Doublethink. Bellyfeel. Unperson. Do these words sound familiar to you? If so, you may be familiar with George Orwell’s Nineteen Eighty-Four, a novel that presents a dystopian society where certain words deemed too provocative are eliminated from the fictional language, known as Newspeak [1]. In Nineteen Eighty-Four, Big Brother—the totalitarian leader of society—reasons that without access to such words, people are unable to understand or even conceptualize the notion of oppression [1]. The backbone of this fictional order rests upon a theory of linguistic relativity also known as the Sapir-Whorf Hypothesis. This theory broadly denotes that the structure of language influences a speaker's cognition and perception [2]. Thus, one’s perception of reality is contingent upon their spoken and written language. The idea of linguistic relativity was first formulated around the 18th century but was notably brought into the spotlight following the work of American linguists Sapir and Whorf in the mid-20th century [2]. While there is much debate over the extent to which language impacts cognition, many studies have sought to investigate how the theory of relativity manifests in different grammar structures and languages. Opponents of linguistic relativism believe in universalism, a theory that states that every language is based on the same underlying universal structure [3]. While leading researchers continue to grapple with where on the spectrum between linguistic relativism and universalism reality lies, the question of cognition is far from settled.
The roots of linguistic relativity can be traced back to ancient Greece in the expostulations of Herodotus, who believed that Greeks and Egyptians thought differently because the Greeks wrote from left to right whereas Egyptians wrote from right to left [4]. In the 18th century, German philosophers J.G. Herder and W.V. Humboldt proposed that people who speak different languages might have different experiences of forming ideas and attitudes [2]. However, it was not until the 1920s, under the influence of two prominent figures, Edward Sapir and Benjamin Lee Whorf, that this hypothesis became a part of the debate around language and cognition. Edward Sapir spoke broadly about the influence of language on the perception of reality. The basic tenet of Sapir’s reasoning supposed that the language in which one speaks and thinks shapes the way that individual perceives the world, and therefore, the existence of different language systems is synonymous with the existence of communities that respectively perceive the world differently [2]. He is famously quoted as saying, “The ‘real world’ is, to a large extent, unconsciously built up on the language habits of the group” [5]. He asserted that different societies experienced distinctly different worlds, “not merely the same world with different labels attached” [5]. Following Sapir’s work, one of his students, Benjamin Lee Whorf, made cross-cultural comparisons—namely between English speakers and members of indigenous tribes—to investigate the connection between language and perception [2]. Whorf conducted an exploratory investigation of the Native American Hopi language and found his efforts to interpret structural characteristics of the Hopi language according to indo-European categories unsatisfactory. Working through this unexpected divergence, Whorf proposed that the way the Hopi tribe perceived reality must differ from the way Europeans perceived reality. He claimed that speakers of native languages, especially those without the constraints of traditional European subject and predicate sentence structure, may be better equipped to perceive abstract states over measurable events in definitive time. Whorf’s more radical conclusions form the bulk of what is commonly referred to as the Sapir-Whorf hypothesis [2].
Despite its initial empirical support, however, the Sapir-Whorf Hypothesis has become a prominent unsettled scientific debate. Its principal claim is that language impacts cognition—whichever language you speak quite literally changes the way you think and perceive. Many view supporting the linguistic relativity theory as a double-edged sword. On one hand, the grammar and lexicon (words and vocabulary) of a language can help us perceive, describe, and communicate in a fashion best suited for our environment. On the other hand, as in George Orwell’s Nineteen Eighty-Four, if we believe that we can only think within the bounds of our language, then we might limit our understanding of processes formed outside of our linguistic framework [2]. Modern researchers have not been so quick to agree with Sapir and Whorf’s claims. In fact, psychologists and neuroscientists have used modern technologies, scientifically-informed methodologies, and new behavioral and molecular lenses to poke and prod at the legitimacy of linguistic relativity for decades.
Since the start of the 21st century, researchers have explored the influence of language on many different forms of human perception, including color, space, time, shape, materials, and events [6-11]. Even as many milestone studies were published in support of the linguistic relativity hypothesis, a number of them were brought into question by opponents unable to reproduce their results [9,13,15].
Several relevant studies from the mid-2000s, for example, reviewed the influence of language on color perception [6, 7]. The basic premise of many of these studies was to examine how long it takes people to perceive a difference between two colors, presented visually as a target color and a background color [6]. People typically take longer to move their eyes between colors within the same color category (e.g., sky blue and teal) in comparison to colors from different categories (e.g., blue and green). The phenomenon is known as categorical perception of color [6]. In studying the effects of language on categorical perception of color, researchers found that it took less time for adults to detect color differences between colors from different categories—only, however, when the target was in the right visual field (RVF) [7]. Consider that the right visual field (RVF) projects to the left hemisphere and the left hemisphere of the brain controls functions related to language and speech. This establishes a potential link between the way our brains process color and the vocabulary we use to differentiate colors. In the same study, however, infants more expediently detected colors from different categories only in the left visual field (LVF). This led scientists to believe that the categorical perception distinctive to the left hemisphere in adults is not an innate feature but rather an acquired development driven by the categories imposed by language. In other words, the more fluent you become in a language, the more this language impacts your perception and brain function. While this study has in more recent years come under scrutiny for methodological discrepancies, neuroimaging studies have shown the connection between color perception tasks and brain regions involved in language, including the medial frontal gyrus, the mid-inferior prefrontal cortex, and the insula [7]. As it stands now, we can’t say with certainty that language determines patterns of cognition related to color, but tools for neuroimaging and theoretical modeling might help us draw a more sound conclusion on the robustness and validity of the Sapir-Whorf hypothesis.
Another area of focus has been the impact of language on the conception of time. A well known study conducted by Lera Boroditsky of Stanford University in 2001 uncovered differences in conceptions of time between English speakers and Mandarin speakers [8]. In English, we refer to time using terms associated with horizontal orientation; for example, “there are good times ahead;” “we push deadlines back.” In Mandarin, however, time can sometimes be referred to using both horizontal terms, like 前（qián; front/before) and 后（hòu; back/after), and vertical terms, like 上 (shàng; up/earlier) and 下 (xiá; down/later). Boroditsky then conducted a series of experiments to examine the differences in time perception between speakers of these two languages, and found that when Mandarin speakers were primed to vertical orientations, they were quicker to answer questions about order of calendar months, and these same results were true for English speakers when primed to horizontal orientations. In both these scenarios, individuals were able to think and process questions faster after having been primed to the orientation corresponding to their native tongue. Boroditsky portrays language as a “powerful tool in shaping thought about abstract domains” although it “does not entirely determine one’s thinking in the Whorfian sense” [8]. Despite her cautious conclusions, however, scientists took issue with Boroditsky’s methodology. Authors January and Kako argued in 2007 against Boroditsky’s claims for lack of replicability in six subsequent attempts [9]. January and Kako contend that there is not sufficient proof that language, rather than simply cultural convention, drives differing internal representations of time [9]. As with research on language and color, research on language and time leaves us with unanswered questions and occasionally unsatisfactory conclusions. But with these questions, researchers are driven to explore novel methodologies and evolving technologies to eventually have a better grasp on linguistic relativity.
Two opaque head silhouettes with gears in the place of brains looking towards one another. An abstracted cloud of bright colors connects the heads, and flows between the two individuals, connect them. Each silhouette has an abstracted circular stream flowing from its lips, surpassing the others head.
Another interesting arena in which to observe linguistic relativity can be found in how the referent type of an object (whether an individual classifies it by shape or material) varies with different languages. One case study compares English with the Yucatec Maya language [10]. In Yucatec, speakers use quantificational unitizers based on material composition for all nouns both discrete and non discrete. In English, unitizers are employed for only non-discrete nouns based on shape. In English, the numbers do not require unitizers for discrete objects (three cars) but the numbers do require units for non-discrete things (three clumps of mud). In Yucatec, all things have unitizers attached to their numbers so the comparative translations would be “three 'long thin' candles" and “three units of mud.” It is interesting to note that in English, the quantificational unitizer is commonly the shape of the noun, whereas in Yucatec, the focus is on the material composition. The implications of this are that Yucatec speakers may focus more on an object’s material composition whereas English speakers may focus on an object’s shape, thus undergoing different cognitive experiences when engaging with the same object. The supporting case study conducted by Lucy and Gaskins involved displaying triads of naturally occurring objects to adult speakers of both languages. Each triad contained a central “pivot” object and two “alternate” objects, one with the same shape as the pivot and one with the same material. In line with expectations, English speakers were found to match the pivot object to the object corresponding in shape, and vice versa for the Yucatec [10]. Lucy and Gaskins’ research honed in on individuals of different language fluencies perceiving objects according to categorically differing identifiers. Through these trials, they shed light on the way that our native language can influence the way we perceive objects and attribute characteristics to people, places, and things.
Lastly, the influence of language on non-linguistic factors provides strong support for the Whorfian hypothesis. A prominent example of this comes in the form of a study by Papafragou, where the eye-movement patterns of native speakers of English and Greek were monitored as they watched motion events (objects moving) [11]. Backing the foundation of this study was the phenomenon that certain languages—including English, German, Russian, and Chinese—tend to encode manner in the main verb (such as jog, roll, or march), whereas other languages—Greek, Spanish, French, Japanese—encode path in the main verb (enters[jogging], enters[rolling], enters[marching]). When participants watched the motion event under instruction to describe it verbally, Greek speakers focused on path over manner overwhelmingly more than English speakers. When told to watch the motion event freely, without instructions to describe it afterwards, eye-movement patterns were very similar for the two groups until the very end of the animation—when English speakers switched to focus on path (instead of the usual manner) and Greek speakers focused on both [11]. In other words, the respective influence of different languages on the brain had implications extending beyond brain function immediately tied to verbal communication—in fact, on a function entirely separate from speech. The big breakthrough was that language impacted thought even when the end-goal was not verbally-oriented (to put into words what happened). The researchers concluded that, in response to visual information, individuals mentally generated “linguistic codes,” which are non-verbal ideas that follow the structure of one’s language. Think of these codes as the kinds of thoughts that form just before you speak [11]. These codes then interfered with participants’ visual processing, thus highlighting the direct connection between language and thought [11, 12].
Fierce opposition to the Sapir-Whorf Hypothesis comes in the form of Universal Grammar (UG), a school of thought formulated by Noam Chomsky that asserts all languages share the same underlying structure [13]. Proponents of UG generally agree that it serves as a common grammatical starting point across all languages, off of which learners of a language build for their respective tongues [14]. UG highlights the commonality among languages, rejecting the premise of linguistic relativity which argues in favor of language-imposed cognitive discrepancies [14]. Many linguists have taken issue with the Sapir-Whorf line of reasoning connecting language to cognition. Steven Pinker widely promoted that there is a specific language module in the mind which deals with processing and production of language [15]. Being informationally encapsulated, this module does not have access to other aspects of cognition in its processing. Similarly, this model of mental architecture would have other cognitive processes equally encapsulated, so language would not be able to affect them in their computational processes. This point is most clearly illustrated when Pinker states, “People do not think in English or Chinese or Apache; they think in a language of thought,” going on to espouse the term Mentalese to refer to this shared language of mental processing [15].
Though most researchers currently reject the strongest form of this hypothesis, which claims that one can only think in accordance with the grammatical structure of one’s language, several researchers have adopted weaker versions of the hypothesis, claiming language ‘‘influences’’ or ‘‘suggests’’ thought patterns or default modes of interpreting the world rather than determining cognition absolutely [4]. Hunt and Agnoli are a prime example of this, who quantify the weaker Whorfian hypothesis and show that thought can be influenced by variations in the lexical, syntactical, semantic, and pragmatic aspects of language [4]. The more moderately positioned studies all point to the shared conclusion: a gap remains in understanding the neuroscience behind different neurological processing. This holds true among not only people who speak different languages, but among people at different levels of development for a shared language [6, 7, 8, 11]. A consensus in favor of more brain scans and scientific trials features prominently among these thinkers [4].
The notion of language impacting brain function has consequences that span beyond the scientific community. There is an ongoing debate in popular culture contemplating to what extent the structure and lexicon of a language impacts the perception and consideration of sex and gender [16]. For example, gendered languages like Hindi, Spanish, French, and Arabic, use masculine as the default grammatical gender [16]. Many scholars have argued that dominant social structures determine the rules of linguistic usage and the meanings attached to them, and since men have historically been in this position, the rules and meanings have been shaped according to their bias [17]. Linguists unsurprisingly differ on this point of contention. Luce Irigaray embraces a strong version of the Sapir-Whorf hypothesis in his response to this question, declaring that “the possibility of another language is the only chance at escaping the ‘mark’ of gender which, for the feminine, is nothing but the phallogocentric erasure of the female sex.”. In other words, within Irigaray’s native tongue (French), the role of the female is inherently constrained by the setup of the language, and in order to escape this hierarchy, a new language must be used. Monique Wittig, on the other hand, determines language to be an instrument that is not inherently misogynistic in its structures, but only in its applications. In her work Gender Troubles, Judith Butler weaves together these conflicting viewpoints, reaching the inoffensive conclusion that even as “gaining recognition for one’s status as a sexual minority is a difficult task within reigning discourses of law, politics, and language, [she] continue[s] to consider it a necessity for survival” [17]. This declaration by Butler serves as license for a more thorough investigation of the impact of language on how sex and gender are regarded.
New research conducted in the last five years aims to resolve the two major points of contention within the linguistics community around the Sapir-Whorf hypothesis, namely the possibility of a common system of human cognition and the difficulty in replicating prior studies investigating the Sapir-Whorf hypothesis [18]. To this end, researchers have turned to the new frontier of neuroscience: computers. Researchers can develop mathematical models that mimic the complex neural processing and dynamics that occur in the human brain, and can even predict to high degrees of accuracy how the human brain will interpret stimuli [19]. These models are developed and constrained by data from previous experiments. Then new experimental data is fed to the model to assess its ability to capture and predict brain dynamics [19]. One such model is the category adjustment model, a model developed by Huttenlocher and Hedges to explain how we perceive the outside world and how we use and develop memories of our perceptions [20]. Huttenlocher and Hedges observed that when these models were presented with stimuli, they increasingly relied upon the category of a stimulus the more unsure they were about the exact nature of the object [20]. Numerous studies have used these models with empirical data to investigate the Sapir-Whorf hypothesis, and the results have been striking. Bae et al. observed that, when human subjects were asked to identify colors with increasing degrees of uncertainty, their assigned color values shifted away from exact values (turquoise or aquamarine) and favored broader categories named by English color terms such as blue or green [21]. This outcome was consistent with the results predicted by the Sapir-Whorf hypothesis, which anticipated that the category system of a language affects how stimuli will be perceived by language speakers [21, 22]. Such models help address some key drawbacks of live models and/or psychological studies, such as replicability and accounting for and limiting external variables through controls [21]. Mathematical models are a novel capacity to predict our perceptions of stimuli by mimicking the process by which we perceive them and signal the dawn of a new age for neuroscience.
An abstracted brain with neon, hairlike projections flowing into and throughout it.
We have paid tribute to the various stages of development of the Sapir-Whorf Hypothesis—from early studies of linguistic relativism, which supported a strong version of the theory, to more recent data, which point toward a weaker, less stringent interpretation. These studies suggest that cognition is not exclusively built upon a universal framework of language processing, nor is it entirely drawn from a language-influenced bias, but is rather a mixture of the two. Determining just where on this spectrum between relativity and universalism cognition truly lies is a work in progress continuously advanced with the help of evolving research techniques. Whether your native tongue is English, Mandarin, or George Orwell’s infamous Newspeak, the one thing we know for certain is that we have a long way to go to fully understand the mysteries behind language and what we call “perception.”
References
Orwell, G. (2021). Nineteen Eighty-Four. Penguin Classics.
Hussein, B. A-S. (2012). The Sapir-Whorf hypothesis today. Theory and Practice in Language Studies, 2(3): 642-646. doi:10.4304/tpls.2.3.642-646.
Universalism. Oxford Reference. Retrieved 28 Nov. 2021, from https://www.oxfordreference.com/view/10.1093/oi/authority.20110804150420139.
Hunt, E. & Agnoli, F. (1991). The Whorfian hypothesis: a cognitive psychology perspective. Psychological Review, 98(3): 377-389. 0033-295X/91/S3.00
Sapir, E. (1929). A study in phonetic symbolism. Journal of Experimental Psychology, 12(3): 225-239. https://doi.org/10.1037/h0070931.
Gilbert, A. L., Regier, T., Kay, P., & Ivry, R. B. (2007). Support for lateralization of the Whorf effect beyond the realm of color discrimination. Brain and Language, 105: 91-98. doi:10.1016/j.bandl.2007.06.001.
Franklin, A., Drivonikou, G. V., Bevis, L., Davies, I. R. L., Kay, P., & Regier, T. (2008). Categorical perception of color is lateralized to the right hemisphere in infants, but to the left hemisphere in adults. PNAS, 105(9): 3221-3225. www.pnas.orgcgidoi10.1073pnas.0712286105.
Boroditsky, L. (2001). Does language shape thought? Mandarin and English speakers' conceptions of time. Cognitive Psychology, 43(1), 1–22. https://doi.org/10.1006/cogp.2001.0748
January, D. & Kako, E. (2006). Re-evaluating evidence for linguistic relativity: reply to Boroditsky (2001). Cognition, 104: 417-426. doi:10.1016/j.cognition.2006.07.008.
Lucy, J. A., & Gaskins, S.W. (2003). Interaction of language type and referent type in the development of nonverbal classification preferences.
Papafragou, A., Hulbert, J., & Trueswell, J. (2008). Does language guide event perception? evidence from eye movements. Cognition, 108: 155-184. doi:10.1016/j.cognition.2008.02.007.
Wolff, P. & Holmes, K. J. (2011). Linguistic relativity. WIREs Cogn Sci, 2: 253-165. DOI: 10.1002/wcs.104.
Steiner, G. (1972). Whorf, Chomsky, and the student of literature. New Literary History, 4(1): 15-34.
Kliesch, C. (2012). Making sense of syntax- innate or acquired?: contrasting universal grammar with other approaches to language acquisition. Journal of European Psychology Students, 88-94. https://doi.org/10.5334/jeps.au.
Pinker, S. (2000). The language instinct: How the mind creates language. New York: Perennial Classics.
Dutta, N. (Oct 6, 2020). The subtle ways language shapes us. BBC Culture. https://www.bbc.com/culture/article/20201006-are-some-languages-more-sexist-than-others
Butler, J. (1990) Gender trouble: Feminism and the subversion of identity. Routledge, New York, 33.
Cibelli, E., Xu, Y., Austerweil, J. L., Griffiths, T. L., & Reggier, T. (2016). The Sapir-Whorf hypothesis and probabilistic inference: evidence from the domain of color. PLoS-One, 11(7): 1-28. e0158725. doi:10.1371/journal.pone.0158725.
Einevoll, G. T. (2006). Mathematical modeling of neural activity. Dynamics of Complex Interconnected Systems: Networks and Bioprocesses in NATO Science Series 11, 232: 127-145. https://doi.org/10.1007/1-4020-5030-5_8
Huttenlocher, J., Hedges, L. V., & Vevea, J. L. (2000). Why do categories affect stimulus judgement? Journal of Experimental Psychology, 129(2): 220-241. DOI:10.1Q37//0096-3445.129.2320
Bae, G. Y., Olkkonen, M., Allred, S. R., & Flombaum, J. I. (2015). Why some colors appear more memorable than others: A model combining categories and particulars in color working memory. Journal of experimental psychology, 144(4), 744–763. https://doi.org/10.1037/xge0000076
Regier, T., & Xu, Y. (2017). The Sapir-Whorf hypothesis and inference under uncertainty. Wiley interdisciplinary reviews. Cognitive science, 8(6), 10.1002/wcs.1440. https://doi.org/10.1002/wcs.1440
תגובות