Research Ideas and Outcomes : Research Article
|
Corresponding author: Andrey Vyshedskiy (vysha@bu.edu)
Received: 25 Jul 2019 | Published: 29 Jul 2019
© 2019 Andrey Vyshedskiy
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation: Vyshedskiy A (2019) Language evolution to revolution: the leap from rich-vocabulary non-recursive communication system to recursive language 70,000 years ago was associated with acquisition of a novel component of imagination, called Prefrontal Synthesis, enabled by a mutation that slowed down the prefrontal cortex maturation simultaneously in two or more children – the Romulus and Remus hypothesis. Research Ideas and Outcomes 5: e38546. https://doi.org/10.3897/rio.5.e38546
|
There is an overwhelming archeological and genetic evidence that modern speech apparatus was acquired by hominins by 600,000 years ago. On the other hand, artifacts signifying modern imagination, such as (1) composite figurative arts, (2) bone needles with an eye, (3) construction of dwellings, and (4) elaborate burials arose not earlier than 70,000 years ago. It remains unclear (1) why there was a long gap between acquisition of modern speech apparatus and modern imagination, (2) what triggered the acquisition of modern imagination 70,000 years ago, and (3) what role language might have played in this process. Our research into evolutionary origin of modern imagination has been driven by the observation of a temporal limit for the development of a particular component of imagination. Modern children not exposed to recursive language in early childhood never acquire the type of active constructive imagination called Prefrontal Synthesis (PFS). Unlike vocabulary and grammar acquisition, which can be learned throughout one’s lifetime, there is a strong critical period for the development of PFS and individuals not exposed to recursive language in early childhood can never acquire PFS as adults. Their language will always lack understanding of spatial prepositions and recursion that depend on the PFS ability. In a similar manner, early hominins would not have been able to learn recursive language as adults and, therefore, would not be able to teach recursive language to their children. Thus, the existence of a strong critical period for PFS acquisition creates an evolutionary barrier for behavioral modernity. An evolutionary mathematical model suggests that a synergistic confluence of three events (1) a genetic mutation that extended the critical period by slowing down the prefrontal cortex development simultaneously in two or more children, (2) invention of recursive elements of language, such as spatial prepositions, by these children and (3) their dialogic communications using these recursive elements, resulted in concurrent conversion of a non-recursive communication system of their parents to recursive language and acquisition of PFS around 70,000 years ago.
Language evolution, hominin evolution, human evolution, recursive language, human language, syntactic language, modern language, Cognitive revolution, Great Leap Forward, Upper Paleolithic Revolution, Neanderthal language, neurolinguistics, neoteny
Association of Wernicke’s and Broca’s areas with language is well-known. Less common is realization that understanding of full recursive language depends on the lateral prefrontal cortex (LPFC). Wernicke’s area primarily links words with objects (
Patients with damage to the LPFC and spared Broca’s area often present with a specific PFS deficit, that affects both their language and imagination. Joaquin Fuster calls their alteration in language “prefrontal aphasia” (
Crucially, PFS disability is not just a receptive language disorder, but a deficit in imagination that can be confirmed by nonverbal IQ tests. Individuals with PFS disability may have normal full-scale IQ, but commonly exhibit a selective and catastrophic deficit in tasks relying on PFS, e.g., matrix reasoning tasks requiring integration of multiple objects (
PFS disability goes beyond problems with understanding recursive language. This is the disability of one of the mechanisms of active imagination. Nonverbal tasks requiring imagining a novel combination of two or more objects is impossible in this condition. Typical IQ test tasks involving PFS of several objects: (A) requires the combination of two objects. The top two rows of the matrix indicate the rule: “the object in the right column is the result of the combination of the two objects shown in the left and middle row” (the solution in the 5th square). (B) shows a question that relies on the PFS of four objects. (C) shows a question in which PFS of two objects has to be conducted according to the following rule specified in the top row: “the object in the middle column goes on top of the object in the left column” ( the solution in the second square). Note that patients with PFC disability commonly have no problem with simpler performance IQ tasks, such as integration of modifiers (
A deficit in imagination in otherwise unremarkable individuals is extremely counterintuitive for several reasons. First, most scientists have never met anyone with PFS disability as these individuals do not frequent university campuses and other privileged institutions. Second reason is that PFS was traditionally grouped with the other components of imagination, such as dreaming, spontaneous insight, and integration of modifiers, making it difficult to discuss the exact nature of a patient’s deficit, whose dreaming, spontaneous insight, and integration of modifiers remain normal. Finally, even when we see individuals with PFS disability, we tend to anthropomorphize, assume their normal PFS, and brush off their shortfall as a linguistic problem, memory deficit, or inattentiveness. However, when considered together with nonverbal IQ test, specifically those items that require integration of objects, there can be no mistake: as many as 18% of modern individuals exhibit PFS disability (
Definitions used in the article.
Term |
Definition |
Prefrontal Synthesis (PFS) |
PFS is the type of imagination that involves conscious, purposeful process of synthesizing novel mental images from two or more objects stored in memory |
Active imagination |
Active imagination includes several neurologically distinct components, such as PFS, Prefrontal Analysis, integration of modifiers and mental rotation. Active imagination completely depends on the LPFC. Active imagination is contrasted to spontaneous imagination, that includes dreaming, amodal completion, and spontaneous insight and does not depend on the LPFC. |
Recursive language |
Recursive language relies on a listener’s PFS ability, to communicate infinite number of novel images with high fidelity. The hallmarks of recursive languages are spatial prepositions, verb tenses, nesting, and other recursive elements that facilitate language ability to flexibly describe various combinations of objects. |
Non-recursive communication system |
Before acquisition of PFS, our ancestors could not mentally arrange objects into novel combinations and therefore their communication system could not include spatial prepositions or recursion. While their non-recursive communication system probably had many words, it was essentially finite as it could not communicate as many novel images as a recursive language. |
PFS is defined neurobiologically as the conscious, purposeful process of synthesizing novel mental images from two or more objects stored in memory. There is no linguistically defined process that is neurobiologically equivalent to this distinct mechanism of imagination. The closest in spirit is Chomskyan Merge, defined as a process of combining any two syntactic objects to create a new one (
Even when language is used to direct PFS in the mind of interlocutor, PFS definition is different from Merge. For example, combination of an adjective and a noun is a Merge operation, but does not fall under PFS that must always involve combination of two or more independent objects. Furthermore, PFS, but not Merge, requires creating a novel mental hybrid object or scene. For example, a sentence ‘ship sinks’ can be understood by remembering a previously seen picture of a sinking ship and thus, completely avoiding the PFS process. Under Chomskyan theory, ‘ship sinks’, however, is considered a Merge operation since the sentence merges a determiner phrase ‘ship’ and a verb ‘sinks’ to create a sentence ‘ship sinks.’
In neurobiological terms, Merge operation is defined in such a way that it utilizes all three brain language regions: Wernicke’s area that primarily links words with objects; Broca’s area that interprets the grammar and assigns words in a sentence to a grammatical group such as noun, verb, or preposition (
The difference between PFS and Merge can be also highlighted by the process of learning a new language in adulthood: when one studies German, Spanish or Italian, one learns new words (Wernicke’s area) and new grammar rules (Broca’s area), PFS, however, does not change a bit. The same PFS abilities can be used to understand German, Spanish, or Italian sentences. PFS, in other words, is universal across all modern human languages, despite different words and grammar rules.
All human languages allow high fidelity transmission of infinite number of novel images with the use of a finite number of words (here and later, words are understood as units of meaning, called sememes by linguists). The magic of using a finite number of words to communicate an infinite number of images completely depends on interlocutor’s ability to conduct PFS. When we describe a novel image (“My house is the second one on the left, just across the road from the church”), we rely on the interlocutor to use PFS in order to visualize the novel image. When we tell stories, we are often describing things that the interlocutor has never seen before (“That creature has three heads, two tails, large green eyes, and can run faster than a cheetah”) and we rely on the interlocutor to imagine the story in their mind’s eye. As Steven Pinker put it, “the speaker has a thought, makes a sound, and counts on the listener to hear the sound and to recover that thought” (
This ability of recursive language to communicate an infinite number of novel images with the use of a finite number of words depends on interlocutor’s PFS capacity and also is facilitated by spatial prepositions, verb tenses, nesting, and other common elements of grammar. Consider, for example, the exponential ability of spatial prepositions to increase the maximum number of distinct images that can be communicated with high fidelity (Fig.
The magic of using a finite number of words to communicate an infinite number of images completely depends on the PFS ability. A person lacking this ability is unable to construct novel mental images according to the rules imposed by spatial prepositions and, therefore, will not understand the meaning of spatial prepositions. In the theoretical example of a language with 1000 nouns and 100 spatial prepositions, the person with PFS disability will be limited in his/her comprehension to the 1000 nouns. Generalizing to other elements of language, we conclude that patients with PFS disability are expected to lack understanding of any recursive elements of language – the symptom reported by Fuster and Luria in prefrontal aphasia patients (
Extending this argument from a single individual to a community of individuals with PFS disability, we note that the communication system in that community must be non-recursive. Similarly, if we could envision a community of individuals who have not yet acquired the PFS ability phylogenetically, we could confidently say that they could not have understood recursive language and therefore could not have used spatial prepositions. They still could communicate, but their communication system must have been non-recursive, void of spatial prepositions and other sentences describing object combinations. Linguistics does not have an established name for such a communication system with thousands of words and no recursion. We cannot refer to it as ‘language’ in order to avoid confusion with recursive language. Accordingly, we will simply refer to it as rich-vocabulary non-recursive communication system.
PFS disability is not limited to individuals with LPFC damage. Individuals without any brain injury exhibit PFS disability if they were not exposed to recursive language in early childhood (
About 90% of all congenitally deaf children are born to hearing parents (
Isolated deaf children who grow up using homesign for communication must be distinguished from deaf children developing in a community of other deaf children, as they are known to be able to independently invent a recursive sign language of their own. In 1980, following the Sandinista revolution, the Nicaraguan government opened several vocational schools for deaf children. By 1983 there were over 400 students in the two schools. The school program emphasized spoken Spanish and lip reading, and discouraged the use of signs by teachers. The program failed and students were unable to learn the Spanish language in such a manner. However, the school provided fertile ground for deaf students to communicate with each other. In this process, children gradually spontaneously generated a new sign language, complete with syntax, verb agreement and other conventions of grammar (
Prelingual deafness is such a serious concern that the US government has enacted laws to identify affected newborns. In 1999, the US congress passed the “Newborn and Infant Hearing Screening and Intervention Act,” which gives grants to help states create hearing screening programs for newborns. Otoacoustic Emissions Testing is usually done at birth, followed by an Auditory Brain Stem Response if the Otoacoustic Emissions test results indicated possible hearing loss. Such screening allows parents to expose deaf children to a formal sign language as early as possible and therefore avoid any delay in introduction to recursive language.
Lack of communication with the use of recursive language is a big concern in children with autism (
Despite the best efforts of therapists, 30-40% of individuals diagnosed with autism spectrum disorder (ASD) experience lifelong PFS disability and the associated impairment in the ability to understand spatial prepositions, verb tenses and recursion (
The PFS ability develops in neurotypical individuals between the ages of three and five (
The next significant decline in plasticity is around the age of five. When the left hemisphere is surgically removed before the age of five (to treat cancer or epilepsy), patients often attain normal cognitive functions in adulthood (using the one remaining hemisphere). Conversely, removal of the left hemisphere after the age of five often results in significant impairment of recursive language and PFS-based cognitive skills (
Further reduction of plasticity occurs by the time of puberty; a lack of experience in recursive dialogs before puberty invariably results in PFS disability (
It is commonly accepted that childhood use of recursive language is essential for normal cognitive development (
As is the case with ontogenetic development of many neurological systems from muscle innervation to the development of all sensory systems, nature’s intent must be complemented by adequate nurture: normal development of vision requires stimulation of retina, normal development of hearing depends on auditory stimulation, normal development of somatosensory cortex is the function of tactile input, etc. What is highly unusual about the ontogenetic acquisition of PFS is that the necessary experience is provided by the exposure to a purely cultural phenomenon: dialogic communication using a recursive language.
For the normal development of vision, light reflected from surrounding objects has to reach the retina, but that occurs whenever it is light, independent of cultural exposure; for the normal development of the muscular system, the trophic factors released by muscles have to reach their neurons, but that occurs whenever a child is moving – the stimulation to neurons comes naturally even when a child is growing alone in a forest (
Two observations have a profound consequence on phylogenetic acquisition of recursive language:
Simply put, it is not enough to be fully genetically modern individual to acquire PFS, one needs to be exposed to recursive language early in childhood. This results in the proverbial ‘chicken and the egg’ problem since neither PFS nor recursive language could be acquired phylogenetically one before the other. This dependency creates an evolutionary barrier, which can be cleared only if multiple factors fall in place within a single generation. The following chapters put forward a hypothesis that resolves this conundrum by proposing that these two processes – the neurologically-based PFS and the culturally-transmitted recursive language – were acquired phylogenetically at the same time.
Once we realize that PFS is a neurologically separate component of imagination, we must then ask when PFS was phylogenetically acquired? Archeological records indicate gradual, piece-meal process of accretion of symbolic artifacts such as perforated shells (
The first definitive evidence of PFS appears in the archeological record around 65,000 to 40,000 and it emerges simultaneously in several modalities:
1. Composite figurative objects. Depiction of composite objects that don’t exist in nature provides an undeniable evidence of PFS. These composite objects such as the Lowenmensch (“lion-man”) sculpture from the caves of Lone valley in Germany (dated to 37,000 years ago) (
2. Creativity and innovation. Improvement of stone tools is our best indication of improving active imagination. Turning an unformed cobblestone into a sharp tool requires an active purposeful imagination of a previously unseen object. According to Ian Tattersall, “To make a carefully shaped handaxe from a lump of rock not only demanded a sophisticated appreciation of how stone can be fashioned by fracture, but a mental template in the mind of the toolmaker that determined the eventual form of the tool” (
What tools can unambiguously signify the presence of PFS in hominins? One of the most obvious examples of tools clearly associated with PFS are bone needles used for stitching clothing. To cut and stitch an animal hide into a well-fitting garment, one needs first to mentally simulate the process, i.e. imagine how the parts can be combined into a finished product that fits the body. Such mental simulation is impossible without PFS. Earliest bone needles are dated to 61,000 years ago (
3. Design and construction. Human dwellings are not built by reflex. An integral part of design and construction is visual planning, which relies on the mental simulation of all the necessary construction steps, which is impossible without PFS. There is little evidence of hominins constructing dwellings or fire hearths until the arrival of Homo sapiens. While Neanderthals controlled the use of fire, their hearths were usually very simple: most were just shallow depressions in the ground. There is almost a complete lack of evidence of any dwelling construction at this period (
4. Adorned burials and religious beliefs. Religious beliefs and beliefs in afterlife are the ultimate products of PFS. An individual with PFS disability cannot be induced into believing in spirits, as they cannot imagine gods, cyclops, mermaids, or any other mythological creatures. The origin of religious beliefs can be traced by following the evidence of beliefs in the afterlife. Beliefs in the afterlife, in turn, are thought to be associated with adorned burials. Therefore, the development of religious beliefs may be inferred by studying the time period when humans started to bury their deceased in elaborate graves with accompanying “grave goods.”
The oldest known human burial, dated at 500,000 years ago and attributed to Homo heidelbergensis, was found in the Sima de los Huesos site in Atapuerca, Spain, and consists of various corpses deposited in a vertical shaft (
Human skeletal remains that were intentionally stained with red ochre were discovered in the Skhul and Qafzeh caves, in Levant and dated to approximately 100,000 years ago (
The number of known adorned burials and the sophistication of the offerings significantly increase around 40,000 years ago. To date, over one hundred graves of Homo sapiens have been discovered that date back to the period between 42,000 and 20,000 years ago (
5. Fast colonization of the globe and migration to Australia. Hominins diffusing out of Africa had been colonizing the Europe and Asia long before the arrival of Homo sapiens: the remains of Homo erectus have been found as far as in Spain (
6. Building animal traps and demise of the Pleistocene megafauna. Without PFS one cannot envision the building of an animal trap, e.g. pitfall trap, which requires digging a deep pit and camouflaging it with twigs and branches. PFS aids trap building in three ways. First, a leader can use PFS to mentally simulate multiple ways to build a trap. Second, a leader could use PFS to think through the step-by-step process of building a trap. Finally, a leader could communicate the plan to the tribe: “We will make a trap by digging a large pit and covering it with tree branches. A mammoth will then fall into the pit; no need to attack a mammoth head on.” In fact, early modern humans are known for building traps; traps for herding gazelle, ibex, wild asses and other large animals were found in the deserts of the Near East. Some of the traps were as large as 60km (37miles) in length (
The correlation of human migration with demise of the Pleistocene megafauna (
Conclusions from paleontological evidence. There is no evidence of the PFS ability in hominins before 65,000 years ago and there is an abundance of clear and unambiguous evidence of the PFS ability in hominins after around 62,000 years ago. Composite objects executed in bone and cave paintings, bone needles with an eye, construction of dwellings, appearance of adorned burials, and steadfast colonization of the planet are all the external manifestations of PFS. The PFS-related artifacts are highly correlated with each other in time and geography and are associated with Homo sapiens diffusion out of Africa around 70,000 years ago. This abrupt change toward modern imagination has been characterized by paleoanthropologists as the “Upper Paleolithic Revolution” (
The articulate speech of humans is unique among primates. The vocal tract of our closest relatives, chimpanzees, is extremely limited in its ability to modulate sound. While there is no theoretical limit on the number of different vocalizations nonhuman primates can generate (
Evolutionary changes in the vocal tract have been extensively studied by paleoanthropologists (
1. The changes in hyoid bone. This small U-shaped bone lies in the front of the neck between the chin and the thyroid cartilage. The hyoid does not contact any other bone. Rather, it is connected by tendons to the musculature of the tongue, and the lower jaw above, the larynx below, and the epiglottis and pharynx behind. The hyoid aids in tongue movement used for swallowing and sound production. Accordingly, phylogenetic changes in the shape of the hyoid provide information on the evolution of the vocal apparatus.
The hyoid bone of a chimpanzee is very different from that of a modern human (
2. The flexion of the bones of the skull base. Laitman (
3. Increased voluntary control of respiratory muscles. Voluntary cortical control of respiratory muscles is a crucial prerequisite for complex speech production (
4. The anatomy of external and middle ear. Modern humans show increased sensitivity to sounds between 1kHz and 6kHz and particularly between 2kHz and 4kHz. Chimpanzees, on the hand, are not particularly sensitive to sounds in this range (
5. The evolution of the FOXP2 gene. The most convincing evidence for the timing of the acquisition of the modern speech apparatus is provided by DNA analysis. The FOXP2 gene is the first identified gene that, when mutated, causes a specific language deficit in humans. Patients with FOXP2 mutations exhibit great difficulties in controlling their facial movements, as well as with reading, writing, grammar, and oral comprehension (
Conclusions on acquisition of articulate speech. Based on these five lines of evidence — the structure of the hyoid bone, the flexion of the bones of the skull base, increased voluntary control of the muscles of the diaphragm, anatomy of external and middle ear, and the FOXP2 gene evolution — most paleoanthropologists conclude that the speech apparatus experienced significant development starting with Homo erectus about two million years ago and that it reached modern or nearly modern configurations in Homo heidelbergensis about 600,000 year ago (
As discussed in the introduction, PFS is not acquired ontogenetically unless children are exposed to recursive language in early childhood. According to our analysis, non-recursive communication system (called kitchensign or homesign, as opposed to a formal sign language) is unable to facilitate acquisition of PFS even in genetically modern children (
Furthermore, since only children can acquire PFS, it follows that around 70,000 years ago young children must have invented the first recursive language. The parents of these children used a rich-vocabulary communication system for millennia. That system, however, contained no spatial prepositions, nesting, verb tenses or other recursive elements of language. The children may have stumbled upon recursive elements of language such as spatial prepositions (development of new words and even complete language is a common phenomena among very young children living together, the process called cryptophasia (
The Romulus and Remus hypothesis attempts to explain the more than 500,000-year gap between acquisition of modern speech apparatus and recursive language by a low probability of an event when two or more very young children living together concurrently
Unfortunately, in its pure form, the Romulus and Remus hypothesis does not survive a simple numerical test. A hominin tribe of 150 individuals spaced linearly from 0 to 30, has 5 peers. Even if we assume that
the model still yields a group of 15 children from two to five years of age per tribe. Fifteen children at the peak of their plasticity is on par with the number of deaf students, who spontaneously invented the Nicaraguan sign language (400 students in two schools) (
To further refine understanding of the number of children, an evolutionary mathematical model of a hominin tribe was generated based on the Australian aboriginal population. Moody (
What may have been the genetic difference that prevented children from inventing recursive elements of language and acquiring PFS for hundreds of millennia? Inadequate vocal apparatus is commonly brought up to explain the conundrum. However, as discussed above, the improvements to the vocal apparatus amassed by hominins by 600,000 years ago must have increased vocabulary by several orders of magnitude, from 20 to 100 in chimpanzees (
From a neuroscience perspective, it is relatively easy to imagine how a single mutation could have increased the brain volume, or the number of synapses, or the number of glial cells, or the extent of axonal myelination by a few percentage points, but those relatively small neurological differences could not have prevented children from PFS acquisition. The one neurological difference that could have a direct effect on PFS acquisition is the duration of critical period. If the duration of critical period in pre-PFS hominins was shorter than in modern children, that would have decreased the probability of invention of recursive elements and at the same time having enough time to train their dialog-dependent neurological networks essential for PFS (
The duration of critical period for PFS acquisition is unknown in hominins, but has been tested in apes, first in terms of language acquisition and second in terms of the rate of PFC development. In many experiments, apes were raised in human environment and exposed to recursive language from infancy. These animals commonly learn hundreds to thousands of words but never acquired PFS (tested linguistically and nonverbally) (
Overall, humans are born with a less mature brain and develop 1.5-2 times slower than chimpanzees: molar teeth erupt three years later and humans become sexually active roughly five years after the chimps do (
Mutations that get selected and fixed in a population are usually associated with some survival benefits (e.g., lactase persistence is associated with the continued ability to digest lactose in milk after weaning). Mutations that decrease an organism’s ability to survive and reproduce are selected against and not passed on to future generations. The "PFC delay" mutation is inherently a strange mutation. By prolonging the critical period, the “PFC delay” mutation increases the chances for acquisition of PFS. At the same time, this mutation carries clear disadvantages. A decrease in the PFC development rate results in a prolonged immaturity when the brain is incapable of full risk assessment. For example, three-year-old chimps often venture away from their mother, but rarely come close to water, their decision-making PFC prohibiting them from doing so. On the contrary, in human children under 4 years of age, drowning is the leading cause of mortality, resulting in over 140,000 deaths a year (
The evolutionary mathematical model was used to study the effect of increased childhood mortality due to PFC maturation slowdown. Decreasing childhood survival by 10% results in the collapse of tribe population to 50% within 150 years. If the “PFC delay” mutation did not lead to an immediate survival benefit that could have balanced the increase of childhood mortality, it would be expected to be weeded out from a hominin population.
How much of a post-pubertal benefit, the “PFC delay” mutation must have resulted in order to balance out the 10% increase in childhood mortality? The population model predicts that a minimum of 26% of post-pubertal increase in survival rate is required to keep a stable tribe population. Slowing down PFC maturation could have theoretically improved PFC-mediated social behavior, working memory, and impulse control, but it is hard to see how these traits could have increased adult survival by 26%. On the other hand, acquisition of PFS with its associated dramatic improvement in hunting enabled by animal trapping, stratagem, and new weaponry can easily explain this dramatic increase in adult survival. We conclude that it is highly unlikely that the “PFC delay” trait have evolved for some other function, persisted in a population, and later, after many generations, was adapted for PFS acquisition. In order to balance the immediate increase in childhood mortality associated with delayed PFC development, PFS acquisition must have quickly followed “PFC delay” acquisition.
Two or more young children with “PFC delay” must have been born at the same time and lived together for several years, so that they could often talk to each other (Appendix 1). These children were in a situation reminiscent of the condition of the youngsters who invented the Nicaraguan Sign Language: they were genetically modern (in terms of the “PFC delay” mutation), but their parents could not have taught them spatial prepositions; children had to invent recursive elements of language themselves. Having invented recursive elements and having acquired PFS, these children would have gained near-modern imagination: a ticket to dramatically improved hunting by trapping animals, proclivity for fast discovery of new tools through mental simulations and the ability to strategize over clever ways to eliminate other hominin competitors. The “PFC delay” mutation and recursive language could have then spread like a wildfire through other Homo sapiens tribes carried by new weapons and an elaborate stratagem made possible by the new recursive language. Improved survival as a result of burgeoning diet and comfortable hunting style can easily explain the unprecedented explosion of human population at the end of the upper Paleolithic (
In this manuscript we presented a Romulus and Remus hypothesis of language acquisition that attempts to explain more than 500,000-year-long gap between the emergence of modern speech apparatus and the abundant evidence of modern imagination 70,000 years ago. We proposed that despite acquisition of modern vocal apparatus by 600,000 years ago, hominin communication system remained non-recursive. Spatial prepositions, verb tenses, and nesting were missing from their communication system not because there were not enough distinct vocalizations, but due to limitations of hominins’ imagination.
Language is always limited by imagination. No animal can ever be taught to follow a command to bring a ‘long red straw’ placed among several decoy objects including other red shapes (Lego pieces, small red animals) and long/short straws of other colors, not because animals cannot learn words for colors, sizes, or objects, but since they cannot purposefully imagine an object in different colors and sizes (
This is not to say that animals cannot imagine objects in different colors or sizes spontaneously or in their dreams. Spontaneous imagination, however, is completely different neurobiologically from active purposeful imagination (
Many scientists make a mistake of assuming human-like imagination in hominins. In the past, many people even extended human-like imagination to animals (e.g., St. Francis preaching to the birds). Anthropomorphism is the natural intuitive fallback for the unknown, but it has to be removed from science based on experimental evidence. Imagination is not a single phenomenon but includes multiple neurobiologically distinct mechanisms (
Once we realize the existence of multiple neurobiologically distinct mechanisms of imagination, the natural question is when these mechanisms were phylogenetically acquired? Obviously, all the distinct mechanisms of imagination could not have been acquired at the same time. The best indication of improving active imagination in hominins is provided by the stone tools record as turning an unformed cobblestone into a sharp tool requires an active purposeful imagination of a previously unseen object (
Since PFS is essential for comprehension of spatial prepositions and recursion, recursive modern-looking language could not have been acquired before PFS acquisition 70,000 years ago. We explained the 500,000 year-long period between acquisition of a modern speech apparatus and recursive language by existence of two evolutionary barriers associated with a critical period for PFS acquisition. One barrier is cultural, the other is genetic. Conversations with the use of recursive language provide an essential training for formation of frontoposterior connections necessary for PFS (
Even in genetically modern humans, this dependence of PFS on recursive language and recursive language on PFS creates an unsurpassable cultural barrier in isolated individuals. Modern children who are not exposed to recursive conversations during the critical period do not invent recursive language on their own and, as a result, never acquire PFS (
The second evolutionary barrier to acquiring recursive language could have been a faster PFC maturation rate and, consequently, a shorter critical period. In modern children the critical period for PFS acquisition closes around the age of 5 (
There are clear similarities of Romulus and Remus hypothesis to Chomsky’s Prometheus.
In some regard, the concept of PFS as uniquely human ability is not new. Uniquely human PFS-like abilities were defined descriptively as “ability to invent fiction” (
Descriptive definitions of PFS-like abilities as uniquely human are not new. The new conjecture proposed in this manuscript is the result of the exact neurobiological definition of PFS. Traditionally, PFS ability is rolled into more general abilities such as executive function, cognition, fluid intelligence, and working memory. None of those traits have a strong critical period since they can be improved well into adulthood (
The entire proposal completely depends on this understanding of the strong critical period for PFS acquisition. (Note that this critical period is different from other language-related critical periods, such as phoneme tuning (
Mental rotation and integration of modifiers are often also defined together with PFS since all three active imagination processes are controlled by the LPFC. Similar to PFS, acquisition of mental rotation and integration of modifiers have critical periods (
Similar to other traits with strong critical periods – monocular deprivation, filial imprinting in birds, and monaural occlusion – PFS cannot be acquired in adulthood. Its neural infrastructure has to be laid down in early childhood. Perhaps this neural infrastructure is related to cortical functional specialization established through competition mechanisms similar to that of monocular deprivation (
The Romulus and Remus hypothesis suggests that the first phase of articulate speech acquisition from around 2 million to 600,000 years ago has to be explained separately from the second evolutionarily fast phase of recursive language acquisition 70,000 years ago (Appendix 3). Articulate speech relies on multiple neurologically distinct mechanisms each of which is the result of complicated evolution and many genetic mutations. Compared to chimpanzees, modern humans improved neurological control of the diaphragm and the tongue, musculature of the mouth and lips, position and control of the vocal cords, hearing frequency range, neocortical processing of auditory stream, and many other abilities. This piecemeal improvement of articulate speech could not have been fast and probably has taken millions of years.
In some sense, the near-modern speech apparatus circa 600,000 years ago can be viewed as a pre-adaptation for recursive language. A classic example of pre-adaptation is bird feathers that initially may have evolved for temperature regulation, but later were adapted for flight. The speech apparatus 600,000 years ago served for the purposes of communication, but not for the purposes of unbounded contemplation. Today, once PFS is acquired in childhood, we are able to use the vocabulary both for communication and unbounded contemplations, just like modern birds use feathers for both temperature regulation and flight.
We concur with Dediu and Levinson who “attribute to Neanderthals modern speech, double-articulation (separated phonology and lexicon), some systematic means of word combination (syntax), a correlated mapping to meaning, and usage principles (pragmatics).” (
It is likely that further parallels could be drawn between individuals with PFS disability and Neanderthals. The contemporary individuals with PFS disability can be social, compassionate, have normal attention and impulse control, artistic talents, and excellent memory. Similarly, Neanderthals and other pre-PFS hominins could have been social and compassionate. They could have taken care of their sick and diseased (
Likewise, Neanderthals and other pre-PFS hominins could have been able to understand the concept of symbol. The symbolic use of objects can be associated with PFS in modern individuals, but PFS is not necessary for using an object as a symbol. Symbolic use of red ochre, production of perforated shells (
Both early modern humans and Neanderthals were using animal hides for warmth, but humans could have been stitching those hides into well-fitted clothes with the use of sophisticated bone needles. Stitching clothing definitely relies on PFS since to cut and stitch an animal hide into a well-fitting garment, one needs first to mentally simulate the process, i.e. imagine how the parts can be combined into a finished product that fits the body. Without PFS, Neanderthals must have been simply wrapping the hides around their bodies like a poncho.
The biggest difference in behavior between Neanderthals and modern humans must have been in their hunting styles. As discussed above, building an animal trap is impossible without PFS. Both Neanderthals and modern humans hunted mammoths, but lacking an ability to invent traps, Neanderthals must have tried to puncture animals with as many spears as possible. This style of attack implies close contact with an animal and must have led to a high frequency of bone injury among hunters. Modern humans, on the other hand, had the capacity to chase a mammoth into a pitfall trap where an immobilized weakened animal could have been easily killed. Comparison of archeological remains between early modern hominines and Neanderthals is therefore expected to show disproportionally larger number of broken bones in Neanderthals that must have resulted from this close contacts with large animals (
An important component of a theory is that it should be testable. A theory must make predictions that were not used in the construction of the theory initially but are now available for inspection. If the predictions are borne out, the theory would be strengthened. If not, then the original theory ought to be modified or abandoned. Here we make several predictions derived from the Romulus and Remus hypothesis.
1. Archeological evidence. The Romulus and Remus hypothesis can be disproved by an archeological finding unambiguously demonstrating PFS ability in hominins significantly earlier than 70,000 years ago.
2. Teaching recursive language to an animal. The Romulus and Remus hypothesis predicts that humans are unique in their genetic predisposition to ontogenetic acquisition of PFS. Thus, the hypothesis can be disproved by demonstrating that other living primates are capable of acquiring PFS.
3. Shortened PFC maturation in humans. If in vivo biomarkers for PFC maturation rate can be established, then duration of PFC plasticity could be correlated to PFS ability. Individuals with increased rate of the PFC maturation and decreased duration of PFC plasticity are expected to exhibit lower PFS ability. From a theoretical point of view such individuals may significantly benefit from an early intensive language therapy.
4. Effect of passive entertainment on children. Lack of dialogs with the use of recursive language during the critical period is predicted to negatively affect the PFS ability. Passive watching of TV and other videos can significantly reduce time available for dialogs and therefore predicted to result in decreased PFS ability.
5. Children with language delay. Children taking no interest in external and internal language can miss the critical period for PFS acquisition. Child’s non-recursive vocalizations alone cannot inform on PFS acquisition. In these children it is important to assess the PFS function directly (
6. An artificial extension of the period of plasticity of the PFC in animals. Recent insights into genetics of the “PFC delay” mutation identified several possible genetic targets enabling this function. Liu X. and colleagues identified four transcription factors that could play a role in regulating the timing of the development of the prefrontal cortex: myocyte enhancer factor 2A (MEF2A), early growth response 1 (EGR1), early growth response 2 (EGR2) and early growth response 3 (EGR3) (
7. Cloning a Neanderthal child. In January of 2013, George Church, a Professor of Genetics at Harvard Medical School, said in an interview with the German magazine “Der Spiegel” that it could be possible to clone a Neanderthal baby from ancient DNA if he could find a woman willing to act as a surrogate: “I have already managed to attract enough DNA from fossil bones to reconstruct the DNA of the human species largely extinct. Now I need an adventurous female human.” While currently it is hard to imagine cloning of a Neanderthal for ethical and legal reasons, history teaches us that eventually intellectual curiosity will win over and the Neanderthal will be cloned. How different will it be? George Church suggests, “Neanderthals might think differently than we do. We know that they had a larger cranial size. They could even be more intelligent than us.” Conversely to Church’s conjecture, the Romulus and Remus hypothesis predicts that the Neanderthals were lacking the “PFC delay” mutation and therefore would not be able to acquire PFS and, as a result, will fail to understand spatial prepositions and other sentences describing object combinations, as well as perform below the score of 86 in nonverbal IQ tests. In terms of the neurological difference, Neanderthal brain is expected to feature a smaller superior longitudinal fasciculus as well as a smaller arcuate fasciculus — the frontoposterior tracts that have been shown to be important for PFS (
8. Human demographic explosion following acquisition of PFS. The evolutionary mathematical model predicts a demographic explosion following acquisition of PFS 70,000 years ago. First, by changing hunting strategy from persistence hunting to building traps, hominins had the capacity to obtain nearly unlimited quantity of food resulting in increased fertility and decreased mortality. Second, tribe’s losses to predation must have come down since hominins could reduce their exposure to predators during persistence hunting and foraging (
9. Animal traps can appear in archeological record after acquisition of PFS. It may be possible to identify archeological artifacts of animal traps after 70,000 years ago; perhaps even pit traps characterized by large quantity of animal dung and bones can be identified.
10. Morphologically-modern versus imagination-modern Homo sapiens. Many researchers consider fossils from Morocco dated to 300,000 years ago to be the oldest known examples of the Homo sapiens lineage (
We suggest that the “Upper Paleolithic Revolution” is explained by simultaneous acquisition of recursive language and prefrontal synthesis (PFS), enabled by a genetic mutation that extended the critical period by slowing down the prefrontal cortex (PFC) maturation. Composite figurative art, hybrid sculptures, adorned burials, proliferation of new types of tools, fast diffusion out of Africa into four continents and demise of the Pleistocene megafauna are all logical consequences of the acquisition of the PFS ability. This event completely separated the pre-PFS hominins, who had a non-recursive communication system, from the morphologically similar but behaviorally different breed of hominins who possessed recursive language and relied on mental simulation to plan war and hunting activities. The acquisition of PFS resulted in what was now in essence a behaviorally new species: the first behaviorally modern Homo sapiens. The newly acquired power for fast juxtaposition mental objects dramatically facilitated mental prototyping and led to a rapid acceleration of technological progress. Humans acquired an ability to trap large animals and therefore gained a major nutritional advantage. As population exploded, humans quickly diffused out of Africa and settled the most habitable areas of the planet. These humans coming out of Africa some 65,000 years ago were very much like modern humans since they possessed both components of full language: the culturally transmitted recursive language along with the innate predisposition towards PFS enabled by the “PFC delay” mutation. Armed with the unprecedented ability to mentally simulate any plan and equally unprecedented ability to communicate it to their companions, humans were poised to quickly become the dominant species.
Who were the children who first invented recursive elements and acquired PFS? Three conjectures are fairly certain:
Other inferences are speculative: These children-inventors were unusually slow compared to their peers. Slow children do best in highly supportive environment, so maybe they were children of a chieftain. They spent a lot of time together and created their own vocabulary, so maybe they were closely spaced siblings or even twins; twin bonding and their propensity for cryptophasia are well documented (
We attempted to reject the least probable scenario of monozygotic twins sharing the dominant de novo “PFC delay” mutation. The following calculations show that the low monozygotic twins birth rate and the low probability of the specific “PFC delay”-causing mutation are not enough to reject the monozygotic twins’ hypothesis. Monozygotic twins birth rate probability is about 0.003 and is uniformly distributed in all populations around the world (
Probability of any birth with a de novo dominant mutation affecting any transcription factor can be calculated as follows. Humans have approximately 100 new mutations per birth (
Romulus and Remus were twin brothers born to Kital, the chief of a tribe of early hominins. After multiple attempts to produce a male heir, he finally got what he wanted, but soon discovered that his children were ‘different.’ When they were born, Romulus and Remus seemed like every other child physically, but odd enough they seemed to lack common sense. They would go near dangerous tar pits, animal dens, and rivers. At the age of 4, they even wandered out into the forest and got lost, something even younger children would not do. Most kids their age already had basic knowledge of their role in society and a simple understanding of what to do and what not to do out in the wild, but Romulus and Remus had no such understanding. While, most children would understand not to go near a rushing river to gather water, the twin brothers almost drowned because they lacked the basic knowledge not to do so. Due to his children’s intellectual inadequacies Kital was extremely displeased with his sons. He wanted great warriors that would grow up to lead his tribe, but unfortunately, he was growing older and after so many failed attempts to produce a male heir, he had no choice but to try and raise his children the best he could. However, as the head of his tribe he lacked the time to constantly look after his sons, so he got the tribe’s medicine man to do so for him in the hope some of his wisdom would rub off on the young boys.
A few years later, while under the care of the medicine man the brothers began to speak the primitive language of their fellow tribe members and slowly picked up on the social cues and common knowledge of others. However, the brothers went a step beyond, they started to add spatial prepositions to their language, something never done before. Even the medicine man did not know what to make of the addition of these new words to the children’s language, and since he could not understand what they meant, he just assumed they were fooling around. The medicine man would also notice strange drawings around the cave where the boys spent most of their time. There were markings carved into the walls of the cave, some of which resembled animals. A normal tribe member would dismiss these drawings as pointless but not the medicine man. He thought these drawings were incredible and illustrated something special about the young boys. Just like the chief, the medicine man did not have much confidence in Romulus and Remus, but their newfound ability changed his mind, and it was not long until the medicine man deemed the boys ready to embark on a hunt on their own, a rite of passage so to speak.
The medicine man and Kital brought them to the hunting grounds and gave them each a spear. It was now time for them to prove themselves worthy successors or die trying. Even though he was the one that agreed to this test, Kital was still terrified that his sons would not be able to catch anything or be killed by wild beasts and his last chance to have an heir would vanish, but what happened astounded him. When he told them that they would have to catch a buffalo on their own he saw them discuss a plan to catch the animal, but oddly enough, despite speaking the same language, Kital did not recognize many of the words they spoke. He could understand words like “buffalo,” “run,” and “rocks,” but he could not understand the relationship between the words. The boys then went to a small path with steep cliffs overlooking both sides and drew a circle in the dirt and dug a hole where the circle was and covered it with leaves and branches so that it was indistinguishable from the surrounding area. Kital was confused by his sons’ behavior, because normally hunters would utilize persistence hunting, which is when hunters would chase animals until the creature would die of exhaustion. After creating the hole, the brothers conversed again, and ran to find buffalo. Around half an hour later, they chased a group of them back to the corridor where they had dug the hole, and as the buffalo passed over the leaves and sticks, they fell down into the pit. Immediately after, the two brothers speared the trapped animals and pulled them out. The brothers had caught not just one wild buffalo, but many. Simply catching one buffalo would have taken a normal hunter, hours to accomplish by use of persistence hunting. It was at this moment that Kital knew he had somehow succeeded in producing successful male heirs.
As years went by, the brothers improved the language they spoke, introducing more spatial prepositions and eventually developing a complete recursive language. Even though their father could not fully understand them, the brothers could work and communicate together to produce miraculous results. This led them to gain the respect of their fellow tribe members as well as their father. Together, Romulus and Remus would end up leading the tribe and conquer all the neighboring tribes with their enhanced intelligence and ability to formulate complex military tactics. Eventually, when Romulus and Remus had children, they found they were just like them when they were young, and they were able to teach them the language they had created. They would take care of their children for a longer time, but after several years, they were able to use the more complex language that their parents had created. These two brothers unknowingly started a pattern that would continue for tens of thousands of years and lead to the modern humans of today.
The Romulus and Remus hypothesis divides the history of language acquisition into two phases: the nearly 2-miilion-year-long period of gradual growth of vocabulary made possible by piecemeal improvements of the vocal apparatus, as described by Dediu and Levinson (
Homo erectus was an adventurous species with the body built for long distance running (
After many generations another leader could get a different mutation that further improved his vocal apparatus and doubled his vocabulary to 300 words. The articulate leader could have been using extra words as nouns to further facilitate job assignment without the need to point to each object: two-word sentences could communicate job assignment: “John flint,” meaning that John is expected to collect flint stones; “Peter sticks,” meaning that Peter is expected to find sticks; “Patrick tubers,” meaning that Patrick is expected to dig tuber; and so on. The leader could also instruct the selected workers in what to take with them: handaxes for cutting trees, spears for hunting, or a sack for carrying throwing stones back to the shelter.
Thousands of years later another mutation may have extended vocabulary of the leader to 600 distinct words and enabled the leader to name more objects, tools, and actions, further improving his ability to efficiently organize the tribe’s productive activities. This leader could have now used more complex sentences: “John, come here;” “Peter, bring the handaxe;” and “Patrick, collect stones.” Different types of edible plants and prey animals could have been assigned their own names and different jobs could have been assigned action words. Various geographical locations, including rivers, mountains, caves, and maybe even individual trees could be named helping adventurous Homo erectus to orient and to describe directions to other tribesmen.
Thus, we envision development of vocal apparatus as a series of beneficial mutations slowly improving control of diaphragm, lips, tongue, chicks, vocal cords larynx position in the trachea, and possibly hundreds of other related mutations over many millennia. When articulate speech mutations originate in a leader, they result in immediate improvement of communication (albeit one-way communication) between tribe members and consequent increase in productivity (aren’t our leaders still more articulate than an average person?). Since leaders also had higher chances to procreate, their “improved vocal apparatus” was slowly fixed in a population over many generations.
Critically, such a communication system with many names, nouns and verbs, while significantly benefiting traveling parties of Homo erectus, does not rely on PFS. In fact, chimpanzees, dogs and some other animals have been trained to follow hundreds of commands, such as “bring the ball/newspaper/slippers,” “leak the hand/floor/bowl,” that rely on memory of nouns and verbs, but do not require purposeful combination of objects from memory into novel mental images via PFS.
The next hypothesized step in the evolution of language is beyond the ability of any living animal, but still does not rely on the PFS ability. It involves acquisition of the type of imagination called integration of modifiers (
Acquisition of integration of modifiers ability could have been also influenced by Homo erectus mobile lifestyle. Whether foraged or hunted, Homo erectus would have exhausted the land around their shelter within a few months and therefore must have traveled regularly from one place to another. For a group of 50-150 hominins, wandering around without knowing the final destination is highly inefficient. Exacerbating the problem would be the fact that most women in the group would have been either nursing or pregnant. Looking for a new shelter in the company of pregnant or nursing women and small children would have been highly dangerous. It is safer and more efficient to decide on an ideal location first, and then gather everybody together to make a quick move. Therefore, it is likely that able-bodied scouts were sent out to look for the next fertile area with a nearby shelter, while the rest of the group stayed behind in safety. Adjectives and integration of modifiers ability would have allowed scouts returning from their trip to compare their observations and decide not only which shelter was better (e.g., a bigger cave, or one with a nearby water source), but also which shelter was best positioned as far as availability of prey animals and edible plants (grasses, herbs, seeds, roots, rhizomes, or tubers). Such discussions and comparisons still do not require PFS. For example, scouts could have measured the size of their caves using their own strides (the size of which is comparable in similarly-built individuals) and then communicate this information to the leader to help choose the largest cave. A tribe leader capable of integration of modifiers would be able to select a better pasture and a better shelter and therefore would have been able to father greater number of healthy children passing down his genes for improved language.
We conclude that Homo erectus lifestyle was conducive to improvements of vocal apparatus. Homo erectus could have used a communication system with thousands of words without PFS. It is possible that Homo erectus may have even had more words than modern humans since without recursion, they needed more words to communicate the same number of concepts.
We wish to thank Dr. P. Ilyinskii, Dr. K. Khrapko, and Dr. D. Gamarnik for productive discussion and Dr. P. Ilyinskii for scrupulous editing of this manuscript.
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.