There was never really much doubt that Donald Trump doubted the science of climate change. He had tweeted to that effect repeatedly before becoming president. He told this publication he wasn’t a “big believer” in human-caused climate change.

That he has now used a bout of cold weather to mock climate change — committing the most basic of errors on the topic, which is to confound any given day’s weather and with long-term trends in climate — should not come as a surprise.

What’s been more surprising, however, is the wide variety of positions taken by the broader Trump administration on the matter — where nobody but the president seems to talk like the president, and where at times, it isn’t even clear whether other members of the administration are representing Trump’s views.

Consider the following three sets of “administration” views on climate change:

Scott Pruitt, administrator of the Environmental Protection Agency, has questioned whether we can measure “with precision” the role of humans on the climate, and he suggested therefore that humans are not the primary driver of global warming.

The U.S. Global Change Research Program, the federal coordinating body for climate change research across the government, recently released a report that is completely incompatible with this view, finding that there is “no convincing alternative explanation” for the planet’s warming other than human causation.

Somewhere in the middle, but perhaps closer to Pruitt, is Secretary of State Rex Tillerson’s view that carbon dioxide is changing the atmosphere but “our ability to predict that effect is very limited.” Tillerson says he knows the change is happening; he’s just not clear on the consequences or how bad they’ll be.

Meanwhile, and as The Washington Post’s Brady Dennis recently chronicled, two of the U.S.’s top Earth-science agencies, NASA and the National Oceanic and Atmospheric Administration, routinely release new analyses and products that document just how warm the planet is, just how much of its ice is melting, just how rapidly its seas are rising, and more.

They, like the federal climate change science program, show little sign of being censored or suppressed.

Indeed, earlier this month, NOAA released a report finding that the Arctic is now warmer than it has been in at least a millennium and a half. The document’s release was embraced by none other than Trump appointee and acting NOAA administrator Tim Gallaudet, who said that the document’s findings “directly relate to the priorities of this administration” — namely, national security goals in the fast-changing Arctic region.

Even at the EPA — which took down a scientifically accurate Web page about climate change, and where rolling back the Obama administration’s climate change policies is a top priority — the planet’s warming is accepted as a premise, at least in technical documents.

The draft Regulatory Impact Analysis that the agency released to analyze the costs and benefits of rolling back President Barack Obama’s Clean Power Plan greatly downgraded government estimates of the economic cost of emitting a ton of carbon dioxide into the atmosphere — but it still said each ton has a cost.

That, itself, is tantamount to accepting that global warming is real and has at least some impact on the world.

At times, it isn’t even clear if top administration officials are actually reflecting the President’s own views on climate change. U.S. Ambassador to the United Nations Nikki Haley said earlier this year that Trump “believes the climate is changing, and he believes pollutants are part of that equation.”

There are at least two questions we should ask about this not-very-consistent approach to climate change in the Trump administration: Why it exists, and how to evaluate its consequences.

As Dennis’s story noted, climate science seems to be far less fettered at science focused agencies than at regulatory agencies — with NASA and NOAA a prime example of the former and the EPA of the latter. At the same time, NASA and NOAA do not yet have permanent administrators in place.

Yet during the administration of George W. Bush — which, like the Trump administration, tried to slow-pedal action on climate change — scientists at NASA and NOAA complained of having their work or ability to communicate with the media interfered with, and scientific documents were edited by political actors.

The difference, really, seems to be that the more tightly organized Bush administration didn’t much appreciate having the president’s message contradicted and potentially undermined by federal scientists. Whereas in the thinly staffed and often chaotic Trump administration, such a push toward message consistency has really never taken hold.

On the one hand, letting scientists and climate science doubters alike say what they think, and then wrestling with the contradictions, seems in many ways consistent with a nation that celebrates free speech but also pays ample tax dollars to employ an army of government scientific experts.

It’s also potentially consistent with a principle that American scientists themselves have stood up for throughout their many decades of marriage to the federal government: Scientists tell it like it is, and then policymakers get to decide what to do about that.

Thus, the Trump administration can go about withdrawing from the Paris climate agreement, and rolling back the Clean Power Plan (these are policy decisions), or even choosing to defund some climate research — but it can’t tell its scientists what to say or think.

So far, so good — but at the same time, there’s that small but nagging matter of intellectual consistency.

After all, the experts do largely have one voice on climate change. The top analyses all do say it’s real, it’s happening, and moreover, the window is closing fast if you want to do anything about it.