Ten Replies to Ten Replies about Advocacy vs Science

Jeff Id, the author of the skeptical (with respect to anthropogenic global warming) blog The Air Vent, has posted several replies to Gavin Schmidt’s “Advocacy vs. Science” post over at RealClimate. Thus, in the name of a good ol’ fashioned blog-o-sphere tumult, I’d like to reply Jeff’s responses to Gavin’s post (which is in response to posts at WattsUpWithThat and Chris Colose’s blog.) We’ll do this block-quote style (all quotes are from Jeff I’ds post above unless otherwise noted):

1. A scientist accepts tough questions, an advocate stifles them.

Gavin Schmidt and the folks at RC are hardly the type of people which refuse “tough questions.” The entire premise of RC is to answer precisely those questions that are commonly flung by skeptics. They published a two-part FAQ on climate models in the past few months, and their entire site is a repertoires of answers to practically every “tough question” that are asked (do I really need to link the posts about G&T, or Tsonis, or anything else?). Of course, the “toughest” questions the authors face don’t actually come from skeptics – they come in peer-reviewed journals. The authors at RC are actively publishing researchers who themselves are asking the “tough questions” necessary for furthering our understanding of the climate system. The allegation implict in this point is that RC moderates comments to remove some of the noise from their discussion threads. Gavin’s final comment here sums the sentiment up nicely.

2. A scientist is interested in the quality of his data source, an advocate adjusts and processes the data until it fits his conclusion.

It’s a bit hazy what exactly is meant to be implied here. If we’re talking about GISSTEMP, it’s a null point; biases are constantly being found, analyzed, and corrected in the temperature record (and other scientific records). It would be bad science to ignore obvious errors. On the other hand, where do we find people constantly obfuscating the true nature of the data sets they’re utilizing by cherry picking beginning and endpoints? The year 1998 and a plethora of graphs come to mind…

3. A scientist using an instrument for measurement doesn’t assume the measurement is correct simply because of producing the intended result. An advocate will accept a tree ring proxy as a thermometer based on simple correlation analysis.

Um, what? For starters, there’s a large literature on dendrochronology (a literature I’m positive Jeff knows about because of his collaboration with the ClimateAudit project) which goes far beyond mere “simple correlation analysis.” This is beside the point though; my response to (2) is just as appropriate here. Climatologists and meteorologists are very aware of biases within our instrumentation and we do everything we can to eliminate them.  This RealClimate post comes to mind.

4. A scientist models data based on measured data, an advocate adjusts the numbers until the final result matches the assumption. How is moisture feedback measured for computer models?

Actually, a scientist models based on theory. Theory is translated into mathematics which can then be translated into a model. Once the model is built, we spend 99% of the rest of the time analyzing how it matches up to real, measured data. Sometimes models capture certain behaviors better than others; based on the large amount of analysis done with the model, it might be tweaked to better capture those anomalous behaviors. Some tweaks are better than others, which is why there is a constant development process to create new formulations of the basic theories of atmospheric chemistry, physics, and dynamics which help us better capture the h0listic behavior of the atmospheric system. 

Moisture feedback isn’t measured for computer models. Feedbacks are part of emergent behavior resulting from the basic chemistry, physics, and dynamics which make the foundation of the model. For instance, some models which will be used to performs runs for the IPCC 5AR will utilize cutting-edge cloud microphysics. This single addition could have dramatic impacts on the model runs – particularly with respect to climate sensitivy and precipitation changes. (Curiously, I don’t know of anyone expecting it to drastically change our estimates of the equilibrium climate sensitivity… I wonder, why is that)


5. A scientist accepts that 50 people in agreement have often been wrong, an advocate declares consensus and stifles dissent.


My 7th grade science teacher had a saying which has proven true much more often than not: “If you’re doing something one way and everyone else is doing it another, you’re probably just plain wrong.” What does “consensus” mean? It means that the people that actually do climate science generally reach the same conclusions and interpretations of the data they work with. In climate science, there is a strong consensus that AGW is the best interpretation of the pertinent data. What this response fails to hit almost entirely is that it’s not the consenus that matters; it’s how that consensus came to be. In the case of climate science, AGW continues to be supported by new lines of evidence. It allows us to answer questions, but also to ask more questions. The important thing is that ain’t nobody done gone and knocked AGW’s knee-caps out! There is a consensus precisely because not only is AGW the most robust explanation of the data out there, but no one has come and falsified it yet.

6. A scientist is concerned about politics in the peer review process, an advocate accepts it silently as long as it supports their project and conclusion.

This is hardly the issue that skeptics make it out to be. You know, when a skeptics scientific thesis can thoroughly be debunked and refuted in a short, few-paragraph blog post, is it any wonder that it can’t pass the muster of peer-review? Peer-review is far from a perfect, flawless process, and it certanly can insulate the scientific establishment from major changes – even changes which prove to be necessary. But the issue isn’t politics in the peer-review process; it’s flawed science wasting peer-reviewers time. And again, this notion of “supports their project and conclusion” is very, very easily turned around on the majority of the skeptic bloggers out there.

7. A scientist reveals their methods and calculatoins to the world for reproduction, an advocate hides (censors) to prevent errors being noticed. 

This is why most scientific paper shave sections on methodology. Now, I’m a huge open-source promoter (it kills me that my research has me manipulating IDL scripts rather than starting from the ground up with Python and R), and I’ve found climate science to be particularly transparent when it comes to code. Most of the major GCM’s out there have their source code published (GISS Model E, NCAR CCSM for examples). NCEP Reanalysis is readily available; initialization files and output data used for IPCC 4AR are all readily available. There is also a huge body of literature explaining how climate models are produced and changing. Jeff is obviously refering to the ongoing ClimateAudit scuffle with scientists like Steig, but it’s still a ridiculous notion. We would all probably agree that transparent code is the best goal to strive for (everything from accessibility to ease of understanding – comments, proper format, consistent style, etc), but science is still about independence in some ways. If you’re going to plow through a paper and verify its results, you should be willing to do more than just copy someone’s scripts and see if they get you the same end data. Science is a process; code is a product.

8. A scientist admits errors and goes back to work, an advocate hides them by concealing data and reproducing similar results with equally flawed methods. Who’s the denier about the latest hockey stick Michael Mann, Gavin Schmidt, Tamino or me.

Is that a rhetorical question? This meme about concealing data really gets tiresome and I don’t feel like dealing with it for the gazillionth time in this post. If someone has a serious issue with the latest Mann paper (or any paper in the climate science literature), then it’s about time that they draft a formal refutation and publish it. I can guarantee that if the refutation is legitimate, it won’t have any problem making it into a respected journal. If skeptics are really sitting on AGW-shattering calculations and have better theories formulated, then they need to start publishing them. The problem is that they don’t.
9. A scientist doesn’t lobby congress for pet political policies to be enacted, an advocate (Hansen) does.

This depends on what discourse you take with respect to governance. Hansen’s actions and rhetoric embody a combination of what Dryzek might refer to as the discourses of Survivalism and Administrative Rationalism. Hansen is an expert in his field, and thus should be involved in the policy process. The extent to which he should participate, on the other hand, is up for debate. Like it or not, the up-and-coming generation of scientists will feature a great many ‘scientist-activist’ – people who are trained to deal with complex scientific issues but also wish to participate in the policy-making process. Personally, I think more scientists should be involved in politics; it can be much to emotional at times, and the cold rationality that science might offer to the policy process would likely be very beneficial. The bottom line is that our society faces legitimate issues which the lay person might not fully comprehend. It is up to scientists and experts to communicate those issues and to actively engage in efforts to help resolve them. 


10. A scientist doesn’t backhandedly and without good evidence (as I have presented) lecture scientists who disagree about being advocates.


I’ve done my best to refrain from turning many of these comments directly back on their author. But this is ridiculous. I’ve said it before and I’ll say it again: if skeptics have legitimate arguments then they need to be engaging their foes in the scientific arena. Sure, on blogs they may reach a large audience, but that audience is not comprised of many people who actively participate in the scientific community – particularly the climate science one. In science, we let data settle our arguments. If you can propose an alternate theory to AGW that doesn’t violate parsimony and is better at a) explaining the data out there and b) provides the field with further possibilities to explore, then that theory will stand on its own. Skeptics have yet to do so. Furthermore, they’ve spent most of their time attempting to refute key elements of AGW theory and have been outrageously unsuccessful. 
I wanted to write my own definitions of “advocate” and ‘scientist,’ but Gavin’s are really spot on. Jeff’s aren’t. 

~ by counters on April 2, 2009.

4 Responses to “Ten Replies to Ten Replies about Advocacy vs Science”

  1. Hi Counters. I’ll comment on a couple of things.

    First, I don’t begrudge the folks at RC and how they choose to run their blog. However, they do seem to use moderation as a filter to control what they choose to deal with. I have seen many of the comments that have been discarded. I don’t think it is fair to chaulk it all up to “noise” (I’m sure they get their share of that though). But hey, it’s their blog and they can do what they want.

    Secondly wrt transparency, I’m glad to here that you are an open source promoter. It is important. There are many important papers which are foundation work and have policy implications (you mentioned Steig) which need open scrutiny. Any healthy system needs checks and balances. It is not as simple as just running data and code and seeing if the results are duplicated. It is about checking on the soundness of the method, data, and intermediate steps. For example, if raw data is processed prior to analysis and the processing method or processed data is not available, then how can conclusions which depend on these intermediate steps be verified? I would not expect you or anyone to trust a paper which subjected solar data to “processing”, and then concludes that the sun, not CO2, accounts for global warming without the ability to examine the processed data and method.

    You’ll be happy to know, then, that I just managed to persuade my research adviser to shift a great deal of our scripting over from IDL to Python 🙂

    There isn’t anything I disagree with in your response. I would just stress caution in what the magnitude of processing is. At the moment, I work with raw output from a GCM, and most of my processing is just re-arranging data to be more convenient; from there, I go straight into analysis. I’m fortunate; if I want to see how dust deposition changes between two model runs, I don’t have to ‘process’ any data – it’s all there in the raw form. But if someone is attempting to reconstruct paleo-climate via a complicated proxy, then obviously some work will need to be done. To be honest though, this isn’t my research focus and I really can’t contribute more than your perfect observation that we should always treat published results with a healthy dose of skepticism.counters

  2. If I get bored later, I might post a reply to this just for entertainment.

    Feel free, although carbon trading is a far more interesting topic.counters

  3. […] Ten Replies to Ten Replies about Advocacy vs Science […]

  4. AGW is the biggest hoax/scam to ever hit the human race. Its easy to see how one sided it is, when well respected climate scientists who doubt AGW are purposely censored. The IPCC are NOT full of scientists but politicians. AGW was purposely fabricated by environmentalists to promote socialism. I will not be runned by fear mongering junk science.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: