To hear my seventeen-year-old son switching effortlessly between Catalan, Castilian, English, French, German and Russian, you wouldn't believe that when he was very small, we wondered if he would ever speak at all. He heard English from me and occasional visitors, Catalan from my husband and his family, including his gasbag of a brother, Castilian from the baby-sitters who helped me out, and from the age of three attended a pre-school where they 'did' English every afternoon. But he didn't talk. This is the point at which well-meaning monolinguals (even psychologists, it has been reported) come out with what you've been dreading, a fear you're maybe even about to buy into: "Learning two languages confuses a child".

It is believed that the child's brain cannot cope and that they will grow up semi lingual, confused or retarded. This is pure myth. Children in bi or multilingual families simply have a lot more to learn. But they can do it. After all, we are constantly being told by personal development gurus that we only use 10% of our brain's capacity. In a loving, supportive environment children learn two or more languages well. Even in a monolingual environment, if it is a stressful one, they may have language development problems.

With the one-person one-language system (OPOL), the child is exposed to both languages from birth. Gradually they build up a list of words from both languages, and begin to use sentences mixing words. It must be stressed that this is a normal stage in bilingual development. Quite soon they learn to associate each system with the appropriate person. Provided OPOL is adhered to, the amount of mixing swiftly decreases. In the third year, many children will have achieved what is called metalinguistic awareness: they are aware that there are two systems and that, for example, Dad says 'shoe' and Mum says 'zapato'. But they are not aware that they are bilingual. The acquisition of separate sets of grammatical rules takes longer and is usually reached in the fourth year.

Because of differences in the amount of input the child gets in each language, one or other may be dominant at different times. In an OPOL family, at first the language of the parent who spends the most time with the child will be dominant. The minority language is likely to be overtaken by the majority language when the child starts pre-school or school, if not before (local childminders, friends, TV and so on). But passive knowledge of the m-language remains and will be activated in the right circumstances. I'll never forget my linguist son's first sentence in English. My mother had just arrived from the UK and sat for hours chatting away. My son uttered not a word. Later, we went to run the boys' bath, and he followed. 'Grandma!' he suddenly proclaimed in perfect accents. 'This is a toilet.'