Did Pauline Hanson really count babies who can't talk to argue migrants can't speak English?

PAULINE Hanson stood at the National Press Club on June 17 and told Australia that 872,000 people in this country cannot speak English properly.

She let that number hang in the air the way she always does — no context, no proportion, just a big scary figure dropped into a sentence about social cohesion falling apart.

What she did not tell anyone is that more than 123,000 of those people were babies and toddlers aged between zero and three.

The woman who wants to run Australia's immigration policy used census data that includes children who have not yet learned to talk to argue that migrants are failing to integrate.

Let that sink in.

The 872,000 figure is real. It comes from the 2021 census. It represents people who reported speaking English "not well" or "not at all" in households where a language other than English was spoken. AAP FactCheck broke down the detailed data and found 14.1 per cent of that number — 123,408 people — were aged zero to three. An additional 829,214 children in that age bracket did not even get asked about English proficiency because their households reported speaking only English at home.

But Hanson did not mention any of that. She presented the raw number as though every one of those 872,000 people was an adult who had chosen not to learn the language.

The broader claim was just as misleading. She cited the census finding that 23 per cent of Australians speak a language other than English at home, framing it as evidence of a national crisis. What she left out was that the vast majority of those people — 84.9 per cent of them — reported speaking English either well or very well. Speaking Mandarin at the dinner table does not mean you cannot speak English at work. Being bilingual is not a failure of integration. It is the opposite.

The ABS itself warns against treating the English proficiency data as a definitive measure of ability. It is self-reported and subjective. One person might say they speak English "not well" because they struggle with workplace jargon. Another might tick "well" because they can order a coffee. The bureau explicitly says the data should be interpreted with care.

Hanson's office was asked by AAP to provide evidence for the claims she made. It provided no response.

This was not an isolated error. The same speech included a claim that 130,000 Australians are "sleeping rough every night" — a figure 17 times higher than the ABS estimate for actual rough sleepers, which conflates rough sleeping with all forms of homelessness including severely overcrowded dwellings. She claimed the largest capital gains tax cohort was people under 35, which is flatly false. She also compared Australia's overseas-born population to the United States without acknowledging the entirely different histories, geographies and immigration systems of the two countries.

Her office responded to none of it.

And then there was the rest of the performance.

Hanson called Guardian Australia reporter Sarah Martin "trashy" for asking a legitimate question about her daughter Lee Hanson's political appointment and publicly declared Martin would be banned from future One Nation press conferences. The International Federation of Journalists condemned the attack. The Federal Parliament Press Gallery committee condemned the media ban. The MEAA called it "bitter, personal and unprofessional".

Hanson's response was to double down.

Meanwhile, GetUp managed to turn the whole event into their own circus. Activists smuggled a drop-down banner into the venue and unfurled it behind Hanson mid-speech, reading: "I opposed a pay rise for workers while I took a $100,000 pay rise for myself." GetUp campaigns director David Sharaz was filmed watching from the audience. The National Press Club referred the matter to the AFP, rejected Sharaz's membership application, banned all GetUp staff from future events, and launched an internal security review. Even Prime Minister Albanese called it "counterproductive".

He was right. The stunt gave Hanson exactly what she wanted — a victim narrative. Instead of the story being about a senator who counted pre-verbal infants to inflate an immigration scare, the story became about GetUp behaving badly at a "high security event". Hanson had already mentioned GetUp in her speech, dismissing the group's $600,000 campaign against One Nation in the Farrer by-election. The banner just proved her point for her.

This is the cycle and it is exhausting.

Hanson manipulates data that does not say what she claims it says. Fact-checkers do the work. Activists pull a stunt that drowns out the fact-check. The media covers the stunt. Hanson's base sees only confirmation that the establishment is out to get her. The actual numbers — the babies in the data, the 84.9 per cent who speak English perfectly well, the self-reported and subjective nature of the entire dataset — never break through.

Benjamin T. Jones, writing in The Conversation this week, put it bluntly: pointing out that Hanson's English figures were inflated by including babies is unlikely to dent her support, because her appeal is emotional, not factual.

That is probably true. But it does not mean the numbers should go unchallenged.

So here they are, one more time.

The 872,000 figure includes more than 123,000 children who have not learned to talk yet. The 23 per cent who speak another language at home overwhelmingly also speak English. The ABS says do not treat this data as definitive. And the woman citing it refused to provide evidence for her claims when asked.

That is not a politician holding the government to account. That is someone who knows the number sounds frightening and does not care whether it is honest.