The crumbs that fall from Instapundit’s table

If I summarized Glenn Reynold’s response to my post on his hyping of a small correction to GISS data, you would not believe me, so I’m quoting the whole thing:

Lamberted! But no Instalanche.

Later: In an update: “Matthew Yglesias links to Tim Lambert, obviously deeming him a reputable source. Hey, this is about politics; not accuracy.” Yglesias has been off his game lately.

More: Brad Plumer has been fooled, too.

Yes, Reynolds is enough of an egomaniac to think that I wrote my post because I was hoping to get an Instalanche. In fact, I wrote it to correct his hype. The change meant that 1998 and 1934 went from being in a statistical tie to being in a statistical tie, and no-one had ever reported that NASA had 1998 as the warmest in the US. The JF Beck post he links to doesn’t dispute that, instead Beck repeatedly calls me a liar. But apparently that’s enough for Reynolds — he seems to genuinely believe that Matthew Yglesias and Brad Plumer are discredited just because they linked to my post. But hey, at least he linked to them. Do you think that in Reynold’s imagination Plumer was at first elated to get OMG! an Instalanche, but then Oh No! shattered to find that Reynolds had refuted his post?

I know this is faint praise, but Ace does a much better job than Reynolds of responding to Plumer’s post. Bradrocket tries to clear things up for Ace with pictures of oranges and apples.

I’ll try to help, too. Ace asks:

0.14 degrees is enough to warrant a scare headline, whereas .03 degrees C is completely “meaningless”? Certainly .14 is a larger number than .03. But the former grabs a big headline and the latter is utterly without any import whatsoever?

Really? What exactly is the cut-off point here? I just want to know so that in the future you guys are on record. What’s meaningful? .05? .07? You tell me; let’s get the threshold for significance stated right now.

In the post Ace is responding to, Plumer quoted James Hansen (my emphasis):

In comparing temperatures of years separated by 60 or 70 years the uncertainties in various adjustments (urban warming, station history adjustments, etc.) lead to an uncertainty of at least 0.1ÂºC. Thus it is not possible to declare a record U.S. temperature with confidence until a result is obtained that exceeds the temperature of 1934 by more than 0.1ÂºC.

I hope the emphasising helps.

Comments

#1 dhogaza
August 17, 2007

We’ll be hearing about this for the next 10 years as oil and coal interests around the world continue to attempt to discredit the science.

BB (Before Blip) it was …

“hockey stick … hockey stick … hockey stick …”

AB it will be …

“hockey stick … 1934 … hockey stick … 1934 …”

Adds a little variety to their dishonest schtick.
#2 jre
August 17, 2007

I dunno, Tim.
As I recall, this and this failed to make it through the Fumento cluewall.

But maybe Ace has a lower clue threshold.

Or was it The Ace? I fully support your Thunderdome proposal, if only to sort out the confusion.

“Two Aces go in! One come out!“
#3 Lazar
August 17, 2007

I did like Glenn Reynold’s writings at one point, I would still do if he weren’t so silly on this issue.

I recall about six months ago he staked out his position on anthropogenic cause as agnostic, ‘uncertainties’ were mentioned, a reasonable position if new to the issue… perhaps better instincts, which is probably the average approach of Joe Public, would suggest starting from a position that 90+% of climate scientists and practically every professional body are highly unlikely to be wrong… but it ain’t too bad a starting point.

In the intervening months Glenn seems to have read a lot of Tim Blair, but no much in the way of books or papers. The constant stream of trivialities or potshots, and the occasional issue of some meat, seem always in support of denialists and their rather strange beliefs… not skeptics mind you, but the ‘local warming on a portion of Mar’s South pole for three [Martian] years explains global century-scale trend on another friggin’ planet’ kind of nonsense.

Does agnostic mean not wanting to know?

Now the latest dustup. Newsweek publish an article on the denialist industry and tactics. So he goes to bat for the crooks and cranks dredging up some very old news and claims which have been debunked.

Glenn, Glenn…
He provides a great news service, and some thoughtful views on other issues.
But I hardly read him anymore.
Many conservatives are fed up with this.

I disagree with practically every word you write on politics, Tim. But on science you’re damn interesting.

And science is more interesting than politics, anyway.
#4 Sparrow (in the coal mine)
August 18, 2007

Tim Lambert,

James Hansen has release a rebuttal to the Y2K crap propagating throughout the web. You should spread the news. Check my blog post for links and summary: (Linky)
#5 Penguin
August 18, 2007

I assume that you saw Michael Duffy regurgitating the whole thing in the SMH today. He says “the hottest year on record was mysteriously altered from 1998 to 1934″.
#6 Kristjan Wager
August 18, 2007

The opinion piece Penguin refers to can be found here (I came here to point Tim in it’s direction).
#7 Peter Bickle
August 18, 2007

Hi all

What a crap title, consensus is politics, not science, science is about fact. Hansen screwed up, that is a fact.
If science was about consensus then we would still think we were in a cube sized planet.

Regards
Peter Bickle
#8 bigcitylib
August 18, 2007

Peter,

So you think there is no scientific concensus that the Earth orbits the Sun, and not vice versa?
#9 Michael Poole
August 18, 2007

NOAA-then and NOAA-now would like to disagree with the idea that no one reported 1998 as the warmest year (in the United States) on record.

Recency and bitrot effects on Internet sites give free online search results — as opposed to trawling through back issues or microfiche — questionable coverage of eight-year-old newspaper and magazine articles, to say nothing of television reports.
#10 Tim Lambert
August 18, 2007

Michael, NASA is not the same as NOAA.
#11 Jay Allen
August 18, 2007

Michael, the 1998 link is for *global* temperatures. About the US, the NOAA report says:

“The United States average temperature in 1998 was 54.62°F (12.57°C), which placed the year in a virtual tie with 1934 as the warmest year in records dating to 1895.”

Which is exactly what Tim, et. al. have been saying: it’s been a virtual tie, and it still is a virtual tie.

What’s your point, exactly?
#12 wildlifer
August 18, 2007

But Jay, didn’t you know the US of A *IS* the world?
#13 Chris O'Neill
August 18, 2007

“If science was about consensus then we would still think we were in a cube sized planet.”

Would that be a planet where 73% is the same as zero?
#14 Boris
August 18, 2007

Cube sized?
#15 dhogaza
August 18, 2007

Cube sized?

Apparently that’s smaller than sphere sized.

Or something.
#16 Peter Bickle
August 18, 2007

Hi all

Cube sized, meaning flat earth, if you go too far you fall off the edge, thats what the consensus was for the earth 2-3000 years ago.
Earth orbits the sun is fact, but GW is not based on consensus, the skeptics are being shut up by the advocates who claim there is consensus. Not all climate scientists claim AGW is caused by CO2, that is a fact as well.
Regards
Peter Bickle
#17 dhogaza
August 18, 2007

Cube sized, meaning flat earth

Your knowledge of geometry seems about as thorough as your knowledge of climate science …
#18 Chad
August 18, 2007

“…thats what the consensus was for the earth 2-3000 years ago. ”
That wasn’t based on science. Nice apples to oranges comparison. Usually you guys use the Aristotelean “consensus” model of the universe to cast doubt on the SCIENTIFIC consensus that exists today on AGW.

“…but GW is not based on consensus,…” Correct, it’s based on science. No one claims that since there is a consensus that AGW happening, ergo, AGW is happening.

“Not all climate scientists claim AGW is caused by CO2,…” What percentage will make you happy? I strongly suspect that it might be 100%. Anything less and you will remain skeptical. I think we have a word for your kind: CRANK!
#19 Chad
August 18, 2007

“cube sized planet” LOL. How did I miss that one?
#20 luminous beauty
August 18, 2007

“Cube sized, meaning flat earth, if you go too far you fall off the edge, thats what the consensus was for the earth 2-3000 years ago.”

Actually, Peter, Greek and Indian astronomers in the first millenium BCE had some pretty good measurements for the circumference of the spherical earth and plenty notions of a helio-centric solar system, too.
#21 Ian Gould
August 18, 2007

“GW is not based on consensus, the skeptics are being shut up by the advocates who claim there is consensus. Not all climate scientists claim AGW is caused by CO2, that is a fact as well.”

It’s also fact that there are scientifically qualified individuals who reject relativity; evolution; the geocentric theory and mainstream geology.
#22 Chad
August 19, 2007

Ian, don’t you realize? The scientific consensus on relativity, evolution, etc is all based on dogma and no evidence! The scientific community needs more Galileos like Bickle, et al to challenge bogus theories like AGW, relativity, evolution, etc.
#23 z
August 19, 2007

“Cube sized, meaning flat earth, if you go too far you fall off the edge, thats what the consensus was for the earth 2-3000 years ago. Earth orbits the sun is fact, but GW is not based on consensus, the skeptics are being shut up by the advocates who claim there is consensus. Not all climate scientists claim AGW is caused by CO2, that is a fact as well.”

Ah, the shotgun argument. “There is no consensus, and if there is it doesn’t matter anyway”.
#24 Sparrow (in the coa mine)
August 19, 2007

If science was about consensus then we would still think we were in a cube sized planet.

Regards Peter Bickle

…..Cube sized, meaning flat earth

There no scientific consensus supporting a flat earth. Even more, there never was a scientific consensus supporting a flat earth.

Please read this response to the flat earth argument.

The scientific method hasn’t been around that long.
#25 Marion Delgado
August 20, 2007

Science is, in fact, consensus. Consensus on what the data is, on what it means. There’s an existing body of scientific knowledge, and every bit of it is based on consensus.

The general consensus in a field is rarely wrong per se, though we remember those cases well. Much more commonly it’s insufficiently generalized, or the field is difficult to obtain data for, or experiment, or it takes a long time. The replacement is more often than not a generalization.

Science is also peer review. We know that progress in science has paralleled the rise of peer-reviewed publications.

Science is collecting data without bribery or coercion. The weather station harassment project is not science – it simply wants people busy collecting weather data to kowtow to right-wing extremists, in the hope that they can use propaganda and politics to pry the surface data away from the other data. Even when they fail, as they did with their hockey stick bashing, with absolutely everything in the GGWS, and so on, they’re going to claim success.

What’s obvious is that they are anti-scientists. Even the scientifically trained fragment are like a person studying medicine to be a better serial killer. To them what makes science science is if a consensus of rich CEOs and market fundamentalists say its so.

What they’re doing now is an obvious bad-faith effort very much akin to creationists. Clearly, like the creationists with Intelligent Design, they’ve learned to tell better lies. Pseudo-science evolves, just as most things do.
#26 Dano
August 20, 2007

The key to the consensus issue is that the denialists infer that consensus shapes the next finding, that no one collects data anymore, they just parrot the party line.

Projection , but effective. We must take Marion’s argument one step further and cut off the “science by consensus” marginalization theme.

Best.

D
#27 saurabh
August 20, 2007

I don’t know that I agree with Hansen – 1934 is the current “highest point” (ignoring the tie with 1998). A statistically significant difference between the first- and second-place ranks in a sample is surely not distributed normally, and wouldn’t be significant if its outside of the uncertainty (std dev?) of the sample. I’m not sure what sort of distribution that has, but I’d guess some sort of extreme value distribution, and that in order to achieve significance the difference has to be much more than 0.1C.
#28 Barry
August 24, 2007

Um, saurabh, a non-normal distribution, in the absence of other data, doesn’t let you say that 0.1 deg C is not a valid cut-off for statistical significance.

In addition, you’re confusing the distribution of temperatures with the distribution of measurement errors.