There Is Authority And Then There Is Authority
As we start this new year and new decade we are witness to one of the most momentous failures of leadership and authority ever seen by humanity. It is global in scope and its consequences are going to be devastating for the vast majority of humans alive today. Happy New Year!
We should all make a New Year's resolution to publicly and profusely question the so-called authorities.
By authority I mean those who possess special knowledge and understanding and whose word may be trusted in making decisions, not those who have some fiat right to order others around. Let's differentiate, then, between two major categories of authority. I make the distinction between authority by virtue of science and that by virtue of opinion and ideology. And let me be very clear here. The former does not mean people in positions of authority (leaders) who pervert and use supposed scientific knowledge to their own purposes. People like Adolph Hitler used their interpretation of science to further their own political goals. They did not base the setting of those goals on science (in this case biology) but they reinterpreted what was known at the time to rationalize their own means and ends. His authority did not derive at all from science. I need to make that clear because all too often I hear from people who want to use Hitler or some similar figure as proof that science isn't the most legitimate source of understanding. I maintain that such people are, themselves, using a bad example to merely justify their own beliefs.
Authority by virtue of science is not absolute nor should it ever be considered so. It is all relative and the knowledge always provisional. But as compared with authority by ideology it is, in my opinion, quite superior. Here is why.
A person's opinion is only as veridical as the evidence they use to form it, and the power of their brain to form it well. The former is at the core of scientific discovery and codification of knowledge. Science is a process. It is not pat formula for exposing truth. Science is conducted by us mere mortal and less than sapient human beings so, from time to time, gets stuck in a backwater of invalid suppositions. People make mistakes. And large groups of people can be mistaken for a while. But the beauty of science, and the reason I hold it in such high regard, is that it is a self-correcting process. Despite the foibles of individuals, eventually the truth will out. A theory that is based on false supposition will not last for long. The beauty of science is that so many flawed people following an algorithmic process that attacks suppositions when the empirical evidence no longer supports them eventually discover more valid understanding. The scientists don't even have to intend to overturn a paradigm. All they have to do is ask a series of little questions that may even seek to verify a theory, but finds contrary indications that force bigger questions to be asked.
In other words, science, without a direct intention, questions authority continually. It is not dependent on particular individuals — the roles of people like Einstein or Newton are generally poorly understood by the lay person. It depends only on a persistent testing of suppositions and previous findings. It depends on a process of replication of results that, when they fail, cause the participants to wonder why.
In contrast are the opinions of political and social pundits, especially those in various media (including blogs!), or religious leaders. They are all subject to belief in dogma. There is a declared generally accepted conventional understanding which they apply over and over to the real world as if to explain what is happening. They know what they know because they are privy to special knowledge somehow. And by virtue of repeated proclamations of their own privileged position, they bombard the rest of us with the 'should's and 'musts' that we are to believe, just because they said so.
What is transpiring in our world is an ungluing of the conventional wisdom of these authorities. Their pat formulas no longer hold up in light of experience. They are plain wrong on most counts because they only know how to apply rules that seemed true during the era of energy flow expansion to our current situation (where energy flow is in contraction). And they are failing right and left to predict accurately what the future is bringing. They no longer have any moral or legitimate governance authority because they simply do not understand what is happening. Worse yet, they are making the whole situation more desperate by holding to their opinions and authority.
As you might gather, I lump neoclassical economists into this category of know-nothing authorities. Their work is not science. For the most part, and the exceptions are significant but rare, they assert they know how things work and provide policy recommendations based on that so-called knowledge. But in reality they refuse to admit that they really don't understand what is going on.
Similarly, political scientists who insist that more old-school sociological approaches to understanding the political process is all one needs have failed miserably to grasp the significance of the evolution of democracy and/or other political forms toward collapsing civilization. Instead it takes people like Joe Tainter ("The Collapse of Complex Societies"), or Jared Diamond ("Collapse: How Societies Choose to Fail or Succeed"), using scientific approaches, to explain large-scale political phenomena.
Finally (for this screed) I grow increasingly dismayed by the positions taken by world religious leaders who may honestly believe that they possess some special power of insight into realms of the non-physical (which they cannot even prove exist) and thus derive some kind of authority over our moral beings. Rare is the religious leader, like the current Dalai Lama of Tibet, who can admit that they might be wrong about the dogma. He has stated that if science shows the ancient beliefs to be in error, then they must be and a new interpretation is needed. Most religious leaders maintain their special knowledge privileges them to pontificate (and I use that term purposefully).
Don't get me wrong. I am not against the idea of a spiritual quest. I feel that I am on such a quest myself. I just don't have any reason to believe that the term spirit need imply a non-material yet causally connected realm. The realm of the mind is complex and mysterious enough. It is fascinating to explore as long as you don't get caught up in contra-scientific beliefs.
So I leave you with this thought. All authority should be questioned, and held accountable for their pronouncements that lead to actions and consequences. That includes scientific as well as opinion-based authority. That includes self-proclaimed as well as governace-based authority. All authority. People have too often been wrong in the past, so the likelihood that some of them are wrong now is pretty high.
Of the two basic kinds of authority I put more stock into that kind that is able to produce evidence for why they make the pronouncements they do. When a scientist says that human caused CO2 is causing global warming I am far more inclined to believe it is likely correct than when a politician proclaims that it is all a hoax. I will strongly question the latter, but still question the former depending on how solid the evidence appears.
Ultimately we all have to live our lives and that means not spending all of our time questioning everything constantly. There are some things we have to accept with some amount of faith simply because it is impossible to be skeptical of everything at all times. We'd never make any decisions, nor would we ever get anything done. The trick is to figure out what kinds of authority you are more willing to accept (provisionally, of course) more readily. The events of my life suggest to me that accepting scientific authority is much more likely to pay off than accepting the blabbering opinionated (talking heads) kind we see in politics and in the media.
You have to decide for yourself, of course. It's your life. But that's my opinion and I'm sticking with it (for now!)
Sincerely, I hope you have as good a year as is possible given all that is happening. I believe your decisions and actions will have a lot to do with whether that comes true or not. Life can still be good despite the stupidity and foolishness of others around us. Make it be the case.
Yet another excellent, thought-provoking post.
"The beauty of science is that so many flawed people following an algorithmic process that attacks suppositions when the empirical evidence no longer supports them eventually discover more valid understanding."
I have doubts about the value of some empirical evidence - which goes back to your previous post on the value of ideology. It seems to me that chaotic systems like the economy and the climate cannot be understood by analysing the empirical evidence alone, nor can their future behaviour be predicted. Therefore a hypothesis cannot be tested. Does this render such systems beyond scientific understanding? I suspect, that this is (at least partially) the case. Which is where ideology comes in...
As I was saying to someone the other day, I have no problem with the world taking the precaution of cutting down on its carbon emissions - we surely want to avoid trashing the planet any further anyway. The difficulty with accepting 'scientific authority' on the climate is that it will logically take us in the direction of launching giant mirrors into space to cool the planet by the scientifically calculated 2.3 degrees C, on the basis of the measured temperature of car parks over a small number of years or decades, that have been 'corrected' arbitrarily... Tell me why this cannot happen!
Posted by: David | January 02, 2010 at 01:50 AM
David,
I guess I need to understand more about your background and exposure to chaos theory and its applications in physical systems. You seem to want to use a popular conceptualization of chaos as an excuse to not pursue scientific understanding of certain phenomena. And I don't think this is valid, if the case.
Briefly, there are many useful things that can be determined in the behavior of chaotic systems. It is true you cannot make detailed specific predictions about the state of a system at any particular time in the future, and the further into the future you are trying to see the less likely that any 'prediction' you make is going to be off. OTOH that isn't the typical approach to chaotic systems. The useful things that you can say about these systems have to do with their macroscopic qualities rather than their details.
It is the same as the difference between temperature and kinetic motion details of molecules in, say, a container of gas. We cannot say anything useful or detailed about the latter. But we can say a great deal about the former and it is very useful information.
Similarly the difference between climate and weather. No climate scientist is attempting to predict weather. They are, however, able to say very useful things about the large-scale behavior of the climate system. In particular, and as it pertains to warming and climate change, they are demonstrating, both in models and with empirical data, that both the mean global temperature (as well as regional means) and the variance are changing very rapidly compared with prehistoric and historic times. This means that the system is non-stationary. The signature of some process contributing to this, other than the already accounted for solar, Milankovitch cycles, etc. is the rate of these changes (or actually the acceleration). The only other phenomenon that we know about that has a similar rate of change is that of human generated CO2 from burning fossil fuels. Unless someone can find another process that provides the forcing that CO2 does and is operating at the same speed, then CO2 is our best candidate.
I don't think you can selectively choose which parts of science you will accept as valid and which parts you won't. Science is self-correcting. If a mistake was made, it will eventually be brought to light. In the meantime you provisionally accept the findings. As I said above, that doesn't mean blind faith or never questioning. But the questions cannot be shotgun pellets based on a general supposition such as chaos renders prediction impossible. It is much more nuanced than that.
Posted by: George Mobus | January 02, 2010 at 06:47 AM
George, do we really know that CO2 is a forcing mechanism? If we assume that it is (because we have demonstrated it in a simple common sense table top experiment) and build it into our models then of course our models will confirm that it is. What do we mean by model? A finite element analysis kind of thing with an approximation of the sun, ocean, land and atmosphere? Good luck with that! But I don't think the CRU idea of 'model' is even that sophisticated.
Personally it would never have even occurred to me that solar wind might have an effect on cloud formation. Maybe it does, and maybe it doesn't, but what else are we not building into our models?
Is the climate really any different from the weather except for timescale? I've seen graphs that show massive swings in temperature in prehistory with incredibly rapid acceleration. Are the graphs accurate? I have no idea because they were probably derived from tree rings or ice cores; no one really knows whether they are a true record or not.
But as I said in my comment, I'm not worried about a theory that says that CO2 is changing the climate. The problem is that "the science is settled" and it's political. The scientists may think that their work is contributing to the world changing its lightbulbs and re-using carrier bags, but in 20 years time it may be being used for completely different purposes e.g. justifying a war with China.
Posted by: David | January 02, 2010 at 11:14 AM
David,
I am not a climate scientist so I will not attempt to go into the points you raise. There are good web sites that address each of the issues and provide better information on the nature of models, etc. It does bother me a bit that you say something like 'good luck with that' without explaining why you think such models are not worthy of paying attention to.
As for 'the science is settled', I promised to be posting a blog soon on this and other misinterpretations of human scientists' rhetoric. Like everyone else, scientists will seek quick phrases that seem to catch their intentions but are too often misinterpreted by non-scientists who don't have the background to understand the differences. I will try to put in my explanation soon.
I have to say that your comments still carry more the tone of an ideologue's rationalization than a true skeptic. What I mean by that will also be covered in the forthcoming blog. On the one hand you purport to have an open mind but then introduce a supposition that goes way beyond the evidence (war with China???) and seem to be implying that that is a good reason to be 'cautious' interpreting the science. Sorry, but I fail to see connections here.
Keep tuned to see if what I have to say makes any sense to you.
PS. I do appreciate your comments and questions in spite of my suspicions about your inner motives ;^)
Posted by: George Mobus | January 02, 2010 at 12:48 PM
George, many thanks for taking the time to respond to my ramblings.
My flippant "Good luck with that!" was intended to imply that I don't believe that it is possible to get anywhere near modelling the climate from first principles in any meaningful way. By that, I mean genuinely construct a physical model from first principles and find that it produces wind, clouds, rain, snow, fog and a mean temperature across the planet of 18 degrees C in 2010 AD (I guessed that value). Sure, we can fit a multi-dimensional surface to historical data (derived from 'proxies') and predict some future values from it, but surely we don't believe that is anything but wishful thinking..? Between those two extremes is a sort of 'hybrid' model where we accept that we can't predict anything in detail, but we think we know the general principles ("CO2 is probably a forcing mechanism", "Progressive taxation results in a more prosperous economy"). But without the ability to test the model against predictions, what is the difference between this and 'ideology'?
I admit that I have nothing to back up my scepticism, any more than I have proof that we cannot construct a model of consciousness. (As an aside, could you, without recourse to anything that resembles ideology, tell me why consciousness cannot be modelled yet the planet's climate can..?!)
I'm not dancing on the head of a pin, honestly: I have a few years experience in pattern recognition and related fields and I am used to the 'panglossian' enthusiasm of the uninitiated; I used to be infected with it myself, but not any more!
I'm slightly perturbed that you don't see any connection between the political appropriation of the now universally-perceived-as- "settled" science of climate change and its use in future trade-, or worse, wars with China. My sense, post-Copenhagen, is that the West is ramping up its rhetoric against China, and I can think of several reasons why it would want to do that. You may know that the science is never "settled", but the average voter is not that astute; tell him or her that you have authoritative proof that Iraq has WMDs, for example, and they will not bat an eyelid as you go to war, as we have seen, and if we take the implications of Climate Change at face value, Islamic terrorism is hardly even worth mentioning. The story is that China is about to submerge several island nations, and turn most of the planet into a desert. The UK government is spending millions in public money to broadcast Climate Change advertisments during prime-time TV, and the BBC is on record as saying that it does not feel justified in airing 'sceptical' views because the science is settled. It's too late now to protest otherwise!
Posted by: David | January 02, 2010 at 06:15 PM
P.S. if I am an ideologue trying to rationalise my prejudices, I don't know what they are. I think it may be something to do with the fact that I am, to some extent, self-taught: I've been tinkering with science and engineering since childhood. At university, and subsequently, I met people who got far by doing things 'by the book' but without any real understanding of what they were doing. My illusions were shattered, perhaps.
When I see someone take a pile of fly-ridden maggoty data, and by the miracle of some statistical formulae turn it into a shiny paper full of precise predictions I want to scream.
My natural inclination has always been to marvel (and worry) that the earth is able to yield the energy and materials to produce all the unnecessary rubbish that people think they want - but I can appreciate the miracles that the free market produces (a secondhand laptop costing $100 can replace all the equipment in a recording studio of 20 years ago! we all now have Dan Dare videophones in our pockets etc.)
It is no surprise to me that the oil may finally be running out, or that the Chinese have managed to corner the market in rare earth metals (oh yes). I have no desire to own a 4x4 or even a new car, nor to possess a granite counter top in my kitchen.
Posted by: David | January 02, 2010 at 07:58 PM
I have been involved in more than a a few arguments with AGW sceptics on forums and my rhetorical attitude now is that I dont care whether AGW is right or wrong because the steps we should be taking should be exactly the same in either case. It is the 20th century version of angels on heads of pins...are the ice caps melting or not?
If CO2 has nothing to do with temperature why did the earths living ecosystem homostatically evolve to pump huge quantities of CO2 out of the atmosphere and deposit them as geological layers of limestone and fossil fuel as the suns output gradually increased by 30%?
And would 750 million cars running around the earth produce about the same amount of CO2 as a constantly erupting volcano?!
I could go on and on...... apologies this is not having a go at you personally David
:-)
Posted by: GaryA | January 03, 2010 at 05:01 AM
GM said "I have to say that your comments still carry more the tone of an ideologue's rationalization than a true skeptic. What I mean by that will also be covered in the forthcoming blog. On the one hand you purport to have an open mind but then introduce a supposition that goes way beyond the evidence (war with China???) and seem to be implying that that is a good reason to be 'cautious' interpreting the science. Sorry, but I fail to see connections here."
George,
OK I've re-read my two post-pub comments from last night in the cold light of day, and from the available evidence, if I were you, I would draw the following conclusions:
"Here is a person who 'tinkered' with science when he was young, which made him feel special and gives him the illusion that he has some special insight gained through experience. However, he was outshone and overtaken by his contemporaries in academia, and in the career stakes, and now bears a grudge against those perceived impostors and the high esteem in which they are held. He is using the potential for future misinterpretation of scientific certainty merely as a prop to dismiss the much-lauded work of much more successful rivals."
Anywhere near what you were thinking? :-)
Posted by: David | January 03, 2010 at 06:04 AM
David,
I appreciate your sharing more background. It does help to understand your arguments better.
As to your last comment re: what supposition I should have derived from your comments, I must say that was much more than consciously entered my brain! My inference that you were (perhaps) ideologically motivated came from your use of key phrases often repeated by acknowledged right-wing ideologues in blogs, newsprint (the WSJ editorial pages), and talking heads (e.g. Tony Blankly on Left, Right, and Center). That rhetoric came from somewhere. Perhaps you unconsciously picked it up, I don't know.
Now regarding your comment re: modeling, etc. First be warned. If you didn't already gather from prior writings I am a modeler myself, so I probably harbor some prejudices that favor modeling efforts to try to get a handle on how reality works.
Since I am also a theoretician I do tend to start from first principles, as you say. If you read my Dec. 18 blog you will have gotten introduced to such a model of economics - an abstract economy or a general theory of economic systems dynamics (http://questioneverything.typepad.com/question_everything/2009/12/economic-dynamics-and-the-real-danger.html ). In such a model there is no (initial) attempt to fit curves to data. Rather you are trying to decipher what the overall behavior of an ideal system would be given the constraints of those first principles and the interactions between the physical entities. Later, after testing the limits of the model using a range of constant values that are at least plausible if not accurate, you then start refining the model by breaking things up into components and attempting to apply more realistic sub-constants to those. This while still being true to first principles. If you have successfully applied the decomposition procedure and your model hasn't yet blown up (become unstable of collapse) then you start looking for real world data that will allow you to estimate the real constant values to see if the model begins to postdict what has been observed empirically. And then continue the refining process, which can include adding variables that represent further decomposed processes, etc.
Those of us who engage in this approach don't think our data is maggety! Adjectives like that probably don't help the discourse along. :^)
Now to some specific points.
You said of modeling:
"...but surely we don't believe that is anything but wishful thinking..?"
Why? Why shouldn't we think that the model outputs might be giving us clues (not facts) that will do two things: 1) show us where we need more data; 2) give us guidance regarding what we might expect from the future -- scenarios? To a an outsider who may have a strong resentment (if your last comment accurately reflects your past) toward scientists (or science as practiced) I can appreciate that it is possible to impute motives to them that imply they are hoping for results that will get them more research money and reputation. I can't say that that isn't at least a little bit of what is going on. But of all the people I know who are engaged in climate modeling or any other kind of modeling like it, I would have to say that is far in the background. You should talk to some of these people privately. They will tell you honestly of their sincere fears about the future. They are not telling the whole gory story in their scientific papers, but they are starting to speak out publicly, driven by that fear. Hardly wishful thinking on their part.
This one particularly caught my eye:
"...we think we know the general principles ("CO2 is probably a forcing mechanism", "Progressive taxation results in a more prosperous economy")."
Basically I think a very strong argument can be made about the differences between a principle from physics and one from economics. Actually I'm not sure where you got that tax proposition but it doesn't matter. Econ theory is largely constructed from a desire to look like physics but were never actually tested against reality until very recently. You're point that "we think we know the general principles..." is valid, but there is a huge difference between what the traditional economists think they know as general principles and what physicists think they know as general principles. And this is why traditional economics is falling apart. So far we don't see physics falling apart, though it does get refined.
And this:
"...tell me why consciousness cannot be modelled yet the planet's climate can..?!"
Apples and oranges! We have a set of first principles and lots of data for the oranges and almost none for the apples.
As a matter of fact we are making inroads toward a better understanding of the phenomenon we generally call consciousness. If you read my series on sapience you would have run into some of this. Among other problems there is the simple fact that there isn't much agreement about what consciousness is, let alone how it might work. But there is growing resolution of these matters. I fully expect that some day we will be able to model consciousness.
Meanwhile there is essentially universal agreement on what constitutes climate (including hydrological conditions). It isn't hard to delineate what the relevant factors are or what the basic physics and chemistry are. That is, in fact, how scientists determined what was happening to the ozone layer.
So I would count this argument as something of a straw man. I hope nothing I said sounds ideological!
As to your last point re: relations with China. I wonder if you haven't inverted the causal chain here. You seem to assume that we want to justify our actions vis-a-vis China and have already decided what those should be. I can certainly understand having this suspicion given the actions of the last administration (WMD and all that jazz). But since we have facts on the ground regarding warming (empirical, not modeled) and CO2 generation (China now surpasses us) I would be very surprised if we are guilty of 'fixing the policy' to achieve a subterranean end. Not being privy to this administration's motives or thinking (other than witnessing their horrible foolishness re: economics) I cannot say more than this. I try not to come to conclusions about motives without substantial evidence.
Well I could have written a complete blog post on all of this. Wait, I think I just did!
Hope some of this clarifies my thinking for you.
George
Posted by: George Mobus | January 03, 2010 at 09:53 AM
GaryA,
I share your reticence to get into rhetorical debates anymore. I too spent no small amount of typing time to trying to argue with deniers (which I differentiate from honest skeptics) and have since given up.
David poses some legitimate concerns that should be thought about in order to form some background framework when one is skeptical and seeks some sort of resolution.
As I said, models aside, it is ultimately an empirical question (AGW or natural warming, or warming at all). We will find out eventually. My concern is that if it is the former, we will find out too late to take any meaningful action. Actually I think that is already the case.
George
Posted by: George Mobus | January 03, 2010 at 10:09 AM
It seems there is much of a climate debate seething beneath the surface in this blog.
It would be nice if someone could clarify the points raised in the following video:
http://www.youtube.com/watch?v=4zOXmJ4jd-8&feature=player_embedded
When Einstein was informed that the Germans had assembled two hundred scientists who had reached a "consensus" that he was wrong, he noted that it would take just one scientist to prove him wrong.
Posted by: Robin Datta | January 03, 2010 at 11:54 AM
George, like a politician I was very careful with the wording of my self-analysis saying that it was the logical conclusion to draw from the "available evidence". I don't think my description of myself is entirely complete or accurate, but it may have a grain of truth in it! I am interested that you equate 'scepticism' with the right; this seems like a generalisation too far. In several comments I, myself, have gone out of my way to state my left wing credentials i.e. I want to see an end to economic growth and to the trashing of the planet (if, indeed, that is an exclusively left wing desire). I may have read some 'denialist' blogs as well as some 'warmist' (can we say 'conformist'?) ones, but not the one you mention.
Today I have been thinking about what unusual "insights", if any, I might have gained from my self-taught dabbling in science and engineering when I was young. It was staring me in the face but I couldn't see it until I recalled a time when I was once involved in patenting a certain invention (of mine) that involved a blend of electronics, software and its interaction with an array of sensors in the 'real world'. I had to liaise with the patent attorney (or agent?) who specialised in electronic devices to help him draw up the documents. Try as I might, I could not make him understand how such a system could work: what the sensors were measuring, how the hardware was 'dumb' without the software, and how the software enabled the system to self-calibrate, and multiplex the sensors and perform the necessary calculations. This continued when the system was sold under licence to an engineering company who sent over their hardware and software engineers to talk to me. As I found out, there was virtually no overlap between their respective specialities which meant that the hardware engineer just could not grasp how the software could transform a bunch of mainly cheap components into something accurate, and the software engineer just couldn't understand the interaction with the hardware and the maths involved. I came away from that experience realising that 'generalists' like me see the world of science and engineering in a very different way from specialists, and what seems plainly obvious to me, they do not see *at all* (and vice versa..?).
So that's my special insight, I think. I might vainly consider myself to be a bit of a 'renaissance man', or should that be "jack of all trades and master of none". Either way, I don't have a resentment of scientists as such; more a distrust of specialists. My intuition is that there are too few generalists capable of *understanding* what the data as a whole means - or more likely what its limitations are. Each specialist can convince themself and other similar specialists that they are finding the truth, the whole truth and nothing but the truth. And the corollary to that is that specialists in other disciplines simply 'take their word for it'. You, yourself started a post saying "I'm not a climate scientist so I can't comment on...". I can't prove any of this of course, but unlike you, I think that the economy is the classic example of when a lack of overall understanding becomes clear much to everyone's embarassment. Each economic specialist seemed able to prove that what they were doing was right. Their models tallied with the past and came up trumps when making predictions, so that they genuinely believed that they had modelled risk, for example, correctly.
You didn't like my comparison between the modelling of climate and consciousness. What about a much simpler one? Supposing I presented you with a PC and asked you to make a reasonable model of it for me from first principles, and predict the output given certain test patterns. How would you tackle it without cloning it? You might deduce that the "general principle" was that of binary digits, and you could suggest that there was a CPU, memory and so on and build models of those. How big? Well, make a guess. But what about the software? And the initial conditions? Get one bit wrong, and the machine doesn't work at all, or does something completely unexpected. From observation, and feeding in test patterns, how would you establish each byte of code? Clearly impossible but only because, and even though, *we know what we're dealing with already*. What proof is there that a climate model is anything but a fantasy?
Would a climate model from 'first principles' (or as close as we could get) contain 'accurate' representations for undersea vents, volcanoes, ocean trenches, cloud formations of different types and at different altitudes? Fish? What of the much-vaunted effect of sunspots? Vegetation dying from drought? Animals dying from drought? Forest fires? The more I think about it, the more my head spins at the futility of attempting to truly model the climate from first principles and expect anything but noise from it. Or are we *sure* that these things can be lumped together into just a few constants and factors? It's my bet that the first line of code would be "// Assumptions: Animals are not considered. All vegetation is assumed to have the same characteristics. One large volcano erupts every 10 years; there are no other volcanoes. Sunspots have no effect on climate" etc. Any reviewer looking at this would say "Hmm. Looks reasonable." and move on. At most they might say, "Hang on a minute, I'm sure it's 8 years between volcanoes". What they won't do, is ask "Just out of interest, what would happen if three volcanoes erupted simultaneously?". They certainly won't ask "Might sunspots actually affect cloud formation?"!
Of course the other approach is the multi-dimensional surface fitted to historical data. How many dimensions do we need? How much data do we need? What level of precision? Does it matter if the data is (are) a bit 'lumpy'? What if we've got very few temperatures from the sea? Again my head is spinning at the futility of it.
Anyway, sorry to go on. Hope you don't mind my posting these rather un-structured comments. Very much enjoying talking to you.
David
Posted by: David | January 03, 2010 at 02:44 PM
I'm afraid dear old swivel-eyed buffoon and professional Obfuscator Lord hawhaw Monckton (a well known British denier clown) Would be taken a tad more seriously if his scientific qualifications amounted to more than a degree in classics and a diploma in journalism....
Posted by: GaryA | January 03, 2010 at 02:45 PM
Robin,
It will have to be someone else stepping up to the plate. I'm afraid I have my hands plenty full. If someone takes up your challenge and posts a non-ideological answer to these supposed questions I will allow it to go through.
But I must say that mere comments about AGW and climate change that are clearly not scientifically based (and I mean with backup data, publications, etc.) I will begin removing because this is not a site for carrying on a political debate. I appreciate good questions, but not politically motivated ones.
So if anyone reading this wants to take on the task, feel free to do so. If your response is science-based and factual it will be published. Otherwise not.
Robin, this is, I think, the fourth time (at least the third) time you have made this terse demand, linking to a video some of us have already found to be unworthy of scientific debate. I will ask you to stop this tactic. If you have anything specific from a scientific paper, then feel free to post it.
George
Posted by: George Mobus | January 03, 2010 at 04:13 PM
David,
Lots there. I will only be responding to a few items. Maybe sometime in the future you can raise a few issues again for discussion. Tomorrow school starts and I need to spend some time with family before the rat race takes off.
Incidentally, I have composed a blog post for tomorrow wherein some of my points relevant to our discussion are made. I even mention you and our discussion in it!
"Each economic specialist seemed able to ***prove*** that what they were doing was right. Their models tallied with the past and came up trumps when making predictions, so that they genuinely believed that they had modelled risk, for example, correctly."
I've highlighted the word prove because, as I say in tomorrow's blog, scientists don't prove anything. They find evidence that strengthens the case for a proposition but their case is always provisional (specific individual scientists given to hyperbole notwithstanding). Also in that blog I will point out that economists have not actually been scientists. I'll explain that. In fact, most of the predictions made by economist have proven wildly wrong over many years. The fact that politicians have relied on their pronouncements relative to policy simply shows how uninformed and (dare I say it) stupid they are.
"Supposing I presented you with a PC and asked you to make a reasonable model of it for me from first principles, and predict the output given certain test patterns. How would you tackle it without cloning it?"
This one is easy. We do this all the time and do not resort to cloning, as you put it. We are able quite well to replicate function of any computer and computer program given the input/output specifications and the first principles of logic gates. In fact there is a theorem that demonstrates that anything that can be specified in hardware can, in principle be built in software (which is how we emulate multiple machines as virtual CPUs) and conversely anything that can be done in software can, in principle, be built in hardware. Then there is the famous universal Turing Machine theorem (essentially the same thing). So I'm afraid your example isn't quite kosher! Had you opened up the nature of input to be stochastic (e.g. not fully specified in detail) then you would be dealing with a different kind of machine, called an Interaction Machine. Then some of your points would be better served. It is impossible, even in principle, to design a machine that is guaranteed to produce specific output when the input patterns are both stochastic in content and sequence. But then you are dealing with a different problem. BTW, a PC is an interaction machine by virtue of the impossibility of specifying all possible user inputs. This is why our computers lock up from time to time.
"...even though, *we know what we're dealing with already*. What proof is there that a climate model is anything but a fantasy? "
As per above, your emphasized phrase is in error. By definition, in an Interaction Machine, we cannot know what we are dealing with. Then you use the word 'proof' again. I will save anything more I have to say about this until tomorrow. But it seems to me that you are awfully anxious to build a case against climate science on the basis of modeling. Yet I don't get that you have actually built or participated in the building of a complex dynamic model. I suspect you would have a little bit different insight into what goes into such activity, and realize that this isn't a case of people constructing a fun fantasy, if you had gotten your hands dirty in the details. Correct me if I'm wrong, but using terms like fantasy shows a lot of disrespect for the work other people do with honesty and good intentions toward understanding how the world works. It sort of wrangles.
You go on to describe something that is, quite frankly, irrational and never even a part of modeling from first principles. If you look at my abstract econ model you will not find anything like details you seem to think should emerge from a model if it correctly models the phenomenon. This is, I hate to say it, plain silly. Nor should you expect any greater prediction capability by adding in such details explicitly. You do know what validation and verification procedures are, don't you. You have heard of sensitivity analysis. Your description is wildly off base. I suggest you re-read it and reflect on what you are trying to say and especially on your motives for saying it.
Sorry to sound like I'm haranguing you. But it does sound to me like to have a priori decided that attempts to understand the climate from modeling are futile (possibly because you do not really understand what goes on in building models in detail) and are then seeking excuses to show why you are right.
Rather than dabbling in science I suggest you plunge into it. Tear apart some aspect of some of the climate models in detail and demonstrate the errors in how they are arriving at their scenarios. Don't just question their methods without doing some real digging into those methods and finding what is really wrong with them. I listen to people (even students!) who raise questions about some specific mechanism in one of my models and then can go on to explicate why its a problem. And that has happened many times. It is then incumbent on me to either do a better job of making a valid explanation, or to change the model as indicated. And I have done both.
But you can't criticize other peoples efforts on vague guesses that something doesn't smell right, or just seems to you to not make sense. You have to go the next step and show why.
George
Posted by: George Mobus | January 03, 2010 at 04:53 PM
Happy New Year George,
Thank you for continuing on your quest and sharing thoughts that matter. There are very few things that I look forward to this year, your thoughts are one of them.
Posted by: Sudeep Bhaumick | January 03, 2010 at 07:53 PM
Hi George. I'll try not to be sloppy in future! When I talk of people seeming able to "prove" something I am talking of how the general public and politicians see it. I don't think the finer points of what counts as proof to scientists are in question here. Please don't feel you have to respond to these rants, but if you spot a juicy morsel don't worry about haranguing me!
My example of trying to model the behaviour of a PC (with Windows XP pre-installed!) was intended to show that even though you may know the general principle of how a system works, you may not be able to predict its behaviour if its behaviour depends on internal 'software' that you don't have access to. I can spend the rest of my life interacting with a PC via its available interfaces (keyboard, mouse etc.) and perhaps deduce enough to replicate some of the basic functions of its software - without necessarily duplicating the way they work internally. Is that a useful model? No, because it doesn't allow me to predict future behaviour in some very un-subtle ways if there is an 'input' I haven't yet tried. My point is that even though we know exactly how a PC is made we still couldn't replicate one by modelling from the outside, in practice. In the case of the climate we're guessing a whole lot more, and we are not in a position to 'prod' it with stimuli.
Which bit of my comment last night was "silly"? I presume it was the bit about explicitly including volcanoes, vegetation, snow etc. in the climate model? I'm still not convinced that it is so silly to include factors that we are told actually do affect the climate, at least in the short term. For example, I thought that flooding the stratosphere with SO2 was a proposed geo-engineering technique, based on the observation of volcanoes..? And yes, I was implying that if our model doesn't actually *produce* precipitation spontaneously then it is just another guess being fed in by the modeller; then there doesn't even seem to be much point in the model, or certainly not at the level of pretty coloured maps that show the temperature of individual countries fluctuating like a 'real' weather map. Why not just show the globe changing from a uniform blue to a uniform orange over time and be done with it? (Sorry, being flippant again)
Please don't bother to respond to the following point if it's a really stupid one, but a model of the economy would not be much use if it didn't factor in some really quite niggly little details such as oil running out/new oil discoveries, increasing longevity, changes in the retirement age, globalization made possible by improved communications, dotcom commerce, China 'creating' a large number of new workers, China cornering the market in rare earth metals, contraception (population falls in Europe), population increases in the third world, terrorism, war, flu pandemics etc. Are these particularly silly details or are they, in fact, extremely significant factors that arise 'spontaneously' and have very 'non-linear' effects?
You say that I should rip apart some climate change models and find fault with them, but the whole point of my rants recently has been to say that it's anyone's guess whether the models are right or not. It isn't a case of proving (in the correct sense) a mathematical formula or running a series of controlled experiments on a physical system.
You say I've never got my hands dirty in a relevant area, but I have, in fact, several years experience in neural networks and pattern recognition in relatively simple, controlled systems like number plate recognition or more interestingly, perhaps, crowd monitoring. The theory says one thing, but the practice is much more difficult - and by that I mean it may never work properly! It doesn't matter whether you have a Cray supercomputer at your disposal and all the training and test data you can specify, you cannot make it work, except by constraining the problem. In the case of the climate we cannot constrain the problem, and we can never know whether our model works into the future (it will take decades for a change in climate to become apparent, and in the meantime it will be buried in the noise of 'weather'). And I still haven't given up on the point that trying to spot temperature changes of 0.5 deg C using a weather station whose average temperature has changed 5 deg C because of the UHI doesn't render all your modelling in vain. (I'm a bit nervous now that you'll say that is right wing thinking...)
Posted by: David | January 04, 2010 at 03:22 PM
George,
I wonder if you have read Jeff Schmidt's "Disciplined Minds - A Critical Look at Salaried Professionals and the Soul-Battering System that Shapes their Lives"
Quote:
Disciplined Minds is a book by physicist Jeff Schmidt [1], published in the year 2000. The book describes how professionals are made; the methods of professional and graduate schools that turn eager entering students into disciplined managerial and intellectual workers that correctly perceive and apply the employer's doctrine and outlook. Schmidt uses the examples of law, medicine, and physics, and describes methods that students and professional workers can use to preserve their personalities and independent thought.
Schmidt was fired from his position of 19 years as Associate Editor at Physics Today for writing the book
end quote
A Radio reading of the book can be found at http://www.inference.phy.cam.ac.uk/sanjoy/schmidt/radio-reading/
A review: http://disciplinedminds.tripod.com/z-magazine.htm
Quote:
The status of “professional” in America indicates to the masses that you have made something of yourself. You have become one of the best and the brightest. But what sort of Faustian deal had to be made to get there? The “best and the brightest” Americans, as historian Howard Zinn has pointed out, are the people who have engineered atrocities like the Vietnam War. More recently, these engineers have been manufacturing the consent of the two biggest American historical events so far in the 21st century: the farcical 2000 U.S. presidential election and the ambiguous terror of the War on Terrorism. And where do these astute professionals come from? They are products of the American education system, of course.
In 1967, an English professor at Cal State L.A. named Jerry Farber declared in his underground classic essay “The Student as Nigger,” “Back in kindergarten you found out that teachers only love children who stand in nice straight lines. And that’s where it’s been at ever since. Nothing changes except to get worse. School becomes more and more obviously like a prison.” If that’s true, then what are the effects on the “inmates” after being there for 12, 16, or even 20 years if they want to become professionals or attain graduate degrees? Jeff Schmidt addresses the tail end of this question and explains what can be done about it in his book Disciplined Minds: A critical look at salaried professionals and the soul-battering system that shapes their lives.
Schmidt received a PhD in physics from the University of California, Irvine, taught physics around the world, and until May 31, 2000, was an articles editor at Physics Today magazine. After publishers Rowman & Littlefield released his book, Physics Today fired him, citing the book’s very existence as evidence that he was not “fully engaged” at work. Of course, that’s the main idea within the book. No sane employee in a hierarchical institution, Schmidt argues, can be fully engaged in his or her work, because the company’s interests are in conflict with the employee’s personal interests.
End Quote
Posted by: Rajiv | January 09, 2010 at 04:36 AM
David,
Look, I'm sorry that I suggested that your rhetoric resembled the right-wing ideological stuff. But, having said that, I read through your rational for why models can't be useful and I'm still perplexed. You seem to want to dismiss a whole area of integrative science in a sweeping indictment. BTW: if you look at my research you will see that my early work was on the simulation of real biological synapses in such a way that it could be used directly as a learning algorithm in a neural network system. What differentiated my work from, say backpropagation (and I knew Rumelhart when he was at U. Calif. San Diego) was that mine was an actual simulation of synaptic dynamics whereas backprop is an overly simplistic weight changing formula. Mine produced results consistent with actual learning behavior whereas backprop and most distributed representation approaches are not much more than simple pattern recognizers. My point is, most of these so-called neural networks are NOT simulations of real systems. Fine distinction.
"You say that I should rip apart some climate change models and find fault with them, but the whole point of my rants recently has been to say that it's anyone's guess whether the models are right or not."
This seems like a cop out to me. Your claim on the basis of what? There are some legitimate climate skeptics who have done some work on finding holes in the models. It has resulted in improvements in the models. That is a good outcome.
Posted by: George Mobus | January 09, 2010 at 05:22 PM
Sudeep,
Thanks.
George
Posted by: George Mobus | January 09, 2010 at 05:23 PM