As our most powerful tool, language not only allows for communication and reading and writing but also for thinking. Languages are hierarchical in that they have phrases, clauses, sentences, etc. Additionally, languages have syntax (grammar), which gives sentences structure. Ants, Bees, and Birds all have primitive languages; pointing and screeching are parts of a language and even emotions are a language. In the 100,000 years of Homo sapiens existence there was no higher thought until the emergence of a complex language occurred, which may have arisen 50,000 years ago, as this time marks the emergence of symbolic representations such as dolls and cave paintings.
Written language, however, only emerged just over 5,000 years ago when Egyptians wrote pictorial hieroglyphics focusing on images of animals, outward anatomy, tools, etc. The meaning of hieroglyphics remained a mystery until one of the greatest finds in all of archaeological history: the famed Rosetta stone was discovered in 1799 by a Napoleonic soldier in the Nile River Delta, and has three sections; Egyptian hieroglyphics, Egyptian Demotic script, and ancient Greek. The ancient Greek allowed the deciphering of the Egyptian hieroglyphics.
I once asked my mother, a poet, English professor, feminist, and lesbian, ‘Who was the best writer?’ She didn’t hesitate, ‘Shakespeare.’ Have we reached the pinnacle in writing? Some think so, but I am reminded that until 1905, for three hundred years scientists thought Newtonian mechanics was the last word, but then came around a 26-year-old upstart named Einstein. Elaboration of Quantum Mechanics would follow and yet there still remains a mystery over the merging of general relativity with quantum theory. So, keep on writing and keep on thinking. Now back to the language of Shakespeare, Yeats, and Woolf.
There are two language areas in the human brain that have received the most attention; the first is Broca’s area, discovered by the French physician Pierre Paul Broca who first recognized the condition known as Broca’s aphasia in 1861. This area when damaged, by a stroke for example, leaves the patient’s speech and grammatical systems severely impaired as well as their use of verbs. Broca aphasiacs still generally retain good comprehension, with the damaged areas located in the left frontal cortex. The second major language area, located in the left temporal lobe, was identified by the German neurologist Carl Wernicke who published his first work on aphasia in 1874. This type of aphasia is characterized by impaired comprehension of spoken words and sentences, while the production of speech may not be very affected. Patients have normal fluency and prosody, but sadly they speak fluent gibberish. The left hemisphere is the dominant hemisphere for language and logic, while the right hemisphere is known as the emotive and creative side.
The question at hand is how did a complex language, complete with a 100,000-word lexicon (vocabulary), and meaning system, as well as our grammar (syntactic) system, evolve from grunts, shrieks and cries? Clearly, we did not evolve in a single step from a grunt to Hamlet, although some leading linguists believe our language abilities did emerge from just such a leap; in a single bound as it were. According to American Noam Chomsky, language is not a product of culture or learning, but a biological feature of the human mind. Chomsky's universal grammar theory implies that language is not a result of natural selection, but a sudden and mysterious emergence in human evolution. He suggests that language may have originated from a single genetic mutation, giving rise to the capacity for complex and creative thought. According to Chomsky, universal grammar is a biological component of the language faculty that allows children to acquire any natural language with minimal input from their environment. Chomsky professes a ‘Merge’ function in which words or groups of words are merged to form grammar, e.g. [the, man] are Merged into {the man} and this Merge could not have occurred partially. That is, there is no state of half merged, hence he argues that grammar emerged in a single step. This would have occurred in Broca’s area, the grammatical area of the brain. He later goes on to write that semantic/meaning aspects of language are mysterious, which leaves a gaping hole in his one step language acquisition model.
Meaning, and associated qualia-based (experiential conscious) comprehension are the ingredients missing from AI. Once an artificially intelligent agent can actually comprehend the qualia that is being said to them, they will have achieved the singularity. It is clear that almost all animals possess some form of comprehension, so that lexical/semantic (i.e. meaning) aspects of language are ancient and evolved over time. That is, emotive warning cries, by nonhuman primates for example, carry meaning and are comprehended as such. In fact, pre-human hominids already had at least one meaning region of their brains, which they used to build complex tools and weapons. One such region of the brain, Wernicke’s, could then be hijacked for a language system, as Broca’s area was later selected for grammar. How qualia-based comprehension occurs is unknown and is one of the most important questions remaining unanswered by science.
Evolution via Punctuated Equilibrium, which American Stephen Jay Gould professed, supposedly occurs after long periods of stasis, say four million years, followed by sudden emergence of new species. In such a scenario the acquisition of language might occur quite rapidly. In favor of Punctuated Equilibrium, over Natural Selection, there are few intermediate forms in the fossil record, whereas Darwin’s gradualism predicted there would be such intermediates. However, not surprisingly, soft tissue does not usually survive eons to make fossils, and most evolution takes place within soft tissue. The human brain most likely underwent Darwinian Natural Selection during those one hundred thousand years without an increase in skull size. Gould proposed that language evolved not for communication but for thought. Gould, somewhat surprisingly, criticized the idea that language had a single origin or a universal grammar, and suggested that linguistic diversity was a result of historical and cultural factors.
Indian neurobiologist V.S. Ramachandran raised an important point about the region in the frontal cortex that houses Broca’s area; it’s rich in mirror neurons. Mirror neurons let us mimic others’ actions as the name states. He argued that the motor areas for the use of tools, for example, could be mimicked by the Broca’s region into step wise grammar evolution. The use of a tool is a grammar in itself; ‘tool user’ (subject) ‘cut’ (verb) ‘twig’ (object). This leaves it to the relatively recent acquisition of Broca’s syntactical (grammar) area for pairing with Wernicke’s area (comprehension), which must have been in place prior to humans. Inner speech has been implicated to Broca's area, and thinking can be seen as a motor activity.
Colombian born neurobiologist Alfredo Ardila argued “that the modern syntacticized language and the development of metacognitive executive functions are simply two sides of the same coin.” That is, the acquisition of grammar was the crucial step in what makes humans human. He may have been right.
A fascinating aspect of noun storage is that different categories of nouns have their own storage locales in the human brain. For example, tools are stored together in the motor cortex. Vegetables/fruits and animals also have their own separate storage locales. It is clear that by categorizing nouns by type, the unconscious brain comprehends the meaning of these words.
Comprehension is a hierarchy from single words, to clauses, onto sentences and beyond. Single word comprehension is averred to be done by the brain calling up word meaning from a lexicon, a mental dictionary. For some time, many cognitive scientists have been under the illusion that sentence comprehension occurred by the simple integration of the individual word’s lexicons. This sounds reasonable, but if you think about what goes into language, that it is laden with emotional clauses, innuendos, inferences to be gleaned that are barely visible on its surface it seems highly unlikely that a complex language is just an amalgamation of dictionary entries. In the human brain, besides the left temporal lobe-Wernicke’s area, there are two top-down frontal cortical regions that are also implicated to be critical for sentence comprehension. The combined network gives the appearance of an assembly line for increasingly more complex processing leading up to conscious comprehension.
With the advent of modern genomic science, the hand-waving associated with linguistic squabbles might be nearing an end. For example, an English family usually referred to as the KE family had members that presented important abnormalities in language development. This disorder was associated with a mutation in the FOXP2 gene. FOXP2 plays important roles in brain development, including the growth of nerve cells (neurons) and the transmission of signals between them. It is also involved in synaptic plasticity, which is the ability of connections between neurons (synapses) to change and adapt to experience over time. Synaptic plasticity is necessary for learning and memory. Mutations in other genes such as ERC1 and BCL11A, have been identified in language pathologies. Research has identified FOXP1 and TBR1, their target genes, such as CNTNAP2, while additional candidate genes such as ROBO1, ROBO2, and KIAA0319 have all been associated with human language deficits. It is clear that the one mutation, sudden emergence of complex language of Chomsky seems unlikely.
Over twenty years ago, I had the insight that the human brain operated under evolutionary control. That is, mechanisms such as duplications, mutations, juxtapositions, inversions, etc. along with feedback selection were at the heart of neuropathways, and higher cognitive function. These evolutionary moves operated via the rapid changing of synaptic strengths and even the turnover of synapses in the brain. Some 18 years later, Belgian Luc Steels and Hungarian Eörs Szathmáry published a paper on “The Evolutionary Dynamics of Language" where they propose that language evolves to “relate meaning with form through the intermediary of syntactic and semantic categories” thereby comparing language function to the evolution of species and the adaptive immune system. I am certain they are correct.
Further reading:
Alfredo Ardila (2015). “A Proposed Neurological Interpretation of Language Evolution.” Behav. Neurol. Jun 1.
Robert C. Berwick and Noam Chomsky (2016). “Why Only Us; Language and Evolution.” Cambridge, MA: MIT Press.
T.G. Bever, N. Chomsky, S. Fong, M. Piattelli-Palmarini (2023). “Even Deeper Problems with Neural Network Models of Language.” Behav. Brain Sci. Dec 6; 46.
S.E. Fisher (2017). “Evolution of language: Lessons from the genome.” Psychon. Bull. Rev. Feb; 24(1):34-40.
Stephen Jay Gould (2002). “The Structure of Evolutionary Theory.” Harvard University Press.
V.S. Ramachandran (2012). “The Tell-Tale Brain.” Random House.
Luc Steels and Eörs Szathmáry (2018). “The Evolutionary Dynamics of Language.” Biosystems Volume 164, February, 128-137.
No comments:
Post a Comment