The relation of language and thought has been a contested issue in philosophy and linguistics for several centuries.  Guy Deutscher’s contribution ( Through the Language Glass: Why the World Looks Different in Other Languages ) sorts through what we know and what we don’t know about that question.

The book is divided into two large sections: Language as mirror and language as lens.  Under the first metaphor, he probes the question of whether language reflects nature or culture.  Initially, he follows a traditional distinction between labels and concepts: Linguistic labels are culturally relative, but the concepts to which the labels are attached are given by nature.  ”Dog” and “chien” and “Hund” are arbitrary, but the range of objects to which they are attached is not.  Deutscher quickly complicates things by showing how “cultural penetrates the land of concepts.”  ”Dog” is an object of nature, but when we move to complex, inherently social realities like “victory” of “Schadenfreude,” it’s clear that the thing is not just out there waiting for us to pin a label on it.

Thus the relativity of cultural labeling affects the way nature is perceived and organized, and not just abstractly in theory but in everyday ways.  We use wee “we” whether we are talking about me and you, me and you and Bob, me and you and my dog and Africans, We the People.  Other languages, though, distinguish between “we two” and “we three” and “me and someone else but not you.”  We know our upper appendages are divided into arm, hand, fingers.  Hebrews use yad for both arm and hand, and Hawaiian uses a single term for the whole assembly.

So much for mirrors.  The lens section - language as the glass through which we look at reality - was the more interesting to me.  Does the lens affect what we see?  If so, how?

Deutscher begins with a deft summary of the debate that about the relation between thought and language going back to the nineteenth century.  Wilhelm von Humboldt launched the discussion.  Studying Basque, and then becoming familiar with South American languages through contact with Catholic missionaries, he recognized that languages cannot be forced “into the narrow rules of Latin grammar.”  The differences in vocabulary and in grammar, he suggested, point to differences of “worldview.”  Language is not merely “the means for representing a truth already recognized but much more to discover the truth.”  It is the “forming organ of thought,” such that thought “is dependent not just on language in general but to a certain extent on each individual language.”  He recognized that, in principle and given enough ingenuity and time, any concept can be expressed in any language, but was intrigued not by what language could do, but by “what it encourages and stimulates its speakers to do from its own inner force.”

Deutscher is sympathetic to von Humboldt’s measured theory, not sympathetic at all with the linguistic relativity of Yale’s Benjamin Lee Whorf, who tried to read worldviews quite directly off of grammatical structures.  American Indian languages do not distinguish between objects and actions; they don’t say “the stone fell” but “it stones down.”  From this, he drew the conclusion that their language betrays a “monistic view of nature.”  From his study of Hopi, he concluded that this tribe lacked all conceptions of time, since their language has no tense system.  Problem was, Hopi is replete with time expressions, as Ekkehart Malotki showed in a 1983 book.  It was too late: By 1983, Whorfism had spread everywhere, and everywhere clamored against the bars of the “prison-house of language.”

Deutscher is as hostile to Whorfism as one can be.  The “prison-house” idea is a “toxic fallacy.”  Yet Deutscher, refreshingly, does not think that the refutation of Whorfism answers the larger and more interesting question of how language might affect and shape thought.  If we can’t read a philosophy of time from the tense system of a language, is there any evidence that language plays the role of lens in more subtle and modest ways?  Here Deutscher returns to a von Humboldtian insight, mediated by Roman Jakobson.  As Jakobson put it, “Languages different essentially in what they must convey and not in what they may convey.”  Languages with grammatical gender, for instance, require speakers to identify the sex of a person about whom they are speaking.  ”I had my neighbor over Monday evening” might describe a romance or a common passion for football.  The French do not have the freedom of ambiguity.  The visitor was either voisin or voisine.

Deutscher spends an entire chapter on gender, citing studies that indicate that speakers of gender-inflected languages frequently associate feminine traits with feminine nouns, masculine traits with masculine nouns.  For Germans, chairs and keys, grammatically masculine, are thought of as “strong” more often than for Spanish, for whom both words are feminine.  Conversely, Spanish speakers find strength in grammatically masculine bridges and clocks more often than Germans, who speak a language in which clocks and bridges are feminine.  Because the languages force speakers to use masculine and feminine endings and articles with nouns, the language affects the experience of the objects.

He also spends a chapter with the Guugu Yimithirr, an Australian tribe most famous for giving the world the word “kangaroo.”  That is an interesting mystery in itself, but Deutscher is more interested in the fact that the Guugu Yimithirr language has no words for “egocentric orientation” - i.e., no left, right, front back - orientation with respect to me .  Instead, they always use absolute or “geographic” orientation - east, west, north, south.  What habits of mind does the language require from its speakers?  Deutscher answers, “In order to speak Guugu Yimithirr, you need to know where the cardinal directions are at each and every moment of your waking life . . . . in order to speak such a language, you have to have a compass in your mind, one that operates all the time, day and night, without lunch breaks or weekends.”  And the members of the tribe display exactly this talent: “They seem to have perfect pitch for directions.”

This has some interesting consequences.  Suppose two identically arranged hotel rooms across the hall from each other.  English speakers walk in Room 1, see the bed on the left and the TV on the right; they enter Room 2 across the hall, and again the bed is on the left and the TV on the right.  The rooms are “identically arranged.”  But for a speaker of Guugu Yimithirr: Room 1 has the bed to the east and the TV to the west, and Room 2 is arranged in exactly the opposite way.  What we think of as “the same reality” will be different for them.  Deutscher makes it clear that he doesn’t believe that we can read off worldview from the gramamtical structure of Guugu Yimithirr, but he concludes that the language “indirectly brings about a sense of orientation and geographic memory, because the convention of communicating only in geographic coordinates compels the speakers to be aware of directions all the time.”

Not only fascinating,  Through the Language Glass is delightfully and wittily written.  And it demonstrates that the question of the relation of thought and language is far from settled.