If you took a moment and considered the prodigious diversity of the world’s languages, you would doubtlessly realize that some languages encode more information than others. For instance, Mandarin has no word with the exact meaning of English uncle; instead, speakers have to use distinct terms to denote which side of the family the uncle is on, whether he is related by blood or by marriage, and, if he is the father’s brother, whether he is older or younger than the father. Other languages, like Vietnamese, use the same term for blue and green, which are regarded as shades of the same color. Even more surprisingly, Pirahã, an isolated Amazonian language, has no numerals, whereas speakers of another language in the Amazonian basin, Matsés, have to denote how exactly they became aware of whatever they are reporting: according to Deutscher (2010), in order to impart the passing of an animal, the Matsés will use distinct verbal forms, depending on whether the event was directly experienced (they saw it pass), inferred (they saw footprints), conjectured (animals usually pass there that time of the day) or hearsay. Examples of linguistic variation could be multiplied endlessly. The fact that we need to observe different rules in speech and encode different aspects of reality in order to properly use our own language makes one ponder.
Does language shape our thoughts or does it simply express them? Do we speak differently because we think differently or the other way around? Or do we all think in the same way? Do we perceive, interpret and remember experiences differently because the languages we speak differ? Such questions are central to our understanding of the mind and have long engendered considerable debate involving linguists, psychologists, anthropologists and philosophers alike. However, little empirical work has focused on these topics until recently. In this article we will tackle three case studies which appear to demonstrate the causal role of language, suggesting that linguistic background has direct effects on our thinking. We will discover that language categories (grammatical gender, absolute directions and the future tense) can also affect non-linguistic representations i.e. the way we see, experience and change the world. Language is a pivotal component of our humanity, therefore understanding its contribution to the mental universe can help us make sense of ourselves, each other and human nature in general.
Gender and object representation
Unlike English, many languages classify nouns into genders: nouns in different gender categories will behave differently from the point of view of grammar. Some languages, like Italian, assign two genders only (masculine and feminine), while others have several genders or noun classes (neutral, vegetative, diminutive, etc.), up to 25 in Fulani. Speakers of such languages have to mark all nouns as gendered by means of articles and pronouns, and may have to alter the form of adjectives, numerals or verbs in order to obtain gender agreement. Boroditsky et al. (2003) consider the influence of grammatical gender on our mental representations of inanimate objects, for which gender assignment is largely regarded as semantically arbitrary. They wanted to learn whether speaking of such objects as either masculine or feminine actually makes people conceptualize them as having a gender. This does not imply that speakers of languages which categorize nouns into masculine and feminine fail to understand that inanimate objects do not have a biological sex. However, the authors hypothesize that, when their language forces them to refer to an object as either masculine or feminine, speakers tend to “selectively attend to that object’s masculine or feminine qualities, thus making them more salient in the representation”. In other words, gender connotations impregnate the mind and make speakers of gendered languages assign to inanimate objects connections and emotional reactions that gender-neutral languages miss. The researchers listed 24 object names that have opposite genders in Spanish and German and asked a group of native Spanish speakers and another group of native German speakers to write down the first three adjectives that came to mind when describing each object. The tests were conducted in English, a language all subjects were proficient in, so that they reflect language-independent thinking rather than thought intended for speaking one’s native tongue. It turned out that, with object names that are masculine in German and feminine in Spanish, German speakers generated adjectives that were rated as more masculine than the terms produced by the Spanish, and vice-versa. For instance, a key, which is masculine in German (der Schlüssel) and feminine in Spanish (la llave), was described by the Germans as hard, heavy, jagged, metal, serrated and useful, and by the Spaniards as golden intricate, little, lovely, shiny and tiny. On the other hand, a bridge, which is feminine in German (die Brücke) and masculine in Spanish (el puente) was depicted as beautiful, elegant, fragile, peaceful, pretty and slender by the German speakers, while the Spanish speakers referred to it as big, dangerous, long, strong, sturdy and towering. Thus, the male/female characteristics assigned to objects matched grammatical gender in the two languages. This experiment confirms the team’s hypothesis: languages which mark grammatical gender differently force their speakers to focus on different features of objects that (stereotypically) correspond to one of the two genders, which thus become more salient in mental representations. In a related study, Boroditsky’s team asked German and Spanish speakers to rate the similarity between pictures of unlabeled objects (all of which had opposite genders in the two languages) and pictures of men and women. Testing involved no linguistic component that might affect the subjects’ judgment. Not surprisingly, the similarity rates provided were consistent with the grammatical gender in each language – speakers of both languages rated an object as more similar to a person when the grammatical gender of the former in their native tongue corresponded to the biological gender of the latter. A convincing explanation is the following: by including items in a category, languages invite us to compare and discover meaningful similarities that may then be highlighted in cognitive representations. To summarize, experimental evidence shows that a cross-linguistically different grammatical class can shape the way in which people describe objects and assess similarities. More generally, the grammatical quirks of the language we speak seem to be consistent with our mental representations, which may have further consequences on our preferences, habits and non-linguistic behavior in general.
Absolute directions and orientation
When most people speak of positions and directions they rely on their own location, using egocentric coordinates, i.e. terms such as right, left, ahead, behind for which the human body is the focal point. However, when the approximately 250 speakers of Kuuk Thaayorre, an indigenous Australian language, talk about positions and directions, conceptualization is not in relation to themselves – instead, they refer to cardinal points, which represent fixed or absolute reference frames. For instance, they will say things like “there’s an ant on your north leg” or “the cup is southeast of the plate”, and if they point to their chest they are probably indicating whatever direction is behind them. Remarkably, the Thaayorre are not alone: about a third of the world’s languages make use of absolute directions, from Polynesia to Mexico, from Nepal to Madagascar. Studies have shown that, as compared to speakers of languages that rely on relative reference frames, for whom orientation seems much simpler since it directly depends on their own bodies, native speakers of such “geographic” languages show a much higher awareness of their environment and remarkable spatial orientation skills (demonstrated, for instance, in navigation). The fact that, from a very young age, these people always know, and can instantly indicate, where the four cardinal points are, even in unfamiliar settings, an accomplishment that prominent scientists in our world can seldom evince, is not surprising. It is their own language which generates such cognitive abilities and imposes a broader perspective of the world, compelling them to train their attention while not regarding themselves as the center – if they did not stay oriented all the time, they could not communicate properly, all the more since the common greeting in Kuuk Thaayorre is “Where are you going?”. The fact that they pick up these skills and learn how to conceptualize space as they acquire the language does not mean that the speakers of such languages are incapable of understanding our representation of space. However, because their own language permanently compels them to convey geographic directions, the Thaayorre or the Guugu Yimithirr, another Australian aboriginal tribe, develop distinct thinking habits engendering unbelievable orientation achievements which for them are a matter of course. According to Deutscher (2010), speakers of geographic languages constantly pay attention to hints in the physical environment, such as the position of the sun or the wind, and have an accurate memory of their orientation at any given time. Moreover, their spatial awareness is instilled very early: children start using geographic directions around the age of 2 and fully master the reference frame by 7 or 8. As a result, the use of this system becomes an undemanding, unconscious habit. The case of Kuuk Thaayorre suggests that our language can teach us how to represent reality. The link between exclusive vocabulary items and the exceptional orientation skills of speakers is thought: the Thaayorre could not display such a sense of direction if they did not cerebrate in the same terms. Evidence comes from a series of experiments designed to determine whether the Thaayorre’s representation of space is reflected in their conceptualization of time (Boroditsky 2009). Subjects were given sets of photos showing temporal progression (e.g. a man at different ages, a banana being eaten) and asked to place them in the correct temporal order. Instead of arranging them from left to right like English speakers would (or from right to left with speakers of Hebrew, which is written in this direction), the Thaayorre set the photos from east to west, regardless of the direction they were facing, i.e. from left to right when they faced south, toward the body when they faced east, and so on. They did this spontaneously, perfectly aware of which direction they faced every time. The fact that the Thaayorre represent time using their spatial orientation indicates a generalized cognitive pattern, prompted by linguistic units.
The future tense and our savings
One of the countless criteria by which languages can be classified is whether or not their speakers make use of specific verb forms when encoding information about future events. Thus, the so-called “futured” languages such as French or Turkish make a clear grammatical distinction between the present and the future and require future marking (e.g. by means of will or be going to in English). On the other hand, futureless languages do not divide the temporal spectrum: they have the same uninflected verb form in both contexts, as in Chinese, or tend to use a present tense form in reference to the future even though there is also a future tense form in the language – one such case is German. Could this difference in language structure have any counterpart in cognitive processes? According to Chen (2013), the way in which our native tongue expresses the future influences our economic decisions, specifically our propensity to save for the future. He argues that futured language speakers are prone to save less than speakers of futureless languages because when the former speak about the future as being separated from the present, they also think about it as distinct, hence distant, so they are less motivated to invest in the future (both financially and health-wise) than if they spoke and thought about it as identical to the present. This theory could explain why countries with apparently similar economies and institutions such as the OECD members display radically different behaviors with respect to their savings, ranging over 25 years from 10% (Greece) to 42% of the GDP (Luxembourg). Most of the languages in the top half of the savings ranking are futureless (e.g. Norwegian, Japanese, Dutch), while futured languages (e.g. Portuguese, Polish, Hungarian) are spoken in countries that tend to save less, with an annual difference of 5% of the GDP indicating contrasting long-term effects on national economies. There are, of course, other possible explanations for these financial differences, the most prominent of which concern the history and culture of the respective nations. However, the link between the futured/futureless distinction and people’s ability to save can be demonstrated. Compelling evidence comes from Chen’s (2013) complex analysis of individual household savings behavior, based on large data sets from various worldwide and regional surveys. The study eventually focuses on nine countries that have significant native populations speaking both futured and futureless languages: Belgium, Burkina Faso, Ethiopia, Estonia, The Democratic Republic of the Congo, Nigeria, Malaysia, Singapore and Switzerland. Statistically matching pairs were formed between families in each of these countries that were virtually identical in every parameter (age, income level, education, family structure, religion) except language. It was revealed that speakers of futureless languages are 30% more likely to save money in any given year than futured language speakers with an analogous background. The former also tend to retire 25% wealthier, are 20 to 24% less likely to smoke, practice safe sex 21% more often, and are 13 to 17% less obese upon retirement. The rigorous matching of the families under analysis can only suggest one justification of this conspicuous concern for the future: the fact that they are forced to think about the future as part of the present influences the economic decisions of futureless language speakers. Consequently, it may be argued that the way in which our mother tongue compels us to conceptualize time affects our propensity to behave across time.
To think and speak about
Our brief survey of recent research indicates that linguistic structures can influence our construction of reality: grammatical gender shapes our mental representations of inanimate objects, the use of absolute directions in speech brings about orientation feats, and the way in which our native language expresses the future affects our saving behavior. Thus, we understand, categorize, locate or build while covertly guided by the vocabulary and grammatical categories of our mother tongue. The fact that one or several languages compel their speakers to think of and provide distinct or additional information does not mean that speakers of other languages are unable to understand it. Our mother tongue does not constrain the mind in the sense of preventing us from having certain thoughts, but allows us not to consider such information whenever we talk, which may have significant non-linguistic effects, as in the case of the futured-futureless distinction. Forced by our own language to specify certain types of information, we pay attention to particular details or aspects of the world that speakers of other languages need not consider all the time. Developing very early, these habits of speech become habits of mind that transgress language, impacting our perceptions, associations, emotional responses, memories or orientation to the world, with potential consequences on our values and beliefs. The rich diversity of languages is a result of human adaptability, our capacity to create and permute notions according to our aims and environments. People who speak so differently also think differently, and there is a strong link between the two. As illustrated by the masculine-feminine experiments above, thinking for speaking a language, which is an internal language itself, can influence thinking in general, so the way we conceptualize is shaped by the constraints in our native tongue. Furthermore, changing the way we talk also changes how we think: for instance, studies have shown that once they learn new color words, people’s ability to discriminate colors evolves. Learning a foreign language means more than memorizing vocabulary: one has to pay attention to novel sets of distinctions and select information that such language includes, thereby unknowingly acquiring new means of regarding the world. When we learn our mother tongue we also appropriate certain frames of mind that forge our experience in meaningful and unexpected ways, as with the incredible abilities demonstrated by the Thaayorre. Although the influence of thought on speech cannot be refuted, the ingenious studies under analysis show that our language shapes the way we think, feel and live.
Boroditsky, L. (2009). “How does our language shape the way we think?” In M. Brockman (ed.) What’s Next? Dispatches on the Future of Science. Vintage Press.
Boroditsky, L., L. Schmidt and W. Phillips (2003). “Sex, Syntax and Semantics”. In: D. Gentner and S. Goldin-Meadow (eds.), Language in Mind: Advances in the Study of Language and Thought. MIT Press.
Chen, K. (2013). “The Effect of Language on Economic Behavior: Evidence from Saving Rates, Health Behaviors, and Retirement Assets”. In American Economic Review, 103(2).
Deutscher, G. (2010). Through the Language Glass. Heinneman