In the fictional world of Orwell's 1984, Newspeak was a language made to serve as a political and social controller. It was a way of thinking that was narrow and deliberately separate from all other paths of imagination. This, of course, is an especially pointed and hyperbolic version of a real feature of our world. We are all accustomed to our own culture's language and normally perceive the world in its light. For example, "time" is seen in one way as an object that can be spent, wasted, or invested, or free. This sort of language programing of society inspired the Sapir-Whorf hypothesis that language limits thought. "All our work, our whole life is a matter of semantics, because words are the tools with which we work, the material out of which laws are made, out of which the Constitution was written. Everything depends on our understanding of them."

In other words, language is the stuff of thought. It is foundational to all mental activity and determines the form of every possible thought. I think this is untrue. I think that language is a program for communication that has been given enough power to shape some thoughts, but is distinct from the "assembly code" of our minds and even as a container it can be stretched. There are many reasons to believe that language is not at the most fundamental level of our thinking, and that we can supervene our semantics.

"Uncritical semantics is the myth of a museum in which the exhibits are meanings and the words are labels. To switch languages is to change the labels." This is wrong. Even under the acceptance of the Sapir-Whorf hypothesis, this is wrong. In SW, different languages would correspond to slightly different museums, that might have mostly the same exhibits but with varying organizations of them. But SW still doesn't capture the full reality here.

Suppose we were to look at language from a reductionistic standpoint. In a child's development, how does the child learn a language? How does a child associate words with meanings if the meanings aren't already in their minds? The child must put a label with a purely nameless concept that works on an even more fundamental level. Also, a child does not take on a language all at once. The child will continuously construct a partial build of a language. That couldn't always be possible under SW, because the later parts of language that are being learned defy expression in the earlier parts of the language. Since language is accommodated into our minds in development, our minds must initially operate on differently without the modes of thinking provided by a particular language.

Now suppose we look at languages from a holistic standpoint. How did language get started in the first place? How do they change and diverge? These issues cannot be resolved with SW, because language is viewed too much as a rigid machine rather than an evolving organism. When Shakespeare set a shining example that effectively redefined the English language, he was inventing words and expressions that had never been seen before. He made metaphors that language could not alone hint at, and that makes it so comparably easy to understand today 400 years later even when other works only 200 years ago are alien to us now.

The idea that language at least influences thought is overwhelmingly accepted by linguists. For instance, the precise meanings behind "life" and "marriage" taken by someone can direct their opinions on abortion or gay marriage. Most linguists accept that there is a reciprocal relationship between language of society and thought of the individual. The nature of this relationship is not precisely spelled out and still has a haze of confusion surrounding it. With a thought experiment I intend to make this relationship a bit clearer.

Suppose we have many artificial intelligence programs that can observe and process information from the real world as well as interact with it. These AI's will learn to use patterns to communicate with each other (call these patterns L, for language), but there is little reason to believe these patterns will correspond exactly to the patterns they use to process information internally (call these patterns M, for mind).

Furthermore, suppose the AI's are hard-pressed for energy to sustain their computer hosts, and so the AIs, by their programming, naturally attempt to internally use efficient ways of "thinking" that will better suit them for dealing with the world and getting energy, however that may be. To this end, once the AIs have developed complex methods of communicating with each other, they will start automatically translating their internal "thoughts" of M into communicable ideas of L, so that they are ready on-demand for discourse about these ideas. Eventually, it will start adopting L itself as the main way of processing the world to further optimize their thinking because they are highly sociable pieces of software.

This scenario, I hope, at least roughly mirrors the real case for us humans. We adopt languages because they are highly useful ways of thinking, and use it as a primary mode for expressing thoughts. However, we can change, add to or subtract from, and otherwise manipulate language as a tool or aid to living.

Would we have thought about gravity if Isaac Newton didn't help identify it? Or how about the concept of "space-time" before Einstein defined it?

Without words, could we have been able to conceive of the following abstract ideas: "soul", "reason", "energy", "subconscious", "conscience", "remorse", "quark", "pi", "square root", "calculus", "absurd", "meaning", "endurance", etc?

Does language shape our views and influence our thoughts? This question was first posed by Benjamin Lee Whorf in the 1930s. The Whorfian Hypothesis, as it is sometimes referred to, claims that the particular language a person speaks, independent of the culture in which he resides, affects the way that he thinks, by determining the framework for his perceptions and thoughts. Whorf and his teacher Sapir argued further that a person's world view is largely determined by the vocabulary and syntax available in his language.

The extreme version of this hypothesis is that all thought is constrained by language. If the word does not exist, then that thought cannot be thought. Of course this is patently untrue. For example, we all experience difficulties in expressing ourselves. We know what we are thinking of, but we are stumped to find the word to express it. Hence we do not need words to think.

At the opposite extreme, to say that language has absolutely no influence on thought, is also false. For example, if a language organized colour names differently, it has been shown that a person's discrimination of similar shades of colors can be influenced.

To explore this question further, Peter Gordon of Columbia University studied the Pirahn Indians. This tribe of hunter gatherers living on the banks of the Maici River in Brazil do not possess a vocabulary for numbers other than two words, "one" which stands for "one or two", and the other meaning "many". Members of this tribe are intelligent, but they have difficulty counting beyond 8 objects. Peter Gordon wrote: "They do not have the word for "number", pronouns do not encode number (e.g. "he" and "they" are the same word), and most of the standard quantifiers like "more", "several", "all", "each" do not exist." Peter Gordon concluded that without having numbers, the Pirahn Indians were handicapped in their ability to conceive of mathematical calculations. This tribe could survive without this concept because their commerce was a form of barter, in which there was no exchange of money. Numbers did not exist as there was no need for them.

Numbers, symbols and words, act as convenient labels for concepts. They help clarify one's thinking. For example, the 19th century physician Langdon Down discovered a subset of intellectually disabled who were later labelled as Down Syndrome individuals. This enabled better delineation of this group and allowed for further studies, such as their the genetic origins.

Words are like building blocks or stepping stones that allow us to advance concepts one step at a time. Armed with the concept of gravity, Isaac Newton took the next step and formulated the inverse square theory of gravity. Similarly, with Einstein’s concept of mass and energy, he could formulate the equation linking the two, which led to further discoveries.

One additional point of interest. Not only do the Pirahn Indians not count, but they also do not draw. Gordon wrote. "Producing simple straight lines was accomplished only with great effort and concentration, accompanied by heavy sighs and groans."

This leads us to the next issue, namely the written representation of words. Benjamin Lee Whorf was acutely aware of the role culture and society played on the development of concepts. Take two cultures, Western and Far Eastern. In the west, words are represented by an alphabet script, whereas in the Far East, words are represented mostly by pictograms. Does the representation of words in alphabet form versus pictograms lead to qualitatively different modes of thinking? Can it explain the differences between, say, Far Eastern philosophy from Western philosophy?

Trying to display abstract ideas in pictograph form is particularly challenging, because it is very tough converting pictures into abstract concepts. The Chinese do it quite cleverly. For example the word for endurance shows a knife directly above the heart, and the word for peace is a woman under the roof of a house.

Whether this pictographic representation of concepts can subconsciously lead to a different way of thinking and a different world view remains largely unexplored. Cross-cultural studies are therefore of immense interest, not only in understanding differences in thought processes, but also in unravelling other mysteries, such as dyslexia.

Well over half a century ago, Benjamin Lee Whorf [Carroll JB (1956) Language, Thought, and Reality: Selected Writings of Benjamin Lee Whorf (MIT Press, Cambridge, MA)] proposed that language affects perception and thought and is used to segment nature, a hypothesis that has since been tested by linguistic and behavioral studies. Although clear Whorfian effects have been found, it has not yet been demonstrated that language influences brain activity associated with perception and/or immediate postperceptual processes (referred hereafter as "perceptual decision"). Here, by using functional magnetic resonance imaging, we show that brain regions mediating language processes participate in neural networks activated by perceptual decision. When subjects performed a perceptual discrimination task on easy-to-name and hard-to-name colored squares, largely overlapping cortical regions were identified, which included areas of the occipital cortex critical for color vision and regions in the bilateral frontal gyrus. Crucially, however, in comparison with hard-to-name colored squares, perceptual discrimination of easy-to-name colors evoked stronger activation in the left posterior superior temporal gyrus and inferior parietal lobule, two regions responsible for word-finding processes, as demonstrated by a localizer experiment that uses an explicit color patch naming task. This finding suggests that the language-processing areas of the brain are directly involved in visual perceptual decision, thus providing neuroimaging support for the Whorf hypothesis.

Ein solcher Ausgangspunkt in die Richtung ist zum Beispiel die Behauptung, die westliche Wissenschaft wäre durch unsere Kultur so geformt worden, wie sie heute ist, und dass darum Wissenschaft eben nur kulturelles Konstrukt sei: aus einer anderen Kultur entstanden, hätte sich auch eine ganz andere Begrifflichkeit und damit ein ganz anderes Verständnis wissenschaftlicher Zusammenhänge ergeben. Nun mag es durchaus sein, dass zum Beispiel christlich-aristotelische Irrlehren über das Seelenleben des Menschen eher Ballast waren für die wissenschaftliche Beschäftigung mit Geist und Gehirn. Aber schon an dem Beispiel kann man sehen, dass die eben nicht auf Dauer die Neurowissenschaften nach ihrem Bild geformt haben, sondern irgendwann durch fortschreitenden Erkenntnisgewinn selbst als überholt betrachtet werden müssen und anschließend auch aus der Kultur verschwinden könnten, zumindest als populäres Konzept.

Sollte sich unsere Kultur irgendwann so verändert haben, dass sie die Bilder zukünftiger Gehirnwissenschaft so radikal in sich verinnerlicht hat; dann würde vermutlich der zukünftige Postmodernist kommen und sagen: "Aha, wer hat's erfunden...? Die Kultur hat mal wider die Wissenschaft geformt." Damit würde er dann nur seine eigene Geschichtsvergessenheit beweisen und weiteres Zeugnis darüber ablegen, dass man im faktenfreien Raum alles behaupten kann, um sich damit dann auf die eigene Schulter zu klopfen.

Unmöglich oder viel zu übertrieben?

Mark Liberman vom sehr informativen (wenn auch für deutsche Interessen öfter etwas abseitigen) Language Log untersucht heute eine alte, aber vielzitierte Veröffentlichung von Bejamin Lee Whorf ("Science and Linguistics"), in der dieser über Sprache und ihre Auswirkungen auf wissenschaftliche Begrifflichkeiten philosophiert. Der indo-germanische Sprachgebrauch sei praktisch wie geschaffen dafür gewesen, die Newtonsche Mechanik hervorzubringen. Demgegenüber hätten zum Beispiel die Hopi in Arizona (ein offensichtlich sehr mysteriöser Indianerstamm, auf den man an allen Ecken stößt: vom esoterischen Ramschladen an der Ecke mit ihren Traumfängern bis zu Erich von Däniken) wohl eine ganz andere Physik geschaffen, gibt es doch in ihrer Sprache kaum vergleichbare Begrifflichkeiten für Zeit oder Geschwindigkeit. Um fair zu sein, ich habe das Paper nicht gelesen, und es scheint auch nicht zwingend so, als würde Whorf persönlich in den postmodernistischen Fehlschluss verfallen, dass die tatsächliche Physik eine andere wäre: nur deren Beschreibung würde sich deutlich von unserer unterscheiden, vermutlich durch einen anderen mathematischen Formalismus. [2]

Erst mal zeigen sich beim Linguisten hier schon ein paar Missverständnisse über physikalische Begrifflichkeiten. Der physikalische Begriff Geschwindigkeit beschreibt nämlich im Formalismus eben das, was Whorf als I ausmachen will: die Geschwindigkeit ist da keine zeitliche Veränderung, sondern die Eigenschaft eines Objekts zu einem bestimmten Zeitpunkt; Whorf umschreibt also nur einen vermeintlich indogermanisch geprägten Begriff mit der eben schon geltenden physikalischen Begrifflichkeit, weil ihn sein indogermanisches Englisch über die wahre Bedeutung täuscht. Und dieses Missverständnis soll dann beweisen, wie gut unsere Physik unsere Sprachgewohnheiten widerspiegelt?

Liberman geht das Ganze dagegen aus der anderen Richtung an, eben der des Linguisten. Und als solcher stellt er dann fest, dass die Hopi-Sprache so unterschiedlich von unserer nicht ist, dass sie sehr wohl Ausdrücke für zeitliche Abläufe und Geschwindigkeit hat. Zum anderen aber untersucht er die Sprachgeschichte der englischen Ausdrücke Geschwindigkeit, Schnelligkeit und so fort, die Whorf als Beleg dafür sieht, dass wissenschaftliche Begrifflichkeit nur Sprache widerspiegelt. Dabei sind die englischen Begriffe auch nur aus ihrer langen Entstehungsgeschichte verstehbar, haben mehrmals ihre Bedeutung gewechselt, und mussten sich am Ende so anpassen, dass die Sprache besser die Begebenheiten wiedergeben kann, die Wissenschaft gefunden hatte. Mithin ist das ein starker Hinweis, dass eben nicht Sprache und Kultur Wissenschaft formen, sondern umgekehrt Wissenschaft die Weltwahrnehmung und damit eben auch Sprache und Kultur verändert.

Vervollständigt wird das durch entsprechende Zitate, etwa von Hobbes, in denen der mit dem alten Sprachgebrauch rang, um die wissenschaftlichen Begrifflichkeiten verständlich machen zu können, weil über die neu erkannten Zusammenhänge die Sprache selbst fremd geworden war. Den ganzen Post gibt's noch mal hier: "Poor, arid, and, in appearance, deformed".

[1] das gesellschaftswissenschaftliche Gegenstück zur Weltformel. Während allerdings selbst die ambitioniertesten Physiker nicht davon ausgehen können, mit der Weltformel alle Vielfalt der Welt erklären zu können, weil es da ja noch diese ganzen emergierenden Phänomene gibt, die wir Alltag nennen, scheint das auf Postmodernisten nicht zwangsläufig zuzutreffen, die fortwährend dahinformulieren, als wären sie nur einen Schritt von aller Wahrheit entfernt. Man muss ihnen das nachsehen: mit einem Theoriekonstrukt, aus dem noch nicht ein nachweisbares Phänomen abgeleitet werden, geschweige denn formuliert werden konnte, kann man natürlich auch keine Erfahrungen über Tragweiten und Begrenzungen von Theoriebildung gesammelt haben.

[2] was übrigens auch nur mangelndes physikalisches Wissen ist: in Wirklichkeit gibt es in der Physik natürlich verschiedene, mathematisch unterschiedliche aber letztlich äquivalente Formulierungen für die Mechanik. Was man Schülern und (Nebenfach-)Studenten beibringt ist allerdings meistens beschränkt auf einen vereinfachten Newton. Eben weil es erst mal darum geht, wissenschaftliche Konzepte möglichst einfach und nach einem gewissen Standard zu vermitteln.

The question of language and though and its reciprocal influence has been discussed countless times. It goes back all the way to the Greeks and probably even further. In the last 30s of last century the businessman and linguist Benjamin Whorf postulated what would be know as the Whorfian-Hypothesis. His hypothesis can be split into a strong version, called linguistic determinism and a somewhat weaker version called linguistic relativism. Linguistic determinism states that our thinking and behaviour causally depends on the structure of our language. Linguistic relativism postulates a relationship between language and thought, but rejects the idea that this relationship necessarily needs to be causal.

Today most scientists, believe that the strong version of the Whorfian hypothesis, as well as the opposite, the claim that language has no influence on thought, is wrong. Even stronger, the theory is thought to be non-scientific, because of circular reasoning, as Steven Pinker (1994). points out in his book The Language Instinct: “Eskimos speak differently so they must think differently. How do we know that they think differently? Just listen to the way they speak!”. In the 1960s, when Noam Chomsky’s work on universal grammar became popular, most scientists turned away from the Whorfian hypothesis, but advances in the field of cognitive psychology in the early 90’s sparked renewed interest. Since then an important shift has occurred in this field of research, as more basic cognitive processes such as spatial orientation and memory have been put on the agenda of the researchers. I am now going to present several pieces of empirical evidence in favour of linguistic relativism.

Memory

Capacity in short-term memory is limited. It therefore makes sense to symbolise complicated constructs in the short term memory and store the complete information of the construct concerned in long term memory (that is what language actually does). The influence of language might therefore for some part be mediated by memory. And indeed there is evidence that language influences information stored in memory. One interesting paradigm to test this are colour perception. Participants were presented with non-prototypical (turquoise) chips and, after a delay, had to pick the right one (Schooler, 1990). Participants that had been forced to label the colour as either blue or green performed worse on the retention task (in the direction of the forced label) than participants that had not labelled the colour of the chips. Interestingly the effect disappeared when participants where forced to use the label blue-green.

Spatial orientation

Whereas in most western languages the dimension of time is semantically situated on a horizontal plane, this is not the case for Chinese. In Chinese the dimension of time spans on a vertical plane and indeed Chinese participants in an experiment by Boroditsky (2001) were faster to confirm that April comes after March if they had previously been primed with a vertical array of objects. On the contrary, English participants were faster, when presented with a horizontal array of objects.

Categorization

Higher order categories seem to be consistent throughout different languages (at least when the objects at hand are equally used in both societies), but this is not the case with lower order categories. In an experiment by Ameel (2005) it was shown that in French and Dutch bilingual speakers (from Belgium, thus sharing one cultural background) the categories used to classify objects (bottles and dishes) shared features of both the naming-pattern used by French native speakers and by Dutch native speakers. The researchers therefore concluded that the classification of objects not only depended on the commonalties between the objects, but also on language-specific factors.

Self-perception

In a study by Kemmelmeier and Cheng, bilingual Students from Hong Kong filled in Singelis’s independent and interdependent self-construal scales in English as well as in Chinese. As hypothesised there was significant difference in the self-construal of the students between the two languages. Describing themselves in English, students attributed a more independent self-construal to themselves, whereas in Chinese their self-construal was more interdependent. Kemmelmeier and Cheng concluded that language can serve as a cognitive cue to prime or reinforce culturally normative self-construals. This is also supported by a lot of episodic evidence from people (including me – especially when talking and thinking French) feeling like a different person, when talking in a different language.

Conclusion

The renewed interest of science in the Whorfian-theory has generated many interesting research findings. New paradigms and techniques, as well as the fast-paced development of cognitive neuroscience, have opened new doors for the researchers. However there is still much room for new research and specification of old findings. Research between different languages is mostly cross-cultural research, inviting numberless alternative explanations. Because of that research findings will need to be replicated in designs using bilinguals as participants. Furthermore, as the research on language and memory has shown, we will need to specify which cognitive processes actually cause the influence of language on though and how these mechanisms work. At the end of this search we might even meet up with the lost Eskimos.

References

Ameel, E., Storms, G., Malt, B. C., & Sloman, S. A. (2005). How bilinguals solve the naming problem. Journal of Memory and Language, 53, 60–80.

Boroditsky, L. (2001). Does language shape thought? Mandarin and English speakers' conceptions of time. Cognitive Psychology, 43, 1-22.

Kemmelmeier, M., Cheng, B. (2004). Language and Self-Construal Priming: A Replication and Extension in a Hong Kong Sample. Journal of Cross-Cultural Psychology, 35, 705-712.

Pinker, S. (1994). The Language Instinct. New York, US: William Morrow and Company.

Schooler, J. W., & Engstler-Schooler, T. Y. (1990). Verbal overshadowing of visual memories: Some things are better left unsaid. Cognitive Psychology, 22, 36-71.