Early Childhood Hub

Children’s pragmatic development

Written By: Elspeth Wilson
12 min read
Elspeth Wilson, University of Cambridge, UK

How do children learn to communicate well? Are they born little communicators or is their communicative development gradual? Children’s pragmatic skills include a wide range of abilities, from how to initiate a conversation and take turns, to understanding a metaphor or ironic comment. Research on child language acquisition shows that these are skills that start to develop early, in the first year of life, and are actually key for language development in general – therefore, children can and should be given a rich language environment in which to use and develop these skills further. But research also reveals that some of these skills tend to develop later and many are being honed throughout childhood – and so children do not always understand everything that adults intend, and not just because they don’t know the meaning of specific words.

This article gives a brief overview of children’s pragmatic development, and focuses on just a few key skills as an example: indirect speech, metaphor and irony.

What are ‘pragmatic’ skills?

Broadly speaking, our pragmatic skills are what we use to understand and communicate meaning in context. Imagine that someone says, ‘I had a big lunch’. Our semantic knowledge gives us the meaning of ‘big’, ‘lunch’ and so on, and our syntactic knowledge helps us to understand how they fit together in a sentence, but it is our pragmatics that enable us to understand what the speaker actually means. If you had asked the question ‘Why aren’t you running?’, then you could fairly easily have understood that it was because they had a big lunch, and infer that they therefore had a full stomach and it was uncomfortable to run, and so on. But if you’d asked instead ‘Do you want a snack?’, you might well have inferred that the speaker meant ‘no, I don’t’ (because they were not hungry after a big lunch): the speaker had not directly answered the question but you had drawn on your world knowledge to link what they said to your question in a relevant way. This shows that the very same words can mean different things when uttered in different contexts. We use our semantic knowledge (vocabulary), world knowledge, what we know about the person with whom we’re talking and what we have already talked about or shared with that person – and we use these sources of information to make an inference about what someone means.

A wide range of skills are classed as ‘pragmatic’; there is no clearly defined set but rather a family of skills, including:

  • understanding and producing indirect speech, metaphor and irony
  • referring to objects and people informatively
  • telling narratives coherently
  • initiating a conversation
  • taking turns in a conversation
  • responding appropriately and relevantly
  • using the appropriate register and politeness.


This article focuses on understanding indirect speech, metaphor and irony, which requires making inferences to ‘read between the lines’ to get at what someone means, although the other skills are no less important. It also takes a look at producing appropriate referring expressions. While in general in language acquisition, comprehension precedes production, it seems that for many pragmatic inferences, children appear to produce them before they fully understand them.

Why are pragmatic skills important?

Pragmatic skills obviously help us to communicate with others effectively and efficiently, which is important for learning and collaborating in the classroom. They also make other important contributions to life experience and wellbeing. The better someone’s pragmatic proficiency, the more likely that they are to be popular with peers and to be able to take part in collaborative learning activities (e.g. Kemple et al., 1992). In contrast, the lower someone’s pragmatic proficiency, the more likely that they are to have socio-emotional and behavioural difficulties (e.g. Helland et al., 2014) – although note that these are correlational and not causal relationships.

When do children develop pragmatic skills?

From birth! Babies’ social cognition, which underpins pragmatics, is nurtured right from the outset, as they take part in dyadic interactions with their caregivers: they respond to each other (dyad means ‘two’) but don’t yet interact about some third thing (like older babies and adults do when they point to or talk about an object, for example). A baby might make a noise, burp or wave their arms, and the adult will respond, often as if it’s a conversation. Really early on, babies also start to show preferences for behaviours that are responsive to them, rather than random, and to show preferences for eye contact (Farroni et al., 2002; for a readable review, see Stephens and Matthews, 2014).

At around nine months, some important developments take place for communication – it’s even been called the ‘nine month revolution’! Babies start to take part in joint attention, learning to interact with someone else and with another object, and both know that they are doing this – this is triadic interaction (Tomasello, 2003). To support this, they learn to track someone’s eye gaze, and to initiate joint attention by pointing to or reaching for something (Carpenter et al., 1998). From around this age, babies also start to react differently when someone cannot do something, as opposed to when they are unwilling (Behne et al., 2005).

As toddlers, at around two years, children continue developing the ability to infer other people’s intentions and to reason about shared knowledge – and then use this in communication. For example, from two years, children point to a toy that a speaker cannot see but they can when that speaker asks ‘Where’s the other toy?’ (Moll and Tomasello, 2006). Even earlier, at 18 months, toddlers try to repair a miscommunication even when they have got what they were trying to ask for (accidentally): if they ask for a ball and the listener turns around to a teddy and says, ‘You want the teddy’, while at the same time distractedly handing the ball to the toddler, the toddler still tries to correct the listener – to repair the communication. That is, they seem to be repairing the failed communication itself, rather than just responding to a failed outcome (Grosse et al., 2010).

So, over the course of the first two years of life, children seem to learn that communication is social and intentional: it’s an integral part of interacting with others. One of its functions is to convey information, and we expect other people to expect that to be its function. These early social cognitive realisations are actually foundational to language learning: children have to work out what words actually mean as they encounter new ones each day, and one way in which they seem to do this is through pragmatic inferences about the speaker’s intended meaning (Bohn and Frank, 2019). But it’s over the Early Years that some pragmatic phenomena really get going, as children develop the capacity to integrate multiple sources of information in context. The next sections take a quick look at indirect speech, metaphor, irony and referring expressions in more detail.

Understanding indirect speech

Remember the example above: ‘Do you want a snack?’ ‘I had a big lunch.’ Adults would usually infer that the speaker means that they do not (because they are still full from lunch) apparently effortlessly – although, in fact, this requires integrating different sources of information, like our world knowledge about the effects of big meals and our knowledge of the speaker (e.g. that they don’t tend to eat too much), with the semantic meaning of the sentence, in order to infer what the speaker intends to communicate. This kind of inference has been called a relevance implicature, because it results from the way in which speakers and listeners assume that they will make relevant contributions to a conversation. Children can make these kinds of inferences aged three, when the inferences are simple enough in terms of things like the background knowledge required (e.g. Schulze, Grassmann and Tomasello, 2013; Wilson and Katsos, 2021).

There are other types of implicature too. Quantity implicatures arise because we assume that we will contribute enough quantity of information to a conversation that we will be informative. For instance, imagine the following exchange: ‘What did you put in your bag?’ ‘I packed a hat.’ Again, from three to four years, children, like adults, will understand that the speaker means that they packed only a hat. In a linguistic experiment, they will choose a picture with only a hat in a bag, and not one with a hat and a book, as the one that matches the utterance – even though both pictures match the utterance in terms of the literal or semantic meaning (Wilson and Katsos, 2021).

A much-researched type of quantity implicature, the scalar implicature, seems to be harder. Consider: ‘I broke some of the plates.’ In many contexts, adults usually infer that the speaker means that they broke some but not all of the plates – even though it would still be true if the speaker broke all of the plates (the literal meaning of ‘some’ is ‘some and possibly all’). Children, however, seem to struggle to make this inference reliably until five years old. One probable reason for this is not yet fully knowing the meaning of quantifier words like ‘some’ and ‘all’; another is not yet knowing how they relate to one another on a scale (some–many–most–all), and then not being able to activate the relevant alternative on that scale (‘all’, in this case). In other words, making implicatures like this can depend on some specific linguistic knowledge, as well as background or world knowledge.

Understanding metaphor

It’s a similar story when it comes to metaphor. Recent research that uses words and concepts with which young children are familiar finds that children as young as three years old can understand simple metaphors. For instance, they can infer that ‘the tower with the hat’ is a way of referring to a tower with a pointy roof rather than one with a balcony (Pouscoulous and Tomasello, 2020). This has sometimes been called a perceptual metaphor, because the relationship between the object actually described (roof) and that denoted by the word used (hat) is to do with perceptual or physical similarities.

However, many metaphors require abstract conceptual knowledge as well as linguistic experience, and children do not develop understanding of this until later. Think of the metaphor ‘my sister is a rock’. For adults, this is probably a conventional metaphor – a well-used one for which they know the meaning, in the same way that they know the literal meaning of ‘rock’. It also requires mapping physical concepts (hard, long-lasting) to abstract concepts of a person’s character: it’s not that she’s hard and grey, but rather that she’s steadfast. This is something that is tricky for young children. In addition, metalinguistic skills – the ability to reflect on language and talk about talking – develop relatively late, at around five or six years old, and so if you ask children to explain how they understand a metaphor, they are likely to explain it in literal terms, even if in some way they had inferred a metaphorical meaning.

Understanding irony

‘Just great!’ we say when it starts raining and we’re caught short without an umbrella. Adults frequently use irony, saying one thing but meaning something else (in this case, just the opposite), and so also conveying an attitude about what was said  – perhaps annoyance, mockery, being funny or praising. Research suggests that children learn to understand irony relatively late: at around six years old they begin to recognise the communicative function of being ironic. Why is the speaker being ironic? Are they being funny or mean? Before that, they are likely to think that an ironic statement is an error or a lie (which, incidentally, children only learn to understand aged four, although they can produce lies sooner!).

But understanding all aspects of irony seems to take a lot longer. There are various reasons why this might be the case: understanding irony is likely to involve more complex reasoning about others’ mental states (higher-order ‘theory of mind’) and recognising that the speaker is being honest but dissociating herself from what she has said. In other words, inferring the speaker’s intended meaning is more complex in the case of irony (Filippova, 2014; Mazzarella and Pouscoulous, 2021).

Referring helpfully

Turning to children’s developing language production, one important skill is learning how to refer to something helpfully in context. Imagine a child wanting a specific cup from a high cupboard: she might point and say ‘that one’ but, given the distance, this is not very informative. Or she might say ‘the green one’, which narrows it down between two green cups. Or she might say ‘the green one on the top shelf’ to be able to uniquely refer to what she wants. Learning to refer helpfully involves several skills, including the requisite language (for example, adjectives or relative clauses), an awareness of what the listener knows (whether he can see what the speaker is talking about, for instance) and sensitivity to what else is in the context (how many green cups). Children begin to develop these skills from the second year of life: even 12-month-olds can point appropriately depending on whether their addressee is knowledgeable or ignorant about the relevant location of something (Liszkowski et al., 2008). Three-year-olds can choose to refer to something lexically (e.g. using the object’s label), rather than just with a pronoun or no word at all, if the listener cannot see it (Matthews et al., 2006), and five-year-olds are more likely to use an adjective-noun phrase, such as ‘the green cup’, to refer to an object where there is another similar object nearby, compared to when there is not (e.g. Davies and Katsos, 2010). However, in the Early Years, these are developing skills, and far from adult-like competence.

A couple of interesting studies have investigated how to help children develop the ability to refer to something in an informative way (Matthews et al., 2007; Matthews et al., 2012). They found that for two- to five-year-olds, the experience of miscommunication – or communication breakdown – and repair was crucial. Children learn from attempting to refer to something and, if it is ambiguous, having the adult listener ask them which object they want. Most effective is doing this in a specific way – for example, ‘Do you want the green cup on the bottom shelf, or the green cup on the top shelf?’ or ‘Do you want the stripy cup or the spotty cup?’. In these studies, this kind of training was even helpful in a different kind of context for older children, indicating that it is transferable.

Pragmatic skills in the classroom

There has been little research giving robust evidence about ways in which to effectively encourage and teach oral pragmatic skills in the classroom, in contrast to the research that has been carried out on improving inferences in reading comprehension, or pragmatic language skills generally in clinical settings. So only some general observations can be made based on all that is known so far about children’s pragmatic development:

  1. Pragmatic skills begin to develop early and are instrumental in language development itself, so we can and should use rich and extensive language in talking with children, in order to give them the opportunity to practise and develop these skills
  2. Modelling, scaffolding and having conversations with children also enable them to practise turn-taking, politeness, contingent talk and narrative
  3. Conversations with adults also support skills like referring: children realise that they have been misunderstood or not provided enough information, and can practise responding more informatively; it helps when adults ask specific questions modelling informative referring expressions
  4. Talking about certain inferences and how we arrive at them may be helpful, particularly for older children – for instance, by using thinking maps to identify the characteristics of two words that are similar and contribute to a metaphor’s meaning (see Kalandadze et al., 2021)
  5. Pragmatic skills are developing in the Early Years, and so children may not understand some figurative language, like metaphor or irony, especially where lots of different sources of information have to be integrated, so providing a paraphrase is useful where clear comprehension is required
  6. Pragmatic skills depend on world knowledge and linguistic knowledge, like vocabulary, and so adopting strategies to build these could contribute to oral pragmatic understanding, just as it does with inferencing in reading comprehension.


Where next? Easy-to-read and accessible articles

Frontiers for Young Minds provides high-quality, plain-language articles about science, written by researchers and reviewed by young people. It is also great for adult readers who want an easy introduction to a topic.

Kalandadze T, Tonini E and Bambini V (2021) When dancers are butterflies: How the brain understands metaphors. Frontiers for Young Minds 9: 606160. DOI: 10.3389/frym.2021.606160.

Lew-Williams C and Weisleder A (2017) How do little kids learn language? Frontiers for Young Minds 5: 45. DOI: 10.3389/frym.2017.00045.

Pexman P (2018) How do we understand sarcasm? Frontiers for Young Minds 6: 56. DOI: 10.3389/frym.2018.00056.

  • Behne T, Carpenter M, Call J et al. (2005) Unwilling versus unable: Infants’ understanding of intentional action. Developmental Psychology 41(2): 328.
  • Bohn M and Frank MC (2019) The pervasive role of pragmatics in early language. Annual Review of Developmental Psychology 1(1): 223–249.
  • Carpenter M, Nagell K and Tomasello M (1998) Social cognition, joint attention, and communicative competence from 9 to 15 months of age. Monographs of the Society for Research in Child Development 63(4): i–174.
  • Davies C and Katsos N (2010) Over-informative children: Production/comprehension asymmetry or tolerance to pragmatic violations? Lingua 120(8): 1956–1972.
  • Farroni T, Csibra G, Simion F et al. (2002) Eye contact detection in humans from birth. Proceedings of the National Academy of Sciences 99(14): 9602–9605.
  • Filippova E (2014) Irony production and understanding. In: Matthews D (ed) Pragmatic Development in First Language Acquisition. Amsterdam: John Benjamins, pp. 261–278.
  • Grosse G, Behne T, Carpenter M et al. (2010) Infants communicate in order to be understood. Developmental Psychology 46(6): 1710.
  • Helland WA, Lundervold AJ, Heimann M et al. (2014) Stable associations between behavioral problems and language impairments across childhood – the importance of pragmatic language problems. Research in Developmental Disabilities 35(5): 943–951.
  • Kemple K, Speranza H and Hazen N (1992) Cohesive discourse and peer acceptance: Longitudinal relations in the preschool years. Merrill-Palmer Quarterly 38(3): 364–381.
  • Liszkowski U, Carpenter M and Tomasello M (2008) Twelve-month-olds communicate helpfully and appropriately for knowledgeable and ignorant partners. Cognition 108(3): 732–739.
  • Mazzarella D and Pouscoulous N (2021) Pragmatics and epistemic vigilance: A developmental perspective. Mind & Language 36: 355–376.
  • Matthews D, Butcher J, Lieven E et al. (2012) Two‐ and four‐year‐olds learn to adapt referring expressions to context: Effects of distracters and feedback on referential communication. Topics in Cognitive Science 4(2): 184–210.
  • Matthews D, Lieven E, Theakston A et al. (2006) The effect of perceptual availability and prior discourse on young children’s use of referring expressions. Applied Psycholinguistics 27(3): 403–422.
  • Matthews D, Lieven E and Tomasello M (2007) How toddlers and preschoolers learn to uniquely identify referents for others: A training study. Child Development 78(6): 1744–1759.
  • Moll H, Koring C, Carpenter M et al. (2006) Infants determine others’ focus of attention by pragmatics and exclusion. Journal of Cognition and Development 7(3): 411–430.
  • Pouscoulous N and Tomasello M (2020) Early birds: Metaphor understanding in 3-year-olds. Journal of Pragmatics 156: 160–167.
  • Schulze C, Grassmann S and Tomasello M (2013) 3-Year-Old Children Make Relevance Inferences in Indirect Verbal Communication. Child Development 84(6): 2079–2093.
  • Stephens G and Matthews D (2014) The communicative infant from 0–18 months: The social-cogntive foundations of pragmatic development. In: Matthews D (ed) Pragmatic Development in First Language Acquisition. Amsterdam: John Benjamins, pp. 13–36.
  • Tomasello M (2003) Constructing a Language: A Usage-Based Theory of Language Acquisition. Cambridge, MA: Harvard University Press.
  • Wilson E and Katsos N (2021) Pragmatic, linguistic and cognitive factors in young children’s development of quantity, relevance and word learning inferences. Journal of Child Language 1–28. Epub ahead of print 28 September 2021. DOI: 10.1017/S0305000921000453.
0 0 votes
Please Rate this content
Notify of
Inline Feedbacks
View all comments

Other content you may be interested in