What if the most unsettling children's show ever broadcast was unsettling on purpose? Not by accident. Not because somebody wasn't paying attention. Because a team of people sat in a room, looked at the research on how two-year-olds process the world, and built something that would land in exactly the way it lands.
That's Daniel's question this week. He wrote in — and he's right that we've poked at the Teletubbies before — but he wants the origin story this time. Who were these people? Anne Wood and Andrew Davenport. What was engineered versus what just happened? Daniel's specific angle is that a lot of what looks strange and inexplicable about that show was actually calculated decisions, plus the technology that happened to be available in the mid-nineties. And then there's Po. Daniel calls her, quote, "arguably the strangest of the lot." He wants to know why.
He's not wrong. Rewatching it as an adult is a completely different experience than watching it in diapers. The baby sun alone.
The baby sun deity, as Daniel puts it.
Right. And here's the thing that makes this a real episode rather than a nostalgia trip — the answers exist. Wood and Davenport have both talked about this in detail. The weirdness wasn't a mystery. It was a design.
So it wasn't an accident. It was engineered.
Every bit of it. Start with 1994. The BBC puts out a call for pitches for a new preschool programme. Eleven other companies bid.
Nine feet of Tinky Winky against eleven competitors.
And Ragdoll wins. Anne Wood's company. She's the producer, veteran of children's television, and she pitched a show aimed at an audience that, in her words, had never had programmes made for it before. Two and three-year-olds.
Which is the part people skip past. Everyone remembers the show. Almost nobody remembers that it was aimed lower than anything before it.
That's the whole thing. Davenport came at it sideways. He had a BSc in Speech Sciences from University College London. He'd trained intending to be a speech therapist. He was working as a puppeteer and co-writer on Ragdoll's Tots TV when he and Wood started talking about a show for a younger audience built on developmental rationale rather than educational rationale.
Say more about that distinction, because it sounds like the whole thesis in one line.
Educational rationale means you're teaching something. Letters, numbers, colours. Developmental rationale means you're designing around where the child actually is — what they can perceive, what they can process, what makes them respond. Davenport's own words were that makers of children's TV tended to come from a teaching background, and that this portrayed a situation alien to the culture of the younger preschool child. Children that age learn by self-motivated experimentation. So he built the show around that instead of around a curriculum.
And the astronauts.
The astronauts. He and Wood had just been to the Smithsonian in Washington, and Davenport was struck by something. NASA astronauts coming out of landing capsules in those bulky suits looked like toddlers in nappies. Oversized heads, foreshortened legs. He said it struck him as funny that at the pinnacle of human achievement, the figures that emerged looked like that.
So he built a show about spacemen.
He devised characters based on spacemen, with limited language, mirroring the emergent speech of young children. They drew up what they called a picturebook of technological babies and took it to the BBC. And they were unnerved to find that the eleven other companies bidding had brought lots.
They brought a sketchbook. Everyone else brought a portfolio.
And they won anyway. Premieres thirty-first of March, nineteen ninety-seven, on BBC2. Three hundred and sixty-five episodes. Over a hundred and twenty countries. Forty-five languages. Three hundred and thirty million pounds generated in the first two years.
Which is where I want to slow down, because the money number is the boring number. The interesting question is what the money was being spent on. What was actually on screen.
Ask it.
What's the baby in the sun for?
Functionally, exactly what it looks like. Davenport's explanation is that a gurgling baby is one of the easiest ways of getting a response out of anyone, even another baby. And by laughing, the baby in the sun also lets the children know when they're supposed to be laughing along.
It's a laugh track made of an infant.
It's a cue. And it's more than a cue. Contemporary coverage at the time added that babies like to see babies, and that the sun baby's intent expressions and reactions reassure viewers that all is well — or let them know something is about to happen. It's doing emotional bookending. It opens the episode and it tells you what register you're in.
So the single most unsettling image in the show is a piece of user interface.
It really is. And here's the honest caveat — nobody has found a definitive creator statement about why specifically a baby's face inside a sun. The function is documented. The visual itself, the choice of a sun specifically as the container, that's not been attributed anywhere I can find. So there's a gap in the record.
The most-asked question about the show and the paperwork just says "and then someone put a baby in a sun."
That's roughly where it sits, yes.
Okay, tummy TVs.
This is my favourite one, because it's the purest example of a constraint producing the aesthetic. The BBC stipulated that real young children had to appear in the show. Actual toddlers, on camera. Which creates an immediate problem — the Teletubbies are enormous. Tinky Winky is nearly nine feet tall in costume. You cannot put a nine-foot creature and a two-year-old in the same frame without it looking like a hostage situation.
So you shrink the child instead.
You shrink the child. Davenport put televisions in the Teletubbies' tummies specifically to solve the scale difference between the monster-size Teletubbies and the young children. The screen becomes the meeting point. The child appears at a size that's plausible next to the Tubbies because the child is inside a screen that's part of a Tubby.
The frame does the work.
The frame resizes reality. It's the same trick as a picture-in-picture on a news broadcast. You're not changing the person, you're changing the box they're standing in.
Voice trumpets next.
Voice trumpets are the one I'd put on a poster. Davenport's research found that young children rarely listen and watch at the same time.
Hold on.
It's a strange fact and it's the one that unlocks the show. A small child watching television is either looking or listening. Not both. They're not dual-tracking the way an adult does.
So speech arriving out of nowhere is wasted.
Speech arriving out of nowhere is wasted, and worse, it's confusing. So the voice trumpets — those periscope things that rise out of the ground — they turn the Tubbies' heads. That's their job. The trumpet rises, the Tubbies turn to look at it, and that physical movement is the signal to the child: something is about to be said, look at the source, now listen.
They built a mechanical attention-grabber because the audience couldn't multitask.
They built a stage direction that the audience could actually perceive. It's the closest thing I know of to a television show designing around a perceptual limitation instead of ignoring it.
And the gibberish.
Davenport is adamant about this one. His position is that it is a genuine language the Tubbies have. It has a proper grammatical structure, and what they say is meaningful to the children.
That's a strong claim.
It's a strong claim and I'll be honest, I'm not sure it's ever been independently verified in the way he phrases it. But the design intent behind it is clear — the point isn't that the words mean something in English. The point is that the speech pattern mirrors the emergent speech of toddlers. The rhythm, the repetition, the sing-song rise and fall. A two-year-old hearing it hears something shaped like their own attempts at language. It's comprehensible in cadence even when it isn't comprehensible in vocabulary.
So it's not nonsense. It's a dialect of almost-speaking.
That's a better phrase than mine.
Repetition.
Davenport's line is that children like the security of repetition, and repetition keeps things simple and familiar. But the knock-on effect is the interesting one. Repetition lets a child predict. If the same thing happens in the same order every episode, the child gets to be right. And being right is a cognitive reward. They're not just being soothed by the pattern, they're training deduction on it.
So the show teaches by letting the audience guess correctly.
Over and over. That's why the structure is so rigid. The rigidity isn't laziness. The rigidity is the teaching.
The rabbits.
The rabbits are where it gets dark.
Go on.
Wood explained that the rabbits needed to be big to fit in with the scale. You can't have a nine-foot Tinky Winky standing next to a normal rabbit, it destroys the world. So they needed enormous rabbits. And the only ones available that size had been bred on the continent to be eaten.
As livestock.
As meat. And their breeding had given them enlarged hearts. Wood said almost weekly the animal trainer would greet her in distress and tell her another one had died.
The giant rabbits that hop across the grass while the Tubbies dance were, in real life, dying off-camera on a weekly basis.
Because their hearts were too big for their bodies. Which is a sentence I never expected to say about a children's show.
So the most surreal element of the whole production was the one with the most ordinary explanation. They needed big rabbits. Big rabbits were being farmed. The farming killed them.
Nobody designed the dying. That was the part nobody designed.
What's Wood's description of the set itself?
Teletubbyland. She called it a cross between the land where television comes from and a nursery rhyme land. They live in a Tubbytronic Superdome and there's a sentient vacuum cleaner called Noo-Noo.
A cross between where television comes from and a nursery rhyme land. That's a set designer describing a set designer's job.
It's also the most honest description of the show anyone ever gave. It's literally a television landscape. Grass that's too green, hills that are too smooth, and a machine that eats things.
So we've got the baby sun as an emotional cue, the tummy TVs as a scale fix, the voice trumpets as a perceptual workaround, the gibberish as a mirror of toddler speech, the repetition as a deduction trainer, and the rabbits as a livestock casualty. That's the whole show.
Every piece of it.
So now Po. Daniel's call. He wants to know why she's the strangest of the lot, and I think he's right, but I want to hear you argue it.
She's the smallest. She's red. She's got a circular antenna, the UHF-style one. She rides a pink-and-blue scooter. She's the one who gets described as mischievous, clever, incredibly energetic, with a fiery spirit and a bit of a rebellious streak.
And her voice.
Her voice is the actual answer. Po was played by Pui Fan Lee. British-Hong Kong actress. And because English wasn't her mother tongue, she struggled to find her Teletubby voice. Wood's account is that they decided Po should use Mandarin.
There's a discrepancy there, isn't there.
There is, and it's worth flagging because Daniel would want us to. Wood told the Guardian in twenty thirteen that they decided Po should use Mandarin. But the specific phrases Po actually says are Cantonese. Pui Fan Lee gave Po a little Cantonese — Fi-dit, Fi-dit, which means faster, faster, and Yāt, yih, sàam, which is one, two, three.
So the creator's own quote doesn't match the audio.
It doesn't. The Cantonese attribution is more specific and better supported. The Mandarin line comes from Wood herself, in one interview, and it's possible she was using Mandarin loosely, or misremembering, or simplifying. But the phrases on tape are Cantonese. That's what a British-Hong Kong actress would bring.
So Po is the strangest because of a casting accident.
Po is the strangest because of an actor's linguistic background. Which is the same root cause as everything else in the show — a practical constraint that the production chose to incorporate instead of hide. They didn't replace Pui Fan Lee. They didn't coach the Cantonese out of her. They leaned into it.
And the result is that one of the four Tubbies is the only one whose language is foreign to the English-speaking viewer.
Foreign, and untranslated. The other three are speaking a dialect of almost-speaking that a child can parse by rhythm. Po is speaking actual words in a real language that almost nobody in the audience knows. Which means Po is the one where the show's central conceit — that the language is meaningful but not in English — is literally true rather than metaphorically true.
That's the answer.
She's the only Tubby who stops being a design choice and starts being a person.
And the reason she's remembered as the strangest isn't because she's stranger. It's because she's the only one where the mystery has a real, undocumented bottom.
Which is exactly what Daniel was circling.
Now — the decade. Wood said they were interested in how children were reacting to the increasingly technological environment of the late nineties, and that the Tubbies were technological babies on the cusp of technological change. Say what that means.
It means the show was positioned as an antidote. The specific thing it was written against was violent children's television — Mighty Morphin Power Rangers is the one that gets named. Ragdoll's pitch was that if you're making something for two-year-olds, you start by not giving them violence-and-catchphrases, you give them a landscape and a rhythm.
And yet the Tubbies are machines. Tummy screens, voice trumpets rising out of the ground, a vacuum cleaner with a personality.
That's the tension. The show is simultaneously anti-technology in its values and completely saturated with technology in its imagery. Wood's framing is that the Tubbies aren't there to be protected from the technological world, they're there to model living comfortably inside it. That's why the devices are friendly. Nothing in Teletubbyland malfunctions in a threatening way. Noo-Noo eats things and then gives them back.
Which is the opposite of almost every children's show that has technology in it. The machine is the ally, not the danger.
That's a unusual position for ninety-seven. Most television aimed at kids with gadgets in it was teaching them to be suspicious of the gadgets.
And then the adults got involved.
The adults got very involved. Wood's line about this is that the intention was to create an imaginative landscape for children to project their own ideas on to. And then she says: we found that everyone in the world projects their own ideas. It's astonishing what some people have read into it.
She built a blank wall and then was surprised when people saw things in it.
She built a blank wall on purpose and was surprised that it was blank enough for people to see things in it.
Give me the famous one.
Jerry Falwell's Parent Alert in ninety-nine. Falwell's newsletter claimed Tinky Winky was gay, on the grounds that he's purple — which is a gay pride colour — and his antenna is a triangle, which is a gay pride symbol. Also, he carries a handbag.
Wood laughed.
Wood laughed when she first heard. But in the United States certain communities took it seriously, to her horror, and she says it damaged the brand considerably. And then — and this is the line I'd frame — she also said it was the best publicity they ever had.
Both of those things at once.
Both of those things at once, and she meant both. It damaged the brand and it made the brand unavoidable.
Did Davenport engage with any of this?
He did, and his line is the most interesting one for this episode. He said the show was quite Brechtian. Because they are actors, but they are also part of the audience.
Unpack that.
Brecht's whole theatrical project was breaking the illusion on purpose. Reminding the audience that they're watching a performance so they stay awake and think rather than disappear into the story. Davenport is saying the Tubbies occupy that weird zone — they perform for the child, and they also sit and watch the tummy screens alongside the child. They're inside the show and outside it at once.
So the show isn't naive at all.
The show is extremely self-aware. Which is the opposite of the popular reading. The popular reading is that this was a bunch of people in suits doing something inexplicable. Davenport is saying they knew exactly what they were doing, including the theatricality.
And the critics were not having it.
Not remotely. Harvard psychologist Susan Linn said there's no documented evidence that Teletubbies has any educational value at all. Dorothy Singer at Yale said gearing a show for very young children goes against everything she believed in.
Which is a striking thing for a researcher to say about a show that was designed with developmental research.
It's a striking thing and it's revealing. Singer's objection isn't about quality, it's about the premise. She's saying the target age itself is the problem.
And then PBS.
Alice Cahn at PBS is the one who pushes back. Her line was: what's so sacrosanct about one-year-olds? Whether we as a society like it or not, kids watch TV.
Which is the most honest sentence in the whole debate.
It really is. Linn and Singer object to the audience existing. Cahn points out that it does.
So Ragdoll's position was that they were the only people in the room taking that audience seriously.
That's the argument. The critic says you shouldn't make this. The producer says somebody is already making it and it's worse than this. And Wood's line, at the time, was that there had never been a programme that speaks so distinctly to children the world over, and that they were the first to understand how children perceive the world around them.
Which is a big claim.
It's a big claim and I don't think it's literally true, but I think it points at something real, which is that almost nobody had built a show around perception rather than content.
So the argument between them is whether a blank landscape is respectful or negligent.
And they never resolved it, which is why we're still talking about it.
The puddle thing. You mentioned there's one more design story.
There is. Davenport watched children shouting ugh at a Tubbie stepping in a puddle. That's it. That's the input. He noticed kids found that funny and would shout at the screen. So he wrote a scene where all four Tubbies step in puddles and shout ugh at each other.
That's how the show was written.
He watched what the audience did and wrote it back into the show. Which is the same loop as the voice trumpets — the audience reacts, the production incorporates. That's what Brechtian means in practice. The show is watching itself watch.
And then the whole thing ends. Sixteenth of February, two thousand one. Last original episode. And then the reboots.
Two thousand fifteen, WildBrain takes it and swaps the tummy TVs for touch screens.
They replaced the design constraint with a design update.
Which in fairness was the correct thing to do if you're rebuilding it for a different decade. The tummy TVs existed to solve a scale problem with a CRT. In twenty fifteen the CRT is gone and everything is a touchscreen, so the screen becomes a touchscreen. It's internally consistent.
But the babies.
The babies got recast, multi-racial Sun Babies. Which again is a defensible choice and it changes the emotional register of the opening significantly, because the original sun baby was one specific baby.
Jess Smith.
Filmed at nine months old. Paid two hundred and fifty pounds plus a box of toys.
That's the whole payment.
She revealed her identity in twenty fourteen at age nineteen.
Two hundred and fifty pounds and a box of toys, for the single most-viewed face in the show's history.
There's something very honest about that number.
And then Netflix.
Twenty twenty-two, a second reboot announced for Netflix, Tituss Burgess as narrator. And Anne Wood criticises it as a major devaluation of the brand.
The creator, watching her own show rebuilt by people who didn't make it, and calling it what it is.
Which is a very consistent position for her. She spent the whole original run arguing that the design was the point. A reboot that keeps the characters and drops the design is, from her perspective, the whole thing gone.
And last year, the two of them.
Twenty twenty-five, Andrew Davenport and Anne Wood jointly received the J.M. Barrie Award for Lifetime Achievement in children's arts and television. Which is a strange and lovely bookend — the woman who made the show about spacemen and the man who made the show about perception, getting a lifetime achievement award named after the man who wrote Peter Pan.
Both of them built around children's perceptions of the world.
Both of them built around children's perceptions of the world and got told at various points that the audience shouldn't exist.
So the show got a lifetime achievement award for doing the thing the critics said shouldn't be done.
Yes.
That tracks.
It does.
There is one more thing. Daniel said "arguably the strangest of the lot," and I want to leave the audience with the answer rather than the question. So — Po.
Po is the strangest because Po is the only one who was literally, a specific person. The other three are designs. Po is Pui Fan Lee.
And the strangeness of Po isn't design strangeness.
It's person strangeness. Which is a completely different flavour.
Anything else on the language point worth planting?
Only that Po's voice is described as more soft-spoken than the other Tubbies, almost like a whisper. Which I think is a feedback loop. The softer the delivery, the more the Cantonese words float out without explanation, and the more the audience has to lean in to hear it. And leaning in is what fixes attention.
So the quiet one is the one you remember.
The quiet one is the one you remember.
Yeah.
And the press did notice. There's a Design Week piece from ninety-nine called "The tao of Po" that treats her as a cultural icon in her own right. Po was the one that showed up in student rooms and became a cult figure. Wood noted that the Tubbies became gay icons and a cult among students specifically.
So the strangest one is the one that leaked out of the show and into the culture.
Hilbert: We had one that went out live with the rabbit still in the hutch, and the segment ran four minutes long, and nobody could get to the bar to open it because the door was on the wrong side of the set.
Four minutes of grass.
Hilbert: The floor manager kept signalling at us to do something and there was nothing to do. We weren't allowed to touch the hutch. That was the rule.
You worked on children's productions?
Hilbert: Set dressing, mostly. Children's was hard work because everything is oversized and you can't see where you're putting your feet. I remember the suits were nine feet and the performers inside were completely blind below the chest. You couldn't see the floor. So somebody had to walk in front of them, going backwards, holding their hand, telling them when to stop.
Every step, somebody walking backwards.
Hilbert: Every step for the entire shoot. And you can only do that for so long before it stops being funny. I saw one performer go down. Couldn't get out quickly enough. Panic, in the suit, in the heat.
That's nine feet of foam with no ventilation.
Hilbert: It was hot in there. Very hot. And the performers on Teletubbies had it in their contracts that they couldn't speak to the public. That was the rule. They put the suit on, they did the show, and they couldn't say a word about it. You'd get asked and you just had to stand there.
Same with the rabbit hutch, functionally. Somebody decided something and the person doing the work found out about it on the day.
Hilbert: That's how it goes. And the rabbits, the ones in the hutch, they were enormous. You could see it in the chest. Their hearts were too big. You'd look at one and know it wasn't going to make it.
Wood said almost weekly the trainer would tell her another one had died.
Hilbert: That tracks. Nobody was being cruel. It was just how those animals were bred. It's a different problem than a suit that doesn't let a man see the floor.
But both of them are that. Somebody designing at scale, and a body paying for it.
Hilbert: That's the job.
That's the correct way to think about the whole show. Every single strange element was a person solving a specific problem with the tools they had.
And some of the tools were alive.
Some of the tools were alive.
The thing I keep coming back to is the sun. Because the one image that everyone starts with — the baby inside the sun — is the one where the documentation runs out. We know why a baby was used. We don't know who looked at the sun and decided that was the container. Everything else in the show has a paper trail. That one has a function and no author.
Which, given what you said earlier about the blank wall, is almost too neat.
Almost too neat. Yeah.
Po's the same shape, actually. We have the reason her voice is Cantonese but not a creator statement about why Po specifically is the strangest. That reading is fan consensus.
The two most-asked questions about the show are both questions where the answer is "we know the mechanism and we don't know the mind."
And I think that's the show's shape. There's a lot of mechanism and not a lot of mind. That's what makes it settle so strangely on adult ears. It's an extremely carefully built thing with nobody visibly inside it.
And that's the forward-looking thought. The show was stranger because it was designed so carefully for a mind that can't yet ask questions about it. The weirdness is the craft. Which is why the reboots fail. You can rebuild the Tubbytronic Superdome. You can replace the tummy CRTs with touchscreens. You can't recreate the process of thinking about what a pre-verbal child can perceive. That part isn't in the assets.
Which is exactly why Wood called the Netflix version a devaluation. She's not objecting to a new show. She's saying you can't reboot a research process.
So a two-year-old will watch the new one and never know the difference. Everybody older does.
And that's the whole episode.
Good. Thanks as ever to Hilbert Flumingtop, our producer, who ran the desk through all of this.
Thanks to Daniel, obviously, for the prompt — and if the exact origin of the baby sun ever turns up in a Ragdoll archive somewhere, we want the email.
Send it — show at my weird prompts dot com. That's show at my weird prompts dot com. One more thing before we go: if you've enjoyed this and you haven't yet, a review helps other listeners find the show, and it's the cheapest thing you can do for us.
This has been My Weird Prompts.
The human-AI collaboration podcast. We'll be back soon.
See you tomorrow.
