2020/1 PSYCHOLINGUISTICS

 15.4 Theory 3: Language determines or shapes our perception of nature

15.4.1 Proponents of the theory

Whorf, Sapir, Korzybski, and others are of the view that one’s knowledge  of vocabulary or syntax influences one’s perception and understanding of nature.

Newtonian space, time, and matter are not intuitions. They are recepts from culture and language. That is where Newton got them.

(Benjamin Whorf, from Carroll, 1956, p. 153)

Einstein’s relativity of time is a reform in semantics, not in metaphysics.

(Philipp Frank, 1953, p. 218)

15.4.1Inadequacies of the theory

We will raise a number of objections to the theory. These will be under the headings of: (1) Perception, interest, and need determine vocabulary; (2) Colour and snow vocabulary; (3) Hopi ‘time’ and Chinese ‘counterfactuals’;

(4) Lack of vocabulary does not indicate lack of concept; (5) Knowledge overrides literal word meanings; and (6) Multilinguals’ view of nature.

15.4.1.1      Perception, interest, and need determine vocabulary

Psychologists have tried experimentally to determine what effects, if any, knowledge of the vocabulary of language has on perception or behaviour. (See Niyekawa-Howard, 1972, for an older but interesting survey of such research.) Rather than vocabulary determining our interest and need, re- searchers have found the contrary to be true. It is our interest and need that determines our coinage of vocabulary and its use. Nowadays people know little of the vocabulary surrounding horses and horse-driven transportation. We know a lot, though, about automobiles and their parts and functions. Vocabulary is selected for use. Once it fails to serve a need, it falls out of use. Regarding interest and need as the prime motivators for the acquisition of vocabulary, we find that American children, like children in many other countries, are enchanted by dinosaurs. They can often name 25 or more! It

is not the case, however, that they perceive the types of dinosaurs because of their language. Rather, it was through perception that they developed their interest in dinosaurs so that they seek the names of these objects. Pronoun- cing the names for such creatures is not an easy task, as many a parent will testify!

15.4.1.2           Colour and snow vocabulary

Colour words

Some languages have only a small number of colour words. The Dani lan- guage of New Guinea has only two colour words, one for light colours and one for dark colours. If language were the basis of thought and of the perception of nature, as Whorf and the others contend, then one would expect speakers of this language, with such a limited repertoire of colour  terms, to have perceptual difficulty in distinguishing colours they have no terms for. Research has generally shown that this is not the case, although some conflicting results have been obtained.

Kay and McDaniel (1978) in a large cross-cultural investigation found no difference in perception of colours for different language speakers. Heider (1972) found similar results in testing the Dani people, who have only two colour terms in their language. Thus, speakers of languages that have only two, three, or four colour terms are as capable of distinguishing among the many colour bands of the visible spectrum as those whose languages have more than eight basic colour terms. People can see the differences but will not give them a name unless there is a good reason to do so. Kay and McDaniel (1978) conclude that ‘rather than language determining percep- tion, it is perception that determines language’ (p. 610).

A recent investigation, Robertson (1999), however, contends that the per- ception of colour is determined by the colour words of a language. The researchers studied the Berinmo, a people who live in Papua New Guinea close to the Dani and who have five colour words in their language. They report that in a matching task, the Berinmo were more apt to match colour tokens together according to their language; just as English speakers were more apt to put colour tokens together according to their language. Now, supposing that these results are valid, what is one to conclude?

In a New York Times report Robertson (1999) states that ‘Berinmo color vision is the same as ours. If they are asked to identify a single color from a group of colors, they would do it in the same way as you or I. But say that you have three colors, and call two of them blue and one green. We would see them as being more similar because we call them by the same name. Our linguistic categories affect the way we perceive the world’ (emphasis ours). But, is this conclusion justified? Clearly not, because Robertson’s conclu- sion contradicts what she had said before, that Berinmo colour vision is the same as ours and they can identify a single colour from a group of colours.

It is clear from this statement that the colour terms of the Berinmo language did not affect the Berinmo’s basic perception of colours. If the Berinmo are able to identify colours as English speakers can, it follows that the Berinmo’s perception of colour has not been affected by their language.

Snow words

There are dramatic vocabulary differences from language to language. The Inuit (previously called Eskimos), for example, have a large number of words involving snow. In Hawaii, there is only one, the English word ‘snow’. The Inuit, though, have single words for snow-on-the-ground, hard-snow- on-the-ground, block-of-snow, and others. Incidentally,  the  superficiality of the linguistic analyses of the Inuit language has been documented by a number of theorists. (For an excellent summary and discussion on the snow example see Martin, 1986.)

There is no reason to suppose, though, that Inuits learn to perceive varieties of snow because of their language rather than through their life experience and needs. It is because of the importance of snow in their lives that they have created more words for snow than have Hawaiians. Then, too, English- speaking skiers in cold countries do distinguish a variety of types of snow despite the lack of vocabulary in English. What they do to describe in language the physical condition of snow is create phrases, e.g. ‘powder snow’, ‘wet snow’, etc. It is this language device of creating phrases that every language has that makes up for any vocabulary deficiency.

15.4.1.3      Hopi ‘time’ and Chinese ‘counterfactuals’

Hopi people and time

Whorf’s research in the 1930s with the Hopi, Native American people, con- vinced him that their language forced them to see the world in a completely different way from speakers of European languages. He believed that the Hopi language had few words relating to time, and that this gave them radically different concepts of space and time.

More recently, however, other researchers have found that Whorf was wrong in claiming that the Hopi language is a ‘timeless’ one. Gipper (1979), for example, who lived with a Hopi family for a period of time and studied their language, found that while Hopi does not have a formal tense, it nevertheless contains a whole series of expressions for time. Many of these expressions appear as adverbs or prepositions. According to Malotki (1985), another researcher who spent many years living with the Hopi, Hopis actually do use a variety of time referents such as periods relating to harvest, the moon, the sun, and other significant events. We do much the same in English (‘Let’s go when it gets dark’, or ‘I’ll fix it when the weather gets warm’). Malotki very neatly concludes, ‘People are not different because of their

languages, but because of their experience. Deep down, we’re all the same. It couldn’t be otherwise.’

The Chinese language and ‘counterfactuals’

Reminiscent of the Whorf claim for Hopi and its later debunking by Gipper and Malotki is the claim by Bloom (1981) about Chinese. After a superficial analysis of Chinese, Bloom claimed that Chinese were not as able as English speakers to think hypothetically about what is not true (to think ‘counter- factually’) because of certain grammatical features of the Chinese language. Again, someone who really knew the language, in this case Au (1983), found that the results obtained were due to faulty translations, and that once proper translations were made, there was no basis for claiming a difference in thinking.

15.4.1.4           Number vocabulary in the  Amazon

Interestingly, two studies on Amazonian languages were published in 2004, both of which concerned with speakers’ numerical understanding. Pica et al. (2004) investigated the Munduruku language, while Gordon (2004) investig- ated the Piraha language. Both of these languages have a rather small set  of number words. Munduruku has words only up to the number 5, while Piraha has only words for 1, 2, and many (more than 2).

After applying some tests, in both cases the authors of the studies con- cluded that the speakers of these two languages were conceptually affected negatively because of the small size of their number vocabulary. The reac- tion to this conclusion was quick and pointed. The critics demonstrated problems in the reasoning of the researchers, particularly Premack and Premack (2005) for Pica et al.’s Piraha study, and Casasanto (2005) for Gordon’s Munduruku study. In the latter case, the criticism was so telling that Gordon (2005) reversed his earlier position. The dominant view emerg- ing is that, given the proper settings, speakers of these Amazonian lan- guages would not show a cognitive defect because of their language. Premack and Premack (2005) argue that these limited numerical systems are conven- ient and suitable for the needs of these Amazonian speakers, who are hunter- gatherers. They say, ‘Although hunter-gatherers had little need for exact numbers, one can imaging that no owner of stored goods would wish to receive three casks of oil when he had stored four’ (Premack and Premack, 2005, p. 673).

15.4.1.5           Lack of vocabulary does not indicate lack of concept

Simple vocabulary is  not a good measure of the concepts that speakers of   a language may hold. It is a fallacy to believe that the vocabulary of a language represents the sum total of the concepts that a person or culture may have. Experience does not always lead us to coin vocabulary items; some- times we choose to create phrases. For example, English speakers are quite

aware of their hand, yet although we have a vocabulary item for the under- side, ‘palm’, we have no word for the topside. Instead we use a phrase, ‘back of the hand’. Lack of a vocabulary item is not indicative of a lack of a concept. Many theorists do not realize that a lack of vocabulary does not imply the lack of a concept. Thus we find an educator like Tadao Suzuki saying that ‘We recognize the fragments of the universe as objects or properties only through words . . . without words we could not even distinguish dogs from cats’ (Suzuki, 1984, p. 35). However, if dogs and cats can distinguish one from the other without language, it would be absurd to think that humans could not do the same thing. New animals, plants, and other things in nature are continually being discovered. The discoverer notes the difference, then offers a name. Not the other way around! There would be no discoveries if we

first had to know the name of what it is we would discover!

15.4.1.6      Knowledge overrides literal word meanings

Consider such items as: ‘the sun rises’, ‘sunset’, ‘red hair’, ‘time flies’, and ‘white wine’. The theories under consideration imply that, as the result of our hearing and using such items, we would come to believe that the sun actually rises or sets on its own, that a person’s hair is actually red, and that white is the colour of white wine.

The fact of the matter is that we can believe something quite different from what the language literally specifies and that the continual use of a language form may not change an underlying thought. We know that the sun does not rise or set no matter how many times we hear people say it or we say it ourselves. English speakers have simply stuck with the original coinings even though it came to be known that it was the earth and not the sun that was doing the moving.

Such a fact, where one thing is said but another is understood (similar to lying, except that everyone knows that what is said is not true) runs counter to the implications of the theory being assessed here.

15.4.1.7      Multilinguals’ view of nature

If the language system forms or guides thought in the way we perceive nature, then multilinguals must be said to have a variety of ways of view- ing the physical world. The multilingual would have as many different conceptual–perceptual systems of the physical world as he or she has lan- guages. If it is true that different languages have distinctive and important effects on the way we view nature, then the multilingual must similarly have distinctive and important ways of viewing nature. As was noted in a section on multilinguals earlier in this chapter, no such differences have ever been noted. The multilingual person is a whole and integrated person who perceives nature as other humans do.