The context of the stolen CRU emails

Earlier I noted the way McIntyre quote mined the stolen CRU emails. But would an honest analysis of the messages have found? Brian Angliss makes the case that it is impossible to understand the emails without consulting with the authors to find out what the original context was.

More like this

Steve Mosher and Steve McIntyre have alleged that the stolen CRU emails prove that Keith Briffa had violated IPCC rules in when working on the 4th Assessment Report. They can't point to any particular IPCC rule and rely on a creative interpretation of an email from Jonathan Overpeck, which Mosher…
The phrase "hide the decline" from the stolen CRU emails has been taken out of context and construed to refer to a decline in temperatures this century when in fact it was a reference to a decline in tree-ring density since 1961. Steve McIntyre knows this, but instead of a correction, he offers…
Deep Climate has been reading the stolen emails that Steve McIntyre didn't mention: Arguing from a cherrypicked selection of quotes from the "Climategate" emails, McIntyre has claimed that IPCC authors Chris Folland and Michael Mann pressured Briffa to submit a reconstruction that would not "dilute…
Peter Hadfield (potholer54) talks on the deceitful quoting of the emails stolen from CRU in 2009 Juliette Jowit in The Guardian puts some more of them in context.

That's a fantastic report by Angliss and the folks at S&R.

So, will Mosher and Fuller finally admit that the major premise for their book is complete and utter BS?

By Former Skeptic (not verified) on 03 Jun 2010 #permalink

So, will Mosher and Fuller finally admit that the major premise for their book is complete and utter BS?

Why would they? What's their incentive?

But [what] would an honest analysis of the messages have found?

By truth machine (not verified) on 03 Jun 2010 #permalink

On this basis we would never be able to understand anything about history because we couldn't consult the original people and understand the context!

Absolute rubbish!

By Dave Andrews (not verified) on 03 Jun 2010 #permalink

>*On this basis we would never be able to understand anything about history*

Reductio ad absurdum

Trying reading the link Dave.

Tom Fuller responded:

"We did provide context in our book, drawing from a variety of sources, including Climategate emails. "

...except of course the scientists involved or a complete look at the relevant studies.

Fuller/Mosher started out with a narrative, then fixed the facts to fit that narrative. Lawyers might approve, but that isn't journalism. Fuller/Mosher did not care to examine the relevant published studies (i.e. on the divergence problem) which puts "hide the decline" in context, or even seek input from the scientists involved who would be most familiar with the discussions. Doing so would not help their narrative.

There is another level of missing context that's detailed by DeepClimate's blog. Sometimes relevant portions of the email text have been deliberately removed. McIntyre and others have engaged in this behavior.

Lastly, I don't think interviews with those involved would paint a 100% clear picture, but it certainly gets us closer. If the topics of discussion are simple, then that would get us closer. Some of the topics also require some level of scientific competency to understand and an additional level of understanding of climate science.

LOL! Angliss is taking Fuller to town - and then some - in the comments over at S&R.

I'm sure I'm not the only one enjoying how Fuller is squirming over there.

Anyone else prepared for the inevitable goalpost moving from Fullersh*t?

By Former Skeptic (not verified) on 03 Jun 2010 #permalink

I'm just looking forward to Fuller explaining his 'Men of Honour don't publish emails without permission' statement on this blog and how it squares with publishing an entire bloody book based on emails that were published without permission.

I'm sure it'll be entertaining, at least. And I'm equally as sure that Fuller will continue not to understand how incredibly unprincipled he is.

"Climategate", inasmuch as it ever existed, is dead. Until the police trace the hackers, that is.

By Tony Sidaway (not verified) on 03 Jun 2010 #permalink

Bud:

> I'm just looking forward to Fuller explaining his 'Men of Honour don't publish emails without permission' statement on this blog and how it squares with publishing an entire bloody book based on emails that were published without permission.

Easy: Fuller never said that he's a man of honour. Stop putting words in his mouth!

Near as I can tell (some TL;DR effect here, holy cow that is a lot of words), Angliss doesn't even address the fact that the "decline" referred to in "hide the decline" is not a decline in temperatures.

By Douglas McClean (not verified) on 03 Jun 2010 #permalink

Reductio ad absurdum

Trying reading the link Dave.

Sorry, but Dave is actually right about this, at least as far as "the case that it is impossible to understand the emails without consulting with the authors to find out what the original context was"; there are all sorts of other context by which one can understand a written record. In the comments, Martin Vermeer refers to this when he writes "âplenty contextâ includes the scientific literature, and knowing first hand how research is done".

holy cow that is a lot of words

It seems to have been authored by the Department of Redundancy Department.

Angliss doesn't even address the fact that the "decline" referred to in "hide the decline" is not a decline in temperatures

Indeed. This article goes to great lengths to create a false equivalence between those who make charges against climate scientists and those who deny them, by implying that no one can conclude anything without a formal inquiry, paying no attention to the wealth of information readily available.

By truth machine, OM (not verified) on 03 Jun 2010 #permalink

"S&R has concluded that the emails do not themselves contain sufficient context to understand what really happened in climate science over the last 13 years.

Many people have claimed that the emails contain all the context needed ...."

Well, duh. Did they actually name names of "many people" there? I actually would be surprised if anyone had truly been that stupid, and am guessing rhetorical excess in that last bit.

In other news, climate models are not perfect, and statistics requires doing math.

Did they actually name names of "many people" there?

Fuller stated "Some of the emails are so egregious as to need no context."

guessing rhetorical excess in that last bit

Yes, but the thesis of the article rests on it.

By truth machine, OM (not verified) on 03 Jun 2010 #permalink

TruthMachine - I disagree that I created false equivalences, but given I'm the author, that's not a surprise. I went out of my way to avoid pointing out the "wealth of information readily available" because it's a double-edged sword. As I pointed out in the comments at S&R, I was writing for a smart but genuinely skeptical technical professional audience, and by NOT relying on the existence of the other information, I believe I made it harder for my target audience to ignore my points. And yes, I used some amount of repetition to drive a few key points home. I tried to keep unnecessary redundancy out of it, however, although perhaps not as successfully as I'd have liked.

Hank - yep, I do name some names, like McIntyre, Mosher, and Geoff Sherrington.

Thank you for noticing the correction, however. When I get something wrong or a judgment call turns out to be wrong, I do try to correct the record as soon and as completely as possible.

Douglas - I wasn't trying to deal with specific issues in this post. That will come later. And it was just shy of 4,000 words when it was all done. I thought about splitting it up (and shortened it by at least 1,000 words from what I had originally planned), but decided that it was best presented in a single, albeit long, post.

I disagree that I created false equivalences

You created the ones you've since struck out (good on that).

By truth machine, OM (not verified) on 03 Jun 2010 #permalink

I went out of my way to avoid pointing out the "wealth of information readily available"

Then you're building a false "case that it is impossible to understand the emails without consulting with the authors to find out what the original context was".

because it's a double-edged sword

Perhaps we have different notions of what "information" is. Let me offer my favorite quote, with emphasis added:

"The desire to be right and the desire to have been right are two desires, and the sooner we separate them the better off we are. The desire to be right is the thirst for truth. On all accounts, both practical and theoretical, there is nothing but good to be said for it. The desire to have been right, on the other hand, is the pride that goeth before a fall. It stands in the way of our seeing we were wrong, and thus blocks the progress of our knowledge." W.V. Quine and J. S. Ullian, The Web of Belief

By truth machine, OM (not verified) on 03 Jun 2010 #permalink

Can someone point me to a defense of the graph itself?

I suppose it's fair enough to argue that it's better to talk to the people involved to get more context, but it's doesn't seem reasonable to just ignore the fact that the stakes are high and everything is public, so people have a large incentive to offer exculpatory context if it's available.

And they actually have been, except that the extent of it is that the "decline" doesn't refer to actual temperatures and that people generally think that the tree-ring-based reconstruction only became unreliable recently (but nobody seems sure about why.)

(in the interest of full disclosure, I last searched for some kind explanation a couple of months ago and I now see that there's been lots of new discussion)

Then you're building a false "case that it is impossible to understand the emails without consulting with the authors to find out what the original context was".

Except that he didn't say that, Tim did (and he was merely paraphrasing Osborn). Brian Angliss likewise was quoting Osborn and that only as being "closer to correct" than Sherrington.

Dave Andrews:

> On this basis we would never be able to understand anything about history because we couldn't consult the original people and understand the context! Absolute rubbish!

Hmm. Never talked to any historians, I take it? Context is everything. Sadly, can't always talk to the people in question, but there are plenty of other ways to get context. No historian would take historical documents on face value; it's the first thing you're taught. They make you practice a lot! And what would you think of an historian who took one sentence of a 19th century document and declared that it proved Darwin was a secret Satanist?

Oh, actually, of course - someone's already done a much better example: Newtongate.

The most remarkable thing about this is the number of hats that McIntyre can don. One minute he is lead blog climate science investigator , then voila , another hat and he's a investigator into goings on at the CRU. Then judge ,jury and executioner.

By Bill O'Slatter (not verified) on 03 Jun 2010 #permalink

#25 Bill O'Slatter

The most remarkable thing about this is the number of hats that McIntyre can don. One minute he is lead blog climate science investigator , then voila , another hat and he's a investigator into goings on at the CRU. Then judge ,jury and executioner.

You omitted McIntyre's most important position: quote-miner in-chief!

By ScaredAmoeba (not verified) on 04 Jun 2010 #permalink

And they actually have been, except that the extent of it is that the "decline" doesn't refer to actual temperatures and that people generally think that the tree-ring-based reconstruction only became unreliable recently (but nobody seems sure about why.)

I'm not entirely sure what your question is, but the post-1960 tree ring data is strongly contradicted by the surface record, the satellite record, and every other available proxy. When you have multiple independent methods of measurement that all give you the same answer, and one method that gives you a completely different answer, you can safely conclude that it's the odd man out that's wrong.

Now, maybe that means that tree ring data shouldn't be used at all. If so, it wouldn't change anything. But McIntyre keeps alleging foul play because they cut-off the portion of the tree ring data that they knew was wrong. That's just stupid.

I'm not entirely sure what your question is

The question is this. There were three options:
1) Not include the tree ring-based reconstruction at all since it's unreliable as evidenced by its disagreement with the instrumental record
2) Include the entire tree ring-based data series, thereby drawing the reader's attention to the fact that it's unreliable
3) Truncate the tree ring-based data series, as was done, or smooth it, or something like that

So the question is, what's the defense of picking option 3?

So the question is, what's the defense of picking option 3?

They evidently have reasons to believe that it's reliable before 1960 (or whenever the cut-off was) but unreliable after. That's a judgment call that I'm not qualified to make, but to the extent that they have good reasons to believe it, it's perfectly defensible.

Dan Olner,

I would not disagree with you. I was talking about Angliss' approach, as reported by Tim, that it was impossible to understand the emails without consulting the authors so as to appreciate the context.

By Dave Andrews (not verified) on 04 Jun 2010 #permalink

They evidently have reasons to believe that it's reliable before 1960 (or whenever the cut-off was) but unreliable after. That's a judgment call that I'm not qualified to make, but to the extent that they have good reasons to believe it, it's perfectly defensible.

The evidence is that they match well with the proxies they were calibrated against (none, however, span the entire time series, which is one reason they want to use the tree proxies, they're able to build a continuous 2,000 year dataset).

They also match well with about the first 2/3 of the 1880-present instrumental record.

Then they go "blooie".

So the belief - unproven at this time - is that the fact they went "blooie" after correlating well with other available data for earlier times, means that something changed that caused them to go "blooie".

The "trees aren't thermometers" claims are clearly a stretch, because the proxies in question *do* match other data for all but the data for recent decades.

WF,

20th century divergence is not evident in all boreal tree-ring series. Nonetheless, those that do show divergence since 1960, show little sign of divergence prior to 1960, neither to the instrumental record, non-divergent tree-ring series nor other regionally coherent proxies such as annular lake bed sediments going back in some cases thousands of years. This is apparent even in McIntyre's [comparison](http://climateaudit.files.wordpress.com/2009/09/rcs_merged_rev.gif) of Briffa's Yamal series and the nearby divergent Khadyta River series. If divergence was a problem in the paleo record, one would think there would be empirical evidence of it.

By luminous beauty (not verified) on 04 Jun 2010 #permalink

Brian,
By not "dealing with" the "specific issue" of what decline we are talking about, you lead your "smart but genuinely skeptical technical professional audience" down the garden path to the inference that it is a decline in temperature being discussed.

By Douglas McClean (not verified) on 04 Jun 2010 #permalink

WF,

20th century divergence is not evident in all boreal tree-ring series. Nonetheless, those that do show divergence since 1960, show little sign of divergence prior to 1960, neither to the instrumental record, non-divergent tree-ring series nor other regionally coherent proxies such as annular lake bed sediments going back in some cases thousands of years. This is apparent even in McIntyre's [comparison](http://delayedoscillator.files.wordpress.com/2009/10/compare.png?w=490&…) of Briffa's Yamal series and the nearby divergent Khadyta River series. If divergence was a problem in the paleo record, one would think there would be empirical evidence of it.

By luminous beauty (not verified) on 04 Jun 2010 #permalink

There is also good (but not conclusive) evidence that the divergence problem is caused by anthropogenic factors, which is further evidence that the tree ring proxies that have the divergence problem (not all of them do) are reliable prior to 1960.

http://www.skepticalscience.com/Hockey-stick-divergence-problem.html

Of course, it's much easier to proclaim that "hide the decline" from an informal email a decade ago is something evil and nefarious, and omitting the full academic context of the issues is very easy to do among an unsuspecting lay audience. It's why I think that talking to the authors is important, but perhaps not absolutely necessary. An astute and careful observer can find the relevant context in the scientific research that has been done.

luminous beauty, can you point me to a reference that analyzes the divergence between the different tree ring datasets? That could justify the truncation.

The points others made (i.e., just that the series matched until they didn't) aren't convincing arguments for option 3. If all we know is that the series match until they don't, that's what the graph ought to show.

Except that he didn't say that, Tim did (and he was merely paraphrasing Osborn).

Do you have a reading comprehension problem? I never said that Angliss said that. But his whole article is about building such a case -- Aranda and Venolia's 4th level of analysis is âdirect accounts of the history by its participantsâ.

Hmm. Never talked to any historians, I take it? Context is everything. Sadly, can't always talk to the people in question, but there are plenty of other ways to get context.

Can't anyone here read? Andrews (it makes my skin crawl to defend him) only objected to the notion that you have to "always talk to the people in question".

By truth machine (not verified) on 04 Jun 2010 #permalink

Brian, By not "dealing with" the "specific issue" of what decline we are talking about, you lead your "smart but genuinely skeptical technical professional audience" down the garden path to the inference that it is a decline in temperature being discussed.

Quite so. It's understandable that people here want to like it because it puts the lie to Fuller, Mosher, et. al., but the article is really pretty crappy.

By truth machine (not verified) on 04 Jun 2010 #permalink

The points others made (i.e., just that the series matched until they didn't) aren't convincing arguments for option 3. If all we know is that the series match until they don't, that's what the graph ought to show.

Only if the graph is about the reliability of tree ring proxies rather than about temperature.

By truth machine (not verified) on 04 Jun 2010 #permalink

Only if the graph is about the reliability of tree ring proxies rather than about temperature.

This makes no sense. The tree rings are proxies for temperature. How reliable the proxies are is very much relevant to a graph "about" temperature.

Actually TM makes perfect sense, within the context.

If the tree ring proxies matched other proxies in past times, and then again matched those proxies and the instrumental record for about a century - which they do - then it can be reasonably inferred that another factor has come into play when they cease to match in recent times.

It's not as if there's some big mystery and we're in the dark about massive post-war global industrialisation leading for example to atmospheric sulphur precipitating out by the mechanism of acid rain which is known to be harmful and eventually fatal to contemporary trees. Doesn't provide some sort of clue (other than to the wilfully clueless), as just one example of anthropogenic causes?

What that factor (or factors) might be in actuality is under discussion by real scientists (and certainly not McinTyres and his support machine) in the literature.
But in terms of establishing a continuous temperature record it's irrelevant.

In any case tree ring proxy-free reconstructions show a notably similar trend, which automatically leads a sensible person to deduce that protests from the denialist corner are nothing but noise and distraction from the main issue which is AGW is real and happening.

Which is pretty much the template that denialists operate under.

I didn't see MarkB's latest comment before posting mine right below. This paper seems good for what we're discussing. It doesn't to me seem to be enough justification for truncating.

Doesn't provide some sort of clue (other than to the wilfully clueless), as just one example of anthropogenic causes?

"Some sort of clue" is not justification enough for truncating the data, to my mind

But in terms of establishing a continuous temperature record it's irrelevant.

Of course it's relevant. If the factor is determined and it's determined it didn't exist prior to 1960, then we can legitimately truncate the series

In any case tree ring proxy-free reconstructions show a notably similar trend

But that's completely irrelevant to whether it's okay to truncate the series.

@32 Dave writes:

I was talking about Angliss' approach, as reported by Tim, that it was impossible to understand the emails without consulting the authors so as to appreciate the context.

I don't see the problem. Sceptics have demonstrated in spades that they haven't the faintest idea what the context of many of the emails was, and were drawing whatever completely erroneous conclusions they felt like at the time. The "trick to hide the decline" has perhaps become the most famous example, being interpreted by sceptics basically as smoking-gun evidence for a worldwide climate conspiracy when it's nothing of the sort.

I might agree that it's not necessarily impossible to understand the context, particularly if you knew about the divergence issue at the time the emails were hacked. However sceptics have certainly demonstrated no ability to understand, or even seek out the context, that's for sure. Not even after it was pointed out to them.

Many sceptics seem to be quite content in their context-proof bubble.

(I feel like I'm in an argument I don't want to be in. It seems like there is a bad argument for truncation (the correlation was good until it wasn't) and potentially some good ones, and someone ought to make all the good ones in one place)

TM writes:

>Andrews (it makes my skin crawl to defend him) only objected to the notion that you have to "always talk to the people in question".

Dave Andrews comment which I critiqued:

>*On this basis we would never be able to understand anything about history because we couldn't consult the original people and understand the context!*

Which was reduction to the absurd. Brian's article makes many interesting points, including describing a spectrum with the above at one extreme. Some points valuable some highly contestable.

But rather than accepting Dave Andrews reductio ad absurdism, I'd encourage a more nuanced critique (such as TM and others use, requiring a more specific critique). BTW Dave wants discussion on the absurd end.

I thought Angliss's bringing in the four levels of analysis made that aspect easer to contemplate and talk about.

I was also interested to see where Angliss found Mosher and Fuller making contradictions.

...and potentially some good ones...

I don't know about you, but I find "don't include the stuff that's obviously wrong" to be a strangely compelling argument. Maybe I'm weird.

pough: note that all the proxy series post 1960 are wrong* to the extent that they don't exactly coincide with the instrumental record, but they're still included in the graph

*Or at least they're more wrong than the instrumental record

re: WF
"(I feel like I'm in an argument I don't want to be in. It seems like there is a bad argument for truncation (the correlation was good until it wasn't) and potentially some good ones, and someone ought to make all the good ones in one place)"

1) The Skeptical Science page is a good start, and ti references the key D'Arrigo, et al (2007) paper.

2) Then, I'd add Mann, et al(2008) PNAS article, in which they reconstruct the last 2,000 years, using a lot more proxies, with or without tree-rings. They get pretty good results back 1300 years without tree-rings, need the tree-rings to get further back. They still think the MWP wasn't as warm as now, and the trees correlate pretty well with the other proxies until the divergence in that subset.

3) Liebig's Law is something (some, at least) farm kids learn by the time they are 10, i.e., growth is limited by the item least available. For temperature studies, people use trees in stressed areas (nearer the poles), rather than ones nearer the Equator.
Many factors are either constant, or evened out by random jiggles, allowing researchers to tease out temperature signals. Serious books get written about this, so nobody is going to boil it down in a few sentences.

4) Scientists being scientists, they aren't going to come out and say "the divergence is anthropogenic" until they are really sure which (combination of) the anthropogenic changes is affecting that subset of the Northern trees.
Given CO2 far outside historical range, polar amplification/heat-stress, possible rainfall issues, and global dimming ... there are plenty of possibilities.

But I doubt you'd knowledgeable people who think the divergence is likely to be natural. [Of course, climate anti-science folks claim everything is natural, including this...]

By John Mashey (not verified) on 04 Jun 2010 #permalink

note that all the proxy series post 1960

All?

but they're still included in the graph

Which graph? So now the complaint is that the proxy series shouldn't have been removed and weren't?

Or at least they're more wrong than the instrumental record

And now thermometers don't work? You seem to be playing a single note: doubt. Doubt doubt doubt doubt doubt doubt doubt doubt doubt doubt. It's getting monotonous.

Which graph? So now the complaint is that the proxy series shouldn't have been removed and weren't?

If you don't know which graph, I can't see how the rest of the conversation makes sense to you at all

John Mashey: thanks.

It seems to me, though, that there still needs to be an explicit justification for the truncation. Your 4, to my mind, argues against ti.

It seems to me, though, that there still needs to be an explicit justification for the truncation. Your 4, to my mind, argues against ti.

The graph was supposed to be an exposition of the science, their "best, informed view".

Including data known to be wrong doesn't really make sense in that context.

As has been pointed out numerous times, the literature was and is full of discussion of the "divergence problem", the data has been included in published papers, etc etc.

"Hide the decline" presumes there's a temperature decline to be hidden. The instrumental record shows that there is not.

WF,

You asked for "[a reference ](http://scienceblogs.com/deltoid/2010/06/the_context_of_the_stolen_cru.p…) that analyzes the divergence between the different tree ring datasets? That could justify the truncation." You apparently didn't catch my [links](http://scienceblogs.com/deltoid/2010/06/the_context_of_the_stolen_cru.p…).

Try [this](http://delayedoscillator.wordpress.com/2009/10/05/yamal-emulation-i/) for starters. Follow the links.

>It seems to me, though, that there still needs to be an explicit justification for the truncation. Your 4, to my mind, argues against ti[sic].

Explicit justifications have been made. It seems that in your mind, hand waving subjective suppositions are a meaningful substitute for reasoned argument.

Please make a reasoned argument for _why_ any of the proffered explicit justifications, but particularly John Mashey's #4, presented are, in your mind, unsatisfactory.

If you can.

By luminous beauty (not verified) on 05 Jun 2010 #permalink

Shorter WF:

I didn't clarify which graph I was referring to. Therefore, I was right.

Oh give me a break. This isn't even that much shorter than what I said

If you don't know which graph, I can't see how the rest of the conversation makes sense to you at all

Right... I was asking which graph because what you were describing didn't seem to match with the graph that was originally under discussion. I wouldn't mind seeing something that shows "all the proxy series post 1960 are wrong to the extent that they don't exactly coincide with the instrumental record". The graph we've been looking at doesn't seem to have the resolution to conclude that. From what I'm able to see, it looks like all the others do follow the instrumental record pretty well, but Briffa's zigs where everything else zags.

Mind you, I'm not expecting you to back up what you're claiming. I haven't really seen you respond to much directly.

dhogaza,

As has been pointed out numerous times, the literature was and is full of discussion of the "divergence problem", the data has been included in published papers, etc etc.

"Hide the decline" presumes there's a temperature decline to be hidden. The instrumental record shows that there is not.

There no dispute here

The graph was supposed to be an exposition of the science, their "best, informed view".

Including data known to be wrong doesn't really make sense in that context.

Again, all proxy data conflicts to some extent with the instrumental record, and still was included, because it did make sense: the "best, informed view" includes the correct view of the accuracy and precision of the proxies. That was the point of showing that they mostly agree with each other and with the instrumental record.

luminous,

Thanks for the links. I'm still reading them. Again, I wish someone had synthesized all of this into one document, preferrably without pough's and frank's input.

That said, I can answer your question about John Mashey's #4 right away.

If there's certainty about some new factor's influencing the rings and it's known that this factor wasn't there before, then it's appropriate to truncate the series. Otherwise, there's a possibility that the tree ring proxy isn't as good as it looks, and so the entire series should be used in order for that to be reflected in the graph

pough, the key word in "they don't exactly coincide with the instrumental record" is "exactly."

(i.e., they do of course track pretty well, but they don't coincide. That's obvious just from thinking about it, but you can see it in the graph at Scholars & Rogues)

Since there seems to have been genuine confusion, I apologize for my snipe at pough in the :48PM post. Frank's input still doesn't look good.

>If there's certainty about some new factor's influencing the rings and it's known that this factor wasn't there before, then it's appropriate to truncate the series.

Yes, yes and yes. Try again.

By luminous beauty (not verified) on 05 Jun 2010 #permalink

If there's certainty

Yes, yes and yes.

Scientists being scientists, they aren't going to come out and say "the divergence is anthropogenic"

WF,

One doesn't need to know the cause of the divergence to know it wasn't there before. __Certainly.__ Knowing it wasn't there before is sufficient evidence to conclude that whatever the cause, it is new. __Certainly.__

By luminous beauty (not verified) on 05 Jun 2010 #permalink

WF:

0) have you yet read the D'Arrigo et al article?

1) Science is filled with correlations or even simple formulae that work perfectly well across large domains ... and then breakdown at the edge of the domain, i.e., "threshold effects."

Example: Newton vs Relativity. Newton is fine for most things, but not for GPS satellites.

Example: transition from laminar flow to turbulent flow.

Example: plant growth, i.e., one may get a linear improvement in growth from increasing some factor ... up to the point where it is no longer a limiter, and then more of it is useless. Higher CO2 helps some plants grow better in some conditions, but no amount of extra CO2 will grow corn in the Sahara.

2) There is a huge difference between:
a) A long-established correlation of a subset of trees and temperature has broken down, we are mystified, we are clueless, this is amazing and shouldn't happen, and this means we know nothing about the usefulness of these tree-rings, ever...

AND
b) (D'Arrigo) ~ The relationship started breaking down circa 1960, we know a set of covarying anthropogenic effects that have by now changed Earth's atmosphere and temperature/atmosphere to be outside the usual parameters for the last 2,000 years. Because they are covarying it is nontrivial to disentangle the *relative* contributions of different effects, so more research is needed to bound teh uncertainties better. Meanwhile, here are a 100 citations showing people working on the problem.

I.e. this is the difference between a) "we know of no plausible reason" and b) "we know of several plausible reasons, and we don't yet know the relative weights."

For some reason, some people *want* to believe the scientists are in state a) on this. They aren't.

I offer an analogy for a):

No medical researcher in the world can predict whether a *specific* 16-year-old smoker:
a) Will eventually die of lung cancer.
b) Will eventually die of heart disease.
c) Or not.
This being the case, researchers must know nothing, and in particular, the existence of c) implies smoking might be safe, so encourage kids to smoke, many want to, and it's good for the economy.

By John Mashey (not verified) on 05 Jun 2010 #permalink

0) More or less, but not in its entirety

Anyway, I seem to again have been drawn into arguing that the truncation was definitely improper, which is not something I even believe. In the spirit of not continuing with this, I'm not going to argue about the analogy, which I don't think is good, and am instead going to be reading the papers or doing other work or something.

Thanks, John, luminous, and MarkB for the links etc.

Which was reduction to the absurd.

The reduction to the absurd is "it is impossible to understand the emails without consulting with the authors to find out what the original context was" -- that's simply ridiculous, and Andrews' comment captured what's ridiculous about it.

By truth machine (not verified) on 05 Jun 2010 #permalink

This makes no sense. The tree rings are proxies for temperature. How reliable the proxies are is very much relevant to a graph "about" temperature.

Of course it is, but we have science concerning the reliability of those proxies, science that justifies the use of those proxies in the graph. The graph is not about the reliability of the proxies, it makes use of the proxies within the scientific perimeters of their reliability to show what the graph is about -- temperature. We do this all the time in science -- we build one piece of science upon another; we do not have to say everything about the underlying science when conversing about the higher level. I'm sorry if that notion of levels of context doesn't make sense to you.

By truth machine (not verified) on 05 Jun 2010 #permalink

Once again I make the point that this entire argument would be moot if certain sceptics had not actually consciously decided they were going to leap to conclusions on what the "trick to hide the decline" actually meant.

Sure, you don't always have to ask someone what their email context was to understand it correctly, but it sure as hell would've helped in this case.

Do sceptics show any interest whatsoever in actually clarifying such things? Nope. Not then, not now, and probably not at any time in the forseeable future.

TM, why do you think are the proxies graphed where the instrumental record is available?

> Again, I wish someone had synthesized all of this into one document, preferrably without pough's and frank's input.

Shorter WF:

I'm not reading the scientific papers, but it's other people's fault for not summarizing them for me!

> Anyway, I seem to again have been drawn into arguing that the truncation was definitely improper, which is not something I even believe. In the spirit of not continuing with this, I'm not going to argue about the analogy, which I don't think is good, and am instead going to be reading the papers or doing other work or something.

Shorter WF:

I'm arguing that there was something wrong with "hide the decline"! Even though I don't actually want to argue it! I promised to read the scientific papers on the subject! I've promised this before, but this time I'll keep my promise! I promise!

> Thanks, John, luminous, and MarkB for the links etc.

Shorter WF:

I promise to read your links. But I won't read them.

> TM, why do you think are the proxies graphed where the instrumental record is available?

Shorter WF:

Here's a dumb question. I still promise to read the papers! I promise!

Frank, how many of the papers have you read?

(the reason I'm asking is anyone who's ever read any scientific papers knows it's hard work, in my experience) And anyway, what's with the abuse? Dumb questions are the easiest to answer, so out with it

TM, why do you think are the proxies graphed where the instrumental record is available?

One point at a time. WF, why did you say that something that makes plenty of sense makes no sense?

By truth machine (not verified) on 07 Jun 2010 #permalink

Shorter WF:

I assure you I'm reading the scientific papers! Even though I'm not asking any specific questions about any specific issues in the papers! Why do you hate me so much?

* * *

> Frank, how many of the papers have you read?

Mann 2008. Douglass et al. 2007. I also wrote a blog post a while back about Zhen-Shan and Xian 2007 which inactivists were using as 'proof' of a coming 'global cooling' (and while I was at it, I also skimmed through Huang et al. 1998, the paper describing Empirical Mode Decomposition).

Generally, the way I tell whether someone's read a scientific paper is to see if he can ask specific questions on the paper rather than spewing general bullcrap. It doesn't take a genius to figure out that you're in the latter group.

And my general way of seeing if someone's read and understood a scientific paper is if they're flinging feces or asking intelligent questions/answering "dumb" questions/doing something other than flinging feces.

And yet here we are, me having read .75 of a relevant paper and you flinging feces.

"it is impossible to understand the emails without consulting with the authors to find out what the original context was."

If you can not understand what "hide the decline" means "without consulting with the authors"; you might as well hand back your Phd.

By Devils Advocate (not verified) on 09 Jun 2010 #permalink

What does it mean, Devil's Advocate?

The decline you're being sold and have bought into Devil's Ad. is the one in the mean intelligence quotient of some of the general population.

"Ain't no other context but the first one you was told, yee haw" as they might say to the gulible.

> And my general way of seeing if someone's read and understood a scientific paper is if they're flinging feces or asking intelligent questions/answering "dumb" questions/doing something other than flinging feces.

> And yet here we are, me having read .75 of a relevant paper and you flinging feces.

Longer WF:

OK frank, so you've actually read and commented on a scientific paper. But that won't fit my talking point! So I'm ignoring it!

You ask me to show that I've understood the scientific papers I'm reading! Well, to show how much I've understood, I'll tell you that I've read .75 of a relevant paper! Look, .75! Point -- seven -- five! That sounds so scientific! Never mind what the name of the paper is, now that I've come up with an impressive sounding number you know how scientific I am!

Therefore, "hide the decline" is evidence of a Warmist Conspiracy, but I never said that.

Wow. Fuller and Mosher continue to dig themselves into a bigger hole over at S&R. I need more popcorn at this rate.

Fuller's comment (#19 over there) is hilarious given how his previous attempts at "honest discussion" panned out here and at Tobis's. What a shame.

By Former Skeptic (not verified) on 09 Jun 2010 #permalink

Oops, #82 and #85 beat me to it. Read it and then STFU, WF.

By truth machine (not verified) on 10 Jun 2010 #permalink

If you can not understand what "hide the decline" means "without consulting with the authors"; you might as well hand back your Phd.

You're right; one need not consult with the authors to know that it refers to [the divergence problem](http://en.wikipedia.org/wiki/Divergence_problem). That's what you meant, right?

By truth machine (not verified) on 10 Jun 2010 #permalink