Meditations on Mentalese

Question: Cognitive Scientist Steven Pinker states that “People think in a language different to any spoken language”. Discuss.

(Alright it was a little moe specific than that but I can’t quite find anything on my desk at the moment. I’ll clean it one day. Promise.)


“I write because I don’t know what I think until I read what I say.” If I had to choose one sentence to represent my argument then it would be this quote attributed to Flannery O’Connor. Mentalese exists across the board, that much is necessary. What we humans have is a ‘levelled up’ form that uses human languages to help to represent abstract terms like love and lust, emotions and ideas, things in faraway galaxies that we have never even seen but we know that they exist from what we know of mathematics and physics. I agree with Pinker that there is mentalese. But I disagree that it is ‘different to any natural language’ in the case of humans.

My argument is against Pinker’s claim that “we think in a language different to any spoken language” because to phrase it as such is deliberately misleading. There are two potential interpretations that we could conclude from this statement. At first glance, Pinker seems to be saying that this thing we call ‘mentalese’ is a language in the sense that it obeys the properties of Universal Grammar, this would be a false assumption as mentalese is not an ‘acquired’ thing, it is how we represent propositions in our molecular minds (Pinker, 2004, p. 82).

The other possible interpretation that one conclude, would be that spoken languages and the language of thought, are fundamentally different things. I agree with the view that language and thought are different things, but not that they are so different that they cannot affect one another as this interpretation of Pinker’s claim suggests.

My stance is closer to that of Nativism than Linguistic Determinism but denies that language and thought are independent systems. I shall be basing my claims off the works of Keith Chen (Could your language affect your ability to save money?, 2012), Monique Fleckon (Flecken, et al., 2014). They, state that the feature of a language to portray the future as a distinct and separate state has an effect on the way in which we perceive the world (Fleckon) and the way in which we spend (Chen).


Our brains are too big. We, as a species, would have a greater success rate if we had a smaller head, and smaller brain, simply because birth would not be as traumatic and life threatening (The Science of Dank Memes, 2016). So why is our brain so big? Human intelligence is the obvious answer here but I wish to look at a more specific aspect of human intelligence. One of the functions that our big brain has is that it seems to set us up for language (Pinker, 2004, p. 32) . Be this through general purpose learning strategies or through something like Steven Pinker’s Universal Grammar. Yet, even though the vast majority of our brains have this remarkable ability to acquire language, there are those who have either grown up without the input needed or are suffering from an aphasia. If language and thought are the same thing then those who have either never acquired language, or have lost language, should not be able to think. The fact that they can proves that there is something like a mentalese. But to say that is not to state the whole story. Mentalese does not account for abstract thought. This is another equally possible conclusion that we could draw from the tip of the tongue phenomena that Pinker claims to be layman proof for mentalese (Pinker, 2004, p. 57).

Language and thought being fundamentally different things implies that there can be one without the other. Although thinking about the nonsense that certain politicians sprout, it really does seem like they can manage without it. Language does require thought. What we say is certainly shaped by our thoughts and to some degree, what I wish to discuss is that our thoughts are aided by our language.

For argument’s sake, let’s assume that there is a Language of Thought. We’ll take Pinker’s idea of mentalese as a name for this ‘language’. What properties and features would such a ‘language’ need to have in order to actually exist in a world beyond our hypothetical imaginings? How does it relate to languages as we know them? Any mature human being able to understand/produce a sentence must have some kind of mental representation of what they said. This follows by necessity. In order to be able to impart meaning to what it is that I say, I need to know that what I am saying has some kind of a meaning, even if I do not know exactly what that meaning is for you, I at least have some preconceived idea that is just so automatic that it is quite possibly innate.

This means that mentalese has to be more inclusive than human language. It needs to be able to account for the recursive ability of language and the simplest explanation for this would be that it too has a recursive system (Aydede, 2015). This is not the only similarity between a Language of Thought and human languages. But recent arguments have been laid forth that suggest that it may just be the recursive ability of our thoughts that allow us to construct grammar and language (see Corballis, M.C., 2014). But this supports Pinker’s claim that language is a universal feature of the human aspect (Pinker, 2004). But, so must a system of representation which acts as a medium of thought be present in all thinking life. Humans have mentalese in common with the likes of dogs, and cows, and wolves, but dogs, cows and wolves cannot do abstract mathematics, so if we have the same kind of mentalese then what makes us able to do more with it than the rest of the animal kingdom?

Taking a closer look at what Fleckon, von Stuterheim, and Carrol (2014) have found, we see that their findings support the claim of a ‘weak’ form of linguistic relativity. This helps to explain how, with mentalese aided by human language, we have a far wider range of possibilities of thought. Their study showed that speakers of German were more likely to focus on the potential end points of a motion event than Modern Standard Arabic (MSA) speakers were (Flecken, et al., 2014). They attribute this to the different aspectual markers found in each language. The speakers of German have greater focus on endpoints in both their expression of the event and for the duration of time that their eyes dwelt on the potential endpoint.

Their findings oppose Pinker and gives support to the gentler version of the Sapir-Whorf hypothesis which states that our language and our thoughts are directly interlinked (Kay, P. and Kempton, W., 1984). Such a theory of Linguistic Determinism is commonly supported by the supposedly massive array of words for snow that Eskimos supposedly have. This originally comes from Franz Boas’ text Introduction to the Handbook of North American Indians (Boas, 1911)and has escalated out of proportion since Whorf himself misreported on Boas’ findings in an MIT technology review (Whorf, 1940). Time and time again has the reported number of Eskimo words for snow been changed and misreported, even by lecturers meant to be perpetuating evidence and not what it has been falsely cited as (Pullmum, 1989, p. 280). Despite the untrustworthiness of some of its claims, the idea that our language determines our thought is a possible one, but not on as strong a level as Sapir and Whorf suggested.

To make the claim that our languages shape our thoughts is to make the claim that one without language is one without thoughts. This is what the strictest version of the Sapir-Whorf hypothesis entails. We have observed feral children being taught languages (Long, M.H., 1990). We saw how difficult it was for them to learn language but to assume that they have no thoughts because they have no language to shape it is absurd. That very idea seems absurd to me. That’s why we have this thing called mentalese.

Keith Chen gives us more ground to stand on with the claim that it is human language that enables us to think abstractly. He matched families on as many levels as he possibly could with the only difference being their language. What he found was that the concept of a future in our mother tongue has some impact on the way in which we spend and save (Could your language affect your ability to save money?, 2012). A handful of families with this difference would mean little as it could be the case that those families who saved more and spoke a futureless language just had a tradition of saving more. But if we replicate this study on more and more groups of people then we can find out just what it is that causes the difference in their saving habits. If it really is the case that futureless language speakers are better savers because their concept of the future is an extended form of the present (Could your language affect your ability to save money?, 2012), then we can conclude that our language definitely has an impact of abstract thought.


Pinker definitely seems to be on the side with the stronger arguments for base thoughts. Thoughts which must be present in animals and language-less people alike. On some level, Pinker and those who stand beside him must be right when they say that there is this so-called mentalese. Where I disagree with Pinker is the level to which this occurs. In my personal experience, I have never been able to ponder about abstract concepts such as love versus lust without formulating my ideas in strings of words. Mentalese exists. It is a base system of representation that enables thought. But this is limited in its capabilities. What studies like those conducted by Fleckon, (Flecken, et al., 2014) and Chen (Could your language affect your ability to save money?, 2012) can tell us is that our language does at least have some impact on the way in which we perceive the world. There is some extent to which our language shapes our thoughts.

However, not enough evidence exists yet for us to conclusively say this. We need more studies along these lines investigating the connection between language and abstract thought. We also need to conclude that it is indeed true that non-human animals are incapable of abstract thought. I stand with Chen, von Stutterheim, Fleckon, and I disagree with Pinker and the schools of Nativism and Linguistic Determinism. ‘I do not think that I think in the way that he [Pinker] thinks I think. (Cole, 1998).


Aydede, M., 2015. The Language of Thought Hypothesis. [Online]
Available at:
[Accessed 30 August 2016].

Boas, F., 1911. Introduction to the Handbook of North American Indians. Bureau of American Ethnology Bulletin.

Cole, D., 1998. I Don’t Think So: PInker on the Thinker. [Online]
Available at:
[Accessed 15 October 2016].

Corballis, M.C., 2014. The recursive mind: The origins of human language, thought, and civilization. Princeton University Press. Vancouver

Could your language affect your ability to save money?. 2012. [Film] Directed by Kieth Chen. Edinburgh: TED.

Flecken, M., von Stutterheim, C. & Carrol, M., 2014. Grammatical aspect infuences motion event perception: fidings from a cross- linguistic nonverbal. Language and Cognition, 6(1), pp. 45-78.

Kay, P. and Kempton, W., 1984. What is the Sapir‐Whorf hypothesis?. American anthropologist, 86(1), pp.65-79.

Long, M.H., 1990. Maturational constraints on language development. Studies in second language acquisition, 12(03), pp.251-285.

Pinker, S., 1994. Mentalese. In: The Language Instinct. s.l.:Harper Perenial Modern Classic, pp. 55-82.

Pinker, S., 2004. The Language Instinct. s.l.:Harper Perennial Modern Classic.

Pullmum, G. K., 1989. The Great Eskimo Vocabulary Hoax. Natural Language and Linguistic Theory, Volume 7, pp. 275-281.

The Science of Dank Memes. 2016. [Film] Directed by SciShow. s.l.: YouTube.

Whorf, B. L., 1940. Science and Linguistics. Technology Review (MIT), 42(6), pp. 229-231, 247-248.


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