All children acquire language in the same way, regardless of what language they use or the number of languages they use. Acquiring a language is like learning to play a game. Children must learn the rules of the language game, for example how to articulate words and how to put them together in ways that are acceptable to the people around them. In order to understand child language acquisition, we need to keep two very important things in mind:

First, children do not use language like adults, because children are not adults. Acquiring language is a gradual, lengthy process, and one that involves a lot of apparent 'errors'. We will see below that these 'errors' are in fact not errors at all, but a necessary part of the process of language acquisition. That is, they shouldn't be corrected, because they will disappear in time.

Second, children will learn to speak the dialect(s) and language(s) that are used around them. Children usually begin by speaking like their parents or caregivers, but once they start to mix with other children (especially from the age of about 3 years) they start to speak like friends their own age. You cannot control the way your children speak: they will develop their own accents and they will learn the languages they think they need. If you don't like the local accent, you'll either have to put up with it or move to somewhere with an accent you like! On the other hand, if you don't like your own accent, and prefer the local one, you will be happy. A child will also learn the local grammar: 'He done it'; 'She never go there'; 'My brother happy' and so on are all examples of non-standard grammar found in some places where English is spoken. These might be judged wrong in school contexts (and all children will have to learn the standard version in school) but if adults in the child's community use them, they are not 'wrong' in child language.

These examples show that different dialects of English have their own rules. The same is of course true of other languages and their own dialects. In what follows, examples are in English, because that is the language in which this article is written, although the child strategies illustrated in the examples apply to any language and to any combination of languages that your child may be learning.

We start with a number of observations about child learning in general, about speech and language, and about how children themselves show us how they learn, before turning to children's acquisitional strategies. These also teach us that children follow their own rules, and that they need plenty of time to sort these rules out.

Like the rest of us, children are individuals. What makes them different from adults, as a whole, is that children are reared in adult worlds according to adult expectations. Children learn to model their behaviour on what goes on around them, be it dress codes, body language, table manners or language uses, usually first through their caregivers and later through peers in their family, neighbourhood or school. That is, children are learning how to function adequately in their environment, and much of this learning takes place through language itself. We talk to children to tell them about our adult world and they learn about the world from what we tell them. But they also learn about our language, from how we use it to tell them about other things. This means that language learning is going on whenever language is used around children.

Children do not simply reproduce as-is whatever they are exposed to, for two reasons: First, they are developing physically. Just as it may take years to be able to develop the fine motor skills needed for sewing on a button, it will take years to be able to use speech organs in equally precise ways. Second, children are developing cognitively. They need to find ways to make sense of their environment, so that they can engage comfortably with it. They do this by progressively adapting the input they receive to their own emerging cognitive and linguistic abilities, and by screening out, as it were, what is as yet too complex for them to understand. Let's see why this difference between physical and cognitive ability matters for child language acquisition.

Speech and language are two quite different things. Speech is a physical ability, whereas language is an intellectual one. The difference between children's language abilities and speech abilities becomes clear from a classic illustration, reported by researchers Jean Berko-Gleason and Roger Brown in 1960. One parent imitates the child's developing pronunciation of the word fish as 'fis' and asks the child: "Is this your 'fis'?" To which the child responds: "No! It's my 'fis'!"

The child recognizes that the pronunciation 'fis' is not up to par, but cannot reproduce the adult target 'fish'. That is, the language item 'fish', complete with target pronunciation, is clear to the child, but speech production doesn't match this awareness. Children of deaf parents give us further proof of the difference between these two abilities: if these children are exposed to a sign language early in life, they will develop that language whether they are deaf or hearing, even though they might not use it. The 'fis-phenomenon' is what explains why children can get very angry at someone who repeats their own baby productions back to them, whether in pronunciation or in grammar.

Since speech and language are independent abilities, emerging language does not reflect emerging speech in any straightforward way, or vice versa. There's nothing necessarily wrong with someone's language abilities if they stutter, lisp or slur their words together, but these features of their speech may need correcting if they impair intelligibility beyond childhood. And there's nothing necessarily wrong with someone's speech if they can't say 'She sells seashells on the seashore' by age 6, although their language ability may need checking if they don't understand what this sentence means, in any language, at the same age.

What speech and language development have in common is that they progress through stages and that their progress takes time. In speech, it is quite normal for English-speaking children, for example, to have difficulties pronouncing the sounds at the beginning of words like 'thank' and 'than' throughout their first 8 to 10 years: the precise coordination of the many different muscles involved in pronouncing any speech sound needs a lot of practice. In language, it is also normal that children have serious trouble throughout many years, for example sorting out the use of pronouns like 'I' vs. 'you' (if people say I of themselves and you to everyone else, what can these words mean??) or following complex instructions (which involve several clauses in one same sentence): children well into their early school years may not have acquired the meaning of words like 'or, before, after', or the cognitive ability to process complex sentences yet. As with the 'fis-phenomenon', in many cases these (typically temporary) child production problems are recognized as such by the child, who can simultaneously understand an adult using the correctly pronounced words in complete utterances. The child chooses to use other forms of expression, or to omit certain forms, so as to avoid using what they know will be badly produced.

Some children will take longer than others to sort out some speech or language issue, or will have difficulties in areas which other children will have a breeze sailing through -- even among siblings, including identical twins. Girls are also known to start their language acquisition process earlier than boys and to proceed at a faster rate of learning than boys. These observations teach us to respect children's learning in two complementary ways: the time it takes, and the individuality of each child's learning.

Language acquisition takes time, a long time, because all learning needs time to digest and make ours what we experience around us; it cannot be rushed. Many caregivers want to see 'results', and in double-time, according to adult expectations, and then worry that 'nothing is happening'. Give yourself some time too: children learn language through natural interaction with caring people in everyday settings. The rule here is 'the more, the better': interact with your children in as many different, engaging, fun situations as possible, so that they realize that language permeates all that we do in all sorts of different ways. Give your children also plenty of time to 'do nothing': let them play around doing nothing in particular, wonder on their own about intriguing things like what the moon eats or why dogs don't wear clothes, or throw serious tantrums to learn how to cope with other people and themselves, and do all this by means of language. It is important to keep in mind that, as the child learns and acquires more and more language, each step in the process becomes less and less visible, to adults as well as to the child. The same is true of adult language learning: it may sometimes seem, to adult learners, that they are progressing very slowly or not at all; at some point they may suddenly realize that they have indeed made progress.

Language acquisition is not a competitive sport either. Children are not aiming to reach or surpass some level of language or some time frame that someone else set for them. They are competing only with themselves, on their own terms. The child whose speech is most advanced at the age of 2 is not necessarily going to be a higher achiever at age 20 than the child who was slower to learn language. Language is only part of what children have to learn and a child who seems slower might be learning in a different way, or concentrating on other things.

Children won't learn anything which they are not ready for -- they may parrot things that you or someone else tries to 'teach' them, but a parrot only learns to parrot. What your child is ready for is not found in books or in someone else's children. It's found in your child, and to learn about your child you must also give yourself -- and your child -- time. Your children are as new to you as everything they are learning about is to them.

Respecting children means learning to understand them. Your child is not you. Children will develop their own strategies for learning whatever they find relevant to learn around them, including language. Children are much more resourceful, resilient and creative than we are often prepared to give them credit for. Besides, and probably most importantly, your worries will reflect on your child. Children are very good at picking up distress signals from adults, and if they learn to associate your worry with their speech, then you may start having a real problem on your hands.

Children have no idea that 'language' is something that adults worry about for its own sake. Language is just a tool that gets things done for them: it's much more effective for a child to ask daddy for a toy that is out of reach than to simply shout in anger because they can't grab it. So let your children experiment with their language(s), their way. They will find the right ways to make language work for them, just as you yourself did when you were growing up. There is nothing to worry about if your child doesn't sound like an adult (which children don't anyway) or like your friend's child or like the 'prodigy' children you may hear about through the media. There may be reason to worry only if your children don't sound like themselves. No one knows this better than you, because no one knows a child better than a caregiver.

Your children have no idea what is 'expected' of them either. Namely, that you may be looking for things that are there, or not, in their language. The truth is that many of us caregivers forget to look for what is there, in our children's language(s), and tend to focus on what we think is missing instead. A lot of people believe that only 'grammatical' language is language, with lots of words and lots of syntactic sophistication. Language is much more than this: your child may prefer to be expressive through intonation, for example, the melody of speech without which no language makes sense. Or may rely on invented words, complemented by expressive body language. Children know that there is a model around them that they must learn to follow. But they don't know what the model looks like, so they approach it by trial and error. Let's see how they do this, with a few examples.

All human beings, young and old, follow two kinds of learning strategy. One, drawing on physical ability, is that we learn in stages (we make sure we can walk before we run). The other, drawing on intellectual ability, is that we generalize from past experience (if you see an insect that you never saw before and that looks like a cockroach you're likely to think it may be a cockroach). These strategies help us explain child productions in the whole of language, from pronunciation through vocabulary and grammar to skills like how to hold a conversation. The following examples deal with pronunciation and word learning because they concern the most common questions received at Ask-a-Linguist. In what follows, we give only rough guidelines for the ages associated with particular stages of development. As said above, a child's physical and cognitive progress is best assessed against the child itself, so that specific age ranges matter less than the child's progress from one stage to the next. Resources detailing milestones in overall child development are included at the end of this article.

The basic insight that we gain from children's developing pronunciation is that there are difficult sounds and easy sounds, and difficult and easy distinctions between sounds. We can tell which are which by looking at what children do, because children cannot articulate what their vocal tracts are not developed enough to tackle yet. We can for example safely conclude that, for the 'fis-phenomenon' child above, the sound at the end of the word 'fish' is more difficult than the sound at the end of the word fis.

Children start using speech sounds when they start babbling. The sounds that they use in babbling are easy sounds and these will be the sounds children will use in their first utterances too. Children usually replace difficult sounds with sounds that are easier for them to articulate, or they may drop difficult sounds altogether. They may call Sam "Tam", for example, and they may want to "pee" potatoes with a potato-"peewah", or ask you why strawberries are "wed" and not "boo". Although sounds tend to be acquired in the same order across languages, we should keep in mind that different children may find different sounds easier or more difficult: each child will have their own individual learning strategies. The important thing is that there is progress in their development. Children's spontaneous play also shows a progression from gross to sophisticated control over their body: they usually start by hitting toys, and hitting things with toys, because it's easier to do this whilst fine motor skills have yet to be acquired. This is also why in virtually all languages the baby-words for 'mummy' and 'daddy' sound very similar. It's not that the children 'know' the words for mum and dad, it's simply that these are the kinds of words that children can say (they say them to us, to the cat, to their toys, to themselves), but parents decided to believe that the children are calling them 'by name', and so reinforced the children's use of these words to them from time immemorial!

Vowels (the sounds usually spelt a, e, i, o, u in English) are easier than consonants and are generally learned first. This is because vowels are the sounds that carry, and that we therefore perceive most clearly. If you want to shout for someone named Eve or Archibald you prolong the vowels in their names, not the consonants. So children are likely to go through some stage where all or most vowels are target-like in their speech, but all or most consonants may still be funny. Since consonants are no piece of cake for developing mouths, it becomes clear that words containing several consonants in a row are young children's worst nightmare. English is particularly child-unfriendly, in that it has words like 'splash', with three consonants at the beginning, or like 'texts', with four at the end (the letter x represents two sounds, 'k' and 's'). If your child is bilingual in a tricky language like English and a straightforward one like Hawai'ian, where only single consonants are allowed before vowels, you shouldn't be surprised if she sounds right in Hawai'ian much earlier than in English. Or if a proud Hawai'ian parent tells you that his monolingual children started 'speaking much earlier' than all the English monolingual children he knows. It's the languages' fault, not the children's. The insights that we gain from cross-linguistic observations like these, by the way, especially among multilingual children, teach us that using what children do in one single language as the benchmark for typical language development across the board is very short-sighted indeed.

This same strategy also accounts for why children leave out certain words and not others in their utterances. They may say things like "Mummy big glass table" but not "My on if the". These are two quite different types of words, the former being more salient to children because they carry stress in connected speech, and therefore much easier to perceive and produce.

Suppose you show a banana to a group of children who are at the one-word stage, when all their utterances contain single words only, and suppose you ask them "What's this?" Some children will say "nana", others will say "mama", others still may say "bana". Child words like these exemplify children's use of generalization: children modify words, replace, add and remove word bits to make them conform to a general pattern that they find easier to tackle. The two-syllable structure of these words and others like them, with straightforward consonant-vowel syllables and a sample of preferred consonants, is typical of children's first words all over the world.

But suppose now one child in the group replies 'moo' to your question. Before you start worrying about this child's linguistic (or cognitive) abilities, try to think about your question and the child's answer on the child's own terms, not yours. You are expecting a word that sounds like 'banana', but how does the child know that? And how do you know what prompted the child to give you this answer? In particular, why should the sound of the word be more relevant to the child than, say, the shape of the object you're holding? It may well be that this child has recently been fascinated by the night sky, and all shiny things in it whose names he's just learned. And a banana does look like a waning or waxing 'moon'. This child is also generalising, though in a different way from his friends. He is besides showing you that he knows how to relate what he learned before to whatever activity is required of him now, which is a very good thing to have mastered indeed. (On a side note, it is this kind of generalization that makes young children, sometimes very embarrassingly, call all adult males 'daddy'.)

Once the first words are in place, children are quick to realize that saying several words together in one same utterance is the next step. So, just as they will attempt to run as soon as they are able to stand up unaided, and will then stumble and fall because of lack of practice walking, they will attempt to say too many words in one go, and will end up jumbling them all together. Many children start stuttering or stammering at this multi-word stage of their development precisely for this reason: lack of practice. Other children may even fall silent altogether for a while, until they've worked out the very difficult skill of coordinating breathing with speaking in long utterances. Professional speakers need practice in this skill too, so that speaking for long periods of time does not wear them out completely, or impair their delivery. Yet other children won't bother at all about the way they sound and will just go on producing unintelligible speech until things fall naturally in place for them, even those children who may have had perfect single-word articulations before.

Other examples of child acquisitional strategies surface in ways that would also appear to give reason for concern, if we didn't know better. Say your child uses so-called irregular past tenses like 'came', drove', or 'slept' with no problem, as well as regular ones like 'baked' or 'cried'. Then one day he starts saying things like "Mummy drived me to school today", or "I sleeped so well". What is happening here is that your child has realized that there is a pattern in some part of the language: some words (linguists call them 'verbs') can have extra sounds at the end to indicate events that happened before the time we are talking about them. Most verbs are regular in this way, so productions like 'catched' or 'swimmed' show that your child has actually learned a general rule and immediately started applying it to any verb -- just as you once learned that cockroaches 'pattern' in a certain way, and so this funny new insect before you must be a cockroach too. The same happens with noun plurals, and your child may start talking about 'foots' or even 'feets' whereas he talked about 'feet' before. Child language researcher Jean Berko-Gleason used an ingenious experiment to show that children are in fact learning rules of language. For example, she showed children a picture of one imaginary, cuddly animal and told the children that the animal was called a 'wug'. Then she showed a picture with two of these cuddly beings and asked the children: "Now there are two of them. There are two ___". The children had to complete her sentence, and they used the correct plural form "wugs", showing that they could apply the plural rule to words that they had never heard before. Apparent 'errors' like 'foots' (or 'catched') thus mean that learning is progressing as it should: the previous, 'correct' production of irregular and regular forms was simply due to imitation. The generalized forms will disappear once your child is ready to learn the next rule, which is that some words follow the general rule and others don't.

All of the child productions described so far are natural, part and parcel of typical language acquisition. But sometimes there can be problems in the process of language acquisition. If a child has not said a recognizable word with a clear meaning (such as 'mama' for mother) by the age of 18 months, it is advisable to make sure that all is well. It is normal for some children not to say their first word until the age of 2 years, but sometimes there are problems that would benefit from early help, so it is worth checking things out a little sooner. If your child has not said the first sentence (such as 'Dere mama', meaning 'There is mama.') by the age of 3 years, it is also worth checking that all is well. If you feel at any time that your child's speech has become less than it was, or that learning has stopped, that is another reason for checking it out. The most common reason for late or deteriorating speech is a hearing problem.

If you suspect that something is amiss with your child's linguistic development, the first thing you need to do is to consult with your doctor to make sure your child's development is otherwise normal. The doctor will establish that your child has (or has not) normal hearing and normal cognitive and motor skills, among other things, and will guide you about required follow-up to any issues arising from non-language related problems which may be affecting language itself. A child may need hearing aids or medication for ear infections, for example. If your doctor finds that everything else except language is as it should be, the usual procedure is to refer the child to a speech-language clinician, who will be able to diagnose and remedy specific problems in language and/or speech. Children with hearing problems will also benefit from help from a speech-language clinician and/or an audiologist.

In case you do need a speech-language clinician, take some time to decide about which clinician to consult with, if you have a choice. Assessing child language in order to diagnose possible problems proceeds through comparison with developmental norms that have been established for children using particular languages and particular dialects. These norms may not match the language(s) or the dialect(s) used by your child. To give a straightforward example, Texan children (and adults) speak differently from New York children (and adults). By the same token, developmental norms that apply to English may not apply to Spanish or Mandarin, and developmental norms accounting for monolingual children will not apply to multilingual children, and vice versa. A competent speech-language therapist, whether monolingual or multilingual, will know this and will be able to address your child's specific problems accordingly.

If you want to find out more about typical language development, these three suggestions may help you.

• David Crystal's book Listen to your Child: A Parent's Guide to Children's Language (Penguin, 1989) offers a very accessible and very entertaining account of what children do with their own language learning.
• Eve Clark's book First Language Acquisition (Cambridge University Press, 2003) is a comprehensive -- and more technical -- account of child language development.
• The Learning Disabilities website lists typical milestones in cognitive, linguistic and social development.