulfulness, we could be more fully appreciating the nature
and resources of physicality and bodies.
A second advantage of an embodied understanding
of mind or soul is that it allows us more easily to understand ourselves
as part of the natural world, and this understanding could help us to love
and preserve that world.
Language is homotypical - developed forms
of language are not found in any other sort of animal - and so the study
of language has carried significant weight in our sense of what we mean
by 'a mind,' and what it means to be human. So what is at stake in the question
of how to think about language is a much larger question of how to think
of knowledge and of the human relation to nature. Should we think of ourselves
as part of nature, or as set against it in some essential way? Also at stake,
in a sort of epistemological politics, are questions of what sort of knowing
should be considered valid and important. In the old ways of thinking about
language and mind, ways of knowing that were least obviously embodied have
been given most prestige. It seems that if we can reformulate language as
thoroughly embodied, more obviously embodied forms of knowing could become
more valued.
3. Basic principle: language is structural
influence
There is a small set of ideas that is central
to all three sessions of this minicourse. They are very simple obvious ideas
and yet they are quite radical in their implications. If we understand them,
we change our whole framework, our whole way of understanding.
Here is a summary. I will repeat it every
session.
1. Embodiment. We feel and perceive
and imagine and think by means of our physical bodies. Every change in our
cognitive state is accomplished (not 'accompanied') by means of a change
in our physical structure, ie in the way materials making up our physical
bodies are arranged. This is true both at the scale of the whole body, and
at the scale of the nervous system and the brain.
2. Consciousness and the unconscious.
Conscious function is by means of a widely distributed network of activity
in the nervous system, probably mostly the brain. Quite a bit of our nervous
system isn't involved in consciousness, but those nonconscious parts can
have an influence on conscious function.
3. Integration and segregation.
Parts of the nervous system that could be, or have been, integrated as part
of the conscious wide net can be made nonconscious: that is, certain kinds
of feeling, memory, perception, and understanding can be cut off, segregated.
This can happen as response to trauma, or just as a result of fairly normal
decisions made in childhood or later. Whatever the reason, a network that
has lost some of its parts is less able: less perceptive, less intelligent.
4. We speak (and write) from the structure
we are at the moment of speaking or writing. This structure may be more
or less integrated, more or less coherent, intelligent, relaxed, sensitive,
balanced.
5. Language has structural effect.
Sometimes the effect is observable - we say something and our hearer blushes,
or laughs, or does something. Sometimes the effect isn't visible, but it
is always there: no one can understand speech without being physically changed.
At the level of the brain we can think of it as a sort of bloodless brain
surgery - we change the other person's wide network of neural activation,
temporarily and sometimes permanently too.
6. Self-talk. When we speak, think
in words, and write we influence our own structure too.
-
In Speaking bodies I, I will talk
in more detail about what is happening in the bodies/brains of people who
are speaking to each other, to make us more able to imagine the embodiment
of language as vividly as possible. An important part of this picture is
the difference between language and non-language hemispheres in the brain.
In Speaking bodies II I will give
a very quick introduction to the ways an embodied approach is changing the
study of language, and what is at stake in this change.
In Speaking bodies III I will talk
about the implications of 5 and 6 above for our practice of language. If
the ways we use language can change our own and other people's physical
structure, what does that suggest about the relations of language to ethics,
art, therapy and spirituality?
Speaking bodies
I. How language happens: the recent science of language
- 1. Bodies able to speak: evolution
of the cortex
a. Perceiving and acting in a single-celled
organism
The protozoan Euglena moves by beating
its flagellum. When the eyespot is shaded, the motion of the flagellum
changes in a way that makes the organism move toward the light source.
[Euglena]
b. Perception-to-action chains in simple
many-celled creatures
In some kinds of sea anemone, sensor and
motion cells are separated by an intervening neural cell that propagates
response.
c. Branching networks in animals with
spines and spinal cords
Long chains of neurons between sensors
and muscles allow all kinds of interconnection and coordination of parts.
Simple coelenterates (like sponges), which
are the earliest phylum to have nervous systems, can have as many as seven
different kinds of sensor. In some of these coelenterates there are separate
neural chanins between sensing and moving parts of different kinds: in
effect, separate nervous systems for separate kinds of perception-action
sequence.
In organisms with more than one kind of
sensor it can also happen that through-lines from different sensors converge
on the same moving part. In such cases contacts between through-lines can
make more complicated kinds of motion possible.
Higher invertebrates have plexes or ganglia
of interneurons, each coordinating a different system. There are crustacea
in which these ganglia are organized into three proto-lobes, which organize
the behavior of eyes, antenna, and alimentary canal separately. These lobes
are proto-brains.
d. Reptiles have the beginning of a
cortex
Although the primitive cortex (neopallium) begins to develop
over ancient smell cortex when early vertebrates emerge from the water,
most sensory-motor behavior is still coordinated outside the cortex. Reptiles
have well-developed vision, for instance, but it is mostly organized at
the retina, so that reptile visual cortex may have little additional visual
work to do. In the submammals motion is still locally controlled from the
spine.
e. Early mammals have large smell brains
with a bit of cerebral cortex developing on top
Brains are like towns where a lot of paths
cross.
In early reptiles, distance perception
isn't important. They mostly smell and feel, and only see and hear what's
near them. As distance vision and hearing become more important than smell,
cortex is built to integrate developing sensory modalities with each other.
[jumping shrew neopallium]
Mammals evolve from the point where most
function is coordinated through the brain. In a primitive mammal such as
the jumping shrew, the new cortex is mostly sensory -- auditory, visual
and tactile -- with an forward part of the tactile area specialized for
motion.
The principle of this development seems
to be that early circuits are maintained but added to. Something like the
same response can be enacted as a reflex organized at the spine, as swift
but integrated behavior organized in midbrain structures, as cortically
organized action, or as a deliberate act extensively mediated by recently
evolved associative circuits. In humans many of these circuits are active
at the same time, and their activity at their many levels must be coordinated.
The kinds of aboutness enabled by widely
distributed networks whose branches are themselves networks, is a multiple
simultaneous aboutness calibrated to the internal needs of the whole organism.
The organism by these means is able to coordinate action relevant to many
things simultaneously present in an environment, is able to be more precise
in relation to particular things whose detailed differences matter, and
is able to stay on top of rapid many-part change, all while continuously
recovering internal equilibrium.
f. In animals that use their forepaws
in more complicated ways, we see more complicated cortex
In mammals that occasionally use their
forepaws to handle objects, bears for instance, and in tree-dwellers like
shrews, lemurs and early primates, the relative position of sensory areas
stays constant, but they begin to be pushed apart by more complex sensory
areas that crowd the earlier visual area backwards, and the earlier auditory
area downwards. In most mammals and even New World monkeys such as the
owl monkey, the new cortex seems mostly to be serving single senses (hearing,
vision or touch) although in more complicated ways. As new sensory cortex
develops there is parallel development of new muscle control areas in the
front of the cortex. Along with these new areas in the back and in the
front of the brain there are large new connections between the two.
g. In Old World primates there are
important new features
Old World primates, from which we descend,
have new multisensory association areas between newer areas of touch, visual
and auditory cortex. [chimp dentist]
i. a new visual system
In Old World primates a new visual system,
ten times more massive than the old system, adds an ability to see in more
detail and in more kinds of detail. The old visual system is specialized
for knowing where something is and moving in relation to it; the new system
is specialized for seeing what something is in particular, and remembering
it. (Rodents and primates don't differ very much in route-finding or object
retrieval capabilities, but they differ spectacularly in their ability
to identify, classify, and remember objects.)
The old system has its own stream of connections
to the front of the brain. The new system has a different stream, lower
down. This lower stream connects it to parts of the brain that have expanded
areas of connection for particularly important parts of the body: the mouth,
the throat and the forepaw or hand. [motor and somatosensory strips]
Close attention to particular objects
is a precondition for attention to other creatures of your own kind, and
close attention, particularly to faces, is one precondition for kinds of
social organization in which communication can develop.
Close attention to objects is also a precondition
for remembering them and talking about them, so it is not surprising that
in humans the lower visual stream's forward connections overlie the stream
used for language function. Knowing what something is, what it is good
for, and what it is called are closely interrelated in humans; "What
is that yellow stuff you put on rice, you know, ... saffron." It seems
that the look of a thing, its use, its name, and other facts about it,
can call each other up because they are connected in the cortex.
ii. forebrain: a new motion control
area
In early mammals the front of the cortex
is simple muscle control cortex. In later mammals such as primates more
associative cortex shows up in front of it. In monkeys and humans there
a whole new area, prefrontal cortex, which is particularly important to
complex action, including deliberate, conscious control of movement.
h. Humans have areas that associate
association areas
[layout of human cortex] [cross-section of cortex]
The organization of the cortex in humans
goes on being anchored in primary sensory and motor areas that exist also
in primitive mammals. These areas are very expanded in humans, and shoved
apart by associative areas built between them, but they retain much of
the character of these structures in earlier animals.
The most significant new development in
human cortex seems to be the elaboration of new areas between existing
association areas. They are in effect associating association areas.
Three large patches of this associating-association
cortex are unique to human cortex: they are found in the frontal lobe,
at the forward tip of the temporal lobe (side of the head), and in the
lower part of the parietal lobe (top of the head). The area at the tip
of the temporal lobe is important to object recognition, naming and memory.
The parietal area is at the junction of visual, auditory and somatosensory
cortex, and it is important to all sorts of complex spatial intelligence,
for instance mathematics. The area in the frontal lobe is important to
deliberate maintaining, shifting and sequencing of attention, either in
perception and action or in imagining.
In the diagram below, dotted and checkered
areas show these new uniquely human areas. [uniquely human]
In this diagram white areas are uniquely human. [uniquesly
human]
i. What it is about these new association
areas
Primary sensory and motor areas that we share with other mammals develop
relatively early (in embryos) and they are relatively mature at birth.
They don't show a lot of variation across individuals.
Sensory association areas in the primate brain develop later in
the embryo, and their connections are organized in the period after birth.
Structure in association cortex is and remains more plastic. In young animals,
injury to these areas will not necessarily impair function. Presumably
as a result of this plasticity, organization of association areas varies
more across individuals.
The very newest associative areas,
those only humans have, are the last to develop, the last to organize their
connections, the most variable part of the cortex across individuals, and
by far the most plastic over time.
i. Overview of evolution of the cortex
In the evolution of vertebrates the cerebral
hemisphere has folded in a horseshoe shape: rat, cat, monkey, human. In
the primates - and to a much greater extent in the Old World monkeys in
our evolutionary branch - burgeoning association areas alter the layout
of cerebral cortex so there is more folding and it begins to have the look
of the human cortex.
[evolution of cortex]
The large increase of association area
in primates occurs in the parietal lobe, the temporal lobe and the forebrain.
- The parietal lobe is a region where vision,
audition, touch, and movement sensation work separately or together to
prepare, monitor, and coordinate many kinds of act. Association tissue
in the parietal lobe integrates vision and touch, so it makes sense that
larger associative area in the parietal accompanies skill in handling small
items. New temporal lobe tissue that supports foveal color and form vision
(among other things) also supports new abilities to make social use of
faces.
-
j. Other aspects of current neuroscience
important to language study:
Joint attention
If perceiving another creature's eye-direction
sets creatures up (in some circumstances) to look where the other is looking,
both creatures will be seeing the same thing, will be about the same thing.
Being about the same thing (perceptually and later simulationally) is a
precondition for much else.
'Mirror neurons' and covert mimicry: action observation and premotor facilitation
Lateralization and manual precision
Hemispheric specialization is likely to
have been important in whatever happened 50,000 years ago. Language hemisphere
swift and flexible, accommodating every kind of transient simulation; real-world
hemisphere keeping a stability it takes from the stability of the world
itself. [cross-section of lateralized hemispheres]
Chimps have quite precise manual control
and use and make tools of a sort. Manual precision is not lateralized,
though, in chimps. There is a good case for lateralized/specialized tool-use
manual skill preceding and allowing the development of language with its
fast fine-scale manual-oral precisions. Tools before language. Syntax of
tool-use gestures before linguistic syntax. See Rizzolatti's paper on mirror
cells, grasp and language. [lateralization in human cortex]
2. History of the development of languages
There are 5000 - 6000 languages from maybe
10 root protolanguages. Half can be considered endangered.
6,000,000 years ago divergence of hominids
from an ancestral species common to both apes and human beings
Language capabilities maybe 150,000 years
ago - in Africa likely - call it 30,000 generations mix of primitive and
derived characteristics, ie structure derived at earlier evolutionary levels
and structure unique to the species
35,000 years ago jump in technology
12 - 20,000 years ago 'Nostratic' hypothesized;
another Dene-Caucasian, mother of Chinese and Na-Dene languages whose speakers
among the first American migrants
10,000 years ago proto-Indo-European
1450 BC Greek
6000 years ago first writing Middle East
Rapidly mutating - 400s Old German, 700s
Old English, 1300s Chaucer. Dutch and Frisian closest to English. Vikings
787, French 1066, Latin
In many languages most of the basic words
are traceable to its ancestor, but in English 99% of words are not from
OE - 62% of words most used are, though.
15 stablest meanings: I/me, two/pair,
you, who/what, tongue, name, eye, heart, tooth, no/not, fingernail/toenail,
house, tear, water, dead - from these extrapolate to proto-Indo-European
3. Language networks: imagine the effect of a sentence
Two important changes in the way we think
about language have resulted from functional imaging studies (PET and MRI)
of brain activity during linguistic tasks.
The first is the discovery that language
is not the product of a confined language module but uses networks very
widely spread through the whole of the left hemisphere and parts of the
right (Damasio 1989). [how to imagine a wide net]
The second is related to the first: it
is that language also uses connections built for basic situational action
and perception.
a. Damasio's three language zones
Damasio thinks of the whole of the language
net as spreading through three zones centered around the Sylvian fissure
(the large horizontal fold under the ear): a language implementation
zone that includes the traditionally recognized language areas; a mediation
zone, used non-linguistically as well as linguistically, adjacent to the
implementation zone; and what he calls a conceptual area spread
throughout primary sensory and motor cortex outside the implementation
zone.
[Damasio's zones]
b. Damasio's implementation zone
Language implementation areas are language
perception and production areas - (usually) left hemisphere areas active
during perception and construction of words. The section of the implementation
zone closer to the back of the head includes areas for perceiving language:
we see written, spoken, or signed words as well as hearing them.
This includes perceiving our own language. So perceptual foci in the language
net would be auditory, visual and somatosensory for spoken language, and
visual and somatosensory for signed and written language.
Linguistic construction can similarly
include oral, written, and signed words. Like other muscular action, speech
articulation, auditory or signed, is instituted from motor cortex. Premotor
and prefrontal (pre-pre-motor) cortex are important to the complex scheduling
needed for sentence production. The forward section of the language implementation
area thus includes large amounts of frontal cortex, as well as the enlarged
mouth, throat and hand areas of motor cortex.
[relation of primary sensory and language areas]
So areas in the implementation network
include auditory, visual and somatic primary sensory areas, sensory association
areas, motor cortex and motor association areas, as well as Wernicke's
and Broca's areas, the two centers that earlier were thought to be the
only language centers of the left hemisphere. The network character of
language function is changing the way we understand these two areas: both
are now being seen as associative centers that set up 'word meaning'
subnetworks both in response to language, and in the process of
producing language.
Wernicke's area, near both auditory and
visual cortex, is a large region of association cortex active in linguistic
function no matter what sensory modality is used. Wernicke's was formerly
thought to be a module for language comprehension. Broca's area is near
motor control areas and was formerly thought to be a module for language
production. Both areas are now known to be active in both production and
comprehension. They are connected by a large bundle of fibers.
[left lang areas side view] [left language areas cross-section]
c. Damasio's mediation and conceptual
zones
Damasio's language mediation zone
surrounds the language implementation areas described above. Injury to
these areas results in subtler damage to language than injury in implementation
zones.
Damasio's conceptual zone includes
non-linguistic perception/action areas whose activity either is triggered
by language, or else organizes language production.
Mediation cortex includes two-way switches
between language perception/production networks and conceptual networks.
Like language implementation cortex, language mediation cortex is normally
left hemisphere cortex, but conceptual or 'meaning' networks turned on
by language forms via structures in mediation cortex are found in
both hemispheres.
d. 'Meaning' and wide networks
The Damasios have found that response
to the name of a thing is similar to response to a thing or to a picture
of a thing. A name or picture of a hammer, like the hammer itself, would
set up cortical structure relevant to "the typical action of the tool
in space, its typical relationship to the hand and to other objects, the
typical somatosensory and motor patterns associated with the handling of
the tool, the possible sound characteristics associated with the tool's
operation, and so on" (1994). Evocation of some part of the potentially
large number of relevant responses would "constitute the conceptual
evocation for a given tool".
What Damasio means by a conceptual
zone, then, is areas of sensory or motor cortex in which object- or
event-relevant structure can be evoked, via mediation cortex, by different
means and for different purposes. He thinks of it as conceptual
because it is not exclusively linguistic but is common to many kinds of
response.
e. Convergence zones in mediation cortex
Hearing or reading language can be thought
of as like a kind of externally-cued dreaming, because when we hear or
see speech, as when we dream, we can be about many things other than the
things we're able to perceive in the world surrounding us.
Convergence nodes in mediation regions
are little networks of neurons from which widespread patterns of this sort
of non-immediate 'meaning' can easily be reconstructed. Activation can
also happen in the opposite direction, from from 'meaning' to language
production.
Convergence zones form near and between
association areas important to aspects of perception and action. There
are many convergence nodes in the left hemisphere. PET studies find that
different categories of names set up activity in different parts of the
cortex. For instance the brain distributes activity differently in response
to nouns and verbs.
i. Nouns
The more recently evolved lower object
vision and memory stream seems to be relevant particularly to nouns. PET
and fMRI imaging studies find multiple small areas in this stream responding
selectively when people are shown pictures of objects, or when they
read or hear their names. Among nouns, there is relative segregation for
common nouns and proper names, and among concrete nouns there are different
areas for animate and inanimate things.
Producing concrete nouns also activates areas of this stream, with different
sorts of object activating different areas. [feature, object and verb areas]
Names of features of objects activate
nodes close to regions responsive during perception of those features.
Generating color words, for instance, activates areas just a few centimeters
away from regions active during color perception.
Object identity networks in the left hemisphere
involve less cortex than on the right, as if object memory during language
use may have access to a more rapid, sketchier sort of sensory memory on
the left.
ii. Verbs
Damasio describes response to verbs, or
response in the process of generating verbs, as evoking the way something
acts and moves. Generating verbs activates interconnected areas at both
sensory and motor ends of the upper (older) through-stream. Response is
found just a few centimeters from regions that are active when we observe
moving objects. In frontal areas near Broca's area there is also activity
near regions that are active when we perceive action or imagine acting.
f. Conscious and nonconscious networks
At any given moment, the comprehensive
wide network active in the cortex will include both conscious and nonconscious
parts, and both linguistic and nonlinguistic parts.
g. Summarizing:
1. Given a linguistic event or artifact,
2. linguistic effect
3. is on the structure
4. of a whole body,
5. and within the body the neural wide
net,
6. within which, in turn, there is a linguistic
subnet
7. and a conscious subnet,
8. which may overlap in different ways
and to different degrees.
9. These structures are dynamically self-organizing
at all scales.
10. Language that we understand evokes
structure.
11. When we create language it runs off
existing structure we have activated in the moment of utterance.
Language plays the instrument we already
are.
It is created by the structure we are at
the moment of creation.
The locus of effect of language (or
any representation) is the body of the user.
In Speaking bodies II I will talk
about the implications of this approach for the study of language as such,
and in Speaking bodies III, about implications for the practice of
language.
-
-
|