Most of the time we adults take language for granted – unless of course we have to learn a new one. Then, things change pretty quickly. We can't get the pronunciation right, and we can't hear the difference between sounds. There are too many new words, and we forget ones that we learned just the day before. We can't say what we want to say, and we can't understand anything either, because everyone speaks too fast.

Then, as if that isn't bad enough, we come across a three-year-old child and watch in envy and amazement as she talks away effortlessly in that impossible language. She can't tie a knot, jump rope, draw a decent-looking circle, or eat without making a mess. But while she was still in diapers, she figured out what several thousand words mean, how they are pronounced, and how they can be put together to make sentences. (I know that I've used “she” all the way through this paragraph, as if only girls learn language. Since English doesn't have a word that means “he or she,” I'll simply alternate between the two. I'll use “she” in this chapter, “he” in the next chapter, “she” in the third chapter, and so forth.)

Children's talent for language is strangely limited – they're good at learning language, but not so good at knowing what to say and what not to say.

Imagine for a moment what it would be like to have words, but no systematic way to combine them. You'd have to communicate in tiny one-word installments – “thirsty,” “water,” “give.” That must be roughly what it's like to be a fifteen-month-old child with something to say.

There's considerable incentive, then, for children to learn how to create sentences, and it doesn't take them long to get started, as we'll see. The development of children's sentence-building skills can be roughly divided into two phases.

The first phase, which begins around the age of eighteen months, sees the appearance of relatively simple two- and three-word patterns. These early sentences are primitive and often incomplete, but they mark the start of something big.

During the second phase, which begins around age two or so, the missing pieces are filled in and there is rapid growth in the ability to produce a wide variety of complex constructions.

Getting started

The basic recipe for sentence building is simple: Combine two words with the right fit (say, an adjective and a noun, or a noun and a verb) in the right order. Repeat the process as many times as necessary, adding one new word or combination of words each time.

Take, for example, the sentence The glass broke. You start with the words the and glass, and combine them to create the phrase the glass.

Children are born into a world full of noises and sounds of all sorts (music, car engines, slamming doors, whistling, coughing, crying, conversation, and so on). Somehow, they have to take the part that is speech, break it down into its smaller parts (words, prefixes, suffixes, etc.), determine what they mean, and figure out how to reassemble them in new ways.

Doing this involves mastering a system of sounds, words, structure and meaning whose intricacy typically defeats even the most gifted adult learners. Yet children get the job done before they learn to tie their shoes. How do they do it?

There's still no real solution to this puzzle, although bits and pieces of the answer are starting to emerge. And as this happens, it's becoming clear that certain widely held ideas about how language learning works are probably dead-ends.

The job of this chapter is to try to sort out the difference between the ideas that make sense and those that don't. We'll start by looking at the popular idea that children learn language by imitating their parents.

Why it's not imitation

Ask the average person how a child learns language and you'll probably be told “by imitating adults.” On the face of it, that makes a lot of sense. The adults in a child's life speak a particular language, and the child ends up speaking that language too.

Imitation of some sort probably is involved in certain aspects of language acquisition. Take words, for example.

A child's first birthday is cause for special celebration in most cultures. It's a sign of survival and growth. By this age, children have their first teeth, they are able to eat solid food, and they're about ready to take their first steps, if they haven't already done so.

Their minds are developing too – they are able to follow the direction of an adult's gaze, they are sensitive to gestures such as pointing, and they tend to pay attention to the same things as the adults with whom they are interacting. Not coincidentally, this is also about the time that they first venture into language.

A child's first word is one of the great milestones in his life – and in the lives of his parents. For most children this happens when they are around twelve months old, give or take a few weeks in either direction. On average, a child has ten words in his vocabulary by age fifteen months and fifty words by age eighteen or nineteen months.

And, yes, it's true that the first words learned by children the world over are usually the names for “mother” and “father.” They get a lot of help with this, though. As we'll see in chapter 6, words like mama, papa, and dada are very easy to pronounce – they consist of very simple sounds arranged into very simple syllables – and they are a natural by-product of children's spontaneous babbling.

Since the first edition was published in 1990 computers have revolutionized geophysics and all of the Earth sciences just as they have our daily lives. Huge datasets can be manipulated, complex computations performed and detailed colour images plotted. An understanding of the fine details of the workings of the many processes taking place within the Earth is becoming achievable and a synthesis of the interactions of the hitherto rather separate disciplines of geochemistry, dynamics, heat and seismology has become possible. This has been an exciting decade for advances in our comprehension of the interior working of the Earth.

I have kept the overall format and structure of the book unchanged from the first edition. Many people have contacted me and seemed to like the approach that I used. I have taken many of their detailed suggestions into consideration when preparing this new edition. Changes have not been made for the sake of using the latest publications, but where there have been significant advances. Older references and sources have been retained in order to lead students to papers of value that they would not otherwise meet, and to show how major steps in understanding were achieved. The major change from the first edition is that the deep internal structure of the Earth and mantle and core processes are now covered in a separate chapter rather than being included in the chapters on seismology and heat.

Plate tectonics

The Earth has a cool and therefore mechanically strong outermost shell called the lithosphere (Greek lithos, ‘rock’). The lithosphere is of the order of 100 km thick and comprises the crust and uppermost mantle. It is thinnest in the oceanic regions and thicker in continental regions, where its base is poorly understood. The asthenosphere (Greek asthenia, ‘weak’ or ‘sick’) is that part of the mantle immediately beneath the lithosphere. The high temperature and pressure which exist at the depth of the asthenosphere cause its viscosity to be low enough to allow viscous flow to take place on a geological timescale (millions of years, not seconds!). If the Earth is viewed in purely mechanical terms, the mechanically strong lithosphere floats on the mechanically weak asthenosphere. Alternatively, if the Earth is viewed as a heat engine, the lithosphere is an outer skin, through which heat is lost by conduction, and the asthenosphere is an interior shell through which heat is transferred by convection (Section 7.1).

The basic concept of plate tectonics is that the lithosphere is divided into a small number of nearly rigid plates (like curved caps on a sphere), which are moving over the asthenosphere. Most of the deformation which results from the motion of the plates – such as stretching, folding or shearing – takes place along the edge, or boundary, of a plate. Deformation away from the boundary is not significant.

Geophysics, the physics of the Earth, is a huge subject that includes the physics of space and the atmosphere, of the oceans and of the interior of the planet. The heart of geophysics, though, is the theory of the solid Earth. We now understand in broad terms how the Earth's surface operates, and we have some notion of the workings of the deep interior. These processes and the means by which they have been understood form the theme of this book. To the layperson, geophysics means many practical things. For Californians, it is earthquakes and volcanoes; for Texans and Albertans, it is oil exploration; for Africans, it is groundwater hydrology. The methods and practices of applied geophysics are not dealt with at length here because they are covered in many specialized textbooks. This book is about the Earth, its structure and function from surface to centre.

Our search for an understanding of the planet goes back millennia to the ancient Hebrew writer of the Book of Job and to the Egyptians, Babylonians and Chinese. The Greeks first measured the Earth, Galileo and Newton put it in its place, but the Victorians began the modern discipline of geophysics. They and their successors were concerned chiefly with understanding the structure of the Earth, and they were remarkably successful. The results are summarized in the magnificent book The Earth by Sir Harold Jeffreys, which was first published in 1924.