A fourteen-month-old sits in her high chair, dinner mostly on the floor, and squeezes her fingertips together in the sign for more. Her mother beams. Her grandmother, visiting for the week, finally says the thing she's been holding in since Tuesday: if you keep letting her talk with her hands, why would she ever bother to speak? It's a fair question, and it has been asked in kitchens for thirty years — ever since baby sign language grew from a small research project into an industry of classes, flashcards, and board books promising earlier words, fewer tantrums, even higher IQs. Here is the strange part: the grandmother's fear and the industry's promise are both wrong. And the true story of what happens when a hearing baby signs is quieter, and better, than either one.

The promise that outran the evidence

In the 1980s, developmental psychologists Linda Acredolo and Susan Goodwyn noticed something charming: hearing babies spontaneously invent gestures. A sniff for flower. Arms out wide for airplane. The researchers began studying what happened when parents deliberately taught babies simple symbolic gestures, and their early work suggested modest language advantages. Their 1996 book Baby Signs turned the finding into a movement, and the movement turned into marketing. Before long, signing was being sold as a cognitive upgrade — a way to raise a baby who talks sooner and tests smarter.

Then independent researchers looked closer. A 2005 review by Johnston, Durieux-Smith, and Bloom went through the published studies and concluded the evidence was too thin and too methodologically shaky to support the developmental claims. In 2013, a more carefully controlled study led by Elizabeth Kirk and colleagues randomly assigned mothers to encourage gesturing with their babies or not, and followed the children's language over time. The result: no overall boost to language development from signing. The miracle, in other words, was oversold.

But notice what none of these studies found, either: harm. Across decades of research on hearing babies who sign, there is no credible evidence that signing delays speech. The grandmother's worry — reasonable as it sounds — has simply never shown up in the data. To understand why it can't show up, you have to look at what language actually is.

Why signing can't delay speech

The worry assumes that speech and gesture are rivals — that a child has one channel for language, and every signal spent on the hands is a signal stolen from the mouth. The science says the opposite: language lives in the brain, not the mouth, and it will pour out of whatever channel is available.

The most striking proof came in 1991, when Laura Ann Petitto and Paula Marentette reported in Science that deaf babies exposed to sign language babble with their hands — producing rhythmic, repetitive, syllable-like hand movements at the same age hearing babies babble "bababa." Deaf children raised by signing parents go on to hit their language milestones — first signs, vocabulary growth, first combinations — on essentially the same schedule as hearing children hit theirs in speech. Sign languages are complete languages, and the brain treats them as such. Language capacity is modality-independent: the hands are not a lesser channel the child might get stuck in. They are simply another door into the same house.

Second, in hearing children, gesture and speech are one integrated system, not competitors. Susan Goldin-Meadow and Jana Iverson's research on early gesture found that it reliably precedes and predicts speech: the objects a baby points at are the words most likely to enter their spoken vocabulary next, and the moment a child starts combining a gesture with a different word — pointing at the cup while saying "mama" — reliably comes just before their first two-word sentences. Gesture isn't a detour around talking. It's the on-ramp.

Third, there is the simple economics of effort. A spoken word works across a room, in the dark, and when both hands are full of applesauce. Once a child can say "milk" and be understood, the sign for milk becomes the slow lane, and toddlers are ruthless about efficiency. Children who sign drop each sign, one by one, as the spoken word comes online. No one has ever met a healthy five-year-old still signing more because talking never seemed worth the trouble.

What a sign actually buys you

So if signing doesn't delay speech and doesn't accelerate it, what is it for? The honest answer: it bridges a gap in timing.

By around the first birthday, a baby's symbolic capacity — the ability to hold an idea like I want milk and know that a signal can stand for it — is running well ahead of their articulation. Speaking requires split-second coordination of tongue, jaw, lips, and larynx, some of the finest motor control a human ever learns, and it matures slowly. Hands are years ahead of mouths. A sign lets the thought out early, through the door that's already open. That's the whole trick. Not a smarter baby — a baby who can tell you she wants more banana a few months before her mouth could have managed it.

And there was one genuinely interesting wrinkle in the Kirk study: while the babies' language didn't differ, the mothers who signed behaved differently. They tended to be more attuned to their babies' nonverbal cues — watching the hands, watching the face, waiting. That matters, because parental responsiveness — noticing a child's signal and answering it promptly and warmly — is one of the best-documented predictors of language growth we have. It raises a lovely possibility: the sign was never really for the baby. It works, when it works, like a string tied around the parent's attention.

The real reason to sign — or to skip it

Which brings us to the practical answer for the kitchen argument. Sign if it sounds like fun. Skip it if it feels like homework. The evidence supports either choice, because the active ingredient was never the handshape — it's the exchange around it. A sign taught from a flashcard, drilled and quizzed, is worth almost nothing. A sign used in a real moment, answered in real words, is one more conversational turn — and conversational turns are the currency language development actually trades in.

Your next moves

  • Pick three to five "need" signs and start today — more, all done, milk, help, eat. Use each sign only in the real moment it applies, and say the word out loud every single time your hands move. The word rides in on the sign.
  • Answer any sign the way you'd answer speech — with sentences. When she signs more, say: "More? You want more banana? Here's more banana." The sign opens the conversational turn; your words are what teach.
  • Never hold the cup hostage. Don't withhold milk until she produces the sign, and don't quiz her for relatives. Signs exist to express wants she already has, not to perform for rewards — pressure turns communication into a test, and toddlers quietly opt out of tests.
  • Watch for the gesture-plus-word combination. When your toddler pairs a point with a different word — pointing at the dog while saying "ball," pointing at Dad's shoes while saying "dada" — that's a milestone worth noticing: those combinations reliably arrive just ahead of two-word sentences.
  • If you never sign a single sign, drop the guilt entirely. Narrate diaper changes, name what's in the grocery cart, pause after you speak and wait for anything — a look, a grunt, a syllable — then respond to it. Responsive talk is the engine. Signs are optional trim.

The string around your attention

The most useful thing buried in thirty years of baby-signs research isn't about hands at all. It's that the interventions that move language are the ones that get a parent to slow down, watch closely, and respond — whatever form that takes. That's the principle Acorn is built on: one three-minute session a day with your 1–3 year old, designed around real words in real moments, with no ads and nothing to upsell — just a small daily reason to sit down, point at the same thing, and talk. If you'd like a gentle structure for those three minutes, you can try it at acorn.lumenlabs.works. And if you'd rather just sign more across the high chair tonight — that works too. It always did.