Thursday, May 20, 2010

Para-normal? Post-science? Post normal? Para science?

Let’s begin with an old paper: ‘Science for the post-normal age’ (Funtowicz and Ravetz, Futures, 1993). While elaborating on the concept of post-normal science it is difficult, from the paper, to ascertain the features of a ‘post-normal age’ in which post normal science would feature. Some 15 years later Ziauddin Sardir welcomes us to ‘post normal times’ in his article ‘Welcome to postnormal times’ (also in Futures, 2009).
Defining our postnormal times Sadir uses the examples of climate change, the recent economic crisis, the pandemic of swine flu*, the energy crisis, the threat of nuclear proliferation and terrorism. Sadir tells us ‘The espiritu del tiempo, the spirit of our age, is characterised by uncertainty, rapid change, realignment of power, upheaval and chaotic behaviour. We live in an in-between period where old orthodoxies are dying, new ones have yet to be born, and very few things seem to make sense.’ Wait a minute! What about the World War I, (or even the Thirty Years War) the 1918 Spanish Flu, the 1957 Asian flu and the 1968-69 Honk Kong flu (not to mention the failure of swine flu to manifest as a pandemic), the Great Depression, World War II, - weren’t these times also characterised by ‘uncertainty, rapid change, realignment of power, upheaval and chaotic behaviour.’ Sadir also tells us ‘In normal times ... we know what to do.’ But he fails to tell us just who the ‘we’ are: politicians, dictators, clerics, scientists, general public? ‘We identify and isolate the problem and apply our physical and intellectual resources to come up with a viable answer.’ Pre postnormal Thalidomide; pre postnormal Hirosima; pre postnormal combustion engine?

Now let’s move to science in this postnormal age. Sadir tells us Ravitz and Funtowicz (1993, again in Futures) ‘noticed that the old image of science where empirical data led to true conclusions and scientific reasoning led to correct policies, was no longer plausible.’ This was unique to circa 1993? Previously ‘scientific reasoning led to correct policies’: as in the eugenics movement for example, as in the Holocaust.? No matter. Funtowicz and Ravitz wrote: ‘Wherever there is a policy issue involving science we discover that facts are uncertain, complexity is the norm, values are in dispute, stakes are high, decisions are urgent and there is a real danger of man-made risks running out of control.’ This was the description of post normal science, albeit preceding the events that seem to define the ‘postnormal times’.  (If facts are ever certain, is there a need for science?)

A little after 1993 (1999 to be exact) Hans von Storch and myself published a paper titled ‘Climate Science An empirical example of postnormal science’ (BAMS 1999). Here we simply employed the milieu surrounding climate science according to a defined x - y axis, that of a high degree of uncertainty and perceived high stakes. We did NOT adopt the concept of postnormal science as the CONSTITTION and new METHOD for science. We simply, empirically, demonstrated that the climate change issue did indeed, as perceived by the climate scientist at the time, include the characterisitics of high risk and high uncertainty.

But there is more to the comprehensive version of postnormal science. It is science now dominated by value judgments; to be moderated by a postnormal extended peer community (extended peer review) . There’s more but that’s enough for the rest of this discussion.

Over the years when we have measurement, 1996, 2003, 2008, there has been little change in the urgency given the issue in terms that the time is always right for immediate policy decisions; that there is enough CERTAINTY that there should be no delay in implementing policy. But in fairness, uncertainty is definitely evident in the scientists assessment of the abilities of models and the accuracy of the depiction of future climatic conditions. Now, about the risk .. . While there seems to be great difficulty in explicitly stating climate change impacts, there is a high level of agreement that the impacts will be detrimental and that climate change is very much one of the leading problems facing humanity. Typically, over the issue of climate change, the public should be told to be quite worried. So, while the matter of uncertainty is no so clear, the risks are still estimated to be quite high. However, between 1996 and 2003, scientists who responded to the 2003 survey claimed that, over all, over the previous 10 years, uncertainty had been significantly reduced.  Here the uncertainty, in all fairness, is equated to the accuracy of the science (in general but not in detail) but not necessarily to the impacts of climate change.

But what about the other defining features of PN Science, the complexity, the role of values, the extended peer review. Well, climate is a complex system so no argument there. But so are astrophysics, neurology, etc., etc., albeit on different scales. In that sense, complexity cannot be a unique defining feature unless astrophysics and neurology are partly postnormal (maybe lower risks), if postnormal science and normal science do not have to be mutually exclusive, if it is a matter of degrees.  But then, can one post normal science be more post normal than another science?  Are we talking science as a single social practise or sciences based on knowledge segregated disiplines?  The debate could be much longer but I think that is enough to raise a few thoughts.

Values: ‘To what degree do you think climate science has remained a value-neutral science?’ A value of 1 = a great degree, a value of 7 = not at all. Mean 1996 = 4.23, mean 2003 = 4.29; no significant change, and neither was there between 2003 and 2008. In brief you could suggest that values play an equal role to objectivity in determining the achievements of climate science. In fact, when asked ‘Given our current state of knowledge, climate change is now mostly a political issue or a scientific issue’ scientists in 2008 responded that it tends towards being a political issue, and politics tends to be directed by values. Hence while the ISSUE might be directed by values it is not necessarily so that the SCIENCE is perceived of as being directed by values.  But what of the role of values in the science of 'weapons of mass destruction'?  What about the values  of the few that are contrary to the values of the many.  How does postnormal science deal with conflicting values? Whose values?  Who decides what is good or what is bad?  Could post normal science exist in a dictatorship? 

Finally, we get to the ‘extended peer review'. For me at least, it is not clear in the PN discourse just what this should entail. In the 2008 survey of climate scientists a number of questions were designed to allude to the notion of the extended peer community. Concerning first adaptation and then mitigation, scientists were asked if priority should be given to the opinions of industry and commerce, political opinions, public opinions or scientific expertise. In all cases scientists perceived that control should remain mostly with science. A post normal issue without post normal science?

In the body of recent literature 'post-normal science' seems to be given the status of a panacea for all that ails the world, a method, a model of science, a manifesto for the conducting of science.  At best, it can be described as an insight, nothing more.  The call for 'dialogue': a sharing of power or the fusion of science and (some type of) religion/political bent?  Is there a risk, if put into routine, that science will simply become a political lacky, watered down rhetoric?

Afterword: Given the role of credit institutes in the recent economic meltdown and the role of scientists in the recent climategate fiasco, perhaps we should begin to consider coining new terms: postmoral science in a postmoral age.  As for the postmoral science I have faith that it would consist of a  small crowd, perhaps those driven mostly by (ill formed) values.

*Annual flu epidemics are estimated to affect 5-15% of the global population, cause severe illness in some 3 - 5 million people and result sin 250,000 to 500, 000 deaths on a global scale. The Spanish flu killed an estimate 20 - 100 million people, the Asian flu about 2 million, the Honk Kong flu about 1 million, Swine flu 8,768 confirmed deaths.


Zajko said...

PNS is a "new" way of thinking about science for policy that acknowledges many of the problems and limitations of the orthodox approach. It is not a recipe for decision-makers, and does not describe some dramatically new state of affairs. What it signals is an attempt to reshape simplistic common-sense conceptions of science with something more accurate, albeit also much more unwieldy. It is the differentiation between PNS and "normal" science which I would take issue with, since it implies that for a long time science proceeded through a process of ordinary Kuhnian problem-solving, but that "this assumption no longer holds" in the age of mad cow and global warming. Of course, this need not be taken as suggesting that science has changed, just that our understanding of it has, but the common view of PNS seems to be something along the lines of, "science has become post-normal".
Science has always just been science, with all of its various forms and without any sort of common essence. What has changed is an appreciation of some of the difficulties in applying science to policy, through the context of several failed or flawed attempts at such.
Perhaps we need a post-post-normal science which might acknowledge that "we have never been post-normal".

willard said...

Kuhn's concept of normal science was not an historical one. Talking about post-normal "times" should not be taken as referring to an historical period, if we are to extend Kuhn's framework while staying without it.

One might find there enough reason to backup the claim that elaborating on the concept of "post-normal science" is difficult. It might also provide a strong incentive to take Sardir's analysis with a grain of salt.

willard said...

Meant "within", not "without":

> Talking about post-normal "times" should not be taken as referring to an historical period, if we are to extend Kuhn's framework while staying within it.

Anonymous said...

My major problem with PNS is the fact that "the characteristics of high risk and high uncertainty" are not independent, in fact quite the opposite.

What we see time after time is that the actual (i.e. scientific) uncertainty creates a perceived high risk - epidemics (flu, BSE or at least CJD, obesity), sky falling in (population - Malthus, Ehrlich, now OPTrust, climate change, coming ice age, nuclear holocaust, terrorism, Y2K, peak oil, WMD etc etc)

This fear, whipped up by the media, seized by politicians (for whom public fear is often a blank cheque) end us up in hysterical times.

Is there an example of PNS which has actually worked? I thought of the Manhattan Project where uncertainties and stakes certainly high, BUT that followed real science albeit with bongo playing.

Anonymous said...

I´m sorry to say that only the title (Para-normal? Post-science? Post normal? Para science?) captured my interest for this post. In short my answer to the post (which I´m sure Eduardo will understand in full), is that we are dealing not with Science at all, but with “Ciencia para-anormales”. PNS is non-science, is just the opposite of Science: it is Cargo-cultism (R. Feynman), it is the corruption of Science, a sophisticated (as coming from a sophist) attempt to destroy Science by non-scientists (just look at Ravetz’s work : is that Science?).

My thoughts: those of us who work in Science better go back to empiricism (resume real data gathering and analysing), stop taking “computer games” too seriously (time up for Nintendo-science), and keep honest to the rest of society about what we really know (which is always very little with respect to the size of our ignorance).

Otherwise, playing the game that the Ravetzs and Hulmes of the world propose will destroy the best tool humankind ever had to understand reality and free herself from the ditactorship of Nature.


Anonymous said...

I think that not just eduardo, but other people in this blog too, must be able to fully understand the "joke" (to say something) by Alfonso, the previous poster.

In Spanish, "anormal" used as a noun is a VERY politically incorrect term that is sometimes applied to intellectually challenged people. It usually refers to people affected by organic sicknesses (genetic or not). It may also refer to people that are unable to think in an abstract/correct way because of other reasons, such as lack of education, but this usage is less frequent.

(3rd acception)

"Anormal" is by no means a neutral word, it is sensibly rejected by associations and relatives of people affected by Down syndrome and other illnesses. The word is (fortunately) being used less often as time goes by, but, still, it is not a word that a sensible speaker would tell to the father of a person affected by a disability.

So, what Alfonso is so happily saying is:

"Ciencia para-anormales" = "Science for intellectually challenged people".

Anyway, the negative connotations of the word "anormal" make the term extremely harsh and insulting for a lot of people.


Anonymous said...

Jon is fully correct, and I apologize to anybody that my selection of words may have offended.

Beyond the politically incorrectedness of the term I used, the same concerns about PNS still remain. How can the PNS concept qualify as Science when it advocates for the abandonment of the search for truth?.

Do not get confused by these people. Science is clearly perceived by Ravetz as a menace "...formerly we asked what science is doing for us, while now we ask what science is doing to us", and they are trying to do something against that perceived threat (destroy it?).

They attempt to convince us that under some scientific circumstances (mainly of great ignorance) we should not pursue the search for "truth" anymore, but instead "quality" will suffice. This is just plain non-sense. What is this "quality" thing they talk about?. Which kind of "quality": good, bad, or anything in between will do?. With the PNS-construct offered by Ravetz, scientist can advice societies to act (even if the consequences are disastrous) under conditions of great ignorance. Is not that a suggestion to go back to the, pre-scientific, dark ages?. Is that an attempt to achieve submission of whole societies and individuals to rules of un-enlightened-authority?.

PNS is a temptation to abandon the hardships of Science. We do not need to know anymore (it is too hard of a work). Let me explain something, when I was at Graduate School I was advised: “If you don’t like the heat, get out of the kitchen”. To me it seems as if Ravetz is attempting to convince us that Science will be OK without all the hardships and heat of the kitchen: He is wrong.

Some people do not belong in the kitchen, and anti-science is not within the realm of Science no matter how it disguises. PNS is non-science.


Anonymous said...

Alfonso, I acknowledge your apologies. Thank you.

Regarding the other points of your comment. I am not a specialist in Philosophy of Science, so, I am afraid I can not comment on that. Particularly, I do not know who Ravetz is, so ... I find it hard to express anything reasonable about that. I do not know what has Hulme proposed, so, in the same way, I prefer not to talk about that. Anyway, I still disagree with some parts of your previous comment, telegraphically.

1) I think that reducing science to empiricism (measure and analyzing measurements) is just a way to abandon very useful branches of actual science. Should society build LHC? Can we live without the measurements that LHC will produce? Is theoretical physics a branch of science even though some of its findings can not (yet) be experimentally proved? I remind old histories about general relativity and perihelion precession of Mercury or some ancient histories about lifetimes of muons and special relativity ... Those were measurements that were useful for proving previous forecasts by theoretical works. I don't think pure empiricism should be embraced. Is data assimilation empiricism? What about the model providing the background field?

2) Computer models. There is no other way in climate science. Is anybody going to track the angular momentum budgets, the energy budgets and so on without computers for a realistic Earth? Apart from very extremely simple theoretical works (Held-Hou model of the hadley circulation) which are affected by strong parameterizations (the meridional profile of potential temperature is prescribed), analytical models are not available. Qualitative explanations can be given to some extent, but quantitative answers rely on computer models of different complexities. You call that "Nintendo-science". You don't like that? Well, your proposed ways give us no other tools to advance. Your proposition is equivalent to "do nothing since we know nothing". We must expect until somebody analytically solves the Navier-Stokes equations before we try to understand the problem. Then, the problem would be to analitically solve the Navier-Stokes equation coupled to the radiation-transfer equations and the solution of turbulence .... I don't share your view.

I appreciate your apologies, in any case.



Hans von Storch said...

Alfonso / 7 - I was surprised about the agressivity of your comment. First, you seem to imply that there is a well-defined terminology, which we are speaking about. It is not. Somehow, I got the feeling that you have never read what Jerry Ravetz wrote (being 80 and beyond, he wrote a lot, and I know only a little), but you seem to have read only what others have written about him. Why not trying to be fair and read the booklet

Ravetz, J., 2006: The no-nonsense guide to science, New Internationalist, Oxford, ISBN 10:904456-46-4, 132 pp.

But all this may be a matter of style, but what is more sigificant is the usage of the term "truth" by you, Alfonso. So Newton's laws are "true" - or only valid in a limited part of the world? So, indeed, the advice “If you don’t like the heat, get out of the kitchen” applies to you - when you are unwilling to reflect about your own practice, and that of your community, about the quality of knowledge claims - apart of naive "truth"-claims, then you should leave the kitchen.

When you state "Ravetz is attempting to convince us that Science will be OK without all the hardships and heat of the kitchen", then you are, indeed, wrong. He is not trying to do what you tell us he does. Jerry Ravetz is a thorough thinker, and he deserves respect - which does not mean that everybody has to accept his conclusions.

The reason why the concept of science in a postnormal situation is relevant is because first such situations exist - more often then we would possibly like (as Zajko notes: it may even ubiquituous). Second, it allows us to think about how we can make use of science even in such a difficult situation, in which it is no longer easy, or even possible, to discriminate between analysis-based and value-based "truth".

Hans Erren said...

The science of paranormal phenomena has many similarities with the science of anthopogenic climate phenomena, both are sitting on the very edge of detectabiltity, the signal-to-noise ratio is huge and the results are significant for the survival of the scientific subject itself.

Unknown said...

The article by Sheila Jasanoff, introduced in another thread (18 May) here, though not using the term "post-normal", seems to describe the same situation as "post-normal" discussed here. Her suggestion to cope with that seems to be similar to "extended peer review" in concept, though not specific enough yet.

As introduced here in February, Jerome Ravetz hoped blogs to facilitate "extended peer review". In my opinion, such blogs as WUWT and Climate Audit provide just "disjoint peer review" to climate science.

The problem seems to be, as Michael Tobis quipped in his blog article:
"Bloggers mostly read other bloggers and write to impress other bloggers. ...
Scientists mostly read other scientists and write to impress other scientists. ...
And so on."

So, our task will be to build an extended peer community, sharing a set of tacit knowledge not exclusively.

Dennis Bray said...

On the matter of extended peer review, has anyone considered the possiblity of on-line first publications for the purpose of peer review, before 'publication'.
Perhaps for a two or three week duration. There is the risk of course of too many reviews and the reviews themsleves would need to be somewhat monitored. Perhaps a registration and password system would limit participation to those who are serious and also provide a mechanism to prevent abuse.

Anonymous said...

There exist several journals which work that way. There are "official" reviewers (which can be anonymous or not) and there is an open discussion. People taking part in the open discussion can not be anonymous. Some of this journals have already entered the ISI data base and some of them have really high impact factors. See, for instance:

I find it a very good idea.