You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Child-directed speech (CDS) — also known as motherese, parentese, or caretaker speech — refers to the distinctive way that adults (and older children) modify their language when speaking to young children. CDS sits at the heart of the AQA A-Level Children's Language Development (0–11 years) topic because it is the single most important arena in which the great theoretical debate of child language plays out: does language come from inside the child (Chomsky's innate Language Acquisition Device) or from outside, through the structured input and interaction caregivers provide? Social interactionists such as Bruner argue that the adult's adjustments are not incidental but are an organised teaching environment that makes acquisition possible; nativists counter that input is merely a trigger for an internally specified system. Because Section B presents you with spoken, written or multimodal data and asks for a discursive essay, you must be able to spot CDS features precisely, name them, and weigh how far they support or undermine each theoretical position.
Key Definition: Child-Directed Speech (CDS) — the systematically modified register adults and older children use when addressing young children, characterised by higher pitch, exaggerated and slowed intonation, simplified vocabulary and grammar, shorter utterances, and heightened repetition. The older term motherese is now avoided because the modifications are produced by many caregivers, not only mothers; parentese and caretaker speech are alternatives.
CDS differs from adult-directed speech (ADS) in systematic ways at every language level — phonology, lexis, grammar, discourse and pragmatics. Crucially, the modifications appear to be largely unconscious and intuitive: most adults sharply deny that they alter their speech when talking to babies, yet recordings show that they reliably do. This intuitive quality is important for the theoretical debate. If adults tune their input without trying to teach, the input cannot be a deliberate syllabus, but it may still be a powerful, naturally occurring support system — which is exactly Bruner's claim. A further point of nuance: CDS is dynamic, not fixed. Caregivers continuously recalibrate the complexity of their speech in response to the child's apparent comprehension, so the register a parent uses with a 14-month-old differs markedly from the one used with a 4-year-old.
It is worth distinguishing CDS (a register defined by its features) from the broader process of infant-directed interaction (the whole multimodal package of gaze, gesture, touch, turn-taking and joint attention in which that register is embedded). Much of the evidence that "input matters" is really evidence that interaction matters — a distinction that becomes decisive when we evaluate the research below.
Examiners reward candidates who can identify CDS features at multiple levels and then do something analytical with them. The tables below organise the features; the discussion that follows explains why each cluster exists.
| Feature | Description | Example |
|---|---|---|
| Higher pitch (raised fundamental frequency) | Adults lift the overall pitch of the voice | Speaking in a noticeably higher register than with another adult |
| Exaggerated, "sing-song" intonation | Wider pitch range with more pronounced rises and falls | "Where's the BALL?" with a dramatic swooping contour |
| Slower tempo and longer pauses | Speech is slowed; gaps are left after utterances | Leaving a pause that creates a "slot" for the child's turn |
| Clearer, hyper-articulated sounds | Phonemes are produced more distinctly than in casual speech | Crisp articulation of each syllable in "ba-na-na" |
| Stress on key content words | The most informative word receives extra emphasis | "Look at the DOG!" |
The phonological cluster has two plausible functions. First, the raised pitch and exaggerated melody are attention-getting and affect-signalling — they engage the infant and convey warmth, which helps explain why even very young babies prefer to listen to CDS over ADS. Second, the slowed, hyper-articulated speech may make the speech stream easier to segment: the boundaries between words in ordinary rapid speech are notoriously hard to detect, and clearer articulation with stressed key words may help the child carve continuous sound into discrete units — a problem sometimes called the segmentation problem.
| Feature | Description | Example |
|---|---|---|
| High-frequency, simple vocabulary | Basic everyday words rather than technical or low-frequency terms | "Tummy" rather than "stomach"; "dog" rather than "spaniel" |
| Concrete nouns for the here-and-now | Words for tangible, present, picturable objects | "Look at the cat" rather than abstract or displaced reference |
| Diminutives | "-y" / "-ie" endings that soften and shorten | "Doggy", "birdie", "horsie", "blankie" |
| Reduplicated forms | Repeated syllables, often onomatopoeic | "Choo-choo", "wee-wee", "night-night", "moo-moo" |
| Adoption of the child's own coinages | Caregiver re-uses the child's invented word | Using the child's "baba" for bottle |
The lexical features keep talk anchored in the immediate, observable context, which matters because young children cannot yet decode reference to absent or abstract things. Diminutives and reduplications are phonologically simple (CV-CV shapes the child can already produce in babbling), which may make early words easier to attempt and remember.
| Feature | Description | Example |
|---|---|---|
| Shorter utterances (lower MLU) | Mean Length of Utterance is well below adult norms | "Look! A dog!" rather than "Can you see the dog running across the park?" |
| Simpler sentence structures | Few embedded clauses, passives or complex constructions | Canonical SVO: "Mummy is eating dinner" |
| More imperatives | A higher proportion of commands and directives | "Come here", "Sit down", "Don't touch that" |
| More interrogatives | Many yes/no and tag questions | "That's a big dog, isn't it?", "Do you want juice?" |
| Frequent repetition and rephrasing | Words, phrases and whole utterances repeated | "Where's the ball? Can you see the ball? There's the ball!" |
| Expansions | Child's utterance rebuilt into a fuller grammatical form | Child: "Daddy car." Adult: "Yes, that's Daddy's car." |
| Recasts | Child's utterance repeated with the error corrected, without comment | Child: "I goed park." Adult: "You went to the park? Lovely!" |
The grammatical cluster keeps utterances within the child's processing capacity (low MLU, canonical word order) while questions and tag questions actively solicit a response, sustaining the conversation. Repetition gives the child multiple, slightly varied exposures to the same structure.
| Feature | Description | Example |
|---|---|---|
| Directives / attention-direction | Telling the child what to do or look at | "Look at this!", "Come and see!" |
| Joint attention | Coordinating focus on the same object or event | Following the child's gaze and naming what they look at |
| Phatic and supportive talk | Utterances maintaining contact rather than carrying information | "Who's a clever boy then?", "There we go" |
| Running commentary / self-talk | Narrating ongoing actions | "Mummy's making dinner now. In goes the pasta." |
| Deixis tied to the visible world | Pointing words anchored to what is present | "Look at that!", "What's this?" |
Catherine Snow (1972, 1977) was among the first to study CDS systematically and remains the anchor reference for this topic. Snow argued that CDS functions as a form of scaffolding — caregivers intuitively pitch their speech a little above the child's current level and adjust it dynamically as competence grows.
Snow's key findings include:
Key Definition: The fine-tuning hypothesis — the claim (associated with Snow) that caregiver input is continuously calibrated to sit just ahead of the child's current linguistic ability, providing an optimally pitched model. The hypothesis is debated: some researchers find input is only roughly tuned, and that children can extract grammar from input that is not finely graded.
Key Definition: Scaffolding (Bruner's term, applied here to CDS) — the temporary, adjustable support a more competent speaker provides so the child can participate in interaction beyond their independent ability; the support is gradually withdrawn as competence grows.
Where Chomsky proposed an internal Language Acquisition Device (LAD), Jerome Bruner (1983) argued that the LAD could only function if it was met by a Language Acquisition Support System (LASS) — the structured, predictable interactional framework caregivers provide. For Bruner, CDS features are the linguistic surface of this deeper support system.
Bruner emphasised formats (also called routines or ritualised interactions): highly repetitive, predictable activity frames in which the same language recurs with the same structure. His classic example is the peekaboo game, with its fixed sequence (concealment → suspense → reappearance → "boo!"). Because the structure is constant and only small slots vary, the child can:
Book-sharing routines work similarly: the caregiver's repeated "What's that?" → child labels → "Yes, that's a ___!" cycle is a ritualised naming format that drills the labelling skill Aitchison places first in lexical development.
Key Definition: LASS (Language Acquisition Support System) — Bruner's term for the structured interactional environment (formats, joint attention, scaffolding, ritualised routines) through which caregivers support language acquisition. It is explicitly framed as a complement to Chomsky's LAD, not a replacement.
Two CDS features are so analytically productive that examiners expect precise handling of them.
Expansions occur when a caregiver takes a child's incomplete utterance and fills in the missing grammatical material while keeping the meaning:
| Child's Utterance | Caregiver's Expansion |
|---|---|
| "Doggy run" | "Yes, the doggy is running!" |
| "More milk" | "You want more milk?" |
| "Teddy bed" | "Teddy is in the bed, isn't he?" |
Recasts occur when a caregiver reformulates the child's erroneous utterance into the correct adult form, again without explicitly flagging the error:
| Child's Utterance | Caregiver's Recast |
|---|---|
| "I goed to the park" | "You went to the park? How lovely!" |
| "Her is playing" | "She's playing, is she?" |
| "Two mouses" | "Oh, two mice! Where are they?" |
The theoretical pay-off is large. Roger Brown and colleagues (Brown, Cazden and Bellugi, 1969) famously found that parents tend to correct the truth value of a child's utterance, not its grammar — so explicit grammatical correction is rare and inconsistent. Recasts seem to solve a puzzle this creates: if grammar is barely corrected, how does the child receive negative evidence (information about what is not allowed)? Michael Saxton's (1997) contrast theory of negative evidence proposes that the immediate juxtaposition of the child's incorrect form and the adult's correct recast ("goed" → "went") supplies exactly this corrective information in a non-disruptive way, and that children are more likely to adopt the correct form after a recast than after no feedback. Note the appropriate caution: recasts correlate with improvement and are plausibly helpful, but isolating their causal effect from the rest of rich interaction is difficult.
Elinor Ochs and Bambi Schieffelin (1984) carried out the most influential challenge to the assumption that CDS is universal and essential. Comparing language socialisation in three communities, they found striking divergence:
| Community | Language Socialisation Practice |
|---|---|
| White middle-class American | High levels of CDS; the infant is treated as a conversational partner from birth; adults simplify and adapt their speech to the child |
| Kaluli (Papua New Guinea) | Little simplified speech to infants; adults consider it pointless to converse with someone who cannot yet talk; instead they use the routine elema ("say it like this"), modelling whole utterances for the child to repeat to a third party |
| Samoan | Limited simplification; status hierarchies mean the lower-status child is expected to adapt upward to the adult, not vice versa; older siblings are major caregivers |
Despite these very different approaches, children in all three communities acquired their language fully and on a broadly normal timetable. The conclusion is carefully bounded: CDS as practised in Western middle-class homes is not a universal prerequisite for acquisition; what may be universal is some form of structured language socialisation, even if it does not take the "simplify and address the baby" form.
Key Definition: Language socialisation — the process by which children acquire not only linguistic forms but also the culturally specific norms for using language. Ochs and Schieffelin (1984) showed that these norms, and the role of simplified speech within them, vary substantially across cultures.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.