How the Language You Speak Influences the Way You Think
The relationship between language and thought is far from straightforward.
Posted Aug 08, 2018
[Article revised on 24 November 2020.]
Time heaved a gentle sigh as the wind swept through the willows.
Communication does not require language, and many animals communicate effectively by other means. However, language is closely associated with symbolism, and so with conceptual thought, problem solving, and creativity. These unique assets make us by far the most adaptable of all animals and enable us to engage in highly abstract pursuits such as philosophy, art, and science that define us as human beings.
Here’s a thought experiment. Imagine what it would be like to live without language—not without the ability to speak, but without language itself. Given the choice, would you rather lose the faculty of sight or the faculty of language? This is probably the first time that you have faced this question: The faculty of language is so fundamental to what it means to be human that, unlike the faculty of sight, we take it completely for granted. ‘Monkeys’ quipped Kenneth Grahame, ‘very sensibly refrain from speech, lest they should be set to earn their livings.’
If rhetoric, the beauty of language, can so bend us, how about language itself? In other words, how does the language you speak influence the way you think? The ostensible purpose of language is to transmit thoughts from one mind to another. Language represents thought, that’s for sure, but does it also determine thought?
Wittgenstein famously wrote that ‘the limits of my language stand for the limits of my world’. Taken at face value, that seems too strong a claim. There are over 7,000 languages in the world—with, by some estimates, one dying out every two weeks or so. The number of basic colour terms varies quite considerably from one language to another. Dani, spoken in New Guinea, and Bassa, spoken in Liberia and Sierra Leone, each have no more than two colour terms, one for dark/cool colours and the other for light/warm colours. But, obviously, speakers of Dani and Bassa are able to perceive and think about more than just two colours.
More subtly, there is no English equivalent for the German word Sehnsucht, which denotes dissatisfaction with reality and yearning for a richer, ‘realer’ ideal. But despite lacking the word, the American poet Walt Whitman (d. 1892) was able, very successfully, to conjure both the concept and the emotion: Is it a dream? Nay, but the lack of it the dream, And, failing it, life’s lore and wealth a dream, And all the world a dream.
The English language has a word for children who have lost their parents (‘orphan’), and a word for people who have lost their spouse (‘widow’ or ‘widower’), but no word for parents who have lost a child. This may mean that parents who have lost a child are less likely to enter our minds, but not that they cannot enter our minds or that we cannot conceive of them. We often think about or remember things that cannot be put into words, such as the smell and taste of a mango, the dawn chorus of the birds, or the contours of a lover’s face or other part of their anatomy. Animals and pre-linguistic babies surely have thoughts, even though they have no language.
If language does not determine thought, how, if at all, does it interact with thought? Russian, Greek, and many other languages have two words for blue, one for lighter shades and the other for darker shades—goluboy and siniy in Russian, and ghalazio and ble in Greek. A study found that, compared to English speakers, Russian speakers were quicker to discriminate between shades of goluboy and siniy, but not shades of goluboy or shades of siniy. Conversely, another study found that Greek speakers who had lived in the UK for a long time see ghalazio and ble as more similar than Greek speakers living in Greece. By creating categories, by carving up the world, language supports and enhances cognition.
In contrast to modern Greek, Ancient Greek, in common with many ancient languages, has no specific word for blue, leaving Homer to speak of ‘the wine-dark sea’. But the Ancient Greeks did have several words for 'love', including philia, eros, storge, and agape, each one referring to a different type or concept of love, respectively, friendship, sexual love, familial love, and universal or charitable love. This means that the Ancient Greeks could speak more precisely about love, but does it also mean that they could think more precisely about love, and, as a result, lead more fulfilled love lives? Or perhaps they had more words for love because they led more fulfilled love lives in the first place, or, more prosaically, because their culture and society placed more emphasis on the different bonds that can exist between people, and on the various duties and expectations that attend, or attended, to those bonds.
Philosophers and academics sometimes make up words to help them talk and think about an issue. In the Phædrus, Plato coined the word psychagogia, the art of leading souls, while discussing rhetoric—which, it turns out, is another word that he invented. Every field of human endeavour invariably evolves its own specialized jargon. There seems to be an important relationship between language and thought: I often speak—or write, as I am doing right now—to define or refine my thinking on a particular topic, and language is the scaffolding by which I arrive at my more subtle or syncretic thoughts.
While we’re talking dead languages, it may come as a surprise that Latin has no direct translations for ‘yes’ and ‘no’. Instead, one either echoes the verb in the question (in affirmative or negative) or expresses one’s feelings about the truth value of the proposition with adverbs such as certe, fortasse, nimirum, plane, vero, etiam, sane, minime… This may have led to more nuanced thinking, as well as greater interpersonal engagement, though it must have been a nightmare for teens—if they even had teens in those days.
As I argue in my new book, Hypersanity: Thinking Beyond Thinking, much of the particularity of a language is extra-lexical, built into the syntax and grammar of the language and virtually invisible to native speakers. English, for instance, restricts the use of the present perfect tense (‘has been’, ‘has read’) to subjects who are still alive, marking a sharp grammatical divide between the living and the dead, and, by extension, between life and death. But of course, as an English speaker, you already knew that, or at least subconsciously. Language is full of built-in assumptions and prejudices of this kind. Here’s another, more substantial example. When describing accidental events, English speakers tend to emphasize the agent (‘I fired the gun’) more than, say, speakers of Spanish or Japanese, who prefer to omit the agent (‘the gun went off’). One study found that, as a result, English speakers are more likely to remember the agents of accidental events—and, I surmise, to attach blame.
Some languages seem more egocentric than others. Many languages forgo the explicit use of the personal pronoun, which is instead built into the verb. For example, ‘I want’ in Spanish is simply quiero. English in contrast requires the explicit use of the personal pronoun in all cases, as does French. What’s more, French speakers often double up on the first-person personal pronoun, as in Moi, je pense que… [Me, I think that] with the stress on the moi. Sometimes, they also redouble on other personal pronouns, Et toi, qu’en penses-tu? [And you, what do you think about it?]. But redoubling on the first-person personal pronoun is much more common: Bon aller, moi j’en ai marre [Whatever, I’m fed up me]. This redoubling, this pleonasm, is more a feature of the spoken than the written word, and, depending on the context, can serve to emphasize or simply acknowledge a difference of opinion. Equivalent forms in English are more strained and recondite, and less commonly used, for example, ‘Well, as for me, I think that…’ The redoubling, in French, on the first-person personal pronoun seems to inject drama into a conversation, as though the speaker were acting out her own part, or playing up her difference and separateness.
In English, verbs express tense, that is, the time relative to the moment of speaking. In Turkish, they also express the source of the information (evidentiality), that is, whether the information has been acquired directly through sense perception, or only indirectly by testimony or inference. In Russian, verbs include information about completion, with (to simplify a bit) the perfective aspect used for completed actions and the imperfective aspect for ongoing or habitual actions. Spanish, on the other hand, emphasizes modes of being, with two verbs for ‘to be’—ser, to indicate permanent or lasting attributes, and estar, to indicate temporary states and locations. Like many languages, Spanish has more than one mode of second-person address: tú for intimates and social inferiors, and usted for strangers and social superiors, equivalent to tu and vous in French, and tu and lei in Italian. There used to be a similar distinction in English, with ‘thou’ used to express intimacy, familiarity, or downright rudeness—but because it is archaic, many people now think of it as more formal than ‘you’: Shall I compare thee to a summer’s day? Thou art more lovely and more temperate… It stands to reason that, compared to English speakers, Turkish speakers have to pay more attention to evidentiality, Russian speakers to completion, and Spanish speakers to modes of being and social relations. In the words of the linguist Roman Jakobson (d. 1982), ‘Languages differ essentially in what they must convey and not in what they may convey.’
In many languages, nouns are divided into masculine and feminine. In German, there is a third, neutral class of nouns. In Dyribal, an Aboriginal language, there are four noun classes, including one for women, water, fire, violence, and exceptional animals—or, as George Lakoff put it, ‘women, fire, and dangerous things’. Researchers asked German speakers and Spanish speakers to describe objects with opposite gender assignments in German and Spanish and found that their descriptions conformed to gender stereotypes, even when the testing took place in English. For example, teutophones tended to describe bridges (feminine in German, die Brücke) as beautiful, elegant, fragile, peaceful, pretty, and slender, whereas hispanophones tended to describe bridges (masculine in Spanish, el puente) as big, dangerous, long, strong, sturdy, and towering.
Another study looking at the personification in art of abstract concepts such as love, justice, and time found that, in 78% of cases, the gender of the concept in the artist’s language predicted the gender of the personification, and that this pattern held even for uncommon allegories such as geometry, necessity, and silence. Compared to a French or Spanish artist, a German artist is far more likely to paint death [der Tod, la mort, la muerte] or victory [der Sieg, la victoire, la victoria] as a man—though all artists, or at least all European artists, tend to paint death in skeletal form. So grammar, it seems, can directly and radically influence thought, perception, and action.
It is often said that, by de-emphasizing them, language perpetuates biases against women. For example, many writers in English continue to use ‘mankind’ to talk about humankind, and ‘he’ for ‘he or she’. Similarly, many languages use masculine plural pronouns to refer to groups of people with at least one man. If 100 women turn up with a baby in a pram, and that baby happens to have a penis, French grammar dictates the use of the masculine plural ils: ils sont arrivés, ‘they have arrived’.
Language changes as attitudes change, and sometimes politicians, pressure groups, and others attempt to change the language to change the attitudes—but, on the whole, language, or at least grammar, serves to preserve the status quo, to crystallize and perpetuate the order and culture that gave rise to it.
Language is also made up of all sorts of metaphors. In English and Swedish, people tend to speak of time in terms of distance: ‘I won’t be long’; ‘let’s look at the weather for the week ahead’; ‘his drinking finally caught up with him’. But in Spanish or Greek, people tend to speak of time in terms of size or volume—for example, in Spanish, hacemos una pequeña pausa [let’s have a small break] rather than corta pausa [short break]. More generally, mucho tiempo [much time] is preferred to largo tiempo [long time], and, in Greek, poli ora to makry kroniko diastima. And guess what… According to a study of bilingual Spanish-Swedish speakers, the language used to estimate the duration of events alters the speaker’s perception of the relative passage of time.
But all in all, with a few exceptions, European languages, or even Indo-European languages, do not differ dramatically from one another. In contrast, to talk about space, speakers of Kuuk Thaayorre, an Aboriginal language, use 16 words for absolute cardinal directions instead of relative references such as ‘right in front of you’, ‘to the right’, and ‘over there’. As a result, even their children are always aware of the exact direction in which they are facing. When asked to arrange a sequence of picture cards in temporal order, English speakers arrange the cards from left to right, whereas Hebrew or Arabic speakers tend to arrange them from right to left. But speakers of Kuuk Thaayorre consistently arrange them from east to west, which is left to right if they are facing south, and right to left if they are facing north. Thinking differently about space, they seem to think differently about time as well.
Language may not determine thought, but it focuses perception and attention on particular aspects of reality, structures and thereby enhances cognitive processes, and even to some extent regulates social relationships. Our language reflects and at the same time shapes our thoughts and, ultimately, our culture, which in turn shapes our thoughts and language. There is no equivalent in English of the Portuguese word saudade, which refers to the love and longing for someone or something that has been lost and may never be regained. The rise of saudade coincided with the decline of Portugal and the yen for its imperial heyday, a yen so strong and so bitter as to have written itself into the national anthem: Levantai hoje de novo o esplendor de Portugal [Let us once again lift up the splendour of Portugal]. The three strands of language, thought, and culture are so tightly woven that they cannot be prised apart.
It has been said that when an old man dies, a library burns to the ground. But when a language dies, it is a whole world that crumbles into the sea.
See my related article, Beyond Words: The Benefits of Being Bilingual.
Winawer J et al (2007): Russian Blues Reveal Effects of Language on Color Discrimination. Proc Natl Acad Sci USA 104(19):7780-5.
Fausey CM et al. (2010): Constructing Agency: The Role of Language. Front Psychol 1:162.
Boroditsky L et al. (2003): Sex, Syntax, and Semantics. In Language in Mind: Advances in the Study of Language and Cognition, ed. Genter D & Goldin-Meadow S, pp. 61-80. Cambridge University Press.
Segel E & Boroditsky L (2010): Grammar in Art. Front Psychol. 1:244.
Bylund E & Athanasopoulos P (2017): The Whorfian Time Warp: Representing Duration Through the Language Hourglass. J Exp Psychol Gen. 146(7):911-916.
Gaby A (2012): The Thaayorre Think of Time Like They Talk of Space. Front Psychol 3:300.