Through the Language Glass: Why the World Looks Different in Other Languages (9 page)

Looking back at Magnus’s theory from today’s vantage point, we cannot but wonder how such eminent scientists could have failed to pick up
on the various rather odd things about it. But we have to put ourselves in the mind-set of the late nineteenth century and remember that much of what we take for granted nowadays, for instance about the physics of light or the anatomy of the eye, was a complete mystery to scientists just over a century ago. The distance between us and Magnus’s contemporaries is even greater in all that concerns knowledge of biological heredity, or, as we call it today, genetics. And, since heredity is the pivot of the whole debate over language’s place between nature and culture, if we are to understand this debate, we need to pause for a moment and try first to jump over the gap of imagination that separates us from the 1870s. This task is far from easy, since the gap is about as long as the neck of the giraffe.

We are all acquainted with the logic of “just so” stories: the giraffe got his long neck because his ancestors stretched and stretched to reach higher branches, Kipling’s elephant got his long trunk because the crocodile pulled his nose until it stretched and stretched, and Ted Hughes’s lovelorn hare got his long, long ears from listening and listening, all through the night, for what his beloved, the moon, was saying high in the sky. Today’s children realize at a fairly early stage that all this is only fireside fable. The main reason why the logic of such stories is confined to the nursery is a truth so universally acknowledged that hardly anyone even bothers to state it explicitly nowadays. This is the understanding that physical changes you undergo during your lifetime will not be passed on to your offspring. Even if you do manage to stretch your neck, like the Padaung women of Burma with their neck rings, your daughters will not be born with longer necks as a result. If you spend hours on end lifting weights, this will not make your sons be born with bulging muscles. If you waste your life staring at computer screens, you may ruin your own eyes but the damage will not be passed on to your children. And training your eye to recognize the finest shades of color may make you a great art connoisseur, but it will have no effect on the color vision of your newborn offspring.

But what—to paraphrase Gladstone—every child in our nurseries knows today was not even remotely obvious in the nineteenth century. In fact, the inheritance of acquired characteristics wasn’t classed as fairy tale until well into the twentieth. Today, under the bright neon
light of the genetics lab, when the human genome has been mapped, when scientists can twiddle their pincers to clone sheep and engineer soybeans, and when children learn about DNA in primary school, it is difficult to imagine the complete darkness in which even the greatest minds were groping just over a century ago in all that concerned life’s recipe. Nobody knew which properties could be inherited and which could not, and nobody had any idea about the biological mechanisms that are responsible for transmitting properties down the generations. Many conflicting theories about the workings of heredity were doing the rounds at the time, but in this great cloud of unknowing, there seemed to be just one thing that everyone agreed on: that properties acquired during the lifetime of an individual could be inherited by the progeny.

Indeed, before natural selection came along, the inheritance of acquired characteristics had been the only available model for explaining the origin of species. The French naturalist Jean-Baptiste Lamarck proposed this model in 1802 and argued that species evolve because certain animals start exerting themselves in a particular way, and in so doing improve the functioning of specific organs. These successive improvements are then passed down the generations and eventually lead to the formation of new species. The giraffe, Lamarck wrote, contracted a habit of stretching itself up to reach the high boughs, “and the results of this habit in all the individuals of the race, and over many generations, was that its neck became so elongated that it could raise its head to the height of six meters [nearly twenty feet] above the ground.”

In 1858, Charles Darwin and Alfred Russel Wallace jointly published papers that outlined the idea of evolution by natural selection, and proposed an alternative mechanism to Lamarck’s evolution-through-stretching: the combination of accidental variations and natural selection. The giraffe, they explained, did not get its long neck by attempting to reach the foliage of higher shrubs and constantly stretching its neck for the purpose but rather because some of its ancestors that were accidentally born with longer necks than usual secured some advantage in mating or survival over their shorter-necked peers, and so when the going got tough, the longer-necked giraffes could outlive the shorter-necked ones. Darwin and Wallace’s joint papers were followed a year
later by Darwin’s
Origin of Species
, and—so most people would assume nowadays—Lamarckian evolution was immediately dispatched to the nursery.

Strangely enough, however, one of the only things that the Darwinian revolution did not change (not for half a century, that is) was the universal belief in the inheritance of acquired characteristics. Even Darwin himself was convinced that the result of exertions in particular organs can be passed on to the next generation. Although he insisted that natural selection was the main mechanism that drives evolution, he actually assigned the Lamarckian model a role in evolution as well, albeit an ancillary one. In fact, Darwin even believed until the end of his life that injuries and mutilations could be inherited. In 1881, he published a short article on “inheritance” in which he recounted reports about a gentleman, who “when a boy, had the skin of both thumbs badly cracked from exposure to cold, combined with some skin disease. His thumbs swelled greatly, and when they healed they were misshapen, and the nails ever afterwards were singularly narrow, short, and thick. This gentleman had four children, of whom the eldest had both her thumbs and nails like her father’s.” From the perspective of modern science, the only explanation for the story is that the man in question had a genetic disposition to a certain disease, which remained latent until he was frostbitten. What his daughter inherited, then, was not his injury, but this preexisting genetic trait. But as Darwin knew nothing of genetics, he thought that the most plausible explanation for such stories was that the injuries themselves were passed on to the offspring. According to Darwin’s own theory of heredity, this assumption was perfectly sensible, because he believed that each organ in the body manufactures its own “germinal material” with information about its own hereditary properties. So it was only natural to conclude that if a certain organ is injured during the lifetime of an individual, it may fail to send its germinal material to the reproductive system, and so the offspring may be born without the proper recipe for building the organ in question.

The belief in the inheritance of acquired characteristics was virtually universal until the mid-1880s. Only after Darwin’s death in 1882
were doubts starting to be raised, at first by one lone voice in the wilderness, the German biologist August Weismann. In 1887, Weismann embarked on his most notorious—and most often ridiculed—research project, the one that George Bernard Shaw lampooned as the “three blind mice” experiment. “Weismann began to investigate the point by behaving like the butcher’s wife in the old catch,” Shaw explained. “He got a colony of mice, and cut off their tails. Then he waited to see whether their children would be born without tails. They were not. He then cut off the children’s tails, and waited to see whether the grandchildren would be born with at least rather short tails. They were not, as I could have told him beforehand. So with the patience and industry on which men of science pride themselves, he cut off the grandchildren’s tails, too, and waited, full of hope, for the birth of curtailed great-grandchildren. But their tails were quite up to the mark, as any fool could have told him beforehand. Weismann then gravely drew the inference that acquired habits cannot be transmitted.”

As it happens, Shaw greatly underestimated Weismann’s patience and industry. For Weismann went on far beyond the third generation: five years later, in 1892, he reported on the still ongoing experiment, now at the eighteenth generation of mice, and explained that not a single one of the eight hundred bred so far had been born with an even slightly shorter tail. And yet,
pace
Shaw, it wasn’t Weismann who was the fool but the world around him. Weismann, perhaps the greatest evolutionary scientist after Darwin, never for a moment believed the mice’s tails would get shorter. The whole point of his perverse experiment was to prove this obvious point to an incredulous scientific community, which persisted in its conviction that acquired characteristics and even injuries are inherited. Weismann’s inspiration for the mice experiment was not the wife in the nursery rhyme but rather a tailless cat that was paraded to great acclaim before the Assembly of German Naturalist Scientists and Physicians in 1877 (the very year in which Hugo Magnus’s book was published). This tailless cat was flaunted as walking proof that injuries can be inherited, for its mother was said to have lost her tail in an accident and it was alleged to have been born tailless in consequence.

The received opinion at the time was that, even if mutilations do not affect the immediate offspring, they will crop up somewhere further down the line. This was why Weismann felt obliged not to limit his experiment to children and grandchildren but rather to curtail generation upon generation of hapless mice. Still, as bizarre as it may sound to us today, even Weismann’s endless genealogies of full-length mice tails did not manage to disabuse the scientific community of the belief in the heredity of injuries and mutilations. Nor did Weismann’s myriad other arguments find much favor, such as his invoking at least a hundred generations of circumcised Jewish males, who betrayed no disposition to be born without the offensive appurtenance and had to undergo the operation to remove it with each generation afresh. Weismann’s remained the minority view for at least two more decades, well into the twentieth century.

Throughout the second half of the nineteenth century, the debate on the evolution of the color sense was thus conducted entirely in the shadow of the assumption that acquired characteristics are inheritable. When Gladstone published his
Studies on Homer
, a year before
The Origin of Species
appeared, the mechanism that he proposed for the refinement of the color sense relied on the only model of evolution available at the time: Lamarck’s evolution-through-stretching. Gladstone’s assertion that “the acquired aptitudes of one generation may become the inherited and inborn aptitudes of another” was simply spouting received wisdom. Twenty years later, by the time Hugo Magnus came out with his anatomical explanation for the emergence of the color sense, the Darwinian revolution was already in full swing. But Magnus’s evolutionary model in 1877 was still identical to that proposed by Gladstone two decades earlier: it assumed that the retina’s ability to perceive colors increased through training and practice and that this progressive training was then passed on from generation to generation. While this reliance on the Lamarckian model seems to us like a great cavity right in the middle of Magnus’s theory, the flaw was not visible at the time. Evolution-through-stretching was not perceived as a direct contradiction
to Darwinism, so the Lamarckian nature of Magnus’s theory did not raise any eyebrows and was not attacked even by his critics.

Nonetheless, a few eminent Darwinists, not least Darwin himself, felt that Magnus’s scenario was problematic on other grounds, principally because of the very short time span it assumed for the development of color vision. It seemed implausible to these scientists that such a complex anatomical mechanism could have evolved so radically in the span of just a few millennia. Critical reviews of Magnus’s scenario were thus not long in coming.

But if—as the critics argued—vision itself had not changed in historical times, how could one explain the deficiencies in ancient languages that Gladstone and Geiger had uncovered? The only solution was to reconsider the question that Geiger had raised in the previous decade: Is it possible that people who could perceive colors just as we do still failed to distinguish in their language even between the most elementary of colors? For the first time, the question was now being thrashed out in earnest. Are the concepts of color directly determined by the nature of our anatomy—as Gladstone, Geiger, and Magnus believed—or are they merely cultural conventions? The debate over Magnus’s book was thus the start of the open war between the claims of nature and of culture on the concepts of language.

The opinion of Magnus’s critics was that since vision could not have changed, the only explanation must be that the deficiencies in ancient color descriptions were due to “imperfections” in the languages themselves. Their argument, in other words, was that one cannot infer from language which colors the ancients were able to perceive. The first person who made this point explicitly was Ernst Krause, one of Darwin’s earliest German disciples. But it was a biblical scholar, Franz Delitzsch, who put it most memorably when he wrote in 1878 that “we see in essence not with two eyes but with three: with the two eyes of the body and with the eye of the mind that is behind them. And it is in this eye of the mind in which the cultural-historical progressive development of the color sense takes place.”

The problem for the critics—whom we can dub somewhat anachronistically as the “culturalists”—was that their proposed explanation
seemed just as implausible as Magnus’s anatomical scenario, perhaps even more so. For how can one imagine that people who saw the difference between purple and black, or green and yellow, or green and blue, simply could not be bothered to differentiate these colors in their language? The culturalists tried to make the idea more appealing by pointing out that even in modern languages we use idioms that are rather imprecise about color. Don’t we speak of “white wine,” for instance, even if we can see perfectly well that it is really yellowish green? Don’t we have “black cherries” that are dark red and “white cherries” that are yellowish red? Aren’t red squirrels really brown? Don’t the Italians call the yolk of an egg “red” (
il rosso
)? Don’t we call the color of orange juice “orange,” although it is in fact perfectly yellow? (Check it next time.) And another example that would not occur to people in the nineteenth century: would race relations between the “dark browns” and “pinkish browns” have been as tortured as between “blacks” and “whites”?