Buy

Books
Click images for more details

Twitter
Support

 

Recent comments
Recent posts
Currently discussing
Links

A few sites I've stumbled across recently....

Powered by Squarespace
« Science hype and overprescribing | Main | Top weather blogs »
Tuesday
Jan252011

Hulme on Nurse

Mike Hulme has published some thoughts on the Horizon programme, little of which will be disputed by sceptics. Here's a snippet:

I do not recognise [Nurse's] claim that “climate science is reducing uncertainty all the time”. There remain intractable uncertainties about future predictions of climate change. Whilst Nurse distinguishes between uncertainty arising from incomplete understanding and that arising from irreducible stochastic uncertainty, he gives the impression that all probabilistic knowledge is of the latter kind (e.g. his quote of average rates of success for cancer treatments). In fact with climate change, most of the uncertainty about the future that is expressed in probabilistic terms (e.g. the IPCC) is Bayesian in nature. Bayesian probabilities are of a fundamentally different kind to those quoted in his example. And when defending consensus in climate science – which he clearly does - he should have explained clearly the role of Bayesian (subjective) expert knowledge in forming such consensus.

PrintView Printer Friendly Version

Reader Comments (59)

Nullius in Verba, John Shade, Jane Coles, mpaul, phinniethewoo - much obliged!

Jan 27, 2011 at 6:26 AM | Unregistered Commenterdread0

Thanks for the links, Nullius. I plan to look at them later today.

I should add a bit to my ramble on statistics earlier. 'analytic' problems are essentially about the future, about things which have not yet happened or about things which do not yet exist. Hence the need to develop theory to support predictions. For statistical purposes, the theory need not convey deep insight into mechanisms, for example, it may be no more than 'tomorrow will be much the same as today'. This happens to be quite a successful 'theory' in weather forecasting, and it is usually used as a reference success-level to beat by other forecasting methods if they are to claim any skill. In industrial processes, sometimes one can demonstrate such a high level of stability in past performance, by gathering data over a period of time, that the required 'theory' is merely that the process is stable, and thus past performance is a guide to the future. We can get short-term predictability in weather variables such as temperature, windspeed, and rainfall - because they are due to a feature such as a weather front which persists for days, or a thunderstorm that persists for an hour. We can do better than that using our knowledge on how features move and otherwise change over time in various circumstances. Much of the benefit in computer models for weather forecasting is because they can capture that, although they may be thrown off by gaps in real data about such things as moisture levels. This is why the 'why?' question can be important in analytic problems - we can benefit from knowing more about the causal influences on whatever it is we wish to make predictions about. In industrial process improvement work, it is commonplace to build simple 'cause and effect' diagrams to display and share ideas about what might be important factors than need to be controlled. In meteorology, we have a high level of control over measurement processes (where to site thermometers, how to calibrate equipment, and so on) (See http://www.surfacestations.org/ for just how sloppy some of that has become in many weather stations, however). But of course, pace the CO2 enthusiasts, we have next to no control over the climate system we are observing.

Jan 27, 2011 at 9:20 AM | Unregistered CommenterJohn Shade

@Jane Coles

And, while you can have a justified true belief that P, you still cannot say that you know that P.

Climate Science has been examined from many different perspectives. Yet, to my knowledge, there hasn't been a single, purely epistemological examination, until Jane's two paragrapher.

I've often been tempted to fill the void but I'm no expert on the matter, having done only a couple of courses way back. My hope was that someone more knowledgeable would take the lead on this.

For the curious, here is a brief overview:

Epistemology is the study of theory of knowledge. The elements that make up knowledge were formulated by Aristotle. Knowledge is justified true belief. To know a proposition P means, 1- the P is true 2- you must believe that P is true 3- you must have justification to believe that P is true. So, Knowledge=Justification+True+Belief (K=JTB). As to what amounts to J or T or B, well, their dictionary definition would be a good place to start reading.

As a field of philosophical inquiry, Epistemology remained unchanged from the times of Ancient Greeks until 1960s. The field was considered dead, exhausted, nothing new to be learned from. Then comes along a scholar by the name of Edmund Gettier. In a simple, elegant, three-page paper, he formulates several logical propositions that show even if you have JTB, you still do not have K. There is an element that is still missing from the formulation. Our knowledge is not as complete as we thought.

In short time, Epistemology is re-invigorated and universities start offering courses again. And this enables yours truly to know something about theory of knowledge, or to know that he couldn't possibly say he knows something with JTB alone. It makes you feel dizzy, I know.

Jan 27, 2011 at 12:09 PM | Unregistered CommentersHx

And if I were to examine Climate Science from and epistemological perspective, where would I start from? There is just so much to be skeptical about the CAGW science.

Jan 27, 2011 at 12:13 PM | Unregistered CommentersHx

Neal Asher asks:

So how can they possibly know which portion of the CO2 rise is anthropogenic and which is a result of the Medieval Warming? How can they even identify anthropogenic CO2? Just asking. Please correct any mistakes here.

WRT the identification of anthropogenic CO2, this is done by isotopic fingerprinting. The official take on the science is laid out at RealClimate here:

http://www.realclimate.org/index.php/archives/2004/12/how-do-we-know-that-recent-cosub2sub-increases-are-due-to-human-activities-updated/

And Roy Spencer voices his reservations about it here:

http://wattsupwiththat.com/2008/01/28/spencer-pt2-more-co2-peculiarities-the-c13c12-isotope-ratio/

Jan 27, 2011 at 1:11 PM | Unregistered CommenterBBD

Philip, Breath,

I like the word Puffball to describe a scientist whose output lacks anything quantitative. This helps distinguish between those who write "a change in motion is proportional to the motive force impressed and takes place along the straight line in which that force is impressed" (Professor Newton of Cambridge Uni, 1687), and "I mean there are some things in science, you know, gravity will remain roughly the same, there will be lots of things in science that remain the same." (Professor Anderson of the Tyndall Centre, 2009).

Would it be rude to refer to the latter as "puffballs"?

Jan 27, 2011 at 8:06 PM | Unregistered CommenterBrent Hargreaves

@sHx

Yes, my formulation was intentionally neutral on the K=JTB equation, although my parenthetical commits me to the view that J and T are necessary conditions for K. And thanks for the web link to the Gettier paper -- I read it many years ago and had retained the skeleton of the argument in memory but not the source.

Jan 27, 2011 at 9:32 PM | Unregistered CommenterJane Coles

Physics World Blog from James Darcy:

"New Royal Society president explores Climategate in documentary"

http://physicsworld.com/blog/2011/01/new_royal_society_president_ex.html

Jan 27, 2011 at 10:58 PM | Unregistered CommenterJonathan Drake

@Jane

B is necessary, too. If you just have Justification and Truth but no Belief, you have no knowledge.

BT without J is not sufficient for K.

JT without B is not sufficient.

JB without T is not sufficient.

And, as Gettier shows, even JTB is not sufficient for K.

Jan 29, 2011 at 2:51 AM | Unregistered CommentersHx

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Post:
 
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>