doi:10.1038/nindia.2013.131 Published online 30 September 2013
This may sound illogical but empirical evidence is now suggesting that our words do shape our thoughts and perceptions.
In a recent interview University of California, Irvine, pshychologist Elizabeth Loftus explains the link. Her research showed that eyewitness testimonies of road accidents varied greatly depending on the questions they were asked. After viewing the same accident videos, people who were asked, "How fast were the cars going when they smashed into each other?" generally estimated higher speeds than those who were asked the question with the verb 'hit'. Lowest estimates were given by those questioned with the verb 'contacted'. Semantics of the question had a profound effect on how people recalled the events.
Words and language could also mislead us into fabricating parts of the story or 'contaminate our memory'. Those asked about cars 'smashing' into one another were more likely than others to report seeing broken glass even though there was none in the video.
Yale University linguist Benjamin Lee Whorf and his mentor Edward Sapir were the pioneers of the theory of 'linguistic relativity'. The fundamental premise of the theory, drawn from Einstein's principle of general relativity, was that linguistic categories or words influence what we see or think. The grammatical and semantic categories of a language provided a frame of reference through which the observations were made.
Whorf's theory of linguistic relativity was, however, based on little empirical evidence and thus fell out with the intellectual community within a decade of his death.
At one end of the classical language-perception debate is this strong-form Whorfian theory and at the other is the Universalist stance, claiming that there is a universal repertoire of human thought and perception that leaves its imprint on all languages of the world. Evidence has swung back and forth but over the past decade, mounting empirical evidence is resuscitating the theory of linguistic relativity.
The effect of language on thought and perception was put to a colour test by Aubrey Gilbert and colleagues . They used the perceived boundary between blue and green for their tests — a boundary that exists in the English language but not in many others. Most languages evolved without a word for blue and acquired it only after the discovery of indigo.
English speakers were shown the figure on right — a cross surrounded by a ring of square tiles, all of the same colour barring one. The participants were to identify the odd-coloured square and indicate its location (left or right of the circle) by pressing a button with the corresponding hand. The target colour was either similar to the colour of the other squares (eg. green against the background of a different green) or a different colour (eg. green against a background of blue). This 'Categorical perception' (CP) test was to ascertain if having a word/label for a colour improved our ability to perceive it, especially when the stimulus straddled the edge of a boundary.
Interestingly, the subjects were able to identify cross-category targets faster than same-category targets but only when presented in their right visual field (RVF), thus suggesting that at least half of Whorf's hypothesis might have been right. Having a word for a colour made it 'pop-out' in our percept. Interestingly, our language faculties are localised in the brain's left hemisphere and the brain functions contra-laterally — the left side of the brain controls the RVF.
Similar tests targeting the 'language centre' of the brain showed that language also had a pronounced effect on perception.
This made scientists question if children, not yet influenced by words, see the world differently. A. Franklin and co-workers compared infant and adult performance in a Gilbert-like study to show a dominant CP in the RVF influenced by the language centre in the left hemisphere of adults. The infants, however, showed no such CP bias in the RVF and instead exhibited a similar bias in the left visual field (LVF). Thus, it seemed that as children grew up and acquired language, their perception migrated from the LVF to the RVF.
They also compared toddlers (2-5 years old) in two groups — the 'namers' who knew their colours and the 'learners' who didn't. The learners were found to pattern colours like the infants while the namers behaved like adults.
The presumed impact of language on thought has long remained an intriguing question in the realm of psychology, philosophy and linguistics with little empirical evidence.
However, over the past decade, researchers like Stanford University neuroscientist Lera Boroditsky have attempted to get definitive answers . With data collected from around the world, she contends that people speaking different languages think differently. For instance, Russians alter the verb to indicate the tense and gender, Spanish and French speakers ascribe gender to common nouns (making a table masculine or feminine), some Australian aboriginal languages have up to sixteen genders that include classes of hunting weapons, canines or things that are shiny.
Speakers of different languages might also attend to, partition and remember their experiences differently. For instance, Pompuraaw, a small aboriginal community in Northern Australia, define space in terms of North, South, East and West as opposed to English speakers who define space relative to an observer with words like right, left, front or back. All space in this community (from a village to a leg) is specified along these cardinal directions. This constant attention to geographical coordinates makes them excellent navigators.
But does the use of coordinates really affect thought? Boroditsky and her group show that representation of abstract concepts such as time, number, musical pitch, kinship, relations, morality and emotions depends on how we think about space. When people were given snapshots of a temporal progression (e.g. a man aging) and asked to arrange them chronologically, English speakers arranged the cards from left to right, Hebrew speakers from right to left akin to their script and the aborigines from east to west.
In representing time, English speakers use horizontal spatial metaphors (ahead or behind) and mandarin speakers use a vertical arrangement (in keeping with their script). Since English speakers talk about time durations in terms of length (short talk, long day), they tend to be confused by distance information — they estimate a longer line to have existed for a longer time, although the two parameters are completely unrelated.
If language indeed influences our perception of the world, then learning a new language should change our view of the world. In one such study, English speakers were taught to use Greek size metaphors to describe duration and Mandarin vertical metaphors to describe event order. Remarkably, once the English speakers learnt these new ways, their cognitive performance began to resemble that of native Greeks and Chinese.
Studies by Boroditsky also showed how subconscious thoughts on gender and abstract concepts in art such as death, victory and sin were influenced by our languages.
A recent study demonstrates the strong effect of language on perception itself, an effect divorced from any higher level cognitive processing. Scientists Gary Lupyan and Emily Ward show that hearing linguistic labels not just affects what one sees but whether one sees something at all. For example, hearing the word 'kangaroo' before trying to find the animal in an otherwise incoherent stimulus increases the chances of success.
Are there any cognitive benefits of multilingualism then? Scientists have found cognitive benefits of knowing more than one language such as superior executive function and protection from Alzheimer's .
Words are the 'moulds' that shape our thoughts and perceptions — the greater the number of 'moulds', the more liberated and adaptive is our perception of the world.