Different functions of children’s language

Each of the child's four utterances serves a distinct one of Halliday's functions, and the context makes each purpose clear.

'want biccy' serves the INSTRUMENTAL function: the child uses language to satisfy a need and obtain an object (a biscuit). Instrumental language is the 'I want' function, used to get something for the self, and the child's stated desire here is a textbook example.

'go away!' (addressed to the dog) serves the REGULATORY function: the child uses language to CONTROL another's behaviour, here directing the dog to leave. This differs from the instrumental function in that the purpose is not to obtain an object but to influence what someone (or something) else does.

'wassat?' ('what's that?', pointing at a clock) serves the HEURISTIC function: the child uses language to EXPLORE and learn about the environment, seeking information about an unfamiliar object. The heuristic function is the child's 'tell me why/what' language, typically realised as questions.

'love you mummy' serves the INTERACTIONAL function: the child uses language to build and maintain a RELATIONSHIP, expressing affection and sustaining the social bond with the parent. Interactional language is the 'me and you' function of relationship-building.

Together the four utterances show the child already using language for a range of purposes — to obtain, to control, to find out and to bond.

This single utterance is MULTIFUNCTIONAL — it performs several of Halliday's functions at once, and the context (the child holding up a drawing to a parent) is essential to reading them.

The most obvious function is PERSONAL: in 'I done a doggy!' the child expresses identity, pride and feeling — announcing an achievement that is the child's own. The personal function is the language of the self, used to express opinions, feelings and individuality, and the proud announcement of 'I done a doggy' is exactly that.

The utterance is also INTERACTIONAL. By saying 'look!' and showing the drawing to the parent, the child uses language to engage the listener and maintain the social bond — drawing the parent into shared attention. The attention-getting 'look!' is directed AT the relationship, sustaining contact between child and adult.

There is, in addition, a REPRESENTATIONAL / INFORMATIVE element: 'I done a doggy' conveys information — telling the parent WHAT the drawing depicts. The representational function is language used to give facts or information about the world, and the child here informs the parent of the drawing's content.

The context decides how these functions are weighted. Because the child is proudly displaying the drawing to a parent, the PERSONAL (pride in the self) and INTERACTIONAL (sharing it with the parent) functions dominate, with the representational/informative function playing a supporting role. The key analytical point is that the utterance is not reducible to a single function: one short sentence does several communicative jobs at once.

This transcript records a child of two years five months (2;5) and, analysed through Halliday's framework of the functions of children's language (1975), the data shows the child already using language across a broad RANGE of communicative purposes — instrumental, regulatory, interactional, personal, heuristic and representational. Organising the analysis by FUNCTION (rather than line by line) reveals not just what the child says but what the child uses language to DO.

The INSTRUMENTAL and REGULATORY functions — Halliday's earliest 'pragmatic' functions — are both clearly present and worth distinguishing. 'I want more juice' is INSTRUMENTAL: the child uses language to satisfy a need and obtain an object, the classic 'I want' purpose. 'No, doggy! get down!' is REGULATORY: the child uses language to CONTROL another's behaviour, directing the dog to act. The contrast is instructive — the instrumental utterance is about obtaining a THING for the self, while the regulatory utterance is about influencing what someone (or something) else DOES. Keeping these two apart, rather than collapsing both into 'getting what it wants', is central to an accurate functional analysis.

The INTERACTIONAL function — language to build and maintain relationships — runs through the exchange. 'Mummy, look!' uses an address term and an attention-getter to engage the parent and sustain shared attention, and 'bye-bye daddy' is a phatic, relationship-maintaining farewell. Both are directed at the social bond rather than at obtaining or controlling anything, and they show the child managing relationships through language.

The PERSONAL function — language to express identity, feelings and opinions — appears in 'me a big girl now', where the child asserts a sense of self and a developing identity. This is the child using language to express who she IS, not to get or control or inform; the personal function is the language of the self, and the proud self-categorisation here exemplifies it.

The HEURISTIC function — language to explore and learn about the world — is realised in 'wassat noise?' ('what's that noise?'). The child uses a question to seek information about an unfamiliar phenomenon in the environment, the heuristic 'tell me about the world' purpose. The fact that the child frames exploration as a question shows the typical realisation of this function.

It is also worth noting MULTIFUNCTIONALITY: 'mummy, look!' is primarily INTERACTIONAL (engaging the parent), but it also carries a REGULATORY edge (directing the parent to look) and, by drawing attention to something, sets up an INFORMATIVE/REPRESENTATIONAL exchange. One short utterance can serve several functions at once, and the context — a child seeking to share attention with a parent — decides their weighting. Dore's finer-grained taxonomy could refine this further: 'mummy, look!' combines CALLING (getting attention) with REQUESTING ACTION, distinctions Halliday's broad interactional/regulatory categories blend together.

Synthesising the functional profile: across just six utterances the child uses language to obtain (instrumental), to control (regulatory), to bond (interactional), to express the self (personal) and to find out (heuristic) — a strikingly broad repertoire of communicative purposes for a two-year-old. This range shows that, well before the child's grammar is complete, language is already a flexible tool serving the full spread of the child's social and cognitive needs. The data is best read not as a list of utterances but as evidence of a child who has learned, in Halliday's phrase, how to MEAN — using language to do many different things in the world.

Read together, these two transcripts show a striking contrast in the RANGE of functions the two children use language for. Child A (1;7) uses language predominantly for Halliday's EARLY 'pragmatic' functions — instrumental, regulatory, interactional and personal — while Child B (3;0) additionally commands the LATER functions — heuristic, imaginative and representational. Comparing them by function reveals how the child's functional repertoire broadens with development, exactly as Halliday's account predicts.

Child A's utterances cluster in the early functions. 'milk!' (reaching) is INSTRUMENTAL — language to obtain an object. 'up!' (arms raised) is also instrumental in purpose (to be picked up) but carries a clear REGULATORY edge, directing the parent to act — a neat example of MULTIFUNCTIONALITY in a single holophrase. 'no!' (pushing food away) is a PERSONAL expression of the child's own will and feeling (in Dore's finer terms, a PROTESTING act). 'dada!' (at the door) is INTERACTIONAL, using language to acknowledge and bond with the arriving parent. Across the four utterances, then, Child A uses language to GET, to CONTROL, to express the SELF and to BOND — the four functions Halliday grouped as developing earliest.

Child B's utterances, by contrast, foreground the LATER functions. 'why the sky blue?' is HEURISTIC — language to explore and understand the world, here a genuine question seeking causal explanation, a more sophisticated heuristic move than a simple 'wassat?'. 'let's pretend I'm a doctor' is IMAGINATIVE — language used to create a pretend world and direct role-play, a function that depends on a level of abstraction Child A has not yet reached. 'the cat is sleeping' is REPRESENTATIONAL / INFORMATIVE — language used purely to convey a fact about the world, with no request, no control and no self-expression; Halliday noted this function develops a little later, and its clear presence in Child B fits that observation. 'I don't like that, it's yucky' is PERSONAL, expressing an opinion and feeling — but, notably, a more elaborated personal statement than Child A's bare 'no!'.

The comparison reveals a clear developmental broadening of the functional repertoire. Child A's language is anchored in the IMMEDIATE and the SELF — getting needs met, controlling the here-and-now, bonding with present caregivers. Child B has retained those functions but ADDED the more cognitively demanding ones: exploring causes (heuristic), creating imagined scenarios (imaginative) and stating context-free information (representational). This matches Halliday's observation that the four pragmatic functions emerge first and that the representational/informative function develops later, as the child becomes able to use language about things beyond immediate need.

In synthesis, the two transcripts capture two points on the expansion of children's functional range: from a repertoire centred on satisfying needs, controlling others, bonding and expressing the self, to one that additionally serves enquiry, imagination and the sharing of information. The comparison shows that acquiring language is not only about acquiring forms but about acquiring an ever-wider set of PURPOSES for which language can be used — learning, in Halliday's terms, more and more ways to mean.

The statement makes a strong claim — that a child's utterance 'rarely serves just one function' — and the transcript of this child of two years eight months (2;8) offers good evidence to evaluate it. Analysing the utterances through Halliday's framework shows that several are indeed MULTIFUNCTIONAL, supporting the statement, while one or two are more single-purpose, suggesting the claim is broadly but not absolutely true. The fairest verdict is that the statement is largely supported, provided 'rarely' is read as 'often not' rather than 'never'.

Consider first the clearly MULTIFUNCTIONAL utterances, which support the statement. 'Mummy, read me a story!' is, on the surface, a request — but it performs several functions at once. It is INSTRUMENTAL (the child seeks to obtain a desired outcome, the story), REGULATORY (it directs the parent to ACT, to do the reading), and INTERACTIONAL (the address term 'mummy' and the bedtime ritual sustain the parent-child bond). One short imperative thus does at least three communicative jobs, and the context — a child managing a bedtime routine with a caregiver — is what lets us read all three. This is exactly the kind of utterance the statement has in mind.

'No, not that one!' is similarly layered. It is PERSONAL — expressing the child's own preference and will (in Dore's finer terms, a PROTESTING act) — but it is also REGULATORY, directing the parent away from one choice and implicitly toward another. The refusal is simultaneously about the self (what I want) and about controlling another (what you should do). Again the utterance resists a single label, supporting the statement.

'I'm the monster, RAAAR!' is dominated by the IMAGINATIVE function — the child creates a pretend role and a fictional world. Yet even here there is an INTERACTIONAL layer (the role-play invites the parent into a shared game) and arguably a PERSONAL one (expressing the child's playful identity). The imaginative function is foregrounded, but the utterance is not purely single-purpose, which again fits the statement.

However, the statement says 'rarely', and the transcript supplies a useful test of that strong word. 'Wassat picture?' is, in context, fairly close to a SINGLE-function utterance: it is essentially HEURISTIC, a genuine question seeking information about the world, with little instrumental, regulatory or interactional load beyond the minimal fact of being addressed to a listener. One could argue a faint interactional element (any utterance to a present listener sustains contact), but the dominant — and arguably sole substantive — purpose is enquiry. This utterance suggests that some child utterances ARE predominantly monofunctional, which qualifies the statement's strong 'rarely'.

'I love you, night-night' is more genuinely multifunctional again — INTERACTIONAL (maintaining the bond) and PERSONAL (expressing the child's own feeling) — and as a routine bedtime formula it does relationship-work and self-expression together.

In evaluation, the statement is largely supported by this data: four of the five utterances are clearly multifunctional, and the analysis shows that reading a child's language through a single label usually loses something. The qualification is that 'rarely' overstates the case slightly — a clear, predominantly heuristic question like 'wassat picture?' shows that some utterances ARE close to single-function. The more precise formulation is that children's utterances are FREQUENTLY multifunctional, especially when they combine getting, controlling and bonding, and that the analyst should therefore weight functions by context rather than assume one label per utterance. Read this way, the transcript strongly supports the spirit of the statement: the safe analytical default is to look for layered functions, not single ones.

This transcript of a child of two years seven months (2;7) at play is unusually well-suited to comparing the two main frameworks for the functions of children's language: Halliday's seven functions (1975) and Dore's finer-grained 'primitive speech acts' (1975). Analysed through both, the data shows that while Halliday captures the broad communicative purposes, Dore's taxonomy accounts MORE precisely for the subtle, conversational distinctions in this particular exchange — so the better framework depends on the grain of analysis the data demands.

Begin with HALLIDAY. The repeated 'doggy' utterances serve a spread of his functions. The first 'doggy!' (labelling a toy) is REPRESENTATIONAL / INFORMATIVE — naming, giving information about what the object is. 'doggy?' (offering it to the parent) and 'you say doggy!' are INTERACTIONAL, drawing the parent into a shared exchange and managing the game. 'where doggy?' is HEURISTIC — a question seeking information about the world. 'no! my doggy!' is PERSONAL (asserting possession and the child's own will) with a REGULATORY edge (warning the parent off). Halliday's framework thus gives a coherent overall picture: the child uses language to inform, to interact, to enquire and to assert the self, often multifunctionally.

But notice where Halliday's broad categories BLUR distinctions that matter in this data. Halliday's single INTERACTIONAL label is doing a lot of work — it has to cover 'doggy?' (offering), 'you say doggy!' (prompting the parent to speak) and the relationship-maintenance running through the game, which are quite different conversational moves. His REPRESENTATIONAL label covers 'doggy!' (naming) without distinguishing it from other kinds of information-giving.

DORE's taxonomy slices these moves more finely, and this is where it accounts for the data better. The first 'doggy!' is, in Dore's terms, LABELLING (naming a referent) — a more precise reading than the broad representational function. 'doggy?' offered to the parent and 'you say doggy!' are REQUESTING (an answer / action) and arguably PRACTISING / prompting — Dore separates the act of soliciting a response from the act of bonding, which Halliday's interactional function merges. 'where doggy?' is REQUESTING AN ANSWER, distinct in Dore from ANSWERING. 'no! my doggy!' is PROTESTING, a named act in Dore that Halliday must reconstruct from a personal/regulatory blend. Dore even has a category — PRACTISING — for the child's repetitive 'doggy … doggy … doggy', which Halliday's purpose-based scheme does not directly name.

Mapping the two frameworks shows their relationship clearly: Dore's LABELLING sits within Halliday's REPRESENTATIONAL; his REQUESTING-AN-ANSWER within HEURISTIC; his GREETING/CALLING within INTERACTIONAL; his PROTESTING within PERSONAL/REGULATORY. Dore is, in effect, a higher-resolution version of the same territory. That is its strength here: a transcript built around subtle, repetitive conversational moves in play (labelling, offering, prompting, protesting) is exactly the kind of data where Dore's finer categories earn their keep.

Discussing 'how far' each framework accounts for the data: HALLIDAY accounts well for the BROAD communicative purposes and is the better lens when the analytical question is 'what is this child using language FOR, in general?' — its seven functions are clear, memorable and widely applicable. DORE accounts better for the FINE-GRAINED conversational acts in this particular play sequence, where the interest lies in distinguishing labelling from requesting from protesting from practising. Neither is simply 'right'; the better framework depends on the GRAIN of analysis the data and the question require. The most sophisticated reading uses Halliday as the organising backbone and deploys Dore to refine the points where Halliday's broad categories blur important distinctions.

In conclusion, the data is well accounted for by both frameworks, but at different resolutions: Halliday gives the coherent functional overview, while Dore captures the subtle conversational texture of this play exchange more precisely. The evaluative point — that the choice of framework should follow the grain of the data — is itself the mark of a mature functional analysis, and this repetitive, move-by-move play transcript is a case where Dore's finer lens adds genuine analytical value alongside Halliday's broader one.