A review
of the second Special Issue of Biolinguistics Vol. 6, No. 3 – 4 (2012) on the
Embodiment of Language
In the editorial ‘Introducing Embodiment of
Language’, the editor, K.K. Grohmann, makes the unusual admission that the
topic of this special issue left his band of reviewers baffled to such a degree
that more suitable reviewers had to be found – and that the original instigator
of the issue, R. M. Allott from Oxford University, pulled out as guest editor.
One wonders why? Hopefully not because of a case of dis-embodiment.
As to the raison d’être of the topic
itself, the editor says being ‘intrigued’ about the ‘emerging field of
“Embodiment of Language”, especially as it includes research on ‘mirror
neurons’ which seem to have something to do with language. This in turn awakened
my interest as our son is currently completing his PhD thesis in linguistics
which proposes, amongst other things, that ‘pragmatics’ is as innate (embodied)
as syntax, mediated via mirror neurons that seem to provide a biological
foundation for neo-Gricean pragmatics. Of course I have lots of interesting
discussion with my son, especially as I subscribe to the Chomskyian school of
thought.
So what is this emerging field all about? A
programmatic summary is one from the original conference at Oxford University:
The embodiment
of language (as well as cognitive embodiment) is a much-debated topic. It offers
a different approach to language function in the brain from the hitherto widely
accepted account in terms of words as symbols, parts of an essentially
psychologically researched conceptual system. Progress in neuroscience, notably
with the discovery of mirror neurons but also with refined neuroimaging
techniques, has opened up the possibility that words in the brain are not
simply labels for concepts but integral parts of perceptual and motor
organisation (embodied semantics).
From a scientific and common sense point of
view it has always been clear that the Cartesian ‘cogito, ergo sum’ arises from
the brain but because hitherto we cannot figure out the mechanics of the brain,
we have no choice but to deal with the output of the brain, namely the
so-called mind – hence the scientific investigation of the mind as psychology
(and linguistics therefore being part of psychology, as Chomsky famously postulated).
Now neuroscience – and neuro-linguistics - seem to make such progress that we
can put our fingers on language itself. Chomsky has of course always maintained
that linguistic theories should be in conformity with what we know about
biological-computational systems.
Equally obvious to me is that we meet with
a paradox here: the snake biting its tail. Everything we know, we know through
language. How can we possibly know language through language? How can we
describe a biological-computational system, namely the human brain, and end up
with language? It’s not that we shouldn’t try, and indeed Chomsky has been the
most convincing linguist of our time in positing a theory of language that at a
basic level conforms to current biological knowledge but at the same time is a
sophisticated model of language as a mental representation. For example the
principle of binary Merge operations seems to conform to basic concepts of
neuroscience in that neurons ultimately operate with “on” and “off” switches.
On the other hand we must be clear about the relative paucity of knowledge we
have about the brain despite what seems to be progress in neuroscience. It
would be putting the proverbial cart before the horse if we now proclaim that
we can deduce language from what we now know about the workings of the brain –
and unfortunately this is exactly what this special issue of Biolinguistics
seems to be about. What seems to me even more unfortunate is the focus on
language evolution via purported human brain evolution – for we know even less
about either, and as such end up with a type of speculation which Chomsky
describes as ‘nothing is impossible but many things are unlikely’.
In the course of these odd pursuits one is
doubly puzzled by the first – programmatic? – contribution to this special
issue, namely Bernard H. Bichakjian’s ‘Language: From Sensory Mapping to
Cognitive Construct’ which opens with the old and baseless refutation of
Chomsky by Evans and Levinson:
[t]he
claims of Universal Grammar … are empirically false, unfalsifiable, or misleading in
that
they refer to tendencies rather than strict universals. Structural
differences should instead be
accepted for what they are, and integrated into
a new approach to language and cognition that
places diversity centre
stage (2009:429)
and then going on to show how embodied
‘perceptual’ language morphed (= evolved) into disembodied ‘conceptual’
language, thereby heaping contradiction upon contradiction, and in a strange
way confirming Chomsky’s innateness theory by trying to disprove it. In other
words Bichakjian first suggests that ‘incipient’ speakers had a very much
embodied language based on perception which then ‘evolved’ to become more
‘efficient’ as ‘conceptual’ language. Apparently languages – at least language
families – evolved in their own peculiar ways so that they end up as pretty
much unrelated by anything resembling UG, quoting Dunn et al, 2011:
We
show that each of these [four] language families evolves according to its own set of
rules,
not according to a universal set of rules. That is inconsistent with the
dominant ‘universality
theories’ of grammar; it suggests rather that language
is part of not a specialised module distinct
from the rest of cognition, but
more part of broad human cognitive skills.
So we are back to cognitive linguistics of
old. Nothing new here. Even so one of the more surprising supporting examples
given by Bichakjian is the ‘evolution’ of writing which moves from the
perceptive pictogram to the conceptual alphabet we have today, and took a long
time to achieve, says Bichakjian, more than two millennia. The word ‘evolution’
is here misused in its Darwinian sense for everything evolves all the time in
the sense of ‘change’ – including of course the surface features of language.
To cite writing as an example of ‘evolution’ of language – starting with the
mythological ‘incipient speaker(s) – is as ludicrous as suggesting that that
the evolution of the Internet is an example of the evolution of human cognition
– starting from the mythological incipient thinker(s). There are also some
outright dangerous implications when suggesting, as does Bichakjian, that the
so-called evolution from the perceptual to the conceptual afforded the human
species with ‘selective advantages’ using the example of changes in measurement
ideas:
… these
anthropomorphic units of measurement, molded on the perception of
the outside world,
have been replaced with the conceptually devised metric
system, which has considerable
selective advantages.
One should remind Bichakjian that the
Americans still use aspects of the ‘anthropomorphic units of measurement’ such
as ‘foot’ and are no worse off than those other highly advanced peoples who use
the ‘conceptual’ metric system. More importantly, one should remind the author
that ‘writing’ does not bestow a Darwinian selective advantage, since as much
rubbish is written as it is spoken. Bichakjian comes close to racism when he
claims that certain ‘incipient’ grammatical features (like certain noun
classes) are still present in aboriginal (sic) languages and ‘survive’ as
certain features in languages such as German and French. If taken this to its
logical conclusion, the speakers of aboriginal languages are stuck at the level
of perceptual evolution while the Germans and French have made the grade as
conceptual speakers. We know for sure that this is false, based on the simple
observation that if we take the aboriginal baby and bring it up in a German or
French speaking family, the child will have no difficulty whatsoever in acquiring
one of these supposedly superior conceptual languages. In other words, the
brain of the aborigine child is as fully ‘evolved’ as that of anyone else on
this earth. The equally unpalatable alternative is to say that the aboriginal
child does have brain power to acquire any language on earth but when exposed
to a ‘primitive’ language the child will regress cognitively, at least in terms
of its measurement terminology. Ipso ergo, Chomsky is right in assuming that
the language capacity is like an organ, situated in the brain, and just like
the human heart or lungs have not ‘changed’ (or become more efficient) in the
last hundred thousand years or so, neither has the language capacity. Therefore
all extant languages are at the same level of competency, for even if there
were some languages more sophisticated than others, Chomsky’s noted ‘poverty of
stimulus’ puts to shame any notions of sophisticated or more evolved input. Of
course humans have ‘changed’ technologically in many ways over this time, and
they keep on changing – some say with increasing speed towards species
extinction, contrary to the expectation that humans and their language is,
according to Bichakjian, ‘an instrument that keeps evolving — becoming ever
more cerebral and, by so doing, ever more efficient.’ On the scale of Darwinian
evolution, humans are only a blip on the clock, and if one shares the pessimism
about the (very short) human history in it, one can always quote Russell:
After ages
during which the earth produced harmless trilobites and butterflies, evolution
progressed to the point at which it generated Neros, Genghis Khans, and
Hitlers. This, however, is a passing nightmare; in time the earth will become
again incapable of supporting life, and peace will return. (Unpopular Essays, 1950)
In conclusion, Bichakjian’s conclusion
below is entirely wrong:
It is our
cerebral nature that explains the developments that were discussed in the
foregoing, and
they in turn support and confirm the view that language is not
an instinct or a steady-state
attribute coded in our genes, an organ as it was
once claimed (Chomsky 1980: 37), but an
instrument that keeps evolving —
becoming ever more cerebral and, by so doing, ever more
efficient.
On the positive side, the author does
unwittingly confirm the status quo by claiming the ‘evolution’ from an embodied
perceptive language to an disembodied conceptual one, echoing the old body and
mind dichotomy and/or the biology versus psychology one, thereby lagging behind
Chomsky and others who have long solved the problem by asserting that the mind
arises from the brain – just nobody at this stage quite knows how, and perhaps
never will (language as paradox). Bichakjian to his credit doesn’t mention
neuroscience once, and thereby seems to negate the whole enterprise of
‘embodiment of language’ in the first place. I don’t want to give the
impression as if the whole quest is hopeless (citing again and again Chomsky’s
dictum that science is like the drunk looking for his keys under a lamp post
because that’s where the light is) hence I repeat one of my ‘conceptual’
speculations about language and mind in the brain: the hyperbolic accumulation
of knowledge and its storage in the brain may have come about by some subtle
changes, namely the collapsing of neurological pathways into cyber-highways,
i.e. if I compute or think about, via my language, a certain problem many
times, I will organise the storage retrieval by the shortest and fastest
neurological pathway, thereby setting up cognitive hierarchies that allow me to
shortcut computation to the point where only new and puzzling information
intervenes. Note that computers can only increase the speed of computation (and
possibly surpass the computation speed of the human brain) but they cannot replicate
the human biology of being able to merge a multitude of neurological pathways
into super-highways – of course I have absolutely no neuro-scientific proof for
this assertion. In passing one should also remind contemporary linguists that
Chomsky is the only one who has contributed a major bit/byte to computer
science, commonly known as the Chomsky Hierarchy, whereby formal languages are
organised as ‘regular, context-free, context-sensitive and recursively
enumerable’. It is worth paying attention to a brain like Chomsky’s. In
exercising his brain he has obviously managed to collapse quite a few neuron
pathways into major highways, thus being able to analyze vast amounts of
scientific data and being able to synthesize a credible theory of language,
allowing him to be the science genius he is generally acknowledged at being.
All human brains are capable of such cerebral feats in principle but few
actually do the rigorous exercises needed to get there, maybe in analogy to the
physical feats achieved by exceptional athletes. Linguists like Bichakjian thus
strike me as slow runners who claim that the exercise regime of the world
champion runner is all wrong. As a silly aside one may add that in sports one
can enhance – à la Lance Armstrong – one’s performance with certain drugs,
while in cerebral contests like the proverbial linguistics wars, nobody worries
about how you stimulated your brain to come up with credible theories of
language. Chomsky’s advice on this aspect of academic life is to peddle as fast
as possible, so you won’t fall off the bicycle.
Having made my point of view abundantly
clear, lets move on to the next article by Valentina Cuccio, entitled ‘Is
Embodiment All That We Need? Insights from the Acquisition of Negation’.
Following Bichakjian’s strategy to propose
a grand theory that is then supported by snippets of data, Cuccio declares:
The aim of this
paper is to present the hypothesis that speaking is a complex ability realized
by means of at least two different mechanisms that are likely developed at
different and consecutive steps of cognitive and linguistic development. The
first mechanism has a neural explanation grounded in the notion of embodied
simulation. The second implies socio-cognitive skills such as Theory of Mind.
In order to fully develop the second mechanism, a symbolic communication and
interaction with a cultural community are needed. This hypothesis will be
tested by looking at the acquisition of linguistic negation.
Negation is of course an important aspect
of any descriptive linguistic enterprise but to use it to support above
hypothesis is like claiming that the mechanics of the left foot is a test case
for the overall theory of human physiology – of course it is. The point is that
there are myriads of other aspects that could still falsify the hypothesis.
Data driven research assembles as complete a picture as possible and only then
a hypothesis or theory is proposed to cover all the known aspects. It is of
course quite impossible to assemble all known research about language(s) – knowing
also that there are still many more gaps in the knowledge – and then arrive at
a credible theory of language. Chomskyian linguistics is often accused to be
overly theory driven which when applied to specific languages fails to account
for certain data. This Popperian obsession is often exercised by unearthing
obscure language data known only to one researcher – e.g. Everett’s Piraha
controversy – and used to ‘falsify’ a major theory. On closer analysis such
data often turn out to be misleading and can in fact be incorporated into the
recognised paradigm. Chinese anaphora, for example, are a more credible
challenge for the Minimalist Program and while many linguists grapple with it,
there are many solutions proposed, some of which fit the model while others
don’t, giving rise to a vigorous scientific debate which is the nature of
science in general. A good starting point is that a general theory covers most
of the known facts but not necessarily all of them.
In any case Cuccio also follows Bichakjian
in this strange case of disembodying embodiment by first stating that
‘cognition and language are embodied’ and then follow up saying that this
‘embodiment’ may not be enough to explain linguistic ability, hence the
additional disembodied necessity of the ‘Theory of Mind’ which is mediated by a
‘symbolic communication and interaction with a cultural
Community’. If Cuccio were to admit that we
simply do not know how exactly the mind arises from the brain, hence in the
meantime (or forever) seek recourse in a sort of disembodied psychology, I
would concur. As Cuccio seems to make a principled distinction in terms of the
old dichotomy, I do not concur, for the putative ‘symbolic communication and
interaction with a cultural Community’ arises from the brain as much as
anything else. Does Cuccio provide a proof for her contention in terms of
negation?
She starts with the not so surprising
observation that in child language acquisition negation proceeds from the
concrete to the abstract. The concrete involves early ‘rejection’ and refusal (“no,
I don’t want to eat this banana”) and the abstract involves ‘denial’ (“no, this
is not a banana”). Cuccio explains that the concrete phase corresponds to the
‘simulative embodiment’ mechanism of language use, i.e. the motor neurons
‘simulate’ the action and as such correspond to the same neural activity that
occurs when the action is performed in reality. Cuccio then contends that the
abstract uses of negation, like denial, cannot be simulated that way, hence
need different neural mechanisms (sic) which seem to be associated with the
above ‘cultural community’ by inferring the meaning via mind-reading of others.
The author goes on to support this idea with observations from autistic
children who seem to have difficulty in ‘mind-reading’, hence perform poorly in
this use of negation. Strangely enough Cuccio ascribes the famous mirror
neurons as belonging to the simulative mechanisms, thereby unable to ascribe
any specific neurons to the mind-reading processes, making it almost disembodied
again. One would have thought that autism as a neurological disability may in
fact involve the pathological lack or damage of mirror-neurons that other
authors ascribe to the ability to ‘empathise’ with other minds.
It is an interesting idea to limit the
definition of ‘language embodiment’ to ‘simulative’ processes, thus associating
motor-neurons with language, and then requiring other neurological processes to
account for more abstract or ‘higher’ cognitive language tasks. The trouble is
that this dichotomy can again be kidnapped for the old body-mind dichotomy,
giving rise to disembodied souls, spirits and other ghostly/ghastly thoughts
and beliefs that defy (or ‘deny’ – excuse the pun) rational thought. Personally
I do not get the point as to why human communication needs something called
higher-order ‘mind-reading’ in order to correctly infer what the other person
is saying or writing. When the child says ‘no, I don’t want to eat this banana’
(where mother’s request was ‘please at this banana’) it doesn’t need to read
the mind of the mother more or less when confronted with the situation of
mother putting an apple in front of the child and saying ‘please eat this
banana’ and the child denying (correcting) her mother by saying ‘no, this is
not a banana’. Sure the mother may be joking or committed a faux-pas but by the
same token the request to eat the real banana may have all sorts of hidden
messages in it as well that need ‘mind-reading’. Straightforward communication is
devoid of the necessity to read the ‘ineffable’ (see last article of this
issue) mind of the other (cf. Gricean rules of implicature). This applies even
more so to scientific communication. I do not have to second-guess Chomsky’s
mind when I read his articles. I do not ascribe weird mental motives to Cuccio
and Bichakjian, thinking they make all this stuff up in order to score brownie
points over Chomsky. I just ‘deny’ the probability of their arguments. Even the
most sophisticated arguments put forth by Chomsky and the like are grounded in
the embodiment of language, arising from the language organ somewhere in the
brain. Trying to figure out what someone is ‘really’ thinking as opposed to
what they are saying is just another occasion of using one’s brain and language
capacity. There are no disembodied language processes. Cuccio may well be
correct though in asserting that there are various different neurological
processes that account for various uses of language (parole) whilst language
itself (langue) is much more likely a confined and constrained computational system
in the brain (hence Chomsky’s metaphor ‘organ’).
The next article by Marco Fenici entitled ‘Embodied
Social Cognition and Embedded Theory of Mind’ makes the same assertions as
Cuccio but testing his thesis via the False Belief Test (FBT) in child
cognitive development. Fenici does go into more detail what exactly is meant by
‘embodiment’ but in the end it comes down to the statement that echoes Cuccio:
I will claim
that early social cognitive abilities
are probably embodied inasmuch as available evidence is consistent with their
implementation by cognitive processes integrating sensory–motor information. On
the other hand, I will argue that late social
cognitive abilities are embedded in social and dialogical practices — and, in
particular, that the ability to pass FBT at age four denotes the acquisition of
a minimal capacity to explain people’s reasons to act.
Fenici concedes that ‘late’ development may
also be partially ‘embodied’ but nevertheless argues for an at least partial
disembodiment which is ‘embedded in social and dialogical processes’. One
wonders why the obsession to cling to at least a bit of disembodied cognition
and language (esp. if one equates language and cognition, as I do). If what
seems like lower-order thought/language being based on bodily simulation, why
should so-called higher-order thought/language be based on something ethereal
and ephemeral? What is so special about ‘social and dialogical practices’ that
elevates them above embodiment? Does the detection of False Belief not entail a
simulation of sorts: “this banana is an apple” as a False Belief is grounded on
simulation as much as “Dilige et quod vis fac” (Augustine). Those who want us
to believe in False Beliefs do so by manipulating our emotional intelligence,
instilling fear and offering a security blanket by way of a community of false
believers (‘the church’) who then must defend their beliefs to the death.
Fenici also concedes that passing the FBT
relies on ‘language acquisition’ but then goes on to say that this isn’t enough,
that there needs to be some ‘dialogical’ practice – which goes without saying.
The common game theory analogy (à la Wittgenstein) is that in order to play the
game you need to know the rules of the game before you can play it. What
happens afterwards is not a function of the rules of the game alone: practice
makes the master of the game. The analogy equates to ‘language competence = langue = grammar = rules of language =
language acquisition device = biolinguistics VERSUS ‘language performance = parole = functions of language = x. The
‘x’ indicates what we don’t know about why one of the many functions of
language seems to be the generation of ‘false beliefs’. That autistic children
seem unable to pass the FTB (presumably the ‘Sally-Ann marble test’) does say
something about the inability to put oneself into frame-of-mind of Sally who
doesn’t know what the observer knows – but does it prove one way or the other
that the whole scenario does or doesn’t arise from a biological brain of all
concerned? Fenici makes the case that certain aspects of language performance
(‘dialogical practice and social enculturation’) take place in an dis-embodied
realm of cognition because we cannot find the connection between these types of
cognitive abilities and the human body (whereas we seem to be able to posit
bodily connections for ‘lesser’ cognitive abilities). Since autism and other
degenerative brain diseases are obviously of yet unknown
biological-neurological origins we cannot simply turn around and say that the
healthy (normal) brain is somehow not the immediate origin of what passes as
‘normal’ behaviour and ‘normal’ cognitive ability such as passing the FTB.
Pathology has always been the prime vehicle for trying to figure out how the
‘healthy’ human body works. To figure out how the ‘healthy’ mind works has
hitherto been the realm of psychology, from Freud’s investigation into hysteria
to the FTB by Baron-Cohen, Leslie, and Frith (1985) in their study of theory of
mind in autism. Fenici no doubt makes a valuable contribution to this field but
contributes nothing to the theory of the ‘embodiment of language’ as a possible
aspect of biolinguistics, as originally conceived by Chomsky.
The next article by Leonardo Fogassi &
Pier Francesco Ferrari entitled ‘Cortical Motor Organization, Mirror Neurons,
and Embodied Language: An Evolutionary Perspective’ is thankfully more down to
biolinguistics as various neurological explanations for language are proposed.
From the outset my main criticism is that ‘language’ seems to end up as a
by-product of various neurological processes, be they motor and/or other sensory
processes. There can be little doubt that language (as a bodily organ in the
brain, as proposed by Chomsky and others) has inputs from many parts of the
brain and the authors come close to this notion when they concede that ‘according
to some linguists, syntax function can be defined as a regulator of language
(Pinker & Jackendoff, 2005)’. Hence ‘syntax’ must reside somewhere in the
brain ‘regulating’ all the inputs and outputs. It’s a bit sad that the authors
cite Pinker and Jackendoff as ‘some linguists’ who in fact are not well known
as defenders of syntax models being central to language. Chomsky as the supreme
syntactician is only cited as part of the Hauser et al. publication on ‘The
faculty of language: What is it, who has it, and how did it evolve?’ But never
mind, let’s have a look at Fogassi & Ferrari’s section on ‘Action Sequences
and Syntax’ which seeks to equate hierarchical action sequences with sentences
being a hierarchical sequence of words:
The hierarchical
sequencing of motor acts into a specific action (for example, (a) grasping a
piece of food, (b) bringing it to the mouth, and (c) biting it) aims to a
superordinate behavioral goal (eating the food). If the order of the motor act
is changed (e.g., biting the food with the mouth, bringing the hand to the
mouth, and grasping the food with the hand: c–b–a) the action goal can change
(take the food out of the mouth). Similarly, the meaning of a phrase is given
by the sequential organization of words. By changing the position of the words
in a sentence, its meaning changes or is lost.
The example given is a bit bizarre,
especially the notion of changing the sequence. It is well known that
computational models of language that parse a sentence as a string from left to
right, will fail. Language is full of disjoint/discontinuous features and
either way it is overly simplistic – and most probably wrong – to say that ‘by
changing the position of the words in a sentence, its meaning changes or is
lost’. Synthetic languages like Latin are famously independent of word-order,
i.e. one can change ‘the position’ of certain words without changing the
meaning at all. What the authors mean to say is that there might be a
correlation between the ‘motor act’ and the ‘verbal expression’ of that motor
act. Such a claim seems eminently reasonable if only because we can visualize
the similarity without difficulty. Indeed it would be most concerning if there
was no match between the two. The real question is how the ‘syntax’ module
manages the input from the motor neurons and transforms it into an equivalent
sentence. As we know that this is a human-specific feat, it seems rather
speculative that the authors claim that the ‘syntax’ organ in the brain
developed from enhanced use of motor activity – surely primates ‘eat’ in the
same sequence of ‘human eating’ and eat as much, if not more than we do but
never got around to verbalizing the process. Surely, the syntax organ must have
developed in an evolutionary process but to put it like Fogassi & Ferrari
do in their concluding remark below, seems somewhat vacuous:
Thus, although
the transition from action to language could have been long and may have
required a complex adjustment of the mechanisms involved in sequence
organization, nonetheless the existence of a motor substrate endowed with a
motor meaning, organized in chunks and accessible by visual and acoustic higher
order input, seems an important prerequisite for both language construction and
its comprehension.
Practically all evolutionary processes are
‘prerequisite’ for the next step, from lower to higher animal species, but what
caused the unique syntax (language) leap in the human brain remains as
unanswered as ever. The best description we have so far, in my mind at least,
is that of the old Marxists who argued for the famous leap from quantity to
quality.
The next article ‘From Gesture to Speech’
by Maurizio Gentilucci, Elisa De Stefani & Alessandro Innocenti is a densely
argued research paper that proves its point that there is a close neurological
connection between speech and gesture but then suffers from an immediate
misconception of language, namely that ‘speech’ equates language. It might
sound quite appealing – as the authors do – to suggest that gestures
(especially to do with ingestion) trigger certain movements on the mouth which
in turn trigger the formulation of syllables which in turn combine to make
words – and voilà, we have speech and language. It has been noted by others
(e.g. Samuels) that vocalisation is common to many species, hence it is
questionable that vocalisation itself leads to human language. It seems far
more logical to assume instead that language/syntax arose as an organ in the
brain which then used the vocalisation ability to turn language into speech. It
may well be that vocalization in itself arose in the ways explained by above
authors but then jump the gun and proceed to explain language evolution as arising
from this process as well. It is well established by Chomskyian linguistics
that the PF interface is the last step of language output as speech,
constraining language in many ways such as motor skills and phonotactics. Gentuluci
et al. make a good case for a feedback loop in that gestures are obviously very
important in the expression of language – as speech – and that speech and
gestures are very much interlinked. Once the language/syntax organ in the brain
was fully established there was no doubt a secondary feedback loop established
between language/syntax and gestures themselves, as evidenced by sign language
as a substitute for vocalized speech. It is noteworthy in this context that
fully fledged sign language is only in minor ways connected to the natural arm
and hand gestures, as say, in the ingestion process as proposed in the previous
article (‘sign languages are not mime – in other words, signs are conventional,
often arbitrary and do not necessarily have a visual relationship to their
referent, much as most spoken language is not onomatopoeic’ says Wikipeadia).
As such language/syntax gives rise to sign-language as much as to speech – not
the other way round.
The next article (one of two on the subject
of colour) by Loïc P. Heurley, Audrey Milhau, Gabrielle Chesnoy-Servanin,
Laurent P. Ferrier, Thibaut Brouillet & Denis Brouillet, entitled
‘Influence of Language on Colour Perception: A Simulationist Explanation’
tackles the intricate and supposedly neural connections between perception,
simulation and language.
While we can readily follow the argument of
‘simulation’ in the sense given by another set of authors (Michiel van Elk,
Marc Slors and Harold Bekkering (2010)) as:
Following the
notion of communicative motor resonance during speech perception several
studies have shown that reading verbs referring to concrete action results in the
recruitment of effector-specific regions of primary motor and premotor cortex,
comparable to the activation observed when moving the effector most strongly associated
with these actions.
We (I) find it more difficult to understand
how this could apply to colour. In the first instance psychology and
linguistics have long grappled with these notions, and indeed biolinguistics
arose from one of these battlegrounds, namely when Eric Lenneberg first
postulated linguistic innateness via colour codification experiments. Since
then countless colour studies have taken up positions between biolinguistics
and cognitive linguistics, now followed by neurolinguistics. Given that
languages are replete with more or less complex colour terminology, one wonders
how this comes to be in the first place. Of course one can summon various
ecological reasons like the well-established notion that Eskimo perceive many
shades of white (snow and ice) while those in luxuriant tropical environments
may perceive the rich hues of green and blue much more extensively than their
arctic counterparts, and so on. While the physiology of colour vision makes no
such distinctions, it is of course conceivable that the actual use of colour
vision very much depends on the environment and as such gives rise to
linguistic labels we are familiar with. It is also conceivable that it is not
so much the colour itself which we perceive, label and memorize but the common
objects imbued with the colours in question – take the orange colour of the
orange as an obvious example. Our authors (i.e. Heurley et al.) do make exactly
this observation when they say:
In short, these
various experiments and also others (see Bramão et al. 2011 for a review)
suggest
that colours stored in memory can facilitate or disrupt perception of
objects presented in colour.
Moreover, they demonstrate that this influence is
produced by reading the linguistic stimuli that
denotes colour related objects,
suggesting an interaction between memory, language, and object
perception.
Unfortunately they then jump to conclusions
that are unwarranted. In the first place it may also be conceivable that
objects can be ‘simulated’ as much as action sequences, assuming a sort of
memory-colour-photo-copy we make of an object like an orange. It is also
feasible that once such an object is stored in the memory, it may then, as the
authors say above, ‘facilitate or disrupt perception of objects presented in
colour’. These are performance issues that have nothing to do with language per se. If and when I have ‘orange’ in
my lexicon, I will of course ‘use’ it in many ways but my syntax-box will no
doubt assign ‘orange’ to either a noun or adjective category, thereby
delineating its uses from a purely grammatical point of view. When I retrieve
‘orange’ from the lexicon to slot it into a sentence and proceed to the
semantic interface I may well light up regions in my brain that store
experiential ‘simulations’ of the term ‘orange’ and as such confuse or facilitate
my semantic interpretation – for example my recent realisation that not all
oranges are orange but can be quite green. Lexical labelling may well be
directly associated with ‘simulation’ storage in the brain, and a feedback loop
may operate subsequent to its acquisition but language as syntax has no
conceivable recourse to such ‘simulation’, precisely because language is an
organ in the brain that is independent of any other system. To operate (use,
perform) the language as syntax system requires input from the lexicon in the
first place and memory and practice in the second place. As such Heurley et al.
should restrict their claims to the lexicon alone and not make claims about
‘language’ as in their conclusions
… the
possibility that language can influence perception through a simulation process
(and also the reverse influence) … According to this approach memory,
knowledge, language and perception function in a coordinated way which can
either alter or facilitate perception.
Language in the Chomskyian sense cannot and
does not ‘function’ in a way that can ‘alter or facilitate perception’ – unless
one defines ‘function’ as the Saussurean parole.
If so, Heurley et al. make uncontroversial observations about the use of
language.
Next in line is ‘Digitized
Fossil Brains: Neocorticalization’ by Harry J. Jerison who thankfully
asserts even in his abstract that he is writing about ‘the evolution of
language as a hominin specialization’, thus saving us from the assumption that
language is an amalgam of various motor and cognitive skills. The author
provides evidence for the growth of the neocortex in primates and humans and
speculates that in the ultimately much larger human brain, it was the neocortex
that provided the blueprint for human language. This article shows both how
much we know about the brain and its evolution and how little – if next to
nothing – we know about the physical evolution of language. My guess is that
neurolinguistics will never rise above the basic tenets of biolinguistics,
leaving us with the black box best explained with mental representations that
have logical links with laws of nature.
Following the paper by Gentilucci et al. on ‘gesture and speech’ it
is perhaps not surprising that this is followed up by Manuela Macedonia &
Katharina von Kriegstein on the topic of ‘Gestures Enhance Foreign Language
Learning’. On the surface of it this seems an uncontroversial claim even though
one generally associates gestures much more with the production of speech
rather than with the learning of a foreign language. A bit more mysterious is
their ‘proposal’ in the abstract that they ‘propose the use of gesture as a
facilitating educational tool that integrates body and mind’. Does the
disembodied ‘mind’ arise again? The funniest comment is then the authors’
assertion that ‘gestures accompanying foreign language items enhance their memorability and delay their forgetting’. First, what are
‘language items’ and second, what on earth is a ‘delay in forgetting’? I always
thought that ‘learning’ is indeed predicated on ‘not forgetting’ – hence a
failure to learn is equal to forgetting. Anyway it is very nice of the authors
to suggest ways and means to ‘delay’ the failure to learn. On a more serious
angle it has been known for ever that various likely and unlikely associations
aid memory, e.g. learning of vocabulary of a foreign language, and as explained
by the authors as ‘over the past three decades, laboratory research has shown
that action words or phrases such as cut
the bread are memorized better if learners perform or pantomime the action
during learning than if they only hear and/or read the words’. It would have
been better to try to explain how this works for actual foreign language
acquisition (as it sure works for native language acquisition), for if I am
learning a foreign language like German and the near equivalent phrase to learn
is schneide das Brot, how does it
help if I mime the action which I already have in my memory from my native
language? Maybe the Germans cut bread differently? The point I am trying to
make that in this instance the accompanying gesture may or may not facilitate
the memory up-take of the foreign phrase. The use of flashcards is widely known
in teaching reading/writing to native speakers but the use of the same aid is
questionable in its effectiveness for foreign language learners, precisely
because the visual image – like a gesture – is already in the memory bank. If
anything we need to disassociate our image cum
gesture if needed as a memory aid: perhaps a bizarre image/gesture will do the
trick better! When Macedonia and Kriegstein discuss the actual strategy of ‘the
body as a learning tool’ in foreign language teaching/learning, they quite
bizarrely accuse Chomsky and Co. as having hindered its development due to:
theories based on a universal grammar
(Chomsky 1959) considered language learning to be an innate process (Fodor et
al. 1974; Chomsky 1975). Accordingly, like mother tongue acquisition, foreign
language was thought to emerge by mere listening and without tools of instruction
because it results from innate processes (Feyten 1991; Krashen 2000). Explicit
explanation and vocabulary teaching by any means, and therefore, also by
action, were considered superfluous. Although there were other opinions in the
field sustaining that child language acquisition and adult foreign language learning
are fundamentally different (Bley-Vroman 1990), the mainstream followed the
mentalistic view of a core grammar present in the learners’ minds. This view
implicitly ruled out the body as a possible learning device …
The authors demonstrate a very poor
understanding of the ‘innate’ acquisition process because the ‘mere listening’
was in fact the ‘poverty of stimulus’ that led Chomsky and Co. to assert that a
native language cannot be leant by input but is acquired via an innate Language
Acquisition Device (LAD). There is also the perennial misunderstanding of
‘universal grammar’ as a ‘mentalist’ construct that has no basis in the biological
make-up of the human species: the very notion of ‘biolinguistics’ means that
language resides in the brain and that any descriptions and explanations
thereof must be constrained by what we know about biological and computational
systems in the brain. Nobody from UG/MP/biolinguistics has ever ‘ruled out the
body as a possible learning device’ whatever the limited merits of the strategy
may be. When I play ‘Simon says’ with my ESOL class, I do get some mileage out
of practising the English names for body parts and movements thereof but I am
equally aware that various proficiency levels will treat the input (language
and body part gestures) differently: beginners will use the translation method
and the combined language+gesture may well aid the speed of translation but may
equally hinder it if the gestured association in their native language is at
odds with what I do, i.e. point to the ‘forehead’ and the German learner of
English (at beginner level) will latch on the ‘head’ and correctly translate it
as ‘Kopf’, thinking I mean the ‘head’ as opposed to the German ‘Stirn’. Such lexical conundrums need
careful ‘verbal’ explanations rather than vaguely pointing in the direction of
the desired object. There is also the problem of the use of gestures as an
idiosyncratic device, or even as a characterization of certain extrovert people
who ‘speak with their hands’, and on top of all that there is the well know
problem of cultural differences in body-language. Even so nobody would argue
against using gestures as a learning aid, and as such it is quite a bizarre
claim by Macedonia & von Kriegstein that their proposal is somehow a new
idea arising from the supposedly new paradigm of the ‘embodiment of language’.
The next article entitled ‘Bidirectional
Influences of Emotion and Action
in Evaluation of Emotionally-Connoted
Words’ by Audrey Milhau, Thibaut Brouillet, Loïc Heurley & Denis Brouillet
again suffers from the this strange assumption that until the recent advent of
the ‘embodiment of language’ campaign everyone else suffered from the delusion
that language and cognition were ‘disembodied’ phenomena, something like ghosts
and spirits dreamt up by religious fundamentalists. In their programmatic
abstract they say that
… the
bidirectional character of influences between language and action will be
addressed in both behavioral and neuropsychological studies, illustrated by the specific
case of emotionally-connoted language. These reciprocal effects are grounded on
the motor correspondence between action and the motor dimension of
language, emerging from a diversity of source such as adaptive motivation, past
experiences, body specificities, or motor fluency.
On one hand one can only agree with the
authors on the uncontroversial claim that ‘emotionally-connoted language’ is
somehow ‘grounded on the motor correspondence between action and the motor
dimension of language’ but on the other hand the authors seem to hark back to
the bad old times of ‘behaviour’ and ‘neuropsychology’. Do we get a new
equation of ‘behaviour = language’ à la Skinner? Such an approach was famously
debunked by Chomsky as a potentially dangerous if not neo-fascist attempt to
ground language as manipulative action - as so enthusiastically embraced by the
advertising industry and all Orwellian state-run propaganda departments. To
employ ‘neuropsychology’ as an intermediary between ‘neurology’ and
‘psychology’ merely attests the problem at hand: how does ‘psyche’ arise from
the ‘neurons’ and how does the proverbial mind arise from the brain? The term ‘biolinguistics’
is also an intermediary between ‘biology’ and ‘linguistics’ and attests to the
Chomskyian research project to explain the latter by the former – and not the
former by the latter as a retrograde project by the likes of Skinner. When it
comes down to the substance of the authors proposals we also see some extreme
claims, like ‘one major contribution of embodied approaches is the redefinition
of memory as a memory of processes and no longer a memory of content’. It
would be wholly uneconomical to claim that lexical memory is purely process
based, i.e. I remember the term ‘hammer’ only via the re-enactment of hitting
myself on the thumb with it and I uttering the emotionally-connoted language
“effing shit!”. Indeed one should correctly assume that ‘process’ is impossible
without ‘content’. That the authors travel down the dangerous road of
‘behaviour = language’ is further evidenced by their descriptions of
‘motivation’ quoting Elliot as follows:
Elliot (2006:
112) explains: Positively evaluated stimuli are inherently associated with an
approach orientation to bring or keep the stimuli close to the organism
(literally or figuratively), whereas negatively evaluated stimuli are
inherently associated with an avoidance orientation to push or keep the stimuli
away from the organism (literally or figuratively).
Such stimulus-response explanations are
fine for the Pavlovian dog but fail miserably for language (as per Chomsky
above). The authors cite many so-called research findings, some of which are
laughable as being evidence for behaviour and language being interlinked on a
one-way street, like this one:
They
were listening to an auditory message explaining this reform, and while listening,
they had
to move their head either horizontally or vertically, under the cover
story of judging the quality
of headphones. Results showed that participants
who had shaken their head vertically were more
convinced by the message than
the ones who had shaken their head horizontally.
This is the
advertiser’s dream come true: subliminal messages will sell the product, like
it or not. There is no doubt that people always try to manipulate other people
by devious means if the message itself is a load of nonsense, like ‘this car is
your dream come true’ because the beautiful blond in the ad shook her head
vertically. Where and what exactly is the loaded and embodied ‘re-enactment’
when the lovers say to each other “I love you”? What emotional highs did
Einstein achieve when he figured out
E = mc2?
To the credit of the authors, they do advocate a ‘bi-directional’ association
between ‘embodiment’ and language inasmuch language can trigger the associated
bodily process. I suppose they would agree that psychosomatic diseases are as
real as the opposite effect of the 60s slogan ‘make love not war’ can have: by
saying it first you can actually find a way of putting it into action. Sadly,
as we all know, the world by and large still operates on the exact,
psychosomatic opposite - make war not love – (distorting language along the way
of Orwellian newspeak) because not enough people seem to have access to the
intelligent political messages by the likes of Chomsky and other syndicalist
anarchists. Hopefully Milhau et al. will from now on proceed in the same
direction, biolinguistically and otherwise.
For the next
article I’ll declare my bias first: I am anti-vivisection and for
animal-rights; I am a vegetarian whenever I have the choice, hence will not go
to war against meat-eaters. As such I don’t really know what the point is of ‘The
Human-Fostered Gorilla Koko Shows Breath Control in Play with Wind Instruments’
by Marcus Perlman, Francine G. Patterson & Ronald H. Cohn, especially as I
also dislike the practice to keep animals in captivity to teach them stupid
tricks (I fully endorse Heathcote Williams for his passionate treatises on
elephants and dolphins, not to speak of pathetic royal blood sports and
President Obama killing a poor old fly with great relish). I do have some
admiration for people like Jane Goodall – who is quoted in the text – who spend
a life-time studying animal behaviour, thus providing insights of animals are
all about. The authors quite bizarrely accuse her of not getting it right when
she observed that ‘even Jane Goodall, after many years observing the chimpanzees
at the Gombe Reserve, came to the conclusion that, “the production of sound in
the absence of the appropriate emotional state seems to be an almost impossible
task for a chimpanzee”, especially as the authors set out to prove that
‘human-fostered’ Koko the Gorilla does have breath control and thus can utter
sounds at will. They even cite the detestable practice of humans giving
primates cigarettes to smoke – as evidence of breath control. The authors then
jump to the conclusion as some others do in this special edition, in that
‘speech’ equates language, and since speech requires breath control, the
primates, especially human-fostered ones, may well have been or are on the way
to acquire speech and language. This is of course total nonsense. The only amusing
point the authors make is to maintain that developing breath control is
adaptive for its flexibility, rather than for any function (or set of
functions) in particular’ (as claimed by Fitch). In other words, evolutionary adaptation
can be for fun (like playing wind instruments) and profit, not profit alone.
With the next
paper entitled ‘Three Ways to Bridge the Gap between Perception and Action, and
Language’ the author Jean-Luc Petit announces his grand scheme in the abstract
as to ‘assess the remaining distance from neuroscience to a science of
language’. I am a bit puzzled by the word ‘remaining’ as it seems to suggest
that there is only a small gap to close. I’d say that this is a wide chasm that
may never be crossed, reminding me of a Leonard Cohen lyric that says something
like ‘we’ve burnt all the bridges so now we don’t have to cross them ever
again’. By way of a bit of extended metaphor we submit is that some linguists
burn the bridges in their enthusiasm to cross them, setting themselves up for a
Sisyphusian task. Petit
doesn’t make it easy for himself either as he demands a three-way highway
across the ‘remaining’ divide, although I have no real idea what he means by ‘from
an eidetic standpoint, one must build the transition between perceptive,
pragmatic and semantic morphologies’. Does he mean that we have a photographic
memory for embodied action sequences which we then translate into and/or
associate with language? What does he mean by ‘morphologies’? Forms in general
or in the narrow context of linguistics? His next question is more to the point
but equally mysterious: ‘from the point of view of subjective experience, one
must understand how it is possible that we move from our sensory and
kinaesthetic experiences to verbal expressions of a
sense that could be shared by others.’ One hopes that Petit concurs with
his friends Milhau et al. in that there is a
bi-directional relationship between language and experience inasmuch we use
language to make sense of our ‘sensory and kinaesthetic experiences’ – rather than the other way round.
Indeed one may answer his question in this way: it is through the language
capacity that we share as humans that we succeed in communicating our
idiosyncratic experiences in a way that is comprehensible to others. Of course
this doesn’t answer the question how this works at the level of neuroscience
nor at the level of biolinguistics, nor at the level of psychology. I am also
worried by Petit’s use of the term ‘verbal behaviour’ which seems to locate his
paradigm in the realm of behaviourism I so decried above.
In his main text he seems to spend quite some time explaining how the
current paradigm of ‘embodiment’ has saved us from the mentalese ‘disembodied’
theories of cognition, again forgetting that biolinguistics has done this job
already and, in my view, a whole lot better than the ‘embodiment of language’
scenario. Petit, as some authors have done above, again cites Chomsky as
somehow being responsible for all the ‘mentalese’ claptrap we have had to
suffer under. Accordingly Chomsky’s ‘strict’ distinction between ‘competence’
and ‘performance’ led to the postulation of a ‘brain-machine’ that is
‘indifferent to its program’. In other words, competence being indifferent to
performance. Petit calls this an ‘ideology’ and goes on to say that ‘a recent
alternative to this ideology, the identification of linguistic information
processing with neural dynamics itself and its laws of association is yet
another form of embodiment of language (cf. Pulvermüller 2002)’. At least Petit
seems to concede that Chomsky’s ‘brain-machine’ is also somehow embodied. The
purported progress seems to be the integration of competence and performance in
this ‘yet another form of embodiment of language’. I think the point Petit
misses is that the ‘brain-machine = competence’ is in fact ‘indifferent’ to the
‘program = performance’, for as by computer analogy the hardware is of course
designed with the potential of being programmed but ultimately the hardware is
totally indifferent to the actual program installed on it. Even if this is
splitting hairs, there can be no doubt that some computer scientists
concentrate on hardware while others specialize in software. Chomsky never said
that ‘performance’ wasn’t worth studying. He simply concentrated on
‘competence’ as a stand-alone module, and rightly so. Petit repeats the old jealousy
of applied scientists complaining about the arrogance of theoreticians who pay
scant attention to the practical applications of their research.
Petit as a philosopher then poses the
question if we are now stuck in a kind of limbo in ‘that everything happens as
if current neuroscience sought a basically inadequate substitute for this
phenomenology in authors who hesitate between behaviorism and cognitivism,
between mentalism and physicalism, between computation and simulation’? It
seems we can solve this mind-body dichotomy only via ‘Merleau-Ponty, the One
Acceptable Phenomenologist’ as he alone seems to build a bridge between body
and mind, perhaps as a kind of Nietzschean ‘beyond good and evil’ concept. All
this philosophizing, I must say, seems to lead down the proverbial garden path,
and whilst, for once, I do not necessarily share Chomsky’s point on such
matters, namely that French post-structuralists/deconstructionalists/etc. tend
to produce a lot of incomprehensible verbiage, I do get the feeling that
Monsieur Petit does on occasion become so dense in his prose so as to become
obtuse. Petit quoting Husserl in German – presumably as a sign of polyglot
sophistication amongst clever philosophers - does the same for me, e.g. ‘Der erste und einfachste Ausdruck
ist der des leiblichen Aussehens als Menschenleib, er setzt natürlich „Sehende”
und “verstehende” voraus” …’ (BTW “verstehende” should be quoted properly with
a capital “V”, denoting a noun like “Sehende”), namely either stating the
obvious by means of unnecessarily complex sentence structure, or else stating
something that is incomprehensible. I do like sophisticated language play in
the style of Nietzsche but am sorry to say that I find Husserl lacking in that
department. Petit does of course credit ‘Husserl’s overcoming of a prior
Cartesian solipsism which posed communication as inessential to thought
promoted body expression to the status of linguistic expression and his subsequent
foundation of subjective experience in intersubjectivity involved the founding of
expression in communication’ thereby giving the Cartesians like Chomsky a bad
name. Again a bizarre notion when we assume we are all engaged in something
called ‘biolinguistics’. When coming back to neuroscience and/or
neurolinguistics Petit is far too optimistic in stating that ‘for the first
time in history of the knowledge of man we see on the basis of data of
empirical research a possibility to trace the uninterrupted course of events
inside the organism that goes from perception and action to communication
through language’ – we are nowhere near that possibility. To cite a few
neurological ‘embodied’ processes that seem to have parallels in language are
far too vague to account for language per se. Petit again brings Husserl to the
rescue by proposing that morphology, syntax and semantics give rise to a sort
of ‘incompleteness-dependence’
grammar that elevates ‘nothing’ to a mysterious binding force between
morphemes, words and phrases, resulting in a sentence with ‘meaning’. That certain
prepositions and spatio-temporal deixis in general can ‘be reconstructed by
equations of differential geometry’, as suggested by Petit, is nothing new in
cognitive linguistics where this is used as evidence of the embodiment of
language. The neuronal simulation of such processes AND its translation into
human language is the crux of the matter – the point being that according to
Chomskyan linguistics at least, it is not the neuronal processes of perception
and action that give rise to language but that there is a human-specific
language capacity in the brain that puts all this and so much more into words,
phrases and sentences.
In the end Petit concedes that all this ‘is
a bet made by a neuroscience of language that would aspire to naturalize our
phenomenological experience of meaning’ – a ‘risky’ business. Using German
metaphors, he says that Lebenswelt is
far more wide-ranging than Einfühlung
(empathy via mirror neurons), hence speech/language conveys far more than
bodily simulated functions. Even basic social speech acts cannot be accounted
by the ‘embodiment of language’.
In conclusion Petit reverts back to an
extended dualism of body and mind, indeed a ‘trinitarian’ approach (no doubt to
be ridiculed as a ‘holy trinity’):
(i)
neurophysiologic investigation
of the organic substrate of the continuous linkage between perception and
action, and language;
(ii)
eidetic-geometric
morphodynamics as norm a priori backing the transformation of forms/schemes in
syntactic or semantic structures; and
(iii)
transcendental constitution of
the Lebenswelt of a community of perceiving-acting personal subjects who
interact by words and
gestures drawing
on bodily capabilities and other operations of meaning-giving.
Petit is
apologetic by imposing item (iii) for the imposition is a phenomenologist
philosopher’s one, treading on the toes of the hardened empirical scientist who
have already nailed item (i) and are supposedly on the way to crack item (ii) –
none of which is true, as far as I can see. The seeming incompatibility between
this tripartite assembly is further excused by Petit as arising from the
philosophers’ Erlebnis (italics added
by me) which he explains as ‘the lived experience of an unresolved tension
between ultimately possibly incompatible approaches which nonetheless impose
themselves as contingent context of the quest for truth’. As a native speaker
of German I am forever hopeful that the Germans will eventually fulfil their
role as Dichter and Denker, hence I do appreciate the French
philosopher’s predilection for German Zeitgeist
terminology but I’m damned to know what Erlebnis
has to do with the price of fish or biolinguistics for that matter, other
than Husserl using this fairly common word occasionally in his treatises. Sure,
for the purpose of translation, languages are littered with lexical gaps, and
occasionally one may elevate a word to a technical term used in the target
language so as to avoid to have to use lengthy paraphrases – Erlebnis seems highly questionable for
this purpose as it is simply translated as ‘lived experience’ as indeed Petit
does in the first place. His added ‘philosophical’ meaning is as idiosyncratic
as anything I have ever read, and as far as I can determine, lacks actual
meaning. Since when do scientists and/or philosophers tell the truth about
language?
Having spent
quite some time on critiquing this paper I must admit that it was the most
enjoyable, if quixotical, read so far.
The penultimate article by Claudia Repetto,
Barbara Colombo & Giuseppe Riva, entitled ‘The Link between Action and
Language: Recent Findings and Future Perspectives’ again opens with a bizarre
anti-Chomsky statement that assigns him to the dustbin of linguistic history.
It is worth quoting the authors’ introduction as it reveals their bias:
Traditional
theories of cognition are based on the idea that knowledge is represented in
the brain
in the form of concepts and stored in memory system as semantic
information. Concepts, from
this perspective, are conceived as amodal, abstract
and arbitrary (Fodor 1975), then independent
from the brain’s modal system of
perception (e.g., vision, audition), and action (e.g., movement,
proprioception). Chomsky’s theory of language (Chomsky 1965) is completely
aligned with this
view: The theory of Universal Grammar considers language as a
corpus of abstract symbols
combined together according to formal syntactic
rules; two properties, among others, are
distinctive of human language, the
generativity and compositionality.
In more recent
years, nevertheless, a radically different conception of knowledge has
been taken
into account, that brings together data from different
methodological approaches such as
neurobiology, brain imaging, and neuropsychology:
the theory of Embodied Cognition (Wilson
2002; Gibbs 2006). According to the
embodied cognition hypothesis, concepts are not amodal
and knowledge relies on
body states and experiences. Therefore, there is a tight link between
concepts,
action, and perception, to the extent that conceptual knowledge is mapped
within the
sensory-motor system. The notion that cognition is grounded in
action and perception is
encapsulated in the term ‘embodiment’.
For a start it
shows how much the authors are out of touch with the biolinguistics program and
the Chomskyian Minimalist Program – by quoting Chomsky’s 1965 ‘Aspects’ volume,
which no doubt was a major treatise in the history of linguistics but has long
since been further elaborated as a theory of language that locates language in
the brain as the modus operandi. The implicit claim that Chomsky’s theory of
language is ‘amodal’ when in fact it should be ‘modal’ according to the authors
is a false claim. Chomsky has quite a modular conception of language even though
he is mainly interested in the syntactic mode of language. It is not entirely
clear if the authors also reject the notion of ‘generativity’ as being
distinctive of human language but if they do, they do so at their peril. For
what is the alternative? A finite set of learnt language behaviours linked to
‘concepts, action, and perception, to the extent that conceptual knowledge is
mapped within the sensory-motor system’? This leads to the simplistic, if not
fascist, behaviourism decried by Chomsky so long ago. The creative aspect of
language – its capacity to generate an infinite number of new sentences –
cannot be constrained by the sensory motor-system which in humans is often far
less sophisticated than that of other animal species. Universal Grammar never
envisaged language as a ‘corpus of abstract symbols’ in the sense of some
airy-fairy mind game. Even Wittgenstein’s game theory of language is far more
sophisticated than that of the authors’. Modern Chomskyian approaches like
binary ‘merge’ as a fundamental syntactic operation are closely aligned to
computational theories that work well for fundamental biological cum neurological processes.
As pointed out
again and again in the previous incarnations of this article, the notion that
language is also ‘grounded in action and perception’ is not disputed by anyone.
Perhaps sadly so, as the daily banality of life and death seems ever more
‘grounded in action’ of the military sort giving rise to Orwellian newspeak. Maybe
this is a human condition we will never escape even though language is creative
enough to envisage the ‘better world’ as also advocated by the likes of
Chomsky.
It seems quite
ridiculous to suggest, as the authors do, that ‘the mind is no longer confined
to the brain but also includes other body parts, such as hands, legs, eyes’
thereby claiming to have discovered something new. It is perhaps instructive to
note that ‘neurolinguistics’ once upon a time referred to the new-age idea of
‘neuro-linguistic programming’ as developed by renegade linguists Bender and
Grinder, to make a big deal out of eye-movement and how it reveals your
innermost thoughts such as hoping your kissing the frog will turn your fortunes
around and set you up with a prince of your dreams. For if motor-movement links
directly to language – as it sometimes no doubt does – we should be able to
predict by the movement as to what the subsequent thought – as expressed by
language – is. When you point the gun at me and demand that I raise my arms,
and I do so, you may well deduce that I am thinking of giving myself up. On the
other hand this may be a trick, reminiscent of the joke whereby three ex-world
leaders (say, Kohl, Blair and Bush) are condemned to death by firing squad by
Mexican rebels, and they ask for their last words, and first-up Kohl shouts
‘revolution’ whereby the rebels panic and let him escape, and second-up Blair
shouts ‘earthquake’ and again the rebels panic and he escapes, and last-up Bush
exclaims ‘fire’! The idea being in this context that language can simulate
dangerous motor-experiences that can trigger either real motor-panic or the
motor-action of pulling the trigger.
When Repetto et
al. go on to discuss the literature of current TMS studies, they attempt to
account for ‘contradictory’ data such as:
For example,
Papeo et al. (2009) reported an increase of MEPs recorded while participants
read action verbs compared with what happened while they read verbs describing
abstract concepts; in contrast, Buccino et al. (2005) described a reverse
situation during language comprehension: MEPs recorded from hand muscles was
lower while participants heard hand-related action verbs compared to
foot-related action verbs, indicating an effector specific inhibition.
Research like this is no doubt very
interesting but the conclusions drawn are far to general in terms of predicting
linguistic patterns as related to MEPs. Consider for example the ancient idea -
as espoused now by theosophical healing methods – that one can generate heat in
one’s limbs by mental control, or for example that artificial limbs can be
controlled by mental activity: these are as yet poorly understood processes and
it is quite unclear if language is involved at all. Do I simulate the limb’s
movement to grasp the egg, or do I tell myself in so many words “I will now
grasp the egg!” (or is it better to use the present tense?). That through
language I may be able to command my motor-movements may well be on the cards
to a certain extent but to claim that the reverse is true is quite illogical.
My actions may be supported by language (e.g. “heave-ho!”) but my actions do
not give rise to my language, especially when it comes to the vast realm of my
language that has absolutely nothing to do with any of my motor skills. To
explain above contradictory results, Repetto et al. revert to speculation with
regards to the many imponderables of experimental design. My son who just
completed a PhD thesis in experimental second language acquisition, just
concentrating on a single item (the acquisition of the Chinese anaphor ziji by learners of Chinese from English
and Korean backgrounds) found that issues like timing can have a significant
effect on survey responses, and the 500 ms quoted by Repetto et al. may well be
a crucial difference to how responses are encoded. Furthermore, as also somewhat
acknowledge by the authors, there are an infinite number of possibilities on
how to frame linguistic tasks, hence an infinite number of possible responses.
Of course the authors prefer to believe that the task variety can be so
severely constrained that one researcher can compare them all:
Tomasino et al.
(2008) compared systematically the effects of different timings of stimulation
during different kind of tasks (silent reading, motor imagery and frequency
judgments) and found that M1 plays a role only during motor imagery, so they
concluded that the recruitment of motor networks during language understanding
is not required, but it occurs only when explicit motor simulation is
requested.
I am sure that Tomasino, as many others,
must be aware that experimental design for language research must be as
multi-faceted as language is itself, as otherwise one reduces language to a
mono-syllabic instrument for communicative action processes (as preferred by
army instructors around the world). Even so Tomasino above found that no
conclusions could be drawn. The authors also present another example which
unwittingly perhaps demonstrates yet another contradiction:
Recently TMS
protocols have been employed to discover the role of morpho-syntactic features
on the activity of M1: Papeo and colleagues (Papeo et al. 2011) compared MEPs
recorded during reading tasks of action vs. abstract verbs presented using the
first or the third singular person (I vs he/she); they found an
increase of MEPs amplitude selectively for the action verbs at the first
person, deriving from these data that motor simulation is facilitated when the
conceptual representation of the verb includes the self as agent. Furthermore,
a sensitivity of the primary motor cortex to the polarity of sentences was
high-lighted: Active action-related sentences suppressed cortico-spinal
reactivity compared to passive action-related sentences, and either active or
passive abstract sentences (Liuzza et al. 2011).
Whilst it seems logical enough that
‘first-person’ action sequences trigger higher MEPs than those of ‘third’
persons, there must be something wrong with the observation of the second part,
namely that ‘active’ sentences ‘suppress cortico-spinal activity’ while
‘passive’ sentences do not. The authors characterise the ‘active-passive’
distinction as ‘sentence polarity’ which is a highly questionable description.
Traditionally the term ‘voice’ is used in grammar to account for this
distinction (and there are not just two voices to be accounted for either, as
is well known to grammarians). The point here is, however, that so far all the
literature seemed to point to the idea that ‘action’ in words triggers
motor-responses in the form of MEPs, so why would active sentences ‘suppress’
cortico-spinal reactivity? Did the authors mix up the results (I haven’t
checked Liuzza et al. 2011)? Or is there something weird going on here?
Jokingly I have sometimes suggested that ergative languages and those speakers
giving prominence to the marked passive voice in English are more sympathetic
people than those action-driven maniacs who speak accusative (sic) languages and who currently rule the roost. Maybe the proverbial
mirror-neurons kick into action if we imagine ourselves in the role of the victim
(the ‘subject’ of a passive sentence) rather than that of the ‘actor’ (the
subject of an active sentence)? Sadly, as we know, speakers of ergative
languages can be as ‘active’ and violent as anybody from the accusative
languages. To prove my speculative assertion we could design an interesting
experiment!
Indeed ‘mirror-neurons’ are the next topic
in Repetto et al.’s paper. This whole idea about sympathetic/empathetic
mirror-neurons is quite fascinating, inasmuch we can read each others minds and
complete half-finished sentences on behalf of our interlocutors, not to speak
of body-language and smiles and tears and the like. To be able to simulate and
replicate an action or skill in our brain – encoded in addition in language
perhaps – seems to be the domain of higher animals, and perhaps the ability to
learn in the abstract is only in the human domain, presumably as it can be
reinforced or is driven in the first place via our language capacity. It is of
course far too early to jump to any conclusions, for neither the brain nor
language – that arises from it – is well understood, as in fact acknowledged in
part by the authors:
Nevertheless,
it is noteworthy that there is not a strong consensus about a
somatotopic
organisation of action words meaning representations, and this fact is
not astounding
considering that the organization of the premotor cortex
is still poorly understood.
For that reason one has to be sort of thankful
for any research results even if they only confirm what anyone could have predicted,
as for example the following observation:
“if
understanding action words involves mentally simulating one’s own actions, then
the neurocognitive representation of word meanings should differ for people
with different kinds of bodies, who perform actions in systematically different
ways” (i.e. right- vs. left-handers): This prediction has been corroborated by
fMRI data which showed a preferential activation of the right premotor cortex
during lexical decision on action verbs for left-handers, and the opposite
pattern of activation for the right-handers.
Maybe at this juncture it is worthwhile to
cite Chomsky again, namely the idea that ‘intuitions’ and deductive reasoning
about language – especially when one is a linguist of the calibre of Chomsky –
should be taken at face value. As such Chomsky related the story of his
erstwhile thesis topic on Hebrew which his then mentor Zellig Harris imposed on
him, requiring him to do fieldwork with Hebrew speakers, only to realize that
he himself (as a speaker of Hebrew) had already all the answers without having
them to be elicited from others. In other words I don’t need to do empirical
research on what is glaringly obvious. One might include in this category the
above topic on right-left handiwork as well.
In the final part of this paper Repetto et.
al suggest that Virtual Reality (VR) scenarios could be great research tools in
determining the relationship between language and action. After all we can
simulate just about any action as VR, from flight simulators to ‘shoot’em’
video games that have now become reality in drone warfare. Indeed the authors
make a rather coy reference to this fact:
Thanks to
different input devices participants could virtually perform any action, even
those typically not performable in an experimental setting (to jump a rope,
kick a ball, or shoot something, for example).
There have already been reports of drone
war-fare operators experiencing the same or similar ‘feelings’ as real soldiers
in combat, even using the same (primitive) language that is associated with
killing the enemy. No doubt the US Army is already providing research grants to
find out what exactly the relationship is between action and language in these
VR contexts. One hopes that Repetto et. al will resist the temptation – which
even MIT couldn’t resist and still doesn’t resist, despite of Chomsky having joined
RESIST a long time ago.
Whilst it is obvious what I think of such
VR scenarios, I will not deny that some VR applications may indeed be useful
research tools in biolinguistics. I seriously doubt however that Repetto et. al
have thereby opened up some new and amazing avenue for research, for VR never
really lives up to the real thing. The idea of remote controlled robots - like
the American drones, and more benignly medical-surgical robots – are not
science fiction anymore but at the same time add absolutely nothing to our
understanding of language. Neither does pure VR.
The last article in this special volume is
by Nicholas Unwin, entitled ‘The Language of Colour: Neurology and the
Ineffable’. I suppose this belongs to the category of solving some sort of
paradox, like Derida’s famous question if the unforgivable may be forgivable –
thus asking if the ‘ineffable’ can be effable? When one compares this article
to the other one above on colour, namely the one by Loïc P. Heurley et al., one
is struck by the latter’s assumption that the language of colour is indeed
linked to neurological events, simulating real colour perception (the
physiology of it), while Unwin revisits the old conundrum, asking if there is a
connection at all, calling it a ‘body-mind’ problem. Since Unwin makes a
convincing case for the ‘embodiment’ of the language of colour, one might as
well congratulate him and leave it that – for there is no real argument to the
contrary, however much Unwin seems to do battle with it. To maintain that the
language of colour (and therefore language itself) is living proof of cultural
– if not cognitive – relativity is of course still a hot topic for the
relativists but hardly one of interest to bio-linguistics where the matter has
been settled ever since Lenneberg (as mentioned in my review of the first
article on colour). Unwin really makes his life more complicated than need be:
his examination of colour-related terminology like as ‘warm, cool, sharp, fresh
and citrusy’ as needing to be confirmed by neurological processes:
I
shall argue that an ideal sort of explanation of why red should look warm is that there be
some
appropriate neurological connections between the visual and tactile
parts of the brain (currently,
the issue is undecided).
Unwin questions the common assumption that
‘red’ is ‘warm’ because of its association with fire when a gas-flame is blue
(the colour of water and ice, hence a cold colour). If the history of language
is parallel to the history of homo sapiens we can assume that the language of
colour was once of the earliest feats of encoding natural phenomena. That a
burning wood fire is red-hot cannot be some vague metaphor alone: it is a fact
deeply ingrained in our consciousness. The relatively recent discovery of
natural gas yielding a blue flame does in no way detract from ‘red’ being a
warm colour. Of course humans and human language change, so why not the
language of colour? Maybe in a thousand years red will have lost all its
association with a burning wood fire – the blue flame having taken over. Quite
obviously there are myriads of possible colour associations, both idiosyncratic
and ecologically determined – and I don’t doubt cultural influences such as
fashion colours – and as such it seems rather futile to want to track down all
the neurological processes in terms of bodily simulation and/or perception. In
fact what Unwin demonstrates is the state of knowledge we have in terms of how
the brain generates language, namely very much in its infancy if anything at
all. That the brain generates language is inescapable – it’s a scientific fact.
Chomsky revolutionized linguistics by suggesting some clever syntax formulae
that can generate language, in analogy to what we know the very basic
biological principles to be. To be really dramatic: Chomsky delivered us from
the past and current intellectual darkness that maintains that the mind – hence
language – is a phenomenon somehow unconnected to the living brain of humans.
The embodiment of language being a new
research trend in linguistics, as claimed in this special issue, is greatly
exaggerated, especially when quite a few authors state that Chomsky and his
school of linguistics are somehow in opposition to it. That language is
embodied is simply a catch-phrase that is easily subsumed in the term of
bio-linguistics. The idea that we are at the stage of pin-pointing neurological
processes in regards to language is also greatly exaggerated by research using
neuro-imaging – however advanced the techniques may be – in association with
language stimuli, especially as this type of research seems to fall back to an
outdated – if not perverse -
behaviourist model of language. At this stage neuro-imaging is still a
very crude instrument to investigate language with, especially as we hardly
know what the basics are of our motor processing skills. To neurologically test
for ‘verbs of action’, ‘negation’,
‘false beliefs’, ‘the language of colour’ and the like (as exemplified in this
issue) seems light-years away from yielding meaningful results vis-à-vis
wholesale claims about language and its syntax. There is nothing wrong with
educated guesses, theories and even speculation in the absence of empirical certainty:
it makes for interesting reading. To jump to vast conclusions in the face of a
paucity of meaningful data, on the other hand, makes a reader like me irritated
and dismissive of the content of this Special Issue on the so-called Embodiment
of Language.