The Cognitive Functions of Language: Dispelling the Discontinuity

Presentation Link

As separate concepts, both language and thought each enjoy a great deal of popularity within the philosophical community, however the specific role that language plays in cognition is a relatively newer question. With few serious inquiries into language’s role in thought being made before a century ago, it seems as if the debate is brand new, but it’s rooted in several older debates about how humans and animals reason, what thought is, and how we acquired our unique set of cognitive abilities.

Like almost any topic in western philosophy, these related debates can be found in Plato, Aristotle, Descartes, and many other great thinkers throughout history, all well known for contributing foundational ideas to the discipline of cognitive science.

However cognitive science has forsaken it’s roots, not in its approach, which is still critical like much philosophy, but in it’s beliefs about language and thought. This is not a betrayal outright, due to the fact that it is a conclusion seemingly supported by evidence that the philosophers of old didn’t have access to.

A Multi Disciplinary Question

Since this is a question of cognitive science, it is really a question of multiple disciplines which must be considered. This is where Carruthers excels. He considers psychology, evolutionary anthropology, developmental psychology, and linguistics to arrive at a view contrary to what he classifies as the “Cognitive science” view.

Carruther’s elucidates differences between all the views, and despite the fact that his goal was not to convince people who stood in opposition, his argument is very compelling, and gives hope to estranged philosophy students who intuitively don’t agree with the social science view. At least one of these students thinks his new beliefs are due to cognitive studies classes completely changing how he approaches questions like this.

Multiple Answers

It is heavily implied by Carruthers that each discipline within cognitive science has its own answer to this question. These discrepancies seem to be dependent on certain biases which thinkers in the discipline are prone to having.

Philosophy’s hands-off nature lends it to elevating language’s cognitive role. What do I mean by hands-off? Simply that philosophers, in comparison to anthropologists, psychologists, linguists and cognitive scientists, tend to read and write, but don’t do much field work. As a discipline, philosophy is based on the powers of analytical argument and rhetoric. The idea that these language-based strategies don’t make up the whole of human mental life hurts the philosophers ego. Furthermore, narcissistic philosophers might just not care about thoughts which aren’t philosophical. Since philosophy is abstract, it requires language. Therefore, philosopher should say that “Thoughts that matter require language”!

The other social science discipline’s have the same problem, but for other reasons. Psychologists, anthropologists and linguists seem to set up tests and observations in order to confirm the widely-held lay belief that language is central to all thought.

Cognitive science and evolutionary psychology seem to escape many of these biases. For cognitive scientists, it is most simply because they have a view of the bigger picture, and are out to analyze data from all of the other discipline’s This external perspective was needed for cognitive scientists to recognize that the social science conception of language was only so ubiquitously held not because it is correct but because of long standing biases in these academic communities.

The Cognitive Revolution

These biases were actually a good thing until about 100 years ago, when serious questions about human cognition and behavior came under scientific scrutiny. One could even argue for the conceptual correctness of the social science theory, given the fact that one common colloquial meaning for the word “thought” is what cognitive scientists would more specifically name “inner speech”. If inner speech is indeed the only mental process which can be considered thought, then it is a purely conceptual truth, a tautology even, that language is the de facto medium for all thought.

This does not seem to be the case, however. Cognitive scientists have a much broader range of cognitive abilities in mind when they use the word “thought”. Carruthers has these same abilities in mind, and is able to explain that one can avoid the exaggeration of the social science view while taking a stronger position which has analytical and practical merit. Carruthers’ heart is in the right place, but he misses a few connections between his analytical framework and the real-world evidence. In actuality, he is closer than any of the other views he explains, and only missed a few details.

Carruthers and his Brilliant New Term

The most insightful point in Carruthers’ article is his description of Fodorian modularity. This is only in an adjacent way to his main view, in that it is necessary for his opinion to appeal to the “nativist” sensibility found in cognitive science.  He introduces a concept that Fodor implies, but never illustrates outright. This is the idea of a conceptual module. Even Fodor admits that certain cognitive processes occupy a conceptual domain somewhere in between modules and central systems. Carruthers names these conceptual modules, and explains that they are slower, more deliberate, and do not receive information from the senses, but other than that are much like a true module. This little term actually clears up many discrepancies in Fodor’s original thinking, while simultaneously offering a catch-all line of reasoning which can be used in many cognitive science inquiries, including the subject of language’s cognitive function.

Conceptual modules seem to signify an official term for something other philosophers and cognitive scientists have described. Carruthers implies that there would be ones which are linked to true modules, and others which are more independent, and can exchange information with true modules, central systems, and other conceptual modules.

This term completely reorganized the way in which at least one cognitive studies student now thinks about intuitive physics, intuitive psychology, and other naive theories. Many once-problematic aspects of these abilities have been redefined and seem to match anecdotal and experimental evidence.

The Trouble With Social Scientists

The success of the conceptual module as a term spells disaster for the social science view. The social science view didn’t depend on modularity at all because as it was previously construed, the modularity view did not mesh with the social science biases and evidence about language.

Now these social scientists have nowhere to hide! Carruthers has crafted a conception of modularity that lends itself to a strong view, but not unrealistically strong. He may not have been trying to get social science to compromise with cognitive science but he did invent a term that actually allows them to.

Gentner also effectively refutes the strong, whorfian hypothesis. Citing research by Steven Pinker among others, Gentner elucidates just how inconclusive the evidence for people with different native tongues perceiving the world differently is. This is followed by analysis of Vygotsky’s work, which makes language out as something which organized innate concepts in order to make them more efficient and powerful. This is consistent with Carruthers’s view.

Just a Few Adjustments

Carruthers was not attempting to be convincing, as he mentions that he is not writing the article to convince people who believe language has no cognitive function. He also foregoes lines of reasoning on the other side of the spectrum, making the implicit assumption that thoughts can in fact exist without language.

This makes the paper more focused. The fact that he essentially dismisses possibly legitimate arguments narrows the article’s scope, and makes it more convincing to anyone who already falls within the spectrum that he’s arguing within. He is trying to convince the less extreme social scientists to accept a more critical view than their current one. He is also trying to rope in cognitive scientists who have been critical and seen the evidence, but have not heard his new conception of the theory.

The only problem with his line of reasoning and analysis is that it may be a bit removed from the data, and is not specific enough in describing how an already existent module became a language module.

Make It Fit The Data

What fits the data and arguments best? Carruthers was on the right track in trying to fit aspects of Dennett’s “Joycean machine” into a weaker form of the hypothesis. What this needed was a way of reconciling the joycean machine theory with the idea of conceptual modules, in a way that adds up in as far as evolutionary and developmental time.

A dedicated module, whether purely input-output, or a conceptual module which is under some sort of central control, could not have just appeared even if it had 6 million years. If this was the story, I think we would have a very different evolutionary timeline. there needs to be a way to explain how it seemed to happen so quickly, when we know evolution is a miser and a tinkerer.

The only way these abilities could manifest themselves as quickly as they did and have the power that they do is if an already existent and fairly developed practical reasoning capacity was augmented by the “colonization” of linguistic concepts. This combines an aspect of Dennett’s view, but appeals to fans of modularity.

In theory, this conceptual module would have been responsible for using information from separate modules in order to solve practical, “big picture” problems. This was where considerations from disparate observations and multiple senses would be combined and collated for presentation to other modules or central systems.

It may seem as if I’m suggesting there hasn’t been a significant biological change. Are humans just apes who have filled their practical reasoning modules with abstract concepts which endow them with their unique abilities?No. On the contrary, this module developed extensively between 6 million years ago and when it started to process language.  This development had only latent effects which are difficult to observe in archaeological settings.

The development of the cross modular practical reasoning capacity steered itself towards modular adaptability. The selection pressures for more complex practical reasoning, and the tracking of and reasoning about objects and events became extreme. These processes made the practical reasoning module more ready to take on pointing, pantomiming, gestures, and eventually language.

The Evidence: Dispelling the Discontinuity

The most useful scientific evidence is from anthropology and evolutionary psychology. These observations provide the real evidence against many of the conceptions of language. Most crucially, insight from evolution limits the scope of any theory. Like so many other cognitive science questions, the role of language in thought must be reconciled with what kind of development was possible in the 6 million years between our common ancestor with chimps, and now.

Data on infants and development helps us look at the little picture. Can prelinguistic infants make practical calculations inconsistent with the social science model? Or do they appear to be thoughtless until they acquire language? This would present problems for the weaker claims

Looking at evidence from aphasia patients, as well as linguistic inquiries into the development and spread of language across the world can give us a better idea of what healthy linguistic capacities are, and how language has changed since it’s invention can tell us where it’s going.

Without any testing, experiments, or observations, there is work to be done in talking about language. It seems that language as a tool helps us gain information which would be impossible to fathom without it. Interestingly enough, language can be used as a tool to gain insight about language! Carruthers points out how a scientific concept like electrons could not be instantiated in a non-linguistic person. The meaning of these concepts depend on abstract explanations, and little connection to sensory data.

Evolutionary Evidence

The evolutionary data has very good evidence to suggest a significant organizational change in human cognition which predates language, but marks the beginning of a process which presumably ended with language. Steven Mithen outlines this process extremely well in a general sense. He puts forward that this change was deeply related to the separate modules of cognition becoming more integrated. He explains the milestone’s of our ancestor’s cognitive adaptations and correlates them with the ease of which we could integrate multiple cognitive domains to solve complex problems. This analysis spans intuitive mechanics, psychology, and biology and even incorporates the crucial ability of metacognition.

Mithen’s arguments and evidence are fairly consistent with Carruthers, and even more so when one considers the idea that many of these changes were in fact occurring in what was then a dedicated practical reasoning mechanism. This very same mechanism is responsible for cross-modular integration, and is what would be processing the first symbols. These symbols had no place in any of the natural modules, for they could represent multiple real world objects or actions, and be related to one or more modules.To be clear, it is quite possible that the reasoning mechanism went through significant development over a long period of time before becoming capable of symbolic thought.This explains several time discrepancies found in the literature. It seems obvious that the practical reasoning module would have been used for this very job.

These conclusions is corroborated by more recent observations. The archaeological data suggests what Tattersall(2014) refers to as a “major developmental reorganization” and even points to the role this change had on our ability to think symbolically.  The study even gives an interesting timeline, which (like Mithen), claims that a significant physical change in head size was a latent factor which resulted in the cognitive reorganization later in our evolutionary development. This lines up with the claim that there was a change in the brain, and an improvement in the practical reasoning mechanism long before this mechanism began using symbolic propositions.

The latency of this brain change is a crucial factor in making sense of the modern day observations and analysis. Many researchers have claimed there is some inexplicable discontinuity between non human minds and human minds. Many scrutinize the archaeological timeline, claiming that language developed too abruptly. They are mistaken it seems, because now we have a claim which explains why it was in fact so abrupt: the mechanism developed gradually in the fashion that most adaptations do, but once it reached the conceptual level to take on symbolic thought, symbolic thought quickly developed and reinforced itself in minds that were already “ready” for it at birth. It’s entirely possible that we actually went past the minimum requirements, but just happened to not start using symbols, and once we did, we were actually more than ready. The “raw” power of this mechanism did not give us our unique cognitive abilities. The organization of the mechanism with symbolic thought harnessed the power gave us the cognitive abilities.

Psychological Evidence

Psychology also can offer us much insight into where non-linguistic thought ends, and language begins. Some of these studies do not address language specifically, but rather have a general analysis of cognitive abilities in development. The investigation of abilities of very young infants is especially crucial. With their little exposure to culture, they give us an idea of what our Homo sapien ancestors were born with, and why the spread of symbolic thought seems so abrupt. Abnormal cases were also researched, including deaf children, non linguistic adolescents, and aphasics.

In “A unified account of abstract structure and conceptual change: Probabilistic models and early learning mechanisms”, Gopnik responds to Susan Carey (Origin of Concepts) with only one disagreement: that our innate abilities support the theory of an evolutionary discontinuity. Where they are in agreement is the fact that babies seem to be born with conceptual abilities. This is the raw power of our practical reasoning mechanism: it has concepts without language. Language organized and structured these concepts to get the most efficiency out of the mechanism.

This is corroborated by studies of non linguistic adolescents and aphasics. Varley (2014) puts forward that while nonlinguistic people may also have trouble with several other higher-order cognitive abilities, It seems as though they do have intact reasoning to an extent. This is highly inconclusive and not extremely helpful, because we still don’t know how to conceptually separate language from a general practical reasoning mechanism, or if we can. If it is like the evolutionary data suggests, it would stand to reason that someone who never learned language would still have a great deal of innate concepts. However an aphasic person may have an injury that only affected part of their reasoning system, leading to reason without language, or compromised most of the reasoning system, leading to linguistic and rational deficits.

Much of the work with aphasics has this pitfall. We simply don’t know enough about how parts of the brain correlate to conceptual modules, or even if they do. Also there is no consistent type of aphasia, with different injuries presenting unique symptoms. What some preliminary findings suggest is that aphasia can effect someone ability to understand general aspects of grammar and syntax, whether in language or not(Zimmermer 2012 & 2014, Rohter 2012). This suggests that the mechanisms which process grammar and syntax may use relational concepts which predate humanity’s use of language.

Analytical support

Many thinkers have weighed in on this issue in the last 40 years. The most prolific is probably Noam Chomsky, who espouses a highly nativist view, referring to language as an input out-mechanism. His work does not necessarily contradict the evidence I have put forward, it would just need some reworking. For example if he allowed the innate grammar mechanism to be a part of a larger practical reasoning mechanism, it would follow that when the reasoning mechanism was populated with symbols, the grammar module just happened to be the part of it that had already been dealing with the difference between objects and events, and integrating the order of concepts when crossing between modules. This allows for grammar to be innate, and even have it’s origins in pre-linguistic practical reasoning processes, but still wield the power it does without it’s adaptation being seen as too quick and therefore impossible.

Summing up

The current evidence is inconclusive, and could fit a number of different models similar to Carruthers’s. Like his own view, these would need to be reworked and redefined to fit the evidence. It seems as if we are incapable of getting evidence which can only be interpreted in one way, and there is no power behind most empirically based theories. What this field needs is people like Carruthers, mixing the evidence with what makes sense

With some careful reorganization and redefinition, it seems as if Carruther’s classification of language as engaging in all cross-modular  propositional thought is fairly sound, despite the caveats I have elucidated. I’m sure if I was to speak to the man myself he would have me fully convinced, But given his paper and my independent research, the clarifications and changes I made were necessary. Language is certainly not the medium of all thought, and even more obviously so, It is only a small part of what makes up our very broad set cognitive faculties. That being said, many of the important abilities humans have seem to be connected with language. It makes some sense to call language the outright cause of these differences. However, I have put forward my reasons as to why it is more practical to look at both language and these other abilities as common effects of a single, separate cause: the development and integration of a highly sophisticated practical reasoning ability, which tailored itself to solving the same types of problems which language helps tackle.

Works cited

Tattersall, Ian. “An Evolutionary Context for the Emergence of Language.”Language Science (n.d.): n. pag. Web.

Zimmerer, Vitor C., Patricia E. Cowell, and Rosemary A. Varley. “Artificial Grammar Learning in Individuals with Severe Aphasia.”Neuropsychologia 53 (2014): 25-38. Web.

If we could talk to the animals. Behavioral and Brain Sciences, Volume 31, Number 2 (April 2008), pp. 146-147, <>

Chomsky, Noam. Language and Mind: Noam Chomsky. New York: Harcourt Brace Jovanovich, 1986. Print.

Language and mind : Noam Chomsky.

Chomsky, Noam

Cambridge: Cambridge University Press, c2006.

Vallila-Rohter, Sofia, and Swathi Kiran. “Non-linguistic Learning and Aphasia: Evidence from a Paired Associate and Feedback-based Task.” Neuropsychologia 51.1 (2013): 79-90. Web.

Alison Gopnik (2011). A unified account of abstract structure and conceptual change: Probabilistic models and early learning mechanisms. Behavioral and Brain Sciences, 34, pp 129-130. doi:10.1017/S0140525X10002438.

Siegal, Michael, and Candida C. Peterson. “Preschoolers’ Understanding Of Lies And Innocent And Negligent Mistakes.” Developmental Psychology 34.2 (1998): 332-341. PsycARTICLES. Web. 8 Dec. 2014.

Zimmerer, Vitor C., Ewa Dąbrowska, Charles A.j. Romanowski, Catrin Blank, and Rosemary A. Varley. “Preservation of Passive Constructions in a Patient with Primary Progressive Aphasia.” Cortex 50 (2014): 7-18. Web.

Varley, Rosemary. “Reason without Much Language.” Language Sciences(n.d.): n. pag. Web.

Representing Inner Worlds: Theory of Mind in Autistic, Deaf, and Normal Hearing Children Candida C. Peterson and Michael Siegal Psychological Science, Vol. 10, No. 2 (Mar., 1999), pp. 126-129

Clowry, Gavin J. “Seeking Clues in Brain Development to Explain the Extraordinary Evolution of Language in Humans.” Language Sciences(2014): n. pag. Web.

Olson, David R., and Janet Wilde Astington. “Thinking About Thinking: Learning How To Take Statements And Hold Beliefs.” Educational Psychologist 28.1 (1993): 7. Professional Development Collection. Web. 8 Dec. 2014.

Chomsky, Noam. “Language and Other Cognitive Systems. What Is Special About Language?” Language Learning and Development 7.4 (2011): 263-78. Web.

De Villiers, Jill. “What Kind of Concepts Need Language?” Language Sciences (2014): n. pag.  Web.

Roth, G., and U. Dicke. “Evolution of the Brain and Intelligence.” Trends in Cognitive Sciences 9.5 (2005): 250­57. Print.

Carey, Susan. The Origin of Concepts. Oxford: Oxford UP, 2009. Print.

Fodor, Jerry A. The Modularity of Mind: An Essay on Faculty Psychology. Cambridge, MA: MIT, 1983. Print.

Chomsky, Noam. “Recent Contributions to the Theory of Innate Ideas.” Synthese 17.1 (1967): 2­11. Print.

Gentner, Dedre. Language in Mind: Advances in the Study of Language and Thought. Ed. Susan Goldin-Meadow. Cambridge, MA: MIT, 2003. Print.


One thought on “The Cognitive Functions of Language: Dispelling the Discontinuity

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s