快猫短视频

BY THE time someone says a dozen words, we know if they are young or old, streetwise or otherwise, from halfway round the block or halfway round the world. How we speak doesn鈥檛 just say what we want it to-it whispers about who we are and where we鈥檝e been, tattles about our origins and ancestors. The same holds true for languages as a whole. The intricacies of their vocabulary and grammar hold clues to thousands of years of human history and prehistory.

That hidden, other voice of language captivated linguist Johanna Nichols at an early age and has dominated her life ever since. For over a decade, she trekked the far reaches of the Russian Federation-Dagestan, Chechnya and Ingushetia-to study and preserve native languages. Now, however, you鈥檙e more likely to find her in her spartan, sixth-floor office overlooking the Berkeley campus of the University of California. You might catch her gazing out the window, but don鈥檛 ask her to describe the students hurrying by, the rolling lawns or the towering eucalyptus trees. Nichols is preoccupied with faraway places and ancient times. Voices from hundreds of languages are clamouring for her attention.

What, she wonders, can modern languages tell us about a great wave of exploration and migration that began more than 50 000 years ago in Southeast Asia and eventually circled the Pacific Ocean? What clues do they hold about the earliest seafarers and about the people who first populated Australia, New Guinea, and the New World? And does the babble of modern tongues conceal a slowly ticking clock that can date the birth of human language itself? No one else would dream of using language to answer such questions. To do so, Nichols had to break out of a mindset that still holds other linguists back. She stopped asking how languages are related to each other and started asking more productive questions, such as how and why languages spoken tens of thousands of miles apart happen to share certain grammatical features.

For two centuries, linguists have painstakingly worked out family trees for languages. They have traced the lineage of most of the 6000 tongues spoken today, along with many that have vanished. When languages share enough similarities, or a regular pattern of differences, linguists can be sure they are related. For example, Spanish, Italian and Portuguese are easily recognisable as offspring of Latin, while English and Dutch are part of a family known as Germanic. Even when there are no written records of an ancestral language, linguists can use the comparative method to recreate much of its vocabulary and grammar. Over time, they have been able to link the remote ancestors of many of today鈥檚 tongues. For example, Germanic, together with Latin, ancient Greek and six other language families, is part of a group of related tongues called Indo-European which has 144 member languages.

But linguists constructing family trees eventually hit a wall. They can trace many language families back 6000 years, and a few up to 10 000 years. Beyond that misty horizon, random changes make it impossible to prove connections. 鈥淓verything changes over time in languages,鈥� says Nichols, 鈥渁nd even the most durable signs of similarity eventually fade out.鈥� Rather than discovering one great family tree with all the world鈥檚 language families and languages branching from it, linguists have found themselves wandering in a forest of 200 to 300 separate language trees, which Nichols calls stocks. Some stocks have dozens of languages branching from them, some have just a few, and some, such as Korean and Basque-a language spoken only in the Pyrenees of France and Spain-stand alone.

While most linguists continue to chip at that 6000-year wall, Nichols has vaulted over it. She set aside the tried-and-tested comparative method of reconstructing language family trees, and has developed a controversial new approach that focuses on identifying and mapping grammatical building blocks. Languages that share these durable features are not necessarily related but they do have 鈥渁ncient affinities鈥� according to Nichols. They may come from the same family, but their shared characteristics could just as easily be the result of neighbouring peoples borrowing linguistic elements from one other. Either way, the similarities can answer questions about the movements of languages and people tens of thousands of years ago.

鈥淯ntil Nichols came along, historical linguists had talked themselves into a straitjacket, in which they saw their major enterprise to be the reconstruction of proto-languages. She loosened that up to show that there are other important projects to be accomplished,鈥� says John Moore, an anthropologist at the University of Florida in Gainesville. 鈥淚 think it was a stroke of genius.鈥�

Nichols鈥檚 breakthrough came when she selected one well-studied language to represent each stock and located these on a map of the world. To her surprise, when she charted where each of her grammatical building blocks appeared, striking geographical patterns emerged. Many language features were neither tightly clustered nor randomly scattered. Instead, sets of markers traced intricate patterns round the globe. While other linguists dismissed these patterns as chance, Nichols was sure they made sense. She has spent the past decade teasing out how the migrations and dispersions of ancient groups created the geographic patterns she now sees. 鈥淲hen people ask me what I do,鈥� Nichols says, 鈥淚 say historical linguistic geography.鈥�

Nichols sorts languages by using several dozen features or typological markers. Some she has borrowed from other linguists studying language types, others Nichols discovered as she was ploughing through the grammars of hundreds of languages. Many are familiar-for example, how a language indicates possession, whether it favours prefixes or suffixes, or puts verbs at the beginning, middle or end of a sentence. Some are arcane, for example ergativity, a language feature in which special prefixes or suffixes modify the subjects of transitive verbs. For the most part, Nichols uses grammatical regularities rather than words or sounds because, she believes, grammar has a tendency to change more slowly over time. But, she has also found several useful sound features. For example, she classifies languages according to how many of their pronouns start with an 鈥渕鈥� or 鈥渘鈥� sound, or whether they use pitch to change word meanings, as do Chinese, Thai and Navajo.

Network of clues

One important, but less familiar marker Nichols plots on her maps is whether a language uses 鈥渘umeral classifiers鈥�. In English and its relatives, a number and what it is counting are linked directly. 鈥淪ix beagles鈥�, we say, or 鈥渘ine daisies鈥�. But most Asian languages, including Japanese, Korean and Mandarin Chinese indicate plurals with a bit of verbal Velcro. They say something like 鈥渟even-classifier-duck鈥� or 鈥渢hree-classifier-drum鈥�. These classifiers usually describe the shape of the object as well. And because things come in many shapes, numeral classifiers come in bunches: Yurok, which is spoken by a Native American tribe, has 15, Korean 26, and Mandarin 51.

Numeral classifiers form one strand of a remarkable configuration of languages, which Nichols calls the Pacific Rim Pattern. On her world maps, languages whose pronouns start with 鈥渕鈥� and 鈥渘鈥� sounds, languages that put verbs first, and languages that use numeral classifiers circle the Pacific Ocean like a necklace. They dot New Guinea and nearby islands, bead the coast of Asia from the south east to the northwest, and trail the length of the Pacific coast of North and South America. Remarkably, these features hardly ever appear in the vast reaches of Africa, Europe, and inland Asia or America.

Since there are no such languages in Australia, Nichols reasons that they must have reached New Guinea no more than 11 000 years ago, when rising sea levels separated the two regions. But other language markers speak of more ancient connections. Features such as 鈥渃oncord classes鈥�-where verbs and pronouns have to be of the same gender as the noun-show up along the coasts of New Guinea and Australia as well as farther inland.

And a third strand of grammatical markers including ergativity is more deep-rooted still. These appear only in the hinterlands-the east, south and interior of Australia, the highlands of New Guinea-and, it turns out, in the eastern reaches of the Americas (see Maps).

To Nichols, the Pacific Rim necklace tells an ancient story. Like gravel dropped by a tsunami, the language markers that litter the Pacific coast of three continents trace the course of a sustained and enormous wave of human migration which started more than 50 000 years ago in a densely populated area somewhere in Southeast Asia. Over tens of thousands of years, successive bands of people spread to the north and south. They could move relatively quickly, Nichols believes, because they could build simple boats or rafts and make a living from the sea. How else, Nichols asks, can she explain that coastal New Guinea and Western America are next-door neighbours when it comes to language type?

Recent research from the Chinese human genome diversity project supports the earliest stages of what Nichols has inferred. The team examined short stretches of DNA called microsatellites from 32 east Asian populations such as the Han, Wa, Hui and Yao. They compared these groups with other Asians, for example, Japanese, Koreans and Taiwanese, and with a variety of African groups as well. The genetic trees the team constructed indicate that early modern humans left eastern Africa and colonised Asia鈥檚 southern coast before spreading into east Asia.

Exodus through Asia

What鈥檚 more, in November 1999, researchers at the University of Pavia in Italy described a group of related mitochondrial DNA variants, called haplogroup M, found in Africa, western India, Tibet and Mongolia. In their findings too, Southeast Asia emerges as the portal through which modern humans spread throughout Eurasia after leaving Africa.

Based on their linguistic similarities, Nichols sees the first Americans and the first Australians as the northern and southern vanguards of the same vast expansion. Archaeologists agree that the first Australians arrived around 50 000 years ago. To reach the continent they would have had to cross at least 100 kilometres of open water, so they clearly possessed seafaring skills. Nichols thinks that at least some of the earliest immigrants to the New World also came by sea, paddling from harbour to harbour along the coast of Beringia-the land bridge that linked Siberia and Alaska-and then down the west coast of North and South America. Markers such as ergativity suggest a very early date indeed.

There was a time when such assertions would have been tantamount to heresy. But in recent years, excavations at Monte Verde in Chile, along with genetic and linguistic evidence have shattered a 50-year-old paradigm which said that the Clovis culture of hunter-gatherers were the first to reach the New World approximately 11 500 years ago (快猫短视频, 17 October 1998, p 24). Nichols assembled much of the linguistic evidence, based on another technique she has pioneered. Her approach goes far beyond glottochronology, the standard technique for dating the emergence of related languages through similarities in vocabulary.

Nichols studied all well-described language stocks (mostly from the northern hemisphere) to find how often they branched to create new families, compared to the rate at which families disappear. She found that some stocks produced lots of offspring, most spawned just a few, and some sired none at all. On average, one-and-a-half new families developed per stock every 6000 years. In effect, Nichols created a linguistic clock, ticking once every 6000 years. She used this to work backwards from the great diversity of Native American languages-more than half the world鈥檚 stocks-to date America鈥檚 first pioneers to between 20 000 and 30 000 years ago, in the depths of the Ice Age.

Linguist Nicholas Evans of the University of Melbourne, however, is not convinced that Nichols鈥檚 clock keeps good time. His research on Australian languages tells him that the hands turned more slowly there. People have lived in Australia for almost 50 000 years, yet it has far fewer language families than the Americas-no more than 24, and arguably just one, called proto-Australian.

Evans suspects that something pushed the clock鈥檚 hands faster in the Americas, explaining its 140-plus language stocks without having to assume that the first Americans arrived during the Ice Age. Nichols, however, points out that Australia is far smaller than the Americas, much less varied geographically, and has been linguistically isolated for most of its history, making it a linguistic anomaly. 鈥淚 could believe that the clock ticks at a different rate for Australia,鈥� says Nichols, 鈥渂ut if so, it鈥檚 Australia that鈥檚 odd-man-out.鈥�

Several studies of human genes appear to support Nichols鈥檚 timeline for the Americas. Douglas Wallace, one of the pioneers of human genetics, used mitochondrial DNA to estimate that humans first crossed from Siberia into America between 20 000 and 40 000 years ago. More recently, Antonio Torroni, a colleague of Wallace at Emory University in Atlanta, narrowed that range to between 22 000 and 29 000 years ago. Similar dates emerge from recent work on DNA sequences from the Y-chromosome, passed on from father to son.

Dating the first words

In an even more audacious move, Nichols has run her linguistic clock backwards to find out when human language was born. The Palaeolithic Explosion of socially shared symbolism-painting, sculpture and personal adornment-clearly implies language, at least 50 000 years ago. But Nichols found that you can only account for the diversity of language stocks by pushing the dawn of language back to 132 000 years ago-soon after our ancestors acquired the brain power and the mouth and throat anatomy to make speech possible.

Even at that time depth, she finds that it is necessary to assume that different languages must have sprung up at about the same time in 10 or more human groups spread out across the East African birthplace of humanity. She points out that modern humans evolved within a relatively large area. Much later on, hunter-gatherers living in ecologically similar areas of a comparable size spoke many tongues representing many different language families. 鈥淥ur species was never so small that only one language would be spoken,鈥� she says. While Nichols would be the first to say that her estimate of 132 000 years is by no means certain, she is convinced that language predates the Palaeolithic symbolic explosion by 50 000 years or more.

To say that Nichols is bold to use linguistic tools to attempt to recreate events 100 000 years ago is an enormous understatement. No one doubts that she is a trailblazer. She is the first linguist to use languages to describe such ancient events. And like most pioneers, she has many critics. Lyle Campbell, a highly respected linguist at the University of Canterbury in New Zealand is not convinced by Nichols鈥檚 analysis of grammatical markers. He finds fault with her definition of stocks and choice of sample languages, the geographic regions she defines, and the stability of some of the features she uses. 鈥淲hile in other areas I think she is one of the smartest and most independent and astute of living linguists, in this area I think she is very, very, very wrong,鈥� he says.

Evans is somewhat less critical, though he too doubts that the building blocks Nichols has identified are solid enough to support her conclusions. 鈥淭here is something out there that requires explanation,鈥� he says. 鈥淚 just don鈥檛 accept the explanation that she is giving. It鈥檚 a very interesting area, but it鈥檚 going too far too fast. The interpretations have got ahead of the data.鈥�

Others, however, praise Nichols鈥檚 approach. 鈥淢ost scholars are hopeful that she鈥檚 right, that the envelope can be pushed back thousands of years by using the methodology she鈥檚 developed,鈥� says Moore. 鈥淭he problem is to have her work corroborated by other scientists and by data from other parts of the world.鈥�

Unfortunately, corroboration may have to wait, since Nichols remains the only linguist doing this kind of research. 鈥淭his is a territory that other respectable, competent linguists haven鈥檛 dared to explore,鈥� says Victor Golla, a linguist who works with Native American languages. 鈥淪o she鈥檚 out on a limb, exploring territory that not very many are willing to follow her into.鈥� Andrew Merriwether, a molecular anthropologist at the University of Michigan, also cheers Nichols on. 鈥淪ure, she鈥檚 out on a limb,鈥� he says, 鈥渂ut that鈥檚 what innovation means. I think she鈥檚 the most innovative person in linguistics today.鈥�