39 Responses to “Rep. McNerney in Wired”

1
1. Andrew Dessler Says:
   March 15th, 2007 at 12:15 pm

   Kevin-

   You might want to ask yourself why the STS community has not convinced the scientific community that the “value-free” model of science is wrong. I suspect it’s because you’ve never presented a very good argument to the scientific community. I’ve asked Roger for evidence/theory in the past, and he always responds by pointing me to 27 papers and a 400-page book and says “the answer’s in there.” Not convincing.

   If someone asks me a question about the climate, I would never tell them to read a book; rather, I’d give them a short answer that references the relevant data and physical argument. Without a response like that, I suspect you’ll never convince anyone outside your community.

   Regards.

2
2. Margo Says:
   March 15th, 2007 at 1:50 pm

   I think we can forgive a *mathematician* for being a little confused about science. (Though, it is true that *in the long run* science is about getting to some sort of truth.)

3
3. Jonathan Gilligan Says:
   March 15th, 2007 at 2:56 pm

   Andrew,

   I’m not a professional STS guy. Just a physicist who went astray somewhere and started wondering when and why the public trusts scientists when they can’t personally verify our research in detail. Maybe these thoughts will help.

   First: value-laden does not necessarily mean that “anything goes” as some of the more anarchistic science studies folk would have you believe. It just means that science is done by scientists who work in a scientific community that agrees on what constistutes proper research techniques and proper ways of resolving disputes. Different specialties are seen to approach both technique and dispute resolution differently, so it’s clear that there is not one unique way to do science.

   Now to dive in and try to give you the best stab I can make at a clear, brief, and empirically grounded defense of the notion that science is value-laden.

   To do this, I want to distinguish between two categories of science. The distinction is not hard and fast, but is useful in thinking clearly about important differences. This taxonomy is stolen from Sheila Jasanoff ["The Fifth Branch" (Harvard, 1989)] and to a lesser degree from various writings by Harvey Brooks and Alvin Weinberg.

   On the one hand, there is what I’ll call “research science,” where the investigator can take as long as necessary to assemble conclusive evidence for or against various theories or hypotheses. Work is largely (but not always) conducted within the framework of an established discipline so even when disagreements become heated, participants understand one another very well and share a mutually agreeable approach to resolving disputes.

   On the other hand, there is “regulatory science.” Regulatory science is conducted with deadlines and ticking clocks. The decision not to take a regulatory or political action is just as much a definite value-laden act as deciding to adopt a particular policy or regulation, so you can’t put decisions off until there is more evidence. Putting off the decision is itself a political decision.

   Moreover, regulatory and political science is inherently trans-disciplinary. The safety of the proposed Yucca Mountain nuclear waste repository depends on biological effects of ionizing radiation; solubility of actinides; colloidal transport in the vadose zone; climatic change over the next half-million years; volcanology; seismology; engineering of crash-resistant transportation casks; and many other fields of study. No one person can master all this well enough to evaluate a comprehensive safety assessment. Hydrological modelers may make very different assumptions from volcanologists about how to treat differences between different stratigraphic zones of tuff. There may not be time or money to settle scientific differences before policy decisions must be taken and different disciplines may have such different ways of looking at problems that scientists may not be able to agree on what would constitute persuasive empirical evidence.

   In such cases, science becomes much more value laden. A good example comes from a study of nuclear engineers conducted in the late 1970s by Manne and Richels ["Probability Assessment and Decision Analysis of Alternative Nuclear Fuel Cycles", in R. G. Sachs (ed.), National Energy Issue - How Do We Decide? (Ballinger, 1980)] who found that the engineers’ estimates of three unrelated matters (future growth of US electricity demand, future cost of photovoltaic cells, and US supplies of high-grade uranium ore) all correlated strongly with the respondent’s preference for or opposition to fast breeder reactors. You could say that the respondents’ scientific views influenced their preferences on FBRs rather than vice-versa, but if that were the case, there’s not a good reason why their scientific views of three disparate questions should correlate with one another (people who thought electricity demand would skyrocket also tended to think that photovoltaics would remain expensive and that there was not much good uranium ore; people who thought electricity demand would grow only slowly tended to think that PVs would drop rapidly in price and that there was lots of uranium ore). This study is discussed lucidly by Harvey Brooks in his excellent paper “The Resolution of Technically Intensive Public Policy Disputes.” [Science, Technology, and Human Values 9, 39 (1984)]

   ——-

   Another direction toward answering why science is value-laden: There is no empirical test to determine whether I should look at diseases using a taxonomy in which I gather rapidly growing tumors in different organs under the category “cancer” or whether I should categorize diseases by the host organ. Different cancers behave very differently, are caused by very different things, and respond very differently to radiation or chemotherapy. The decision to create a category called “cancer” with which to think about myelommas, astrocytomas, and carcinomas of the breast can’t be proved right or wrong by experiments. It’s a manifestation of researchers’ preferences and values. It has real political implications too, as demonstrated by the targeting of research money toward this artifact called “cancer” in competition with money targeted toward diseases of specific organs.

   ——-

   A third consideration: Is there any empirical reason why fruitless searches for axions or supersymmetric particles should be taken seriously by the scientific community while fruitless searches for extrasensory perception or telekinesis should not? I see the community’s embrace of the one and dismissal of the other as an example of community values, not objective empirically grounded evaluation of the merits of the research. (Don’t get me wrong. I think these values are good ones and that axion searches ARE much more valuable than ESP searches. I just want to acknowledge that this is a statement of values).

   In a related matter, if science is really objective, should anyone care about whether the President (a Democratic or Republican one) dismisses members of scientific advisory committees and replaces them with people all of whom have clear ties to similar industry or environmental activist groups? Wouldn’t they all come to the same conclusions about the objective science? But think of the furor over Bush’s packing the CDC childhood lead poisoning panel with friends of the lead industry or what would happen if a Democratic president tried to pack EPA’s CASAC with scientists all of whom had close ties with the NRDC and Sierra Club. At some level, don’t we believe that people’s political or economic interests have some bearing on how they will view research on environmental toxins and other matters in the public interest and don’t we all prefer that scientific advisory panels have some sort of balance on values that are not universally shared?

   ——-

   I may not persuade you with these arguments, but I offer them in hope that they may be useful.

   ——-

   P.S., I’ve read and really enjoyed your climate change science and policy book. It’s an excellent contribution to the literature and I am hoping to use it in my teaching in the future.

4
4. bud Says:
   March 15th, 2007 at 4:27 pm

   ” …the stakes of getting an equation slightly wrong or interpreting a figure incorrectly aren’t really that high.”

   Another way to define the difference between “science” and “engineering”.

5
5. Andrew Dessler Says:
   March 15th, 2007 at 6:58 pm

   Jonathan-

   Thanks for your reply. That was a far better explanation than Roger has ever supplied.

   I agree with many of your points, but I would question the implications. For example, I completely agree that individual scientists have all sorts of biases: political, personal, professional, institutional, etc. But when you consider the scientific community as a whole, I would argue that biases tend to cancel out. I think it’s a whole lot harder to argue that the entire scientific community is biased … but Iwould be interested to hear arguments in that direction.

   I also agree with your argument about taxonomies and categorization. I would just add that often times those taxonomies are provided by funding agencies, whose responsibility it is to make intrinsically normative decisions about science and reserach. But I’ll have to think about this one some more.

   I guess what I really object to in STS is the assertion that all knowledge is relative. In his paper “How science makes environmental controversies worse,” Sarewitz argues:

   This condition may be termed an “excess of objectivity,” because the obstacle to achieving any type of shared scientific understanding of what climate change (or any other complex environmental problem) “means,” and thus what it may imply for human action, is not a lack of scientific knowledge so much as the contrary—a huge body of knowledge whose components can be legitimately assembled and interpreted in different ways to yield competing views of the “problem” and of how society should respond. Put simply, for a given value-based position in an environmental controversy, it is often possible to compile a supporting set of scientifically legitimated facts.
   http://www.cspo.org/ourlibrary/documents/environ_controv.pdf

   What he’s arguing here is that there’s enough science out there that, depending on my values, I can support any claim.

   That’s crap. There is only one truth to the question “is the climate warming?” and “are humans to blame?” And that truth does not rely on values. Different people with different values might disagree, but they are not both right.

   For research where well-developed ideas are not available, Sarewitz’s ideas might seem correct at first glance. For example, a case can be made both for and against linking hurricanes with global warming. But this is just uncertainty, not values. Again, there is only one answer, and it does not depend on values: either hurricanes and global warming are linked or they’re not. We just don’t know it yet.

   As I see it, this is the fundamental problem with STS. This point was the point made by Mooney and Sokal in their LA Times Op-Ed (http://scienceblogs.com/intersection/2007/02/mooneysokal_oped_in_la_times.php).

   Finally, thanks for your kind words about the book!

   Regards.

6
6. kv Says:
   March 16th, 2007 at 3:50 pm

   JG — thanks for the excellent comment.

   Andrew — I’m not an STS scholar either, but what I’ve seen from that community has resonated with my experiences. The point I was making is that in McNerney’s contrast he’s saying that science is not influenced by “people, relationships, and rules” and that “there are [not] a lot of rules” in science, meaning not just science rules, but cultural rules that determine how science is done and what science is done. I think it’s clear that those rules do exist, as Jonathan illustrated with the ESP vs. axions example.

   As far as why the “STS community has not convinced the scientific community that the “value-free” model of science is wrong” …. have they even tried? It’s not like there is a bunch of interaction between the STS world and the standard science world. That might be part of the problem. The science world is diffuse and there is no real peg to hang on. Not sure how STS could even convince the science world of anything.

   I think you’re missing Sarewitz’s point. He’s not talking about something — ok, let’s call it “linear” b/c I don’t have a better word — like “what is the trend in T over 100 years?” or “are humans influencing that trend?” He’s talking about the system, which is so intricately complicated that values are intermeshed within it. He’s not talking about something like deriving how much heat input is required to raise the temperature of 1g of water 1C.

   I also have to disagree with your emphasis on taxonomies and categorizations. Scientists have been doing that since day 1 to make their lives a lot easier — long before funding agencies came along. They categorize and then move on. It’s not difficult to imagine that in certain ways seeing things along a pre-defined categorization scheme puts the blinders on.

7
7. LDilling Says:
   March 16th, 2007 at 4:03 pm

   Hi Andrew,
   Like you, I was educated to believe, and still do believe that science is about a search for truth, using the scientific method. However, whether we actually ever know “the truth” — and why we know it– at any point in time on a given subject is open to debate. Scientific consensus on many different topics has shifted over time and history, and consensus is not always built through definitive knowledge of the “truth” but by indirect or multiple lines of suggestive evidence.

   Naomi Oreskes has a nice article on this point illustrated by the history of scientific consensus on plate tectonics, online here: http://historyweb.ucsd.edu/oreskes/Papers/science%20and%20policy%20.pdf).
   It shows that the scientific community works both by scientific method and by community norms, and that what is considered truth changes. We hope that the scientific process will evolve our notion of truth over time. But social processes also heavily influence what truths are accepted. One particularly heinous example from history is that of the “science” of craniometry, which was used to reinforce the prevailing social prejudices of the late 19th century and early 20th century in the US.

   Naturally there are points in the evolution of scientific theory when the theory does get firmed up to the point where it isn’t challenged any more. I think Sarewitz’s point is that often in environmental controversies, we just aren’t there yet, and it’s often hard to get any kind of definitive “proof” in a timescale that is useful for public debate. Take for example questions of the appropriate water level in the Klamath river to ensure survival of an endangered fish– there is good, robust science that still fundamentally disagrees. In that case, what to do about the fish fundamentally gets down to values since the truth is still elusive. And, as Jonathan shows for the nuclear engineers, values can and do influence the choice of research problem, and the conclusions that one can reach. When scientific results have direct consequences for social decision making, the stakes are much higher and unresolved differences and uncertainties can be amplified, as Oreskes also states.

   I very much appreciated Jonathan’s comments, and I tend to agree with him that science is value-laden, by virtue of the fact that any observations, theory, or experiments that we do still must be initiated and interpreted by a human mind, which is not free from bias. So much as we might try to seek truth, we can, at best, only approximate our knowledge of it, and often perhaps have to acknowledge either incomplete or competing possibilities for what we accept as truth.

8
8. PrajK Says:
   March 17th, 2007 at 12:32 am

   Hi Andrew:

   Though I, like Jonathon, am just a (future!) physicist who dabbles in STS, I believe that the Mooney/Sokal piece misrepresents the field. I am pretty sure that no serious scholar claims that science “is nothing more than politics in disguise.”

   There may be a few who adhere to such extreme relativism. But identifying them as representative is analogous to selecting the handful of scientists who disagree with AGW. Sokal’s own website has a response that points out the straw man argument used: “The truth is that none of his targets would ever make such statements. [That facts are a social construct.] What sociologists of science say is that of course the world is real and independent of our observations but that accounts of the world are produced by observers and are therefore relative to their capacities, education, training, etc.”
   (http://physics.nyu.edu/faculty/sokal/fish.html)

   I’d also like to take a stab at the bias of the “entire scientific community.” I think we would agree that science does not compel a specific policy outcome. Given the empirical fact of AGW, society can choose any combination of adaptation, mitigation or geoengineering based on underlying values. Yet many sides, especially conservatives in this example, routinely distort the science to support their preferred solution. We then cry foul and ask why those politicians sully the purity of science to promote their agenda.

   But who imbued science with so much authority? Who created the situation where people believe a body of empirical data determines how we should live our lives? It was us of course! For the past 300 years scientists, more than anyone else, have asserted that science/scientific facts/scientific reasoning have direct implications for social policy.

   Craniometry, which Lisa highlighted, is not the only case. Max Born, for example, argued for a connection between quantum mechanics and economics. (Interestingly enough, also a response on Sokal’s website: http://www.mathematik.uni-muenchen.de/~bohmmech/BohmHome/sokalhoax.html). Scientists routinely say that we have the solution to the world’s problems. How often do you hear scientific leaders say that economic growth and global poverty cannot be addressed without more science funding? We can’t have our cake and eat it too. If we insist that science has special authority and provides special solutions, we shouldn’t be shocked when people use it to advance their own agenda.

   So my point is that the “entire scientific community” is biased towards viewing science as the way to resolve social problems (itself a value judgment). As a community, we have done everything possible to guarantee that people will debate the facts of global warming rather than what type of world we want to live in. We have guaranteed that federal science funding is seen as a necessary component of economic growth, and we have silenced any meaningful discussion on other approaches or negative consequences such as income inequality.

   Given our attitude integrated over time, I’m not surprised that everyone wants to have science on their side. Instead of complaining about people distorting science, we in the scientific community should take a look in the mirror.

9
9. bob koepp Says:
   March 17th, 2007 at 7:20 am

   Discussions about the value-laden nature of science aren’t going to get very far unless the discussants make a simple distinction between how science _should_ work and how science _does_ work in the hands of mere humans who hold many values besides the epistemic values which nearly all of us recognize as intrinsic/internal to the scientific enterprise. If you collapse the distinction, you undercut your ability to approach science with a properly critical attitude.

10
10. John Fleck Says:
    March 17th, 2007 at 11:23 am

    JG – Brilliant, thanks.

    Andrew – I think your examples – “There is only one truth to the question ‘is the climate warming?’ and ‘are humans to blame?’” – miss the point of what Sarewitz is arguing. In what Jonathan has described as “regulatory science,” the definition of both the questions and what counts as acceptable evidence in answer – becomes a point of contention. Scientizers try to then settle the debate in their favor by controlling the definitions.

11
11. Andrew Dessler Says:
    March 18th, 2007 at 12:11 am

    All-

    Thanks for your comments. I could try to respond to everyone’s points individually, but my experience is that, when you do that, the conversation rapidly drifts off topic.

    So let me first acknowledge that there are situations when values do enter into science. Lisa’s example is certainly apt, as is the famous study of insanity by Foucault. Similarly, I acknowledge that Jonathan’s example of regulatory science would certainly include values.

    But let’s cut to the chase — climate change. I will challenge you all to point out where values enter into the conclusions of the IPCC WGI.

    I would argue that the answers one gets for the questions addressed are independent of the values one holds. But I’m interested in your thoughts.

    Thanks!

    PS: Don’t try to pull the ‘ol “the selection of question shows values” excuse. The questions addressed by the IPCC are put together NOT by the scientific community but by the IPCC member governments (admittedly in consultation with scientists). So the normative decision about what questions to address is properly determined by the governments.

12
12. Mila Says:
    March 18th, 2007 at 10:02 am

    What’s the biggest difference between science and politics? Science is all about truth. You gather your evidence and logically prove your claims. Congress is all about people, relationships, and rules. There are a lot of rules.

    I think the biggest difference between science and politics is the characteristics that successful practitioners of each discipline tend to posess.

    “Science is all about truth”. I think rather to practice good science, one needs intellectual honesty.This means being able to face the facts as they emerge. Whether they are congruent with your intentions/ambitions or not.

    “Congress is all about people, relationships and rules” In essence, highly subjective. Laden with individuals’/groups’ vested interests. This is altogether not conducive to intellectual honesty on a grand scale, because mistakes or inconvenient truths tend to be equivocated about.

    Being nice and keeping people happy are completely irrelevant to science.

13
13. Roger Pielke, Jr Says:
    March 18th, 2007 at 4:32 pm

    Andrew- Can you support the following claim?

    “The questions addressed by the IPCC are put together NOT by the scientific community but by the IPCC member governments”

    Thanks!

14
14. Andrew Dessler Says:
    March 18th, 2007 at 10:00 pm

    Hi Roger, welcome to the discussion. I’d just as soon not get into the debate over the source of the IPCC’s questions right now. I feat that’ll divert the discussion.

    What I’d really like is to hear where you think values enter into the conclusions of the IPCC WGI.

    If you think it’s the definition of the questions, then I’ll back up my claim.

    Regards.

15
15. Harry Haymuss Says:
    March 19th, 2007 at 5:12 am

    Andrew – “There is only one truth to the question “is the climate warming?” and “are humans to blame?”" are two questions, not one.

    The answer to the first, everyone agrees, is “yes”, although the magnitude varies depending on who you listen to – e.g. “When was the warmest year”.

    The second question you have oversimplified, intentionally or not. It should be broken out into three: “How much is attributable to human activities”, “How much is attributable to CO2″, and “How do we know it will be bad, as climate is always changing anyway?”

    Also, I think Roger deserves an answer to his question, which you are evading.

16
16. Roger Pielke, Jr. Says:
    March 19th, 2007 at 9:53 am

    Thanks Andrew … a few replies.

    1. My understanding is that the questions addressed by the IPCC and its chapter outline are indeed developed by scientists. They are certainly developed without anyinvolvement of government officials with responsibility for mitigation or adaptation decisions. Since the IPCC claims to be “policy relevant” the there must be some values-based criteria of relevance used to decide what science to include and what not to include. E.g., Jim Hansen has criticized the TAR IPCC for neglected black carbon, and RP Sr. criticized the same report for neglect of land use change.

    2. More specifically, and perhaps more of what you were looking for, on p. 16 of the recent SPM the IPCC states, “Based on current understanding of climate carbon cycle feedback, model studies suggest that to stabilise at 450 ppm carbon dioxide, could require that cumulative emissions over the 21st century be reduced . . .” Why 450? It is not in the FCCC, nor has this number been agreed to outside the EU … there is no scientific reason for 450, rather than 426 or 473 or 511.6 (see Richard Tol on this point). The IPCC’s focus on describing the consequence of 450 reflects a bias toward this particular stabilization level versus other possible levels, which reflects a choice made inside the IPCC, as there is no intergovernmentl consensus on this number.

    3. Along these lines more generally, the SRES scenarios that WG1 relies on are “storylines” tha reflect a wide range of values. IPCC WG1 accepts each of these in parallel fashion, and while the recent report distinguishes a range for each scenario it provides no further information on the scenarios, their probabilities, or their desirability.
    Simply being mute does not mean value-free. The entire WG1 exercise is based on the work of WG3 which could not be any more values based.

    4. Finally, the IPCC focuses on quantitative predictions, or projections if you like, which suggests that decision making ought to be made based on this particular approach to knowledge. There are other possible approaches, such as the robust decision making apporach suggested by Lempert et al., adaptive management (Kai Lee), backcasting (John Robinson), vulnerability reduction (India Declaration), technological development (Hoffert et al.), geoengineering (WIgley et al.) and so on.
    Each of these approaches to the climate problem would imply a very different IPCC, mos notably by much reduced focus on quatitative predictions. The very structure of the IPCC reflects a commitment to prediction and control.

    I hope that these examples provide a sense of the many areas where values and science are of course intermixed in the IPCC. The more important question should not be “Is the IPCC value-free?” as you suggest, but “Is the IPCC useful?” Value-free is not only impossible in practice, but we wouldn’t want it as usefulness depends upon some connection to things we care about.

    Thanks!

17
17. Andrew Dessler Says:
    March 19th, 2007 at 2:49 pm

    Roger-

    Thanks for your comments. Here are a few responses:

    1. In a public lecture, Susan Solomon said that the outline for the WGI went through *two* iterations with member governments to get the outline acceptable. In addition, the outline was formally approved by member governments in 2003. Unless you argue that the member governments are idiots who cannot stand up for themselves, you have to conclude that the outline represents the questions member gov’ts want to know in order to respond to climate change.

    2. I don’t know where this 450 number came from, so I’ll give you the point.

    3. Where are the values of the SRES? The question: “what are the possible emissions futures we could see?” sounds like a well-posed positive question to me.

    4. Given the member gov’ts buy-in of the outline, I conclude that predictions is what they want. So while it’s perhaps a value judgment, it reflects the values of the member governments.

    5. I agree that the WGI is not value-free, but rather that it reflects the values of the member governments that requested the report. This is as it should be because, as you suggest, its usefulness depends on it. However, I continue my argument that it does not reflect the values of the scientific community.

18
18. James Annan Says:
    March 19th, 2007 at 4:42 pm

    I don’t particularly have a dog in this fight but it seems more than likely that the choice of 450ppm is directly linked to the 2C UNFCCC “dangerous” threshold. I’ll let you fight over whether this was chosen by scientists or politicians…

19
19. Harry Haymuss Says:
    March 20th, 2007 at 11:13 am

    Andrew, your statement: “That’s crap. There is only one truth to the question “is the climate warming?” and “are humans to blame?”" is actually two questions.

    The first is not even in question.

    The second is counterproductive and simplistic. It actually breaks down into at least three questions:

    1. How much is anthropogenic CO2 to blame for the current warming?

    2. How much is due to other causes, both anthropogenic and otherwise?

    3. Is there a real need to be alarmed?

    Can you answer those three? I think not…

20
20. Jonathan Gilligan Says:
    March 20th, 2007 at 12:02 pm

    Pardon my long digression to set the stage, but there’s simple example of value-ladenness in IPCC FAR SPM at the bottom of my ramblings.

    We use the words “value-laden” in many different ways. I reject many of the stronger arguments about value, which suggest that Voodoo has equal validity to modern Western medicine [Feyerabend, "Against Method" (Verso, 1975)]. I prefer a weaker version, which simply states that although we trust a huge amount of natural science as true and reliable, when we look at why we believe it we find that even for the most solid natural laws there is no algorithm for testing their validity. There is no clear rule for distinguishing anomalies and outliers from data that violates the law. Any good scientist has a good intuition that can tell the difference, but you can’t easily explain how to do so, or spell out an algorithm for telling when a natural law has been proved wrong. Simple conceptions of scientific method, such as Popper’s notion of falsifiability or the idea of decisive experiments just don’t hold water when you look at the historical record.

    In natural science validity seems to come through building consensus among the scientific community. The success and cohesion of the scientific community shows us that relying on values is not necessarily bad, nor does it condemn us to anarchy. Jacob Bronowski points out [Science and Human Values (1956)] that the scientific community has been around and functioning well far longer than any government on earth, so political scientists could learn a lot about what works by studying how well the scientific community governs itself.

    From that digression, value-ladenness in climate science: In some areas, such as the theory of blackbody radiation, there is such universal acceptance of the theory that the value-laden aspects are not interesting to most of us because everyone shares the relevant values. If a problem emerges, there is widespread consensus how to approach it.

    On the other hand, there is no such consensus about what constitutes adequate proof for many scientific propositions about climate. One test that’s been proposed for distinguishing consensus-seeking science, where the value-laden aspects are uninteresting even in vociferous debates, from controversial science, where value-laden aspects are crucial, is whether the parties in disagreement can agree on a mutually acceptable set of evidence that would make either party change its mind.

    What I don’t see is a clear set of criteria that all parties to debates over climate change would accept as grounds for changing their minds. Could those who fear catastrophic deglaciation and those who doubt it agree on what sort of data it would take to settle the matter short of waiting to see what the ice does over the next several centuries? Consider similarly the debate over anthropogenic contributions to hurricane intensity: can everyone agree on what would constitute persuasive evidence one way or the other?

    Where we don’t have consensus on standards of evidence and proof, we have a debate where values are very important in choosing what counts as evidence and what its weight is.

    The judgment of what counts as solid evidence and what’s an outlier or anomaly appears to me to be very value-laden in this discussion. Michael Crichton’s execrable State of Fear exploits this by presenting a straw-man version of the anthropogenic warming hypothesis and then citing lots of isolated data out of context as though it clearly falsified this hypothesis. I’ve seen Fred Singer do much the same both about ozone depletion and climate change.

21
21. Lab Lemming Says:
    March 20th, 2007 at 6:38 pm

    What is STS?

22
22. Paul Says:
    March 21st, 2007 at 8:04 am

    Roger:

    ” My understanding is that the questions addressed by the IPCC and its chapter outline are indeed developed by scientists. They are certainly developed without anyinvolvement of government officials with responsibility for mitigation or adaptation decisions.”

    A quick visit to the IPCC will confirm the following:

    The “scientists” are chosen by governments (any politician worth their salt knows you can chose the agenda or chose the person to set the agenda to similar effect). And I quote the IPCC web site:

    “IPCC reports are written by teams of authors, which are nominated by governments and international organizations”

    Once selected, how do the scientists proceed? I would hesitate to claim as you do, that it is without any political involvement. Again, the IPCC state:

    “Policymakers and other users of IPCC reports are consulted in order to identify the key policy-relevant issues.”

    So there is a definite scientific bent to the IPCC, but it is under the mandate of politicians, under the guidance of politicians and answerable to politicians. Is anybody really surprised, given we are talking about an INTERGOVERNMENTAL Panel here?

23
23. Jonathan Gilligan Says:
    March 21st, 2007 at 10:33 am

    I tried to post something before on this, but it didn’t take. I hope this won’t become a double post.

    Andrew asks about value-laden content in the latest WG1 SPM.

    Throughout, assessments are made of confidence and likelihood with numbers attached (“very likely” and = 90%+ probability). Despite the numbers, these statements do not derive from objective statistical calculations, but from “expert judgment.” Nor is there any process to validate or test these assessments of confidence and likelihood. They are completely statements of the values of the authors.

    Consider this thought experiment: could the community of hurricane scholars potentially achieve consensus on an objective procedure to test the SPM statement “it is likely that future tropical cyclones (typhoons and hurricanes) will become more intense, with larger peak wind speeds and more heavy precipitation associated with ongoing increases of tropical SSTs.” (I am emphasizing the use of the word “likely” with its associated definition “>66% probability of occurrence”). Or is the question of likelihood too value-laden for the different sides to agree on standards of evidence and proof?

    The heavy use of subjective expert judgments of likelihood, certainty, and confidence are both value-laden and very useful components of the WG1 SPM. If the experts only presented their observations (Here is the output of a particular model run; Here is a table of raw satellite observations of IR radiance) the report would be completely useless to the public and policymakers. What we want and need is expert judgment presented in the context of the data and theoretical framework that supports it. But let’s be clear that despite all the expertise that goes into them, these are statements of value.

    In the broader picture, values are essential to the important scientific task of sorting conflicting data into those that are reliable and support a particular hypothesis or theory; those that are reliable and pose serious challenges to the theory; and those that are largely irrelevant artifacts or outliers. This is one place where subjective expert judgment comes in and WG1 relies heavily upon this judgment.

    Good scientists have great eyes for telling the difference between an artifact or outlier and a real problem, but this is a value-laden judgment and not something for which you can spell out objective rules or algorithms. Michael Crichton’s State of Fear is a great example of failing (perhaps deliberately) to make these distinctions and thus confusing outliers and artifacts with serious observational challenges.

    It would be a mistake to conclude that because much of science is expressions of the values of the values of the scientific community that this somehow implies intellectual anarchy or that the value-laden results aren’t true. On the contrary, one of the great strengths of the scientific community is the way it’s managed to maintain a functioning and coherent set of values over the course of more than four centuries—longer than any system of government has done. Jacob Bronowski points this out in “Science and Human Values” (1956) and advises that political scientists would do well to look to the scientific community for lessons in what values and modes of governing make societies work well.

24
24. Nosmo Says:
    March 21st, 2007 at 12:27 pm

    Jonathan Gilligan,

    If some scientist said it is likely that the Casini spacecraft we will find some chemical on some specific moon of jupiter, would that be a value judgement.
    Or concider Prof. Phil Marcus of UC Berkeley prediction several years age of a major change in the climate of Jupiter (Ten degree Kelvin change in the differenece between the pole and equator temerature). If he attached a probablility to that prediction, would that have been a value judgement? (I’m not up on the the latest results but as of last year the indications were that he is correct).

    Are these different from your examples?

    Nosmo

25
25. James Annan Says:
    March 21st, 2007 at 7:52 pm

    Jonathan,

    Nosmo beat me to it but I don’t agree with your use of the term “value laden”. All such probabilities are essentially statements of the strength of belief of the researchers. But that shouldn’t be directly dependent on their values, any more than tomorrow’s weather forecast is.

26
26. Andrew Dessler Says:
    March 21st, 2007 at 8:39 pm

    Jonathan-

    I think that you’ve made a mistake when you explicitly equate expert judgment with values.

    Values are something that are not right or wrong in an absolute sense — e.g., murder is wrong, taxes should be low, spreading democracy is a morally legitimate use of American military power. In our book, we call these as normative statements.

    However, the question of whether a particular theory is valid, whether a data point is an outlier, etc., does have a right answer. You and I might have different expert judgments on these questions, but only one of us is right.

    As far as your hypothetical question goes, I don’t know what the hurricane community would agree to (they seem to be particularly combative). But I would expect that wherever they are now, as future data comes in, we can expect the consensus to evolve. Eventually, when enough data has been obtained, virtually everyone will agree on a single answer to the question of connecting AGW and hurricanes. This is another reason to suggest that the differences are not value-based — data is of little or no help in resolving value-based disagreements.

    Thus, I still don’t see values in the WGI. If you still disagree with me, then I’d be interested in having you identify specific values that scientists would hold that would cause them to reach different estimates of, say, our confidence that humans are responsible for the recent warming.

    Thanks!

27
27. Andrew Dessler Says:
    March 21st, 2007 at 8:52 pm

    All-

    This has been in illuminating discussion. I’m afraid that I’m still far from convinced about the merits of STS. Here are a few problems that remain.

    1) I’m still not sure what the theory is. Everyone seems to have a different idea of how values enter into science. Roger has one idea, Jonathan G. another, Sarewitz yet another, etc. These ideas all seem quite different and none are all that convincing to me.

    2) I certainly recognize that values enter into science in some instances. Jonathan’s example of regulatory science and Lisa’s example of craniometry are both convincing. However, Roger in particular seems to be trying to make this some kind of universal law, like F=ma, that applies everywhere. It takes a lot of evidence to elevate a theory to that level, and frankly Roger’s argument (as articulated in his comment up above) is, well, weak.

    I hope this doesn’t come off as too harsh, but if your goal is to convince mainstream scientists that these STS ideas have merit, you need to refine and sharpen your ideas.

    My expert judgment is that simply telling scientists, “you’re biased but you just don’t know it” is not going to work.

    Regards

28
28. PrajK Says:
    March 22nd, 2007 at 2:01 am

    Hi Andrew-

    Thanks for your comments. I’m afraid that I’m still far from convinced that you understand STS. Here are a few problems that remain.

    1) I’m still not sure what the argument is. You accept that values sometimes enter science, but reject the notion in other instances. How do you draw the distinction? You also seem to have an incomplete idea of what STS scholars actually do–values in science is just one part of it. They also study, among other topics, models of science communication and literacy, expanded definitions of science-related expertise, and the democratic and social ramifications of technocratic decision-making.

    2) I certainly recognize that some academics have made extreme statements. However, you seem to be trying to make this some kind of universal law, like F=ma, that applies to all of STS. Frankly, your argument (as articulated in your comment up above) is, well, weak. Consider, for example, Bryan Wynne and Robert Paine. They demonstrated that scientists’ neglect of situated expertise led to mishandled treatment for farm animals after Chernobyl. Others have shown that parents of autistic children can offer meaningful, reasoned input on their children’s treatment, and that improving scientific literacy involves much more than scientists lecturing so-called laypeople. Seems to me that these ideas do have merit.

    I hope this doesn’t come off as too harsh, but if your goal is to convince mainstream STS scholars that you can critique their work, you need to refine and sharpen your ideas. I guess I am just a lowly physics graduate student and I am quite far from being an expert in anything. At least in this stage of my career, I find that I still have to do my homework.

    My non-expert judgment is that simply telling STS scholars, “I’ve read a very, very, very small part of your work and I think it’s crap, and therefore your entire field is without merit” is not going to work.

    Regards

29
29. Nosmo Says:
    March 22nd, 2007 at 11:55 am

    PrajK, Andrew,
    Seems like the problem is that we really don’t know what the consensus on STS is, or may don’t understand what value-laden means My questions to Jonathan Gilligan were not rhetorical. I truly do not understand what he means.

    Andrew says at the top “You might want to ask yourself why the STS community has not convinced the scientific community that the ‘value-free’ model of science is wrong. I suspect it’s because you’ve never presented a very good argument to the scientific community. I’ve asked Roger for evidence/theory in the past, and he always responds by pointing me to 27 papers and a 400-page book and says ‘the answer’s in there.’ Not convincing.”

    So far I would have to agree with him. My inclination is to believe that the practice of science is very value ladden but the arguements I’ve read so far are do seem rather week. I’ve been reading Prometheus for almost two years and I very often don’t understand what Roger is trying to say.

    So let me ask again. We have a scientist who made a major climate prediction with many specifics (about Jupiter) which is clearly correct. (Taked to someone from his group after I posted yesterday. He has several papers in press.) Can someone explain how making these predictions reflected his values? If they do why is this different then a prediction about climate about earth?

30
30. Jonathan Gilligan Says:
    March 22nd, 2007 at 2:16 pm

    Andrew,

    You’re absolutely right that there’s no single STS perspective, but a wide diversity of opinions on how culture and values shape science. I find STS accounts that try to cram all of science into one cultural picture completely unpersuasive. My experience working in a number of different areas of science, from quantum optics to surgery, is that different scientific communities have very different ways of doing science and very different standards of evidence, so I am wary of attempts to write grand unified theories of how science works.

    From my perspective, trying to force Newton’s laws into a picture that says they’re culturally relative or possess normative content is possible, but only in a tendentious, trivial, and pointless sense. I’m happy to say that Newton’s laws are the truth (let’s leave aside details about relativity and quantum mechanics). But Newton’s laws are settled science and it’s not very interesting to start a new research project to test them once more.

    I agree that the vast majority of science is uncontaminated with any significant (or interesting) normative issues. But often (although not always) in politically sensitive matters, such as climate change and the health effects of pollutants, the answers to the scientific questions we care most about are those most likely to be caught up in and influenced by partisan or normative concerns, whereas in those places where science is most reliable, it may “merely sound the [same old] alarm with greater precision.” [Sarewitz, Frontiers of Illusion, p. 86]

    I agree, though, with your criticism that my previous comment about expert judgment confused scientific norms (judgments about what’s good and bad science) with social and political norms (judgments about how we ought to behave). I was sloppy in my thinking and writing. I think there is a connection between scientific and social/political norms, but I did not support that assertion. I will get back to that in a future post.

    For me, the merits of STS are not that it has much to tell scientists about how to do science—it doesn’t and besides, scientists are doing very well at their research without any help from STS—but that it might help the political process make better use of science. Thus, I have absolutely no desire to convince working scientists that their laboratory work is somehow “biased.” What I want is to contribute to better ways for the fruits of that work to be useful in helping a democratic public participate in making informed decisions.

31
31. Mark Bahner Says:
    March 22nd, 2007 at 3:58 pm

    Hi,

    I’m getting in late on this discussion, but I’m troubled by the people (e.g. Jonathan) who seem to be grouping what I’d call “scientific judgment” into “value-laden statements.”

    For example, suppose I say that I think that global warming will help cockroaches, seals***, and shippers using the Northwest passage, and say that global warming will hurt polar bears. That’s a “scientific judgment.” It’s got nothing to do with my values.

    ***Seals are judged to benefit, because there will be fewer polar bears to eat seals.

    http://www.stemnet.nf.ca/CITE/pb_p_diet.PDF

    It’s only when I get into saying that bad resulting from hurting polar bears exceeds the good coming from helping cockroaches, seals, and shippers using the Northwest passage that I’ve slipped into a “value-laden judgment.”

    P.S. Aha! Judging from Jonathan’s just-posted comments, I think we’re all in agreement. That’s good.

32
32. Andrew Dessler Says:
    March 22nd, 2007 at 4:50 pm

    Jonathan-

    I think we basically agree on this now.

    I also find myself in the odd — perhaps even unprecedented — situation of agreeing with Mark Bahner.

    However, I somehow doubt that Roger or Dan Sarewitz would agree with the three of us. Perhaps Roger will weigh in and let us know.

    Regards

33
33. Andrew Dessler Says:
    March 22nd, 2007 at 5:20 pm

    PrajK-

    I appreciate the symmetry argument you are making in your post. However, I think you miss one thing: one goal of the STS community should be to convince scientsts like me as well as policymakers that STS ideas have merit.

    Without buy-in from scientists and policymakers, STS will have little or no practical impact in the larger world.

    I, on the other hand, do not care at all what STS scholars think of my climate research. It does not change the impact of my work whether Roger et al. think it’s crap or not.

    Thus, the symmetry is broken: the burden of proof is squarely on them to convince scientists, not on scientists to understand STS. As I said in my original message, STSers need to come up with better arguments if they want to convince the wider world.

    Of course, if they don’t care whether scientists and policymakers understand their theory, then they should just keep doing what they’re doing.

    Regards

34
34. Nosmo Says:
    March 22nd, 2007 at 5:32 pm

    “However, I somehow doubt that Roger or Dan Sarewitz would agree with the three of us. Perhaps Roger will weigh in and let us know.”

    It does seem odd that we are very uncertain what Roger thinks about this. Which I think supports Andrews first comment about the STS community (or at least Roger) doing a bad job of convincing scientists.

    (Roger you will be happy to know that I broke down and ordered your book. I’m hoping it helps.)

35
35. PrajK Says:
    March 22nd, 2007 at 8:27 pm

    Hi Andrew.

    Thanks for your response. But I made more than a symmetry argument. I cited specific examples of STS scholarship that have implications for how we scientists do our work. I guess I’m still confused about your point. A few yes/no questions would clarify: are you sweeping aside all of STS because you disagree with some scholars who argue that values enter climate science? (If so, you’re quite bold!) Are you arguing that parents of autistic children have nothing useful to offer? That lay-citizen input has have never improved clinical drug trials? That issues of trust and authority don’t matter as long as we produce objective facts? That science appreciation and literacy will increase if people only learned more scientific facts from professional scientists? Again, these are also the types of issues that STS scholars research. Whether scientific facts are embedded with values is just a small part.

    Scientists get very emotional and protest strongly whenever anyone suggests that science and values intersect or that science in practice isn’t the same as science in principle. In the process we distort a field that has much to to offer. The examples I listed earlier are just the beginning. If you believe, as I suspect you do, that science should serve the public good, then i don’t see how we can avoid the values-based questions that our research impacts. If STS is to help both science and society, then at the very least scientists should be willing to listen. I get the impression that you’re not ready to do even that.

    Perhaps we need a new way of discussing this topic because I think we’re speaking past each other. The question of whether science is truly objective and value free is a wonderful academic question and great for the purpose of intellectual masturbation. I think that Roger and Dan would both agree you that there is a difference between scientific and value-laden judgments. But they would add that oftentimes the distinction is uninteresting, meaningless and distracts from more important issues. When they speak of values and science, I take it to mean more the scientific enterprise as a whole and its role in democratic decision-making and discourse. In this analysis, values are very much a part of science whether or not they affect specific observations or measurements. When scientific research has social implications, and if we want to contribute to our country and our world, we have to confront broader values that exist. I think you missed this point in Sarewitz’s essay.

    It’s important to note the article is titled “Excess of Objectivity”, not “Excess of Subjectivity.” In the introduction itself Dan highlights that “the success and impact of science is argument enough for the validity of its method and results.” These controversies are intractable precisely because objective scientific facts can be mapped onto divergent values. In such situations, facts hinder rather than assist.

    Roger, Dan: I hope I’ve represented your views correctly. Please correct me if I was wrong.

    Regards,

36
36. Andrew Dessler Says:
    March 22nd, 2007 at 10:10 pm

    PrajK-

    Here are a few answers to your questions:
    I acknowledge that values sometimes enter into scientific inquiry. Jonathan’s example of regulatory science and Lisa’s example of craniometry are both convincing.

    What I object to is the sweeping statement that this is a universal law, like F=ma, that applies everywhere.

    I do think science should serve the public good. The way society interfaces its values with a value-free scientific community is through the funding agencies. The funding agencies make the normative choices of what science is most important, and then deploys resources such that the scientific community is directed to answer those questions.

    In fact, much of what I suppose you would identify as “values” in science I would argue come not from scientists or the scientific community but from the funding agencies. I study climate because it’s interesting, sure, but also because NASA pays me to study it. If the money disappeared, I’d pick something else to study.

    I also agree that we seem, to some extent, to be talking past each other. I think this represents a fundamental limitation of blogging — it’s inability to easily resolve nuanced positions. Nuance tends to be lost rapidly.

    Thanks again.

37
37. Jonathan Gilligan Says:
    March 23rd, 2007 at 8:16 am

    Mark,

    Regarding the connection between scientific judgment and normative values: My example of the nuclear engineers illustrates that when there is a strong connection between the scientific question and a political question, scientific judgment can be strongly influenced (possibly unconsciously) by normative values.

    There is an extensive literature on problems with expert judgment being biased (though not necessarily by normative considerations) and overconfident [See, e.g., B. Fischhoff et al., "Lay Foibles and Expert Fables in Judgments about Risk," American Statistician 36, 240-55 (1982) and P. Slovic et al., "Decision Processes, Rationality and Adjustments to Natural Hazards." in G.F. White (ed.), "Natural Hazards: Local, National and Global," (Oxford, 1974), 187-204]. I’m with Andrew on the thought that ultimately hard data can overcome the effects of bias, but I don’t share his faith that while uncertainty prevails the invisible hand will operate in the scientific community as it does in the marketplace to ensure that individual biases almost always cancel one another.

    I am working up a post that goes into this in greater depth, but it’s way too long and needs to be edited down to manageable size. I was sloppy in eliding the two kinds of judgments in my earlier post and Andrew caught me out at it, but I’m not backing away from my opinion that the prominent role for scientific judgments in WG1 is an open door for normative values to influence the scientific content. I do owe you who disagree with this a solid argument for you to challenge or refute.

38
38. Roger Pielke, Jr. Says:
    March 24th, 2007 at 11:40 am

    Just back online after one week off …

    Nosmo- Thanks! Let me know what you think …

    Praj- Thanks, keep up the good work!

    Andrew- I don’t recall saying that any of your work was crap, however, your statements about STS might be interpreted in the opposite manner.

    Thanks all!

39
39. Roger Pielke, Jr. Says:
    March 24th, 2007 at 11:46 am

    Andrew- Sorry that you don’t find my efforts here convincing, sorry about that. Praj, Lisa, and others on this thread have made convincing arguments. Even so, we have a vigorous and growing certificate program in science and technology policy here that involves scientists and engineers from about a dozen disciplines (And several other schools do as well, focused on training scientists, not STSers). So some folks apparently see some value or worth in this area of scholarship. Andrew will probably argue that this occurs despite my participation, c’est la vie Thanks!!