There’s an interesting recent paper, that Willard will probably appreciate, called why Popper can’t resolve the debate over global warming: Problems with the uses of philosophy of science in the media and public framing of the science of global warming. The basic argument seems to be that people are using the philosophy of science arguments to try and justify their positions, but without really understanding if they’re appropriate, or not. It concludes that
studies of the science of Anthropogenic Global Warming debate would benefit from taking greater interest in questions raised by un-reflexive and politically expedient public understanding(s) of the philosophy of science of both critics and supporters of the science of Anthropogenic Global Warming.
What I found interesting about the paper is that, even amongst those who study the philosophy of science, there doesn’t appear to be broad agreement about the relevance of Popper. My own view is closest to something quoted in the paper
The classic Popperian approach to science, in which potentially refutable hypotheses are defined and tested is not well suited to the challenges posed by an Earth System that is characterised by high degrees of complexity, non-linearity and a lack of definable cause-consequence relationships. (Oldfield and Stefffen, 2014)
In fields dominated by observations, such as Astronomy and Climate Science, scientists don’t – on a daily basis at least – think in terms of Popperian falsifiability. Typically, the goal is to try and explain some system for which you have observations; in some cases, observations that you can’t really improve. You consider what physics should apply to the system and if – using that physics – you can explain the observations. Of course, in doing so, you would typically see how your assumptions influence your results, what you might be leaving out, and how this might influence your analysis.
However, if you were unable to match the observations, you wouldn’t necessarily immediately assume that your model had been falsified. Firstly, what would be falsified if your model is based on fundamental physics that has already essentially been through that test? It’s much more likely that you’ve missed something out of the model, or misunderstood some aspect of the system you’re considering. You don’t just throw it all away and start from scratch; you would consider if there is something relevant that you haven’t included. You can’t even necessarily rule out that there might be problems with the observations.
Of course, I’m not suggesting that Popperian ideas have no role to play. In fact, even something like anthropogenic global warming (AGW) could be falsified (see, for example here and here) but it would take a great deal more than a model not being able to match some observations.
I guess my basic view would be that we should avoid assuming that science advances by applying simplistic rules. The goal of science is to gain understanding of whatever it is that we’re studying. Being aware of these different philosophical ideas is very useful, and blogging about this topic has certainly improved my understanding. However, these ideas are intended to guide how we do science, but are not really intended to impose a set of rules that we never violate. In most cases, the systems we’re considering are too complex for a set of simplistic rules to be applicable.
...and Then There's Physics 
My experience is that people invoking Popper do so b/c they don’t have legitimate scientific arguments. It is used as a smokescreen, not a bona fide philosophical argument.
Andrew,
Certainly seems to be the case and seems to be essentially what this paper is suggesting.
The classic Popperian approach to science
For me this is a strawman.
Poppers philosophy was not an “approach to science”. He was working on the demarcation problem: how to distinguish between scientific hypothesis and a non-scientific idea (which can still be valuable). How determine whether quantum mechanics, Freud’s psychoanalysis, string theory, electro-magnetism or Sheldrake’s morphic fields are science?
I think, it is a valuable suggestion that it should be possible to falsify a scientific hypothesis. That means that you need to write up with sufficient clarity what your idea is.
When someone has a mental illness you can always make some story about his/her childhood. Without numbers Rupert Sheldrake’s fields are new age bunk. When you do not find that the first time an idea/molecule is developed is “harder” than the next time, you can always say that the morphic field was not strong enough. If it is true that string theory can be made to fit any observation, it is not science.
That does not mean that science should be done by naive falsificationism. That is a strawman. Popper does give such naive examples in his books, but I would see that as illustrations of the principle and not as guidance on how to do science. That was not the problem he was interested in.
My main problem with Popper’s demarcation is that it is not complete. The idea that there is a red cube on the back side of Pluto is falsifiable, but does not contribute to science. I would not call this a scientific hypothesis. If we only had Popper we would. Something is missing, but I find it hard to formulate what that is.
Victor,
What you say at the end of your comment seems a key point. My thoughts were that we have to careful of assuming that we can simply apply a set of simple rules, which seems similar to your suggestion that something is missing.
I agree with ATTP, and not just in astronomy and physics, but every single area of science without exception. Science is probabilistic decision making. Always has been, always will be. There is always a non-zero probability that any observations are wrong, any assumptions are wrong, that there are unrecognized assumptions,….
One irony is that so many of those who appeal to the authority of Popper (Feynman, etc.) so frequently accuse others of appealing to authority.
included in the irony, of course, is how selective their appeals are. They just pick and choose what confirms their biases…often in contrast to the philosophy the reference. (One of my favorites is Judith’s bastardization of Nasim Talebs’s ideas).
ATTP writes: “In fields dominated by observations, such as Astronomy and Climate Science, scientists don’t – on a daily basis at least – think in terms of Popperian falsifiability.”
And for good reason. Victor writes good things on this topic. It’s a type of “gotcha” to make you dance to the questioner’s tune. You could refuse to play that game, but as Willard writes, the only way to lose is not to play.
Falsifiability kicks in because of inductive logic. If you are using deduction then there’s nothing to falsify. The process of induction requires many guesses, most of them wild guesses, and scientists will spend quite a lot of time disproving these guesses. What is left after all that stands by itself. It still isn’t proven, but more importantly, it is unchallenged.
In the case of climate science, what you have is observations (mostly). One then uses induction to speculate on the causes of what is observed. Then one attacks these speculations. If little evidence exists that these speculations have been vigorously challenged, then every unchallenged speculation or idea or explanation of the observation is still on the table.
More to the point; these alternatives might not be on your table but will be somewhat unpredictably be found on other people’s tables (questions in their minds).
An example question on the table: How do you know, with certainty and exclusiveness, that humans are the majority driver of global warming over the past 150 years? The answer might be on SkS, but an advocacy website does not challenge its own assertions. That would be ridiculous in an advocacy website. It will sometimes pretend to it; like Chris Matthews having a weak conservative challenge a strong liberal. The outcome is pre-determined; everyone knows it and so its pointless.
ATTP writes “However, if you were unable to match the observations, you wouldn’t necessarily immediately assume that your model had been falsified.”
It is if the claim was made that the science was “settled”. To then declare that the science isn’t settled pretty much yanks the rug out from under everything that had been claimed to be settled.
I never believed it was settled; it is certainly more settled now than 10 or 15 years ago when these kinds of claims were being made. But I recognize the political utility of reminding people of these certain (but wrong) claims made 10 to 15 years ago. If they were wrong then why shall I accept they are correct now?
Thanks Victor, good comment.
re: something missing , I think Richard Feynman said something along the lines that a ‘proper’ scientific hypothesis should make a [set of predictions] that is in some way novel. I can’t find the exact quote at the moment. I’m not sure the distinction is very significant though IMO i.e. a scientific hypothesis might make trivial predictions, in which case it’s probably not a very interesting hypothesis.
I don’t understand the difference between a model being falsified and the model is missing something.
If the model is missing something, which is what makes it wrong, why is it not falsified.
Would you not change the model to add the missing something and create a different model?
I get that observations can be wrong and the issues that might create.
For example, if the speed of light is not a constant but varies with the expansion of the universe (just making that up as an example) – that would really mess with all of our observations – and I get that.
But it seems to me that if a model is found to be missing something – it is falsified by definition.
What am I missing?
Really? I’d like to see an example of this since it simply doesn’t ring true, especially in the contemporary world. Inductive scientific reasoning is always made within a knowledge framework. It doesn’t require “guesses” (let alone “wild” guesses). The interpretations of a set of observations consistent with validated theory aren’t “guesses”.
But maybe you have a different notion of inductive scientific reasoning – an example or two of what you mean would help.
Inductive reasoning involving lot of observations interpreted within a knowledge framework. Of course we don’t know with “certainty” (and “exclusiveness” ??? not sure what you mean by that), but we know with a very high degree of probability since we have a very large evidence base (that involves detailed knowledge of temporal development of [CO2], solar output, volcanic eruptions and stratospheric aerosols and so on). That knowledge exists independently of any specific repository and so SkS (for example) in presenting the evidence and scientific arguments isn’t necessarily engaging in advocacy, unless it is advocacy in support of informed and honest interpretation of the scientific evidence.
The novel-prediction concept maps pretty well to training/test sets in machine learning. If you train a machine learning algorithm on the whole data, it will overfit. If you train it on a subset, then test it on data it didn’t learn from, then the predictions it makes are novel — and if the model it learned was good, the predictions will be good, which means you have a good learning algorithm (or an easy problem).
That assertion (“the science is settled”) is catnip for those who wish to cause mischief ! It depends what one means by “the science”. The fact that mankind is releasing massive amounts of greenhouse gases into the atmosphere, that this is enhancing the greenhouse effect and that this is (and will continue to) cause the earth to warm, glaciers to melt, sea levels to rise, oceans to acidify is all settled science in the context in which the term is normally used. This is to correct misrepresentational assertions that mankind isn’t the cause of enhanced [CO2], or that enhanced [CO2] doesn’t cause the Earth to warm significantly etc.
The science is settled. But it’s not all settled nor completely settled. If one is interested in addressing this honestly, then one needs to be clear about which particular aspects of the science we’re considering.
> The science is settled. But it’s not all settled nor completely settled. If one is interested in addressing this honestly, then one needs to be clear about which particular aspects of the science we’re considering.
We could push this further:
http://plato.stanford.edu/entries/truthlikeness/
Popper’s treatment of that problem rests on a non-inductive approach. As an optimist Realist-with-a-big-R, he wasn’t fond of inductivism. His falsificationnism has always been rooted in the modus tollens instead.
Yet statistical hypothesis testing and machine learning won, which may imply that falsificationnism has indeed been falsified. But then does it mean it has been corroborated?
One needs to be able to test a given hypothesis given whatever the hypothesis assumed.
If the hypothesis is rejected, under testing, was it the hypothesis itself or it’s underlying assumptions.
You can test for the underlying assumptions, if those are shown to be valid then it is on to the hypothesis itself.
Science suffers most from assuming too much IMHO. Hmm, err, Hansen16.
All roads lead to Jim.
…”lack of definable cause-consequence relationships”? surely not!
I would have thought that the main difficulty with a “classical Popperian approach” is that we can’t do multiple controlled experiments on a series of Earths to test Earth scale hypotheses at large. Otherwise “high degrees of complexity” and “non-linearity” are not inherent limitations to a Popperian approach (and we have plenty of definable cause-consequence relationships). In any case one can apply a more “classical” Popperian approach to important sub-elements of the system. e.g. (hypothesis) sea water acidification causes undersaturation of calcium carbonates and this affects the ability of specified marine organisms to build and maintain skeletons/shells. Or (hypothesis) increased greenhouse warming under real world conditions reduces productive yields of specified plants. etc. Some of the fundamental hypotheses about the Earth system are perfectly amenable to a “classical Popperian approach” e.g. (hypothesis) enhanced atmospheric CO2 will cause the Earth to warm and sea levels to rise [note the “definable cause-consequence relationships” btw], even if the potential falsification may require long term observations (so maybe this makes it non-“classical”???)
In any case (again) the “classical Popperian approach” (“approach” isn’t the best word) of potential falsification isn’t really how science is done at the coal-face even if we all work under the expectation that our hypotheses are falsifiable. In general hypotheses are tested by exploring their consequences and any falsification occurs at a sort of second-remove (rather than by directly trying to falsify the theory which may only occur if you happen to hate the hypothesiser, or the hypothesis has consequences that interfere with your interests, and here you’re not so much trying to falsify the theory as to trash it. 🙂 )
ATTP,
thanks for the heads up – I have just written a paper on philosophical approaches to AGW and the climate wars, which was almost ready to submit. I will have to hold off submission to summarise the arguments in this new paper, which is really useful for its comprehensive documentation of where Popper has been invoked. (I was relieved to find this paper does not cover the same ground I have)
Victor upthread asks a question that Lakatos tried to answer more comprehensively than Popper did, in that something can be falsified is not necessarily science, through progressive science research programs (although Popper was quite clear about the difference between falsification and science).
But for me, the interesting areas come in with new experimentalism, particular the ideas of severe testing developed by Deborah Mayo. They invite severe testing in an inductive research program and I think the more rigorous application of such methods would improve some research considerably.
Also, what Ethan said about assumptions generally (though Hansen to his credit is talking more about risk, than the underlying science, where the benchmark is plausibility, not the higher level that the theory itself requires).
The aforementioned paper of the OP starts out (after two quotes) …
“The two quotes above, both from supporters of the scientific thesis of Anthropogenic Global
Warming (AGWS), help set the stage for the discussion to follow: Lawson’s quote captures a dominant theme that appears in numerous popular commentaries on the climate change debate, that is, Popper’s philosophy of science offers an epistemological ‘fix’ that should be used to settle the issue.”
AGWS? I get the AGW part, I don’t get the “S” part, “S” stands for Signature or System or Signal or Scenario or Situation or Sumption or …
OK I got it … Science. Nevermind.
“References to other figures in the History and Philosophy of Science: most notably T.S. Kuhn, Galileo, Richard Feynman and appeals to images of science drawn from sociology of science, and critiques of post-modern views of science also appear but are less prominent (Douglas, 2009; Kuntz, 2012; Ravetz, 2012).”
Michael2:
The portion of the science that’s most often being argued, the portion that relates to the big picture and near-term policy, is settled, and has been settled for a few decades. This is that atmospheric CO2 is rising, the globe is warming as a result and will continue to do so, and that this is likely to be dangerous and expensive, with rather small chances of “no sweat” and “apocalyptic”.
New evidence is still pouring in, of course, but very little of it contradicts the very well-established science. The physics lines up, the observations line up, and the computer models do a very good job for what they were designed for. Which is to calculate the long-term changes in the climate given a respective set of forcings. It feels like a lot of people still forget that “climate” is 30 years, and there’s quite a lot of literature about why models can’t yet do better, but how it doesn’t affect their long-term value.
Anyways. The point is “consilience”. We have a lot of evidence from independent ways of attacking this problem, and they all agree: humans are causing lots of climate change, and that might be unfavorably dangerous and expensive.
Popper is often raised in the climate debate because there are some who claim that everything is caused climate change, although a more common position is that everything bad is caused by climate change.
Until such day that climate science, or more precisely the media whores of climate science, come up with more precise forecasts of what climate change will and will not do, Popper will be there to haunt us.
“A clear example is that if the average world temperature drops one degree, back to values before 1900 and stays there for a long time without there being other reasons for the temperature decrease (e.g. volcanoes, sun, aerosols) the theory would be falsified.”
–
Agree or disagree? [chorus of complaints or a better Popper please]
–
Chris says: May 28, 2016 at 12:32 am
” the “classical Popperian approach” (“approach” isn’t the best word) of potential falsification isn’t really how science is done at the coal-face even if we all work under the expectation that our hypotheses are falsifiable.”
–
How would you really do science then?
–
Both serious questions.
One of the problems I see with arguments on both sides is the use of the word “might”, as in sea level might rise 6 meters by the end of the century for a warmist example.
Best and worst case scenarios are endlessly played out to reassure or scare people even though the chances may be vanishingly small.
As in the case of refuting AGW.
Very few people will commit to a falsifiability index in print.
It is as if an admission of such means one has lost the argument rather than or by defining the boundaries of the argument.
Richard,
I guess you haven’t bothered reading the paper that the post discusses?
Agree. It would have to drop by more than 1 oC to get to values before 1900 though, but I agree with the point – if the extraordinary amount of extra energy accrued in the climate system under enhanced greenhouse forcing were to “leak” away with no discernable cause then this would, I suppose, falsiify AGW.
I say “suppose” since it’s such a ridiculous prospect.. it would be interesting to try to think of a more realistic/likely set of circumstances that could falsify AGW – the fact that this is difficult (to me anyway – have a go!) is an indication that the theory is pretty rock solid. It’s a little like trying to think of a set of circumstances that would falsify gravity or the theory of evolution (the latter seems easier actually)!
Yes, if something like this were to happen, it would – IMO – falsify AGW.
Why is this a problem? We can’t say with certainty what will happen, hence we should be using words like “might”, “could”, etc. Ultimately, the policy side of this debate is about risk. How do we balance the possible risks associated with continuing to emit CO2 into the atmosphere and the risks associated with trying to avoid doing so. If you want certainty then science is the wrong tool.
Roger,
That sounds interesting. I look forward to reading it. If you were interested in a guest post here about it once you’ve published it, you’d be most welcome.
OK, this really requires quite a long answer, but trying to be concise:
In general most of science isn’t done to address grand theories of the type that formed the examples of for example Kuhn’s “Structure of Scientific Revolutions” (some is, especially in theoretical physics and efforts to explore the fundamental nature of the universe). So most scientists are trying to find stuff out, to make things work in a practical sense (find a druggable molecule that will slow development of Alzheimer’s; explore means of improving earthquake predictability; improve ability to identify Earth-like planets, etc.). These efforts might result in new fundamental theories that overturn previous ones but that’s not the aim which is much more positivist in outlook. We’re trying to find stuff out. In the UK we now have to include sections in our grant applications that define the real world impacts for society/human health of our research.
Of course within these seemingly more prosaic efforts, there are uncertainties that might lead to different interpretations (hypotheses which might be competing with one another). For example, some scientists may consider that a particular drug has its effects by diffusing into cell membranes and altering their physical properties whereas others might prefer evidence that the drug interacts directly with a hormone receptor in the cell membrane. In general, these scientists would not be trying to falsify each other’s hypothesis (they might!) – more likely they would be pursuing what they considered to be productive lines towards their particular research aims. In time one of the two competing interpretations may well prevail, but this will happen because evidence will tend to support one or other interpretation, not because someone has set out to falsify the other’s interpretations.
Actually this is all in accordance with Popper which (as Victor points out) is not about an “approach”, but is about categorising what is and isn’t a valid scientific theory/hypothesis/interpretation. I’m really making the point that science is unlikely to be productive if scientists were permanently engaged in trying to falsify theories! It’s interesting that the few rather vociferous individuals that oppose climate science have suggested just that – that money should be provided for those that wish to pursue efforts to falsify the science of AGW (can’t find an appropriate link unfort)!
Regarding a sustained cooling as falsification …
part of the paper argues that atmospheric warming as we are observing cannot be used to either falsify or confirm AGW. The core theory is radiative forcing of greenhouse gases will trap infrared energy and the positive feedbacks that lead to the resulting atmospheric response measuring larger than the direct radiative component.
It may be that cooling (and particularly ocean cooling at the equator) would be a sign that all is not well, but to falsify the core theory, firstly, it would have to be shown that ghgs are not trapping IR radiation and/or that there is a negative feedback instead of a positive temperature response. Secondly, the counter process – what actually does happen in those circumstances – would have to be explained.
Neither of these tasks can be carried out using statistical inference on atmospheric temperature time series. The theory can only be overturned by a better one that explains the data better, in such a way that the previous theory is highly unlikely as an alternative. This requires proper scientific process and experimentation on aspects of the core theory, not arguments about what fluctuations on graphs might mean if viewed differently.
So the argument is that a cool period is not enough in itself, but that a better account is required for falsification to occur. Even Popper argued that refutation is necessary but insufficient.
Roger,
Yes, I agree with what you say in your second paragraph. We would want to show that something fundamental (CO2 traps heat, feedback positive) is not operating and we’d also need to come up with an alternative that does explain the new observations.
These points are a little complex, but this is what I think we should be explaining to the public about how falsification proceeds in science (and it is usually a long, drawn-out process).
It is needed to counter the dogmatic falsification that insists one observation can disprove a theory, or even that a statistical inference of any kind can do so.
Roger,
Absolutely. That’s why I was trying to highlight that simplistic ideas about how this should/does work are probably wrong.
And that’s why I *liked* your post 🙂
Einstein is quoted as saying “No amount of experimentation can ever prove me right; a single experiment can prove me wrong”. Strict Popperian falsification apparently.
However he did not live up to this in practise. Some of the early experiments that tested relativity by measuring light-speed produced results that appeared to deviate markedly from the predictions of GR.
Einstein cited the uncertainties and difficulty of the measurement and said he would wait for ‘better results’…
My understanding is that even though Eddington’s 1919 measurement of gravitational lensing during a solar eclipse is seen as a confirmation of GR, the uncertainties were so large that it was consistent with no shift.
Roger Jones “to falsify the core theory, firstly, it would have to be shown that ghgs are not trapping IR radiation and/or that there is a negative feedback instead of a positive temperature response.”.
–
Um, I think we are all agreed that GHG’s trap infrared otherwise they would not be called GHG gases. I see this as core science.
The second point is the one that might need to be considered.
–
Chris
Thanks for the answer as to how to really do science, appreciated.
Perhaps I could rephrase your comment though to include where Popperism works, even if it is not needed at the coalface.
” the “classical Popperian approach of potential falsification is how science is done if we all work under the expectation that our hypotheses are falsifiable.”
It is good that you are making an attempt at the concept of falsification, doing so and finding that you cannot is one of the ways we strengthen our understanding.
Angech: ““A clear example is that if the average world temperature drops one degree, back to values before 1900 and stays there for a long time without there being other reasons for the temperature decrease (e.g. volcanoes, sun, aerosols) the theory would be falsified.”
It depends. First, which theory. Anthropogenic warming is not a theory, but rather a prediction of the consensus theory of Earth’s climate, or more generally planetary climates. Ceteris paribus, if we were to see sustained cooling or a lack of warming, that would indicate some shortcoming in that theory. The issue is that one must determine which ceteris ain’t paribus, and whether this completely overthrows the theory, or whether only minor modifications are required.
The situation is further complicated by the fact that planetary climate theory has deep roots in other branches of physics (e.g. thermodynamics of gasses and other fluids, statistical mechanics and even quantum mechanics) that are quite well established. One is not free to simply change the theory in ways that affect these other areas of science.
Popperian falsification works best for relatively simple theories with few moving parts and without deep roots in other areas of science. The best way to falsify complicated theories is to build a better theory (better in this case meaning more consistent with the facts and having better theoretical underpinnings in those other areas of physics on which it relies).
Two paragraphs of straw man argumentation from the bored economist trying to stir up a rise? DNFTT(ol).
As for the main topic, my basic take (simplistic or unintellectual though it may be) is that many perfectly competent research scientists go through their entire careers without ever reading or pondering Karl Popper. Science is run according to scientists’ own working rules, not by philosophers’ abstractions, even if the latter may at times provide a useful high-level ‘view from a distance’.
Richard Tol wrote “Popper is often raised in the climate debate because there are some who claim that everything is caused climate change, although a more common position is that everything bad is caused by climate change.”
a straw man and a non-sequitur, all in one sentence!
Michael2 wrote “The answer might be on SkS, but an advocacy website does not challenge its own assertions. That would be ridiculous in an advocacy website.”
LOL.
I agree with what ATTP and Victor say, I would add though that not only is falsification not a complete solution to the demarcation problem, but there are branches of science that are not currently falsifiable, but are nevertheless science (e.g. multiverse/eternal inflation, which isn’t falsifiable AFAICS, but then again I am not a cosmologist!).
Popper and falsificationism is a bit like Godel’s theorem or frequentist p-values, in that they tend to be bandied about in discussion of science/maths, but most of those doing so (including me, although I’m probably O.K. on p-values ;o) don’t fully understand all of the consequence, and their significance tends to be overstated (e.g. IIUC Godels theorem does not mean mathematical systems cannot be proven to be both complete and consistent).
snarkrates says:
”
Anthropogenic warming is not a theory, but rather a prediction of the consensus theory of Earth’s climate, or more generally planetary climates. Ceteris paribus, if we were to see sustained cooling or a lack of warming, that would indicate some shortcoming in that theory. The issue is that one must determine which ceteris ain’t paribus, and whether this completely overthrows the theory, or whether only minor modifications are required.
”
And then “the issue” becomes whether the ‘minor’ modifications are ‘ad hoc’ or not – where ‘ad hoc’ is any modification that makes the theory less falsifiable.
We’ve been down this path before:
wherein ATTP doth quote David Rose:
”
how much longer will the so-called pause or hiatus have to continue before you would begin to reflect that there is something fundamentally wrong with the models?
”
It is logically possible that “the models” could be falsified by an extended “pause” – but as ATTP pointed out in that previous OP, so much physics would need to wrong that the likelihood of this occurring is vanishingly small.
How long would a roller-coaster have to coast up-hill before you would begin to reflect that there is something fundamentally wrong with our understanding of gravitation?
“Skeptics” loves them some Popper because falsification is a serial-killer of scientific theories – a methodological apex predator that devours all but the rare, highly adapted hypotheses that can escape its fangs. And because Popper’s account is a strict;y critical one – not concerned with justification – only with the survival-of-the-fittest competition of hypotheticals.
Natural scientists, however, abhor a theoretical vacuum. If the “pause” were ever to kill the ’cause’, the ’cause’ would be replaced by a more falsifiable ’cause’ – and the new ’cause’ could not be killed by the ‘pause’.
Of course, even if there is “something fundamentally wrong with the models” (and at some level, there surely is, since no climate model satisfies the constraints of both QM and GR), the models may still be… useful.
[Drive-by. -W]
Rev. Jeb,
The stronger the basis of the science on which a model is based, the harder it is to kill–as long as the fundamental science is applied correctly. The basis for climate theory is pretty fricking strong. .
I’ve finally got the Round Tuit to take a quick look at the article. After the commonplace fluff, here’s the first sentence that speaks to me:
This states the author’s task. It’s a problematic statement. First, I have no idea how the author will succeed in assessing motivations. Second, the “more symmetrical” is strange wording – think about it. Third, paying lips service to Bloor 1976 may tell very little to the casual reader, but reveals the author’s preemptive strike to his target audience.
Here’s the background that the author presumes with this project statement:
https://en.wikipedia.org/wiki/Strong_programme
The missing ingredient from the author’s declaration is the notion of explanation. The Strong Programme seeks to explain how scientists proceed without begging the question as to why they produce Sound Science ™ or not. Whether it’s parapsychology or psychophysics, the explanations should remain the same. In other words, they let go of the demarcation problem altogether.
This missing ingredient is important insofar as the Strong Programme does not seek to “assess claims” or “assess motivations.” The very notion of assessing might go against the Strong Programme. The Strong Programme should seek sociological explanations, not argumentative prescriptions.
***
That sentence alone indicates a few things. First, Warren and his Nottingham gang should like this article, because Strong Programme – I can almost predict they will soon cite him if they haven’t already. Second, the author will raise concerns about tone in his discussion, because of his own understatements that his declaration of intent contains. Third, we should pay due diligence to what the author says of Popper, because he comes from a research programme that is a natural predator to Popper’s objectivism.
***
In the following comments, I will try to compile ideas for a Popper for Bloggers series I had in mind for a few years now.
I still remember reading Alan Chalmers’ excellent little book “what is this thing called science?” over 30 years ago when, as a life sciences undergraduate, I unwisely attempted a course on history and philosophy of science. The Aristotelian stuff seemed a bit naive to me but once we got to Popper, I thought yes, that’s what science is. Then, of course, the book goes on to introduce the ideas of other people who have thought far more deeply about the issue than I ever could, starting with Kuhn and Lakatos…
Dikran:
See, I wouldn’t call these science. Such ideas are worth exploring, sure, and maybe with enough exploration we’ll find ways to test / falsify some of them, and then they’d be science. But such speculation about how physics might work, by itself, isn’t really “science”. It doesn’t quite meet that bar yet.
> I wouldn’t call these science. Such ideas are worth exploring, sure, and maybe with enough exploration we’ll find ways to test / falsify some of them, and then they’d be science.
This would be stricter than falsifiability, which only stipulates that scientific hypotheses should in principle be able to be falsified under operationalized testing. Or something along these lines. I can find a quote in Popper later if anyone’s interested.
That could be an entry for Popper for Bloggers.
Speaking of string theory, we’re almost two months late to stop the press:
http://ucsdnews.ucsd.edu/archive/newsrel/science/stringtheory07.asp
AT’s – where you’ve heard it almost two months too late.
Given the recent large amount of discussion on the role of consensus, I thought it might be helpful to call attention to the following article at the Internet Encyclopedia of Philosophy, since it relates consensus to Popper’s ideas:
“Karl Popper: Philosophy of Science”
http://www.iep.utm.edu/pop-sci/
Quotes from “c. Basic Sentences and the Role of Convention”:
“In order to resolve this apparently vicious regress, Popper introduces the idea of a basic statement, which is an empirical claim that can be used to both determine whether a given theory is falsifiable and thus scientific and, where appropriate, to corroborate falsifying hypotheses.
…
In order to avoid the infinite regress alluded to earlier, where basic statements themselves must be tested in order to justify their status as potential falsifiers, Popper appeals to the role played by convention and what he calls the “relativity of basic statements.” He writes as follows:
‘Every test of a theory, whether resulting in its collaboration or falsification, must stop at some basic statement or other which we decide to accept. If we do not come to any decision, and do not accept some basic statement or other, then the test will have led nowhere… This procedure has no natural end. Thus if the test is to lead us anywhere, nothing remains but to stop at some point or other and say that we are satisfied, for the time being.’ (1959, p. 86)
From this, Popper concludes that a given statement’s counting as a basic statement requires the consensus of the relevant scientific community-if the community decides to accept it, it will count as a basic statement; if the community does not accept it as basic, then an effort must be made to test the statement by using it together with other statements to deduce a statement that the relevant community will accept as basic.”
I’ve never thought Popper very useful for anything but sinple cases.
I prefer:
0) Models ( hypotheses/theories) are approximations to reality, of varying qualities
1) George Box: all models are wrong, some are useful.
2) Model hierarchies are widely employed, where simple ones give good enough approximations for some uses, but not others.
Ex: Newton is not as accurate as Einstein, but is good enough for many uses… Although even without relativistic issues, can be sufficiently incomplete, as in artillery ballistics.
Ex: semiconductor designers need models for circuits, logic, timing, snd power/heat.
Ex:
Flat earth
Flat earth, with local elevations
Spherical earth
Oblate spheroid
Oblate spheroid with local differences in gravity
As above, but with changes
When driving, flat earth is usually good enough, although cyclists care about elevations, at least in some places.
Angech,
A bounded estimate is falsifiable. It’s entirely appropriate to use the word “might” to express uncertainty.
Here’s the boundary of the argument: we don’t want to find out how much SLR there *will* be IF CO2 levels continue to rise.
Richard Tol,
It is certainly more appropriate to add the qualifier “some” when invoking human factors’ contribution to a particular negative event. Best is to quantify to the extent that is possible.
Let’s not allow what *some* media whores do detract from the actual problem, or the need to deploy solutions to it.
There is an alternative to falsification.
It is “detection and attribution”
A process or event is first detected by observation and measurement and then attributed to the cause which best fits.
This the recent LIGO results demonstrated the simultaneous detection of a signal by two separate experiments, which then passed all the noise detection tests.
The chirp was then compared with the modelled output of a variety of astronomical events and attributed to a merger between black holes.
This how you have to work when controlled experiments are not possible, as with climate change.
> A process or event is first detected by observation and measurement and then attributed to the cause which best fits.
Many call it inference to the best explanation:
http://plato.stanford.edu/entries/abduction/
If we accept that scientists use abductions (just like everybody else) and considering that Popper is the father figure of deductivism [1], then we should conclude that Popper’s philosophy of science does not describe what scientists do.
[1]: http://plato.stanford.edu/entries/induction-problem/#KarPopVieInd
Wouldn’t a Popperian say that abduction is one method for forming theories, which become “scientific knowledge” to the extent that they survive subsequent attempts at falsification?
Anyway I remember a visiting prominent German Popperian saying that to one of my logic lecturers in an argument. My lecturer responded by giggling and snorting to himself for a few minutes, probably because he & everybody else in the room, apart from the German, were completely stoned.
Szilard,
I don’t know what a Popperian would say. How you describe abduction also sounds like how I understand the principle of parsimony — which is also invoked to form hypotheses, which upon further testing may eventually reach a consensus status of theory based on consilience of evidence and consistent supporting argumentation.
Abduction also sounds like what hypothetical aliens do, but let’s not go there. Or maybe let’s go there, it could illustrate what scientists very probably should NOT do … nor anyone else for that matter.
The argument is about what scientists do to go from hypothesis to theory. Willard makes a case for it not necessarily being Popperian falsification. Anders’ argument is that Popperian falsification isn’t appropriate to all conclusions formed by climate scientists, but he also doesn’t rule it out as useful in all climate cases.
The message I’m reading thus far is: let’s not constrain scientists to our personal philosophy(ies) of science as we may not (if not likely don’t) understand the particular domain well enough to second-guess them. Perhaps a better way of putting it is, let’s attempt to understand their methods and conclusions on their own merits (or lack thereof) and not quibble so much over philosophy. Popper himself has no answers to particular climate studies any more than the oft-invoked Mr. Feynman does — neither of them are present to make specific comments.
I’m not sure purist Popperians would include abduction in their framework, Slizard. As far as I can recall, it’s either deductions or pure liberal genius. If I may indulge in my inner Strong Programmer, Popper teached at the London School of Economics, and he believed in the freedom to fabricate myths about Occidental superiority.
More importantly, and at the risk of being blunt, Popper always had a problem with the concept of explanation. If you look at his “Science as Falsification,” you’ll see two occurences of the term, both related to what he does not consider science. The patch is obvious here:
http://plato.stanford.edu/entries/popper/
According to this, an explanation is better because it allows us to make more successful predictions. The idea is exquisite but is easier said than done: it’s far from clear that we really can measure the empirical content of theories, which makes the model strangely non-scientific. If I may indulge again my inner Strong Programmer, when someone appeals to content or to power, reach for your wallet.
So the connection between prediction and explanation may deserve due diligence. The difficulty here is that falsification is related to prediction, not explanation. That science tells us that reality is such and such and so and so may crucially matter, but that’s far from being enough. The main reason why science rocks is that it gives us explanations, i.e. an idea how the world is like it is. We do science to understand the world a bit better more than to anticipate what will happen next if we do this or that.
If we seek scientific explanations, falsifying predictions is just one part of the whole game plan.
Except that scientists rarely attempt to falsify theories. IMHO it’s important to stress the distinction (Victor has already done this) between the essential element of Popper (his idea that a sensible and useful scientific theory is one that is falsifiable) and the real-world processes by which scientific knowledge advances.
In fact if rather formalized attempts at falsification (of hypotheses) are made these are most prominently done by the hypothesisers themselves as part of the normal process of investigation (and especially preparation of the work for wider publication). Here’s an example:
Scientists Emma and Jon receive a grant to investigate their hypothesis that molecule X interacts with chromosomal DNA to enhance production of a putative tumor suppressor protein (TSP) in prostate cells to suppress prostate tumor development. This is a fundable study since if the primary target of X can be defined then drugs based on modified versions of X may be developed by rational design. Emma and Jon do some experiments with rats and show that X does in fact suppress prostate tumor growth, and this is associated with increased amounts of TSP in prostate cells. Their hypothesis is supported and since this hasn’t been reported before they consider writing the study for publication.
However before doing so they will certainly consider alternative explanations – they will do other experiments to attempt to support and falsify their hypothesis. They will attempt to determine whether blocking production of TSP results in tumor suppression anyway (i.e. the effect of X has nothing to do with TSP-hypothesis falsified), that X does interact directly with chromosomal DNA in tumor cells (rather than some other protein, for example, that itself interacts with the DNA to enhance TSP levels-if this were the case designing drugs based on X-DNA interactions would be futile)…and so on. These supplementary experiments will be included in their publications as experimental controls and supporting evidence.
Why do they go to this extra effort to attempt to falsify their hypothesis. This do this because:
1. They want to find out something important.
2. They want to get it right.
3. They want to publish the paper in a good journal to make a strong impact
4. They expect this work to develop into a longer term program of research and they want the foundations of this work to be strong. Etc. etc.
It’s worth thinking about these more prosaic examples of how science is done and the context of falsification, since there are some apparent attempts at theory falsification which are not what they seem (will post on this later).
Note that there may be other groups with alternative hypotheses of how X works. Will they attempt to falsify Emma and Jon’s hypothesis? Very probably not. Of course they will inspect the paper carefully and see whether there may be flaws. But more likely they will attempt to provide more evidence in support of their own hypothesis – in time one of these is likely to prevail, but this will be because it provides better explanatory power, and more productive experimental lines, and not because any of the other hypotheses are formally falsified.
For it is very difficult to falsify a theory, and (usually) very wasteful of effort to attempt to do so.
brandonrgates says:
“A bounded estimate is falsifiable”.
That is why very few people will commit to a falsifiability index in print.
No-one wishes to get knocked off their perch.
And if they sit on their perches, all happy, avoiding Popper, they think they may be right.
–
” It’s entirely appropriate to use the word “might” to express uncertainty.”
Brandon, No it is not.
It certainly is not.
“Of course it might be” shows how the word might does your comment in. It is expressing uncertainty at the thought of your comment being in any way appropriate or right.
–
“Here’s the boundary of the argument: we don’t want to find out how much SLR there *will* be IF CO2 levels continue to rise”
– You do it again just to stir.
The SL might rise if I stamp my foot. You of course would not want to see how far it rises when I stamp my foot.
But we both know that it will not rise very far.
Similarly we both know that SLR does not pose a risk next week or next year or in 10 years to Miami going 6 foot under water.
We disagree on how far SLR will go with CO2 levels rising .
But no matter how high it goes over tens of thousands of years land dwelling creatures will have buckets of time to adapt [and hopefully lots of buckets!].
angech,
I don’t your point about not using “might”. In my view it’s important to try and express uncertainty.
Angech writes: “But no matter how high it goes over tens of thousands of years land dwelling creatures will have buckets of time to adapt [and hopefully lots of buckets!].”
Angech, you do realize this is nonsense and you’re only arguing out of habit, right? Where are these land-dwelling creatures going to go? Are there vast stretches of presently uninhabited land near to them? If there are – then why haven’t they already inhabited them?
Climate change is not the simple ‘we’ll all adapt’ problem you think it is. Try reading some actual scenarios already playing out – like How Arctic Spring Is Killing Birds In Africa. Note these changes have occurred over 30 years – not thousands.
Perhaps you’ll fund a bird education program to teach them how to adapt before they go extinct.
Angech – for SLR and extinction you might like to consider: Environmental determinants of extinction selectivity in the fossil record, S.E.Peters, Nature 454, 626-629 (31 July 2008) | doi:10.1038/nature07032
TL;DR — Over the past 540 million years, every increase in the rate of extinctions—including the five so-called mass extinctions—has been linked to environmental changes wrought by changing sea levels.
Makes “But no matter how high it goes over tens of thousands of years land dwelling creatures will have buckets of time to adapt look rather ill-informed.
I think you may be misunderstanding Popper angech! In what sense is someone that provides a well-defined set of bounds for the likely progression of a phenomenon “avoiding Popper”? And what has Popper got to do with it anyway?
Are you suggesting that scientists should guess at the value of a predicted parameter? That in consideration of predicted sea level rise by 2100, rather than providing a range of estimates based on particular greenhouse gas emission scenarios, they should state a blunt value (“the sea level rise will be 86 cm by 2100”).
Should a doctor considering a prognosis of a lung cancer tell her patient “analysis of a large body of epidemiological evidence indicates that you’ve got a 60% chance of survival through the next year”? …..or would you prefer something more specific so we can make a “Popperian analysis” of the doctor’s hypothesis (e.g. “you’re going to be dead in 6 months” or alternatively: “you’re going to be alive in 6 months” )? Is the doctor “sitting on (her) perch, avoiding Popper”, if she doesn’t make a guess at the prognosis?
> For it is very difficult to falsify a theory, and (usually) very wasteful of effort to attempt to do so.
A very important point, which may explain why Popper stressed more falsifiability than falsification later in his career. (It’s just an impression; I have not attempted to falsify it, as I would not care less for Popper were it not for that personality cult from contrarians.) Popper’s demarcation criteria requires that a scientific hypothesis needs to be falsifiable in principle – we should be able to build up a test that would make us see that it’s false. It’s not a moral imperative.
One practical reason why implementing this systematically would be wasteful is that we usually don’t need to build that test to see when an hypothesis leads nowhere interesting or is fruitful enough to be worth the effort. Even chessplayers, who work with a 8×8 model with 32 figurines, will usually drop lines from their analysis before it leads to mate. There’s little point in investing more in checking a move when we see it loses a piece or a pawn for no good reason. Analyzing with powerful Chess engines still requires us to use our common sense, most of the times without satisfying any real test. And that’s just a chessboard, now imagine sequencing DNA or running climate models.
As Herbert Simon might say, we’re satisficing engines more than model checkers.
While the practical reason should be enough to realize that orthodox falsificationnism may very well be farfetched, there’s a logical reason too. It is even stronger – in the end, holism wins. That’ll be for another entry of the Popper for Blogger series.
”
If we accept that scientists use abductions (just like everybody else) and considering that Popper is the father figure of deductivism [1], then we should conclude that Popper’s philosophy of science does not describe what scientists do.
”
What a Popperian way of falsifying Popper, Willard.
We should accept that some scientists abduct.
We should accept that some scientists also use Popperian deductive falsification.
We should accept that some scientists do both. Others do neither.
The time-space-dependent Venn diagram of “science”, “what scientists do”, and “X’s philosophy of science describes” is very, very complicated. Mysterious even. Sometimes I think it might best be drawn on a non-countable set of Möbius strips.
I’ll have much more to say on this when I’ve finished demarcating the demarcation problem.
Then try to falsify falsificationnism, Very Reverend. Report.
Willard said on May 29, 2016 at 1:38 pm,
“I have not attempted to falsify it, as I would not care less for Popper were it not for that personality cult from contrarians.”
Well, one of their problems is that they refuse to properly address the background assumptions (auxiliary hypotheses). They refuse to accept that when even just one such assumption is found to be false, the whole falsification test breaks down in terms of further deductive inference – no more deductive inference is possible after that finding. They refuse to accept that so far, when such assumptions found to be false are replaced with such assumptions that are true or closer to the truth, the models do better at projecting what they are designed to project. (Garbage in, garbage out…duh. They very wrongly expect models to do well even with garbage in.)
For those that might read this and not know what my last point refers to:
The models are designed to project global warming paths given various emissions scenarios, but these paths may find themselves buried *underneath* the fluctuations in the data we should expect, cyclic or otherwise. See all those published studies in recent years starting around half a decade ago with such studies as Foster and Rahmstorf (2011) that address this problem of finding these paths underneath these fluctuations. Here’s a nice article that explains that Foster and Rahmstorf study:
Foster and Rahmstorf Measure the Global Warming Signal
http://www.skepticalscience.com/foster-and-rahmstorf-measure-global-warming-signal.html
Quote:
“This is a very similar result to Huber and Knutti (2011), who estimated that approximately 100% of the observed surface warming since the 1950s has been caused by human effects.”
This is essentially what the IPCC meant when it said that it’s most probable that 110% of the warming since 1950 has been caused by humans, which is a technical way of saying that 100% of it was caused by humans but it would have been 10% worse were it not for some cooling effects from air pollution from humans. See the next article for more on this, especially the graphs (in one of which OA denotes other anthropogenic forcings (which has a cooling effect which would mean aerosols from air pollution):
IPCC attribution statements redux: A response to Judith Curry
http://www.realclimate.org/index.php/archives/2014/08/ipcc-attribution-statements-redux-a-response-to-judith-curry/
I’ll take this opportunity to share again an explanation of these background assumptions that is quite nice because it is written for the general public and succeeds at being very easy for the general public to understand:
Bundle up your hypotheses
http://undsci.berkeley.edu/article/bundle
In other words, the general public should know that the climate scientists are doing really good work that so far, clearly would be able to pass any *properly designed* falsification test thrown at it.
All roads lead to Karl.
At least that is what I got out of this paper. As in, it dealt solely with Popper and uses thereof in popular Anglophone cultures with respect to AGWS,
From the last sentence of the paper …
“By focussing on both AGWS critics and supporters, it was shown that there were a number of similarities in the way pop-versions of Popper’s philosophy of science and appeals to the authority of science appear widely across the AGWS debate and that the use of politically strategic expedient models of good science and unreflexive views of science are not the monopoly of AGWS critics and political conservatives.”
There is, of course, an asymmetry in invoking Popper by contrarians, by O(10) to O(100), versus AGWS proponents.
Philosophers, sociologists, psychologists, psychiatrists and economists. In that order in their lack of utility (philosophers being, of course, Mostly Harmless). Five fields of homo sapiens endeavours that Mother Nature don’t give a fudge about. That, and well, Mother Nature doesn’t give a fudge about homo sapiens to begin with in the first place.
I sort of left out religion as that does not pretend to be science.
More excerpts from the paper …
“Numerous studies have identified a pattern in traditional mainstream media in many Anglophone nations for AGWS to be framed as controversial and in ways that disconnect with mainstream scientific views.”
Disagree. It’s always been the domain of less than traditional mainstream media sources (e. g. contrarian blogs).
“Studies of the AGWS debate suggest that there are important differences between national settings so the observations I make in the following discussion are more applicable to Anglophone contexts involving nation states with strong cultural and economic and political affinities to the United States, such as Australia, New Zealand, the United Kingdom and Canada (Grundman and Scott, 2014; Mann, 2012); this point will be returned to in my conclusion.”
Leaving out the other 90% of homo sapiens for the moment.
Anglophone, zOMG hit me with a dumb stick. You make your bed you get to sleep in it.
I’m curious though, how long have living homo sapiens put new thoughts into dead homo sapiens minds?
The wicked Popper is dead …
Kuhn is melting …
Earlier today I posted on my observations that scientists rarely attempt to directly falsify theories, but usually subject their own hypotheses to critical examination (which may include attempts at falsification) especially before disseminating their hypotheses more widely. They do this because they want to present a convincing argument (to themselves as well as everyone else), and they’d prefer not make any cock-ups – if there are any errors in their arguments they’d rather find these themselves before publication, rather than someone else doing so.
Curiously, there is a sort of mirror image of this normal scientific behaviour which is particularly apparent in efforts to “falsify” (aka “trash”) established science. Climate science is one arena in which these can be found, and there are several instances, for example, of attempts to “falsify” the hypothesis that increased atmospheric [CO2] results in surface and tropospheric warming that is enhanced by positive feedbacks. These efforts are characterised, amongst other things, by an apparent lack of the self-critique that scientists normally engage in. To avoid being overly tedious I’ll give just one example, but there are quite a few of these we could look at:
A very long (15 year) misanalysis of tropospheric temperature estimates from satellite microwave sounding units (MSU) was used to advance the idea that the theory of enhanced greenhouse gas-induced warming must be incorrect since the troposphere wasn’t warming or only marginally so, as atmospheric [CO2] was rising. The deficiency of standard self-critique in this attempt at falsification is apparent in the published comments of the group that uncovered the latest of the misanalyses:
angech,
This is a fun game. And *IF* they know something you don’t and *seem* to avoid Popper, you *may* argue that they’re up to no good.
Ok then. It’s entirely appropriate to use the word “could” to express uncertainty.
Dirt is worth money. Its value is determined by a number of things, including future expectations of its market value … which is obviously contingent on it not being under water. Dual meaning intended.
I’ll put it another way. People don’t buy land with the expectation of it being a wasting asset.
Anders,
*IF* he doesn’t like the implications of the possibility being raised, he *could* consider “might” to be weaseling. But I’m *probably* uncertain.
KeefeAndAmanda,
Some of those background assumptions can be falsified because that’s just what they are – assumptions. There is no theory saying that externally-forced and internally generated variability are independent of each other but most practice treats them as they are. The theory would actually suggest the opposite (in the literature, both are held as being possible). That means that models like Foster and Rahmstorf’s work very nicely in stripping out a monotonic curve but in doing so they also remove some of the nonlinear forced change that operates over decadal timescales.
Because the climate wars are largely being fought over temperature trends, both sides are right and wrong. Over the long term, climate follows a complex trend, but over the short term (a few decades) the change is governed by storage and release mechanisms that are decidedly nonlinear. This is a competing hypothesis that says the core theory of AGW is sound but the auxiliary assumptions are not. So the shifts that contrarians claim to be there and disprove AGW might actually be there as part of AGW (and help validate it).
Popper cannot be used as a working method but is best applied in hindsight (so say Mayo and many others, including Chalmers). Severe testing that uses inductive, experimental methods (error-based testing) is capable of testing single hypotheses drawn from more complex theories – these methods are non-Popperian but may be able to deliver the outcomes he sought. One of the key questions that needs to be answered for climate is “what is the role of variability in change?” I think it’s interactive.
Anders, ” I don’t your point about not using “might”. In my view it’s important to try and express uncertainty.”
brandonrgates says:
“*IF* he doesn’t like the implications of the possibility being raised, he *could* consider “might” to be weaseling. But I’m *probably* uncertain.”
Worth bottling that comment, so many thanks Brandon.
Explains the whole concept of the problems of using might and possible in future scenarios.
Expressing uncertainty is fine and should be done wherever needed.
The trouble with the word might is that it has an infinitely long tail and some people use extremely impossible examples or scenarios where they just should not.
If one used a reasonable estimate of uncertainty or placed it within bounds then our arguments are more meaningful and sensible.
If no-one objects I could point out a couple of examples if they occur in the next few threads purely as examples.
Chris says: May 29, 2016 at 1:32 pm
“I think you may be misunderstanding Popper, angech! In what sense is someone that provides a well-defined set of bounds for the likely progression of a phenomenon “avoiding Popper”? ”
–
If you provide said well defined set of bounds you are not avoiding Popper, you are providing a basis for your your belief set to be challenged and sometimes falsified.
[This is what Popper got to do with it anyway ]
Now no-one likes to be wrong so one way of avoiding this is to refuse to set out bounds.
This happens on both sides of the argument.
You did agree with some bounds above so you are genuinely interested and thinking it through.
Road to Damascus situations apart all of us will have to wait until one or more of our basic beliefs is overthrown to accept falsification.
–
Scientists must consider falsification whenever they put forward a hypothesis before they put it forward, If they knew it was not going to work why would they do it?
They can of course be proven wrong in the course of the experiment or afterwards by factors they had not known about or taken into account a priori.
angech,
It was more intended to illustrate the problem of arguing hypothetical conditionals, e.g.:
Now we just have to agree on the definition of “reasonable”. One *could* undermine any scientific conclusion ever made using that construction *if* that’s what one *wanted* to do.
Examples *can* be a good idea, but I *might* caution against making general conclusions on the basis of anecdote … that’s one of the classic problems of induction. Here I’m not objecting, I’m just saying.
Willard says: May 29, 2016 at 1:38 pm
“Even chess players, who work with a 8×8 model with 32 figurines, will usually drop lines from their analysis before it leads to mate. There’s little point in investing more in checking a move when we see it loses a piece or a pawn for no good reason.”
–
There are numerous moves, times and occasions where chess masters sacrifice pieces for what would appear on the surface to be losing a pawn or piece for no good reason. It is only by weighing up all the possibilities that they produce their great games.
–
” Analyzing with powerful Chess engines still requires us to use our common sense. ”
agree.
“most of the times without satisfying any real test”
disagree, common sense is a real test.
> There are numerous moves, times and occasions where chess masters sacrifice pieces for what would appear on the surface to be losing a pawn or piece for no good reason. It is only by weighing up all the possibilities that they produce their great games.
When I say “no good reason,” I’m not speaking about what would appear on the surface, but about what masters use to justify them spending time considering a move, including a sacrifice. The word they would use to justify material sacrifice would be compensation, which can be decomposed in properties like piece activity, an exposed King, weaknesses in the pawn structure, etc. Good Chess players can develop an intuition about these things, which may sometimes fail, but rarely for the kind of position I have in mind. A pawn is a pawn, and you don’t give pawns for no reason. One does not simply waste time looking for ways to give up material without any compensation.
Besides, good luck with weighting up all the possibilities in a Chess position. To give you an idea, Stockfish analyzes more than 5,000 nodes per second on a quad core. Also, consider these numbers:
https://en.wikipedia.org/wiki/Chess#Mathematics_and_computers
Now, compare with the state space climate modulz need to estimate.
angech,
I still don’t really get your issue. If someone said “we might warm if we continue to emit CO2” that wouldn’t really provide any real information. On the other hand if someone said “sea levels might rise by 1m by 2100” does provide information.
angech wrote “The trouble with the word might [might] is that it has an infinitely long tail and some people use extremely impossible examples or scenarios where they just should not.”
However if you are interested in falsification it doesn’t matter how long the tail is, if some contingency has any plausibility, no matter how small, under the thery/model then it doesn’t provide an opportunity for falsification.
While most science aims at reduction to the best explanation, quite a lot of papers discuss what might occur and the reason they do this is to place bounds on what can be explained by the theory/model. If a scientists says “might” then they will generally be talking about a boundary case of that nature. IIRC WUWT recently had a post about bad science where this sort of cautious language was used, but that doesn’t mean it is a valid criticism.
In short, if a scientists write “might” they are probability just telling you that something is not ruled out by the theory/model, they are not saying that it is likely. In cases where the impact of such a happening might be substantial, we may want to take it into account even though it is just within the bounds of possibility (so there is a good practical reason to use “might” as well as a scientific one).
Got round to reading Mercer’s commentary eventually! A major theme is that everyone using Popper equally poorly (those that support the science as well as those that attempt to trash it). He highlights some apparent contradictions in statements within the different groups. (Whether he believes these to be real contradictions or apparent contradictions isn’t clear since “contradictions” is in quote marks)
Here’s one of his “contradictions” [Mercer; bottom of p9]
However I don’t believe these two statements are incompatible and this isn’t a real contradiction. It really comes down to the strength of the evidence [and also the motives of the hypothesis proposer, though unfortunately we can’t really discuss this except in a positive light as I did earlier in the context of the motivation of scientists to explore their hypotheses carefully ( https://andthentheresphysics.wordpress.com/2016/05/27/popper/#comment-80188 ) ]
Anthropogenic warming resulting from the direct radiative effects of enhanced greenhouse forcing coupled with positive feedbacks is supported by a very large range of evidence from theoretical knowledge, direct observations, paleodata and modelling. It’s perfectly correct to state that ”Popper would have required more efforts at falsification than a few simple claims” in support of the use of Popper [*] in arguing for a more robust attempt to cast significant doubt on this theory.
On the other hand to state that ”a small number of papers by climate sceptics have been ‘refuted’ and therefore their position falsified” is also acceptable since it is objectively true. I gave an example of the MSU tropospheric temperature data in a post above. If the assertion that climate sensitivity must be low since the troposphere isn’t warming (much) is based on an objectively faulty analysis, then this assertion in that context is refuted rather simply.
Similarly with Lindzen’s assertion that climate sensitivity will be low since the upper troposphere will dry in a greenhouse warmed world (no positive water vapour feedback) [**], or his similar assertion based on ERBE measurement of radiation fluxes (refuted since the analysis was over-dependent on selection of a particular set of time-periods), or Chylek’s interpretation of low climate sensitivity based on analysis of radiative forcings from aerosols through the last glacial-interglacial transition (again, analysis over-dependent on a particular non-representative set of data points ( http://www.clim-past.net/5/143/2009/cp-5-143-2009.html )…and so on.
The reason that these papers are straightforward to refute is partly due to the fact that the practitioners seem not to have engaged in the self-critical analysis of their data that is normally done by scientists (see my post earlier linked above!).
[*] Discussing this doesn’t constitute an endorsement of Popper’s ideas which I personally consider to be not particularly helpful, though the notion of falsifiability does crystallize a phenomenon that is already second-nature to scientists IMHO.
[**] This is an example where a theory has been falsified. However the fact that we endorse the theory that greenhouse warming will induce a positive feedback from enhanced tropospheric water vapour is not because Lindzen’s theory has been falsified (a la Popper) but because the evidence inescapably supports enhanced tropospheric water vapour in a warming world (a la Sokal, perhaps).
Roger Jones said on May 30, 2016 at 1:35 am in reply to my comment on May 29, 2016 at 5:36 pm,
“Some of those background assumptions can be falsified because that’s just what they are – assumptions………Over the long term, climate follows a complex trend, but over the short term (a few decades) the change is governed by storage and release mechanisms that are decidedly nonlinear. This is a competing hypothesis that says the core theory of AGW is sound but the auxiliary assumptions are not. So the shifts that contrarians claim to be there and disprove AGW might actually be there as part of AGW (and help validate it).”
It may be that if these auxiliary assumptions in question are false, then their negations (which then would give true statements or more accurate information) will provide more support for AGW (a test hypothesis). And it also may be along the line of what you said in that they should be defined to be part of AGW rather than defined to be auxiliary or in the background. To address this further:
One of the points I argued is this: Given that an assumption can be a statement or a conjunction of statements, it’s important to know whether a given assumption (that’s relevant to the issue – see further below on ad hoc) is defined to be a background or auxiliary assumption or is defined to be part of the test hypothesis. This distinction is crucial as to whether a test hypothesis is falsified given certain measurements. On this question of which way it is defined, note that an assumption or set of assumed information can be defined as an auxiliary or background assumption even if it is fed into the models – being fed into a model does not imply that it must be defined to be part of the test hypothesis. (Example: An assumption that there will be a certain amount of CO2 emitted by humans or a certain amount of aerosols emitted from, say, volcanic activity should not actually be considered part of what is to be tested – testing such things as the veracity of the formulas is what this is about rather than such things as the psychic abilities of scientists to predict the future.) If an assumption is defined to be a background assumption rather than part of the test hypothesis, and this assumption turns out to be false, then it becomes logically impossible from there to deduce the falsity of the test hypothesis. This fact is what those who claim to falsify AGW trip over again and again.
For any reader who would like me to put it in symbolic form, perhaps making it clearer for some:
Let T be the test hypothesis, let A be the conjunction of all the background assumptions or auxiliary hypotheses, and let P be a prediction or projection of T, or if I may, what is implied by T given A. (Each variable could cover a rather large set of statements, and so could be a conjunction of many statements.) The setup implication is
A -> (T -> P)
(I like this form, since it makes explicit that A is in the background, hence “background assumptions”) or any one of its equivalent forms such as this implication with a conjunction antecedent,
(A & T) -> P.
Suppose a measurement gives the negation of P, which I symbolize as
~P.
Then we have
~(A & T)
or any one of its equivalent forms such as this disjunction,
~A v ~T.
Now suppose we find that one of the background assumptions or auxiliary hypotheses is false. This gives us
~A.
At this point, it becomes logically impossible to deductively infer ~T (using propositional logic). It’s as I said in my above prior comment, “Well, one of their problems is that they refuse to properly address the background assumptions (auxiliary hypotheses). They refuse to accept that when even just one such assumption is found to be false, the whole falsification test breaks down in terms of further deductive inference – no more deductive inference [that concludes with ~T, a false test hypotheses] is possible after that finding.”
(To some who might not know: Positing a statement as part of a disjunction or implication is not the same as positing it as a standalone statement. This is about where we can deductively infer ~T [as a standalone statement] given ~P, which we cannot do once we have ~A.)
Chris said it on May 29, 2016 at 9:11 am:
“…it is *very* difficult to falsify a theory,…”
One reason this is true is that it seems in practice to be very easy to find a false background assumption or auxiliary hypothesis when confronted with a measurement that says ~P.
To anticipate that some might object based on ad hoc: It’s of course not true that any old assumption qualifies as a background assumption or auxiliary hypothesis (or as part of the test hypothesis). I think the general public can rest easy that when fine tuning their models into more and more accurate models given more and more accurate information fed into them, climate scientists are not committing some logical fallacy based on ad hoc.
Popper had something to say on ad hoc:
Karl Popper: Philosophy of Science
b. Auxiliary and Ad Hoc Hypotheses
http://www.iep.utm.edu/pop-sci/#SH2b
Others have different views, with perhaps a different definition of this term ad hoc such that it may or may not be problematic or fallacious, depending on how it’s used. Here’s an example:
Popper’s Explications of Ad Hocness: Circularity, Empirical Content, and Scientific Practice
http://bjps.oxfordjournals.org/content/44/2/335.abstract
> Now suppose we find that one of the background assumptions or auxiliary hypotheses is false.
When building a model, this obtains trivially. We could always find at least one assumption in need of correction. The modulz are stupid and wrong, after all.
This kind of indetermination even obtains for simple hypotheses like “All swans are white.” Let a proverbial zoologist Z go to Australia and find his first black swan and return to his alma mater, claiming having falsified the hypothesis according to which all swans are white.
(Note, in passing, that falsification works at the hypothesis level – one does not simply falsify a theory as a whole.)
So our Z goes to a conference and lays out his finding by producing a live specimen in front of them. The other zoologists share with him a neverending list of concerns: are you sure this is a true swan, are you sure it was really white in Australia, do you have all the code and the data that produced the analysis of the whiteness, who cares if all swans ain’t white, why are philosophers always using the darnednest examples, etc.
To make matters worse, once Z finds a non-white swan, he and his community has a choice between rejecting the hypothesis (in this case a natural law about a natural kind), tweaking what it is to be a swan, asking color theorists to revise their concept of whiteness, adjusting the theorical claim by adding circumstancial caveats, or waiting for more studies on that matter before reaching any firm conclusion. In other words, there’s always (at least in principle) enough wiggle room to infer almost anything from an experiment.
Once you accept that no single experiment ever refutes a theory, then you reject what we can call the myth of the experimentum crucis. This is a powerful myth: even Newton held it. It was a keystone of good ol’ empiricism.
But in the end, holism wins.
Not sure what you mean by this Angech, (or many of the things you say since you seem to like single sentence assertions!). First of all I pointed out earlier that scientists generally do incorporate the idea that their hypotheses are falsifiable. However nobody has to accept falsification (explain what you mean by “falsification” please). If you mean (a la Popper) that science advances by falsifying several of competing hypotheses, or that (a la Popper) the purpose of scientific observations and experiments is to falsify theories [*], then I don’t accept falsification, since I consider these to be very poor descriptions of the way science progresses in the real world.
And why the fixation on Popper? There’s nothing privileged about his ideas – he was a rather extraordinary chap that formulated a perspective on the progression of science, but there’s no reason why his ideas should hold precedence over any others except to the extent that someone finds them useful. I don’t particularly, but maybe you do – in which case you should tell us why. Personally I prefer Fayerabend or (I’ve just discovered), Sokal.
As I’ve said earlier, I don’t think scientists engage in refuting hypotheses/theories, that it is very wasteful of effort to try to do so, and that theories prevail not because their competitors are falsified, but because they are increasingly supported by evidence and are more useful/have greater explanatory power etc.
And I consider that we can be much more positive about the status of our knowledge than Popper would allow. I’m happy to accept that from a philosophical perspective theories are never proven and will always be superceded (don’t actually accept the latter actually but we would have to consider specific examples). However in the real world we really do know stuff, and many of our hypotheses/theories are so well supported as to be essentially proven (in context). We really do know the atomic resolution structures of proteins, that human immunodeficiency virus causes HIV/AIDS, that greenhouse-induced warming results in enhanced tropospheric water vapour and so on.
I think that there are some very interesting questions about the nature of our knowledge in 2016 compared to (for example) 1963 when Popper published Conjectures and Refutations, but I’ll leave this for now…
Yes, I said this, more or less, earlier. Someone should tell those guys who messed up as described in my earlier posts:
what’s your opinion of their efforts?
[*] “Tests proceed partly by way of observation, and observation is thus very important; but its function is not that of producing theories. It plays its role in rejecting, eliminating, and criticizing theories.” Popper (1963) Conjectures and Refutations. p. 128
Willard,
Duhem be damned, One Model to Rule Them All. /sarc
Thanks for experimentum crucis and epistemological holism. It’s nice to know that these things were thought about before AGW became unfashionable.
Willard said on May 30, 2016 at 2:52 pm in reply to my comment on May 30, 2016 at 12:51 pm,
“>Now suppose we find that one of the background assumptions or auxiliary hypotheses is false.
When building a model, this obtains trivially. We could always find at least one assumption in need of correction.”
For a reason, I wrote, “It’s of course not true that any old assumption qualifies as a background assumption or auxiliary hypothesis (or as part of the test hypothesis).”
That is, it’s not true that we could *always* find at least one assumption that *reasonably* qualifies as an auxiliary hypothesis and that is in need of correction. Note 1: I did not say that if we cannot find at least one assumption that *reasonably* qualifies as an auxiliary hypothesis and that is in need of correction, then such an assumption does not exist. Note 2: Granting that such an assumption could still exist even if we cannot find it does not preclude us from taking the inductive step of affirming A as true “for the time being”. I put this last phrase in quotes because these are Popper’s words that I related in my comment on May 28, 2016 at 6:26 pm. He said, “Thus if the test is to lead us anywhere, nothing remains but to stop at some point or other and say that we are satisfied, for the time being.” (1959, p. 86)
“In other words, there’s always (at least in principle) enough wiggle room to infer almost anything from an experiment.”
For a reason, I wrote, “I think the general public can rest easy that when fine tuning their models into more and more accurate models given more and more accurate information fed into them, climate scientists are not committing some logical fallacy based on ad hoc.”
That is, it’s not true that there’s *always* enough wiggle room to *reasonably* infer almost anything from an experiment without committing a fallacy or putting forth some problematic argument that would not meet certain higher standards of reasonableness. Take for instance the highest standards of reasonableness we find in the highest courts of law – apply them, and watch what happens. Examples: See those two attempts here in the US by those who deny biological evolution to prove their cases in federal court. They failed terribly each time. Now imagine what would happen if those who deny climate science tried to prove any of their cases in federal court.
> That is, it’s not true that there’s *always* enough wiggle room to *reasonably* infer almost anything from an experiment without committing a fallacy or putting forth some problematic argument that would not meet certain higher standards of reasonableness.
That’s reasonable enough.
Granted that we can reasonably infer something under whatever standards of reasonableness, this kind of inference has little to do with formal deduction. Which leads me to a point that needs to be said about Popper’s usage of “logic” (e.g. his Logic of Scienfic Discovery) – it’s more a meta-logic than a logic. While he characterized scientific discovery as following a “logic,” the higher standards of reasonableness scientists are using are simply not enough for formal deductions. Popper’s point was to explain our error correction mechanism by way of modus tollens.
Logicizing Popper led for instance David Stove to go a bridge too far in his criticism of Popper:
http://philpapers.org/rec/DCSPOS
Rowan & Smithson convincingly argues that this kind of argument forgets to take the context of the statements into account:
http://philpapers.org/rec/MICSOP
It thus seems that salvaging what’s good from the falsificationnism project relies in part on the work of philosophers of language, whom Popper denigrated time and time again, something which may or may not be related to the firepoke incident:
https://en.wikipedia.org/wiki/Wittgenstein%27s_Poker
perhaps this can help the holism holdouts
““A clear example is that if the average world temperature drops one degree, back to values before 1900 and stays there for a long time without there being other reasons for the temperature decrease (e.g. volcanoes, sun, aerosols) the theory would be falsified.”
The hypothesis is this
If we increase c02 and hold all else constant the temperature will increase.
There are some additional hypotheses
1. We know all additional forcings
2. Internal variation sums to zero over the period of interest
Such that.. if we saw temperatures drop. If we looked at all known forcings and saw that only c02 increased. if this happened for some time…. THEN
we have a choice which is largely pragmatic
1. Change the assumption that increasing c02 will warm
2…Admit that there may be negative forcings we dont know about
3. Assert that some long term internal variation is the cause.
each of those decisions “follows” from the outcome. it would be an unsettling experiment
but there is no logical reason to choose 1
An interesting approach to falsifiability
Perhaps someone who understand this issue could explain to me whether making assertions of changes over time, without quantifying uncertainties or providing longitudinal evidence, meets the requirement of falsification.
Brandon, you might * be interested in this example of if and might and the power of wishful thinking
–
KeefeAndAmanda says: May 30, 2016 at 12:51 pm
Roger Jones said on May 30, 2016 at 1:35 am
“Some of those background assumptions can be falsified because that’s just what they are – assumptions…… This is a competing hypothesis that says the core theory of AGW is sound but the auxiliary assumptions are not. So the shifts that contrarians claim to be there and disprove AGW might actually be there as part of AGW (and help validate it).”
–
“It may be [might**] that if these auxiliary assumptions in question are false, then their negations (which then would give true statements or more accurate information) will provide more support for AGW (a test hypothesis).”
–
Stretching things to the limit to include the possibility of the opposite of what is meant.
Otherwise known as having your cake and eating it as well.
Here the use of the word might* enables one to say if* something in our theory was proven wrong it makes the rest of our theory even more right.
Hence the use of boundary conditions for falsification.
Whats the point of even trying to argue if an essential part of the theory, falsified], makes one say the theory is even stronger?
In Australia our politicians have core promises which automatically become auxiliary on being elected.
There’s socio-politico-economics:
And then there’s PhiloBall ™:
The viewpoint according to which “all life is problem solving” (if I recall Sir Karl correctly) begs for a psychological explanation, if not a psychoanalytic one.
Steven Mosher says: May 31, 2016 at 12:47 am
“re if the average world temperature drops one degree etc”
The hypothesis is this, If we increase c02 and hold all else constant the temperature will increase.
Such that.. if we saw temperatures drop. If we looked at all known forcings and saw that only c02 increased. if this happened for some time…. THEN
we have a choice which is largely pragmatic
–
1. Change the assumption that increasing c02 will warm
2…Admit that there may be negative forcings we dont know about
3. Assert that some long term internal variation is the cause.
–
each of those decisions “follows” from the outcome. it would be an unsettling experiment
but there is no logical reason to choose 1
–
I agree with every point.
There is a logical reason to not choose 1 as you said.
In an ideal atmosphere increasing CO2 must increase the temperature.
Since we assert initially all else is constant, 3 is falsified or non consequent, take your pick.
This leaves 2.
There may [arrgh!] be negative feedbacks we have not considered.
Note, not that “we don’t know about”.
When the pause was in vogue over 20 plausible [to some] negative feedbacks were postulated.
Correct?
Denial of the possibility of negative feedbacks is a boundarum crucis.
Willard says:
“”The viewpoint according to which “all life is problem solving” (if I recall Sir Karl correctly) begs for a psychological explanation, if not a psychoanalytic one.”
Having witnessed a series of tragedies in life around me recently I feel the purpose of life is to die.
Very depressing.
“Staring at the Sun,” psychiatrist and Stanford University professor Irvin D. Yalom might be right after all.
– Chris And why the fixation on Popper?
there’s no reason why his ideas should hold precedence over any others except
— “to the extent that someone finds them useful.”
Self answered I see.
“I don’t particularly, but maybe you do – in which case you should tell us why.”
See Willard says: May 31, 2016 at 1:55 am above for an explanation.
–
“Personally I prefer Fayerabend or Sokal. ”
Just guessing but you seem to dislike the idea that falsification of an idea or theory proves that theory is wrong’.
Willard has provided good examples to KeefeAndAmanda above, there are philosophers and scientists who argue that
Science is wrong
Science is useless.
Science cannot prove anything.
Science has multiple answers in parallel universes.
Now they are actually right but on this blog [ATTP] we have a scientific model and viewpoint and all the ifs and might’s that deny scientific procedure in this current universe have to produce proof of falsification of the scientific method first.
I will look them up on Wiki to see if my theory is correct.
Chris
Feyerabend became famous for his purportedly anarchistic view of science and his rejection of the existence of universal methodological rules.
Sokal is best known to the general public for the Sokal Affair of 1996.,” Sokal submitted a grand-sounding but completely nonsensical paper entitled “Transgressing the Boundaries: Toward a Transformative Hermeneutics of Quantum Gravity.
Seems I was right on both counts.
Even better would be Pierre Duhem (1861–1916) was a French physicist and historian and philosopher of science recommended? by Willard [ see holism link]
. As a physicist, he championed “energetics,”. In philosophy of science, he is best known for his work on the relation between theory and experiment, arguing that hypotheses are not straightforwardly refuted by experiment and that there are no crucial experiments in science.
> Seems I was right on both counts.
You were right about what, and what does it have to do with the price of tea, Angech? Quote fests (a quote, a response, another quote, another response, etc.) are hard to read. They also lead exponential exchanges. Think about it.
That there’s no specific method we can dub scientific does not imply there’s no science. It just means that (at least according to Feyerabend) looking for a method in the scientific madness is a fool’s errand. We’re an opportunistic specie, all up to our knowledge gathering ways. And whatever the demerits of Duhem’s beliefs, it is his idea that
http://plato.stanford.edu/entries/duhem/#AgaNewMetDuhThe
that is relevant here and should be accepted or discussed. If you prefer, you could call both Feyerabend and Popper irrationalists, like Stove (an Aussie, to boot) did, and be done with it.
There’s also Duhem’s view on teh stoopid modulz a bit below in that Stanford entry if you are interested. Spoiler: he’d drop them all. Scientific practice showed he was wrong. Big deal.
Even Popper cannot provide rigidly defined areas of doubt and uncertainty
Has Popper been falsified? Was Kuhn just a passing paradigm?
Philosophers may suffer existential angst over an inability to achieve epistemic closure on the scientific method. I am tempted to suggest that at the level of social significance the best measure of the science is its utility. But that would seem to relegate our blog hosts’ efforts into the inconsequential!
Has Popper been falsified?
“Popper” is philosophy, not science. (Still useful)
Falsifiability is for scientific hypothesis.
When you say ” the demerits of Duhem’s beliefs,” you indicate agreement that his ideas are not sound.
proof of which is his quote
“To seek to separate each of the hypotheses of theoretical physics from the other assumptions upon which this science rests, in order to subject it in isolation to the control of observation, is to pursue a chimera.”
whilst epistmologically true we live in what he calls a chimera so his words can never make sense to us.
Unless you wish to put forward arguments saying that when your logic is wounded on one front one merely moves to another front. Mosher explained the true scientific way in his 3 steps above. You seem to be using the then a miracle occurs approach.
1 CO2 warms. 2 is it warming, great. 3 If it is not warming what else should we be looking at ?
angech at 1:54 am
You said this, then left it hanging:
“Stretching things to the limit to include the possibility of the opposite of what is meant.
Otherwise known as having your cake and eating it as well.
Here the use of the word might* enables one to say if* something in our theory was proven wrong it makes the rest of our theory even more right.
Hence the use of boundary conditions for falsification.
Whats the point of even trying to argue if an essential part of the theory, falsified], makes one say the theory is even stronger?”
You are treating this as rhetoric. That’s not how theory (or my argument) works. There is core theory that sits at the centre of a proposition. Auxiliary theory if shown to be wrong does not threaten core theory. A substantial amount of climate theory is nonlinear. So if I show the change is nonlinear in line with both the auxiliary and core theory, how does that equate to politicians’ core and non-core promises?
oh well, never mind angech. I thought you might have something interesting to say but seemingly not… you’re just going to dump Wikipedia cut-n-pastes as non-responses…
A couple of things though:
1. earlier you were castigating an apparent reluctance of individuals to make well-defined (“falsifiable”) statements/predictions: you said “And if they sit on their perches, all happy, avoiding Popper, they think they may be right.”
But a careful description of one’s thoughts on a subject is a type of hypothesis wouldn’t you say? My description of how I consider science is done in the real world, or why I consider Popper’s ideas not to be a very good science description, and so on, are a type of hypothesis – they define my view on these things – it’s possible that my viewpoint is incorrect, and so the things I say in my posts on this thread are falsifiable. In fact they are particularly falsifiable in the Popperian sense since I’ve been as explicit as I can be.
I was really inviting you to do the same. Since you’re being a little contrary I am genuinely interested in your viewpoint to see how this might or might not differ from mine. However you seem reluctant to offer me the courtesy that I offered you in response to your requests. it seems you prefer to “sit on your perch, avoiding Popper, and think you might be right”: angech:“Seems I was right on both counts”
2. You may not have discovered this but writing a carefully though-out description of your viewpoint is really valuable. In doing so you actually help yourself know what you think about a topic and why. You’ll likely find you’re viewpoint is reinforced in doing this and you may be rather pleased with what you’ve come up with (it’s a creative process).
3. Obviously I mentioned Sokal in the context of his ideas on the way that science is done. I didn’t really know about him (i remember the spoof article) but I was interested to discover on Sunday that his view on the progression of science is (at a superficial reading) similar to mine (science advances largely through the success of theories rather than the falsification of competing ones, to make a blunt precis). On the other hand I think the spoof article was an excellent idea.
As Victor pointed out at the beginning, Popper was concerned with a question in philosophy more than an analysis of science. The big question of the time was how to determine true from false. What statements could be made that would have a water-tight epistemological foundation.
Willard has reminded us that Popper was developing his ideas in the context of a conflict with competing ideas such as those of Wittgenstein.
There is a third name in this web, about the time that Popper and Wittgenstein were grappling with what can meaningfully be determined, a mathematician who is thought to have attended some Wittgenstein lectures was working on a similar problem in mathematics. There is a (apocryphal?) story that Wittgenstein chided this young mathematician for his pure maths research, decrying it as the ultimate in ‘playing with symbols’ and contrasting it to the utility of applied maths that could at least calculate the strength of components in a bridge to ensure it did not fall down.
However it was that mathematicians’ paper:-
On Computable Numbers, with an Application to the Entscheidungsproblem.
That has proved to have much greater utility and impact on science and society than Popper or Wittgenstein.
Popper’s work now seems to be regarded as an elegant idea that works in theory but not in practise.
Wittgenstein has apparently been relegated by many philosophers as not a useful contribution, a mess of a theory, but perhaps in practise…
Meanwhile the idea that you can construct a machine to decide has given us devices that can beat humans at GO.
@-angech
“whilst epistmologically true we live in what he calls a chimera so his words can never make sense to us.”
This is not a meaning that can be derived from the Dunhem quote-
“To seek to separate each of the hypotheses of theoretical physics from the other assumptions upon which this science rests, in order to subject it in isolation to the control of observation, is to pursue a chimera.”
A chimera is a Greek mythological beast made up of the arbitrary conglomeration of disparate animal parts. I would suggest that Dunhem is pointing out that theoretical physics is NOT a chimera, it is a unified whole of interlocking theories and axillary hypothesis that gain validity from the interactional consilience of the theories or explanatory narratives that are used.
The idea that it can be artificially separated into independent component hypothesis that can be individually assessed in isolation is the mistake, it is the pursuit of a chimera and attempt to divide into parts a subject that is a cohesive unified whole. Albeit one that develops and progresses. And grows.
I like KeefAndAmanda’s comment at 12:51 and most of Willard’s. Was impressed in that KeefAndAmanda took on my challenge and expanded on the big picture.
I think there is a better way, though, to test the auxiliary hypothesis outside of a full-blown research program. Define H and not-H (the null hypothesis). Develop a structural test that describes both alternatives and lend this to an experiment that yields data x. The experiment has to show that x is approved to the level that not-x has no appreciable likelihood of being true. If the outcomes are ambiguous, then H is not validated and, if initial assumptions favour either H or not-H in any way, then the experimental results cannot be taken up as conclusive evidence. Materially, within the ‘scientific method’ there are many potential ways to do this. Ideally, this is a collaborative discussion in a research team because no one person has a mortgage on this.
What people aren’t taking into account of sufficiently in this discussion (thanks ATTP & W for moderating it) is that the theory is networked and cannot be dealt with as a linear deductive process all of the time. Following through a theoretical process in a complex system is decidedly non linear and is one reason why logical positivism had to be discarded. If it was amenable to that type of reasoning, it would have been sorted long ago.
The anarchy described by Feyerabend is accurate but is a description in hindsight that takes in both the social process of doing science and the many methods (mostly inductive) of carrying it out. Experimental process and reasoning attached to hypothesis testing can be improved.
There isn’t a single philosophy of science. Some is retrospective and tells us more about the process. Some gives social insights about the processes of how a scientific community focuses on a particular paradigm (or not). Some is experimental (and is an improvement on Popper).
I got through this without mentioning Latour (Don’t start).
Willard. I clicked through to the Jim Holt review of Wittgenstein’s Poker. It’s pure unadulterated bullshit.
> When you say ” the demerits of Duhem’s beliefs,” you indicate agreement that his ideas are not sound.
I said And whatever the demerits of Duhem’s beliefs. By beliefs, I was referring to his other beliefs like the one you cited (energetics), which are independent from his holism, i.e. the quote. Your idea of soundness may have forced scientists to disregard Newton’s Principia because it was both from a jerk and an alchemist.
***
> I clicked through to the Jim Holt review of Wittgenstein’s Poker. It’s pure unadulterated bullshit.
I’m sorry to hear, RogerJ.
If you could acknowledge that for Popper we falsify hypotheses, not directly theories, let alone “core” theories, that would be great.
W, it’s just that it undermines what both of them were about. Wittgenstein’s philosophy was about observing responses because he didn’t think we could see the substance of an object in itself. This has been enormously important for subsequent thought. Popper was looking for a logical, deductive frame that had more to do with ideal than practicality..
This – Popper, by contrast, was a homely, ordinary-seeming fellow whose most important work, ”The Logic of Scientific Discovery,” had yet to appear in English and whose chief intellectual attribute was — unexcitingly — common sense. Whereas Wittgenstein was homosexual (the authors decline to join the controversy over just how active he was), Popper had an adored wife, albeit one whom he could never bring himself to kiss on the lips. – FFS, totally irrelevant.
Popper, common sense not, and what is with the homo-wife thing??
But the second – happy to acknowledge – are you familiar with Mayo? Not a problem.
Yes that’s a distinction that should be made (theory/hypothesis). Is it the case that Popper’s falsification referred explicitly to hypotheses rather than theories? That seems at odds with the following quotation but perhaps one needs to read it in context (rather than pasting it from Wikipedia!)
Real world examples (apologies for doing this yet again!) suggest that things can be a little messy. An often-used example is chemiosmosis as an explanation of how the free energy from oxidation of food (fats/sugars/protein) is transformed into “high energy” phosphate bonds (ATP) to power cellular functions. Two competing hypotheses in the 1960’s were: (a) an as-yet-to-be-identified high energy intermediate in mitochondria stores and transfers the free energy into ATP synthesis; (b) the free energy of oxidation is converted into a proton gradient across the inner mito membrane and the free energy stored in this gradient is used to power ATP synthesis (chemiosmosis).
Chemiosmosis “won” even if the first hypothesis was never formally falsified (very difficult to establish the lack of existence of something) – high energy intermediates do in fact exist (e.g. phosphocreatine) but these are now known not to be direct intermediates in free energy transfer. Chemiosmosis provided good experimental possibilities and increasing explanatory power and so it became the dominant paradigm (it’s basically and rather undeniably a true explanation). It’s now a theory (the Chemiosmotic Theory).
From the perspective of Popper I would say that (a) the competing hypotheses were perfectly falsifiable (one to Popper); (b) that experiments were done largely to attempt to provide evidence to support each of the theories rather than to attempt to disprove them (minus one to Popper); (c) chemiosmosis “won” because it was “better” rather than because the high-energy-intermediate hypothesis was formally disproven (minus one to Popper) and (d) we can be much more secure in our knowledge that the Chemiosmosis Theory is true (in context) than Popper would allow (minus one to Popper).
That’s my accounting – others may have a different view.
Something else comes to mind. When I say (just above):
“experiments were done largely to attempt to provide evidence to support each of the theories rather than to attempt to disprove them.” …
This is making a statement about the motivation of the experimenter. We could notice that an experiment done to attempt to provide evidence in support of a hypothesis might in fact not provide this support, and so it could turn out to cast doubt on the hypothesis. if we were to pursue scientific experimentation using automatons we might remove motivation, and experiments would be designed solely to explore the consequents of a hypothesis in a disinterested fashion (to be honest scientists actually do this to some extent). It’s possible that Popper meant something like this in the quotation I pasted in the post just above.
in other words when he says that Tests (experiments/observations) play their role in rejecting, eliminating, and criticizing theories, he may not mean that experimenters design their tests to do this, but that if one were to look at the progression of a scientific field in hindsight, one might surmise that the tests (however motivated) played this role.
RogerJ,
I agree that Popper’s and Wittgentstein’s backgrounds may not be relevant to judge their viewpoints. That’s neither what the Poker’s authors nor what its reviewer did. The biographical details matter insofar as you’re interested in the story that leads to the Poker clash. While we may suspect that the account beautifies the opposition between the characters, the conflict was manifest in their philosophical approaches.
Popper didn’t buy the first Wittgenstein (the author of the Tractatus), whom is one source for (the Carnapian version of) the logical empiricist program, because there’s no correlation between meaningfulness and scientificity. According to him, it’s quite possible for metaphysical statements to carry meaning. He sneered over the second Wittgenstein (the one who spoke of “language games”) even more forcefully, for he always considered that philosophy, like science, was about solving problems, not dissolving mundane utterances with abstruse analyses.
I totally buy the “homely” characterization of Popper. His books are quite readable. He never really formalized his doctrine, and when formalized it doesn’t work. He was amazed that people took at face value his “logic” – he never thought that we could discuss “basic statements” outside their context and their framework. Here’s my reading of this: he’s trying to understand how we settle scientific questions without having to gerrymander over meaning and other linguistic preoccupations. At the very least, if you compare Popper with Wittgenstein, in whom Baron-Cohen sees so much Aperger traits that he has no qualms in putting him in the autistic spectrum, it’s quite obvious that Pop himself has everything to please the conservative establishment.
I should comment on the (somewhat exaggerated) rise and fall of positivism soon enough. I should also try to clarify what I have in mind about the problem-solving mantra.
Popper cannot survive random noise or “natural” variation. Eli hates Popper because he screws up how science is taught and thought about. If Willard wishes, Popper is the dead end of Russell and Whitehead’s Principia Mathematica beloved by certain 16 year old bunnies (ok we were weird), it looks useful till you try to use it,
Wait a minute.
We know that we can’t predict accurately what will happen in 2100 because of, among other things, the instability of the solutions to the equations of motion of the atmosphere.
However, falsifiable predictions can be made and have been made regarding the tendency of the atmosphere to warm with increased radiative forcing. Global warming is falsifiable and that’s why I tend to believe it: within the infinite range of possibilities, it is a theory that has not been falsified.
Now, other things have been falsified within the range of our available observations.
Warming at a rate of 2C/century, for any trend of 17 years or more through 2015 has not occurred, so theories which postulated such warming are false.
Increasing tropical cyclone energy, increasing drought, et. al. are also falsifiable predictions.
But we have confidence in the theory of global warming not in spite of Popper, but because of Popper.
TE,
I didn’t say it couldn’t be falsified (try reading the whole post). What I did say was that applying simplistic rules to something as complex as this is unlikely to be a valid way to assess this topic.
It is of course possible to simply wait and see what happens, both in terms of our future emissions and in terms of our predictions, given those future emissions. However, that doesn’t mean that the projections have no merit until such time as they have either proven correct, or not.
Nonsense. According to AGW the rate at which we warm will – on average – depends on the rate at which we’re emitting GHGs. That we haven’t warmed at 0.2C/century or faster for any period of 17 years or longer (and I haven’t checked to see if this is true or not) does not mean that any projection suggestion that we could warm at this rate has been falsified.
We also only live in one actual realisation of what could have happened. That we did not have a period where we warmed at 0.2C/century of greater for a period of 17 years or more, does not mean that such a warming rate was not possible.
No. This is a category error (addressing a scientific problem at the wrong scale). The relevant scale is that of energy balance and this is pretty well predictable (i.e. the temporal variation in radiative forcing as a result of particular greenhouse gas emissions). The parenthesised latter provides a set of uncertainties, likewise the uncertainties relating to feedbacks. So we can set likely bounds on the warming to be acccrued by 2100 according to particular emission scenarios.
No.. In fact warming since mid-late 20th century along with careful attribution studies is perfectly consistent with at least 2 oC of warming per doubling of [CO2].
No not really. Our understanding of all the elements of global warming exit entirely independently of Popper (or any other philosopher of science, however interesting their philosophies might be).
“Now, other things have been falsified within the range of our available observations.
Warming at a rate of 2C/century, for any trend of 17 years or more through 2015 has not occurred, so theories which postulated such warming are false.”
This is rather ambiguously worded, so it is hard to be sure what the claim actually is, but has anyone actually promulgated such a theory?
Increasing tropical cyclone energy, increasing drought, et. al. are also falsifiable predictions.
“But we have confidence in the theory of global warming not in spite of Popper, but because of Popper.”
I don’t think that is looking at Popper in the right way. Falsification is a way of distinguishing between science and non-science, but to determine the degree of confidence [IIRC Popper talked of “corroboration” rather than “confidence”?] we should have in a theory/hypothesis depends on how often attempts at falsification have failed and how sharp those tests have been (how risky the prediction, how easily the hypothesis could be falsified); it isn’t a yes/no issue. Also it is not the only source of confidence, there are also issues such as consillience (e.g. we know CO2 is a greenhouse gas, we know there is a lapse rate, both are observable facts, so if there isn’t global warming, there needs to be some additional factor to cancel it out as global warming is a direct consequence of the two observed properties), thus Occam’s razor gives us confidence. As ATTP suggest, it isn’t as simple a matter as it is often made out to be.
The point I was making is that the 17 year thing leads to a very very blunt test of falsification as the chance of seeing a 17 year apparent pause/hiatus as a result of internal climate variability is not that low.
This is rather ambiguously worded, so it is hard to be sure what the claim actually is, but has anyone actually promulgated such a theory?
The IPCC did, of course, in the AR4. We’re not yet sufficiently into the period they indicated to test, so that will be something to watch.
A “theory” isn’t the same as a prediction/projection.
17 years plus one month
I *believe* that TE has silently thrown the “But Santer Said” card and *may* need to be “reminded” what the abstract actually says: Our results show that temperature records of at least 17 years in length are required for identifying human effects on global-mean tropospheric temperature.
TE “The IPCC did, of course, in the AR4.”
I don’t recall AR4 making decadal scale projections. As Easterling and Wehner (2009) showed it isn’t unexpected to find periods of a decade or two with a low rate of warming or even cooling in the model runs, even during a long term (centennial scale) warming. so we shouldn’t be that surprised to see it in the observations either.
I notice that you didn’t disambiguate the claim, even though I pointed out that it was ambiguous. The discussion might be more fruitful if you explained what you meant rather more clearly.
“Our results show that temperature records of at least 17 years in length are required for identifying human effects on global-mean tropospheric temperature.”
Indeed, some also have difficulty understanding what is meant by “at least” (hint: in benign conditions 17 years may be enough, but a super-El-Nino spike may confuse things and you might need more). Sure 17 years is enough to make the apparent pause worth studying, but the idea that it approaches a sharp test for falsification is rather overstating things.
For the next two decades, a warming of about 0.2°C per decade is projected for a range of SRES emission scenarios. Even if the concentrations of all greenhouse gases and aerosols had been kept constant at year 2000 levels, a further warming of about 0.1°C per decade would be expected. …
I studied at the knee of Girma.
JCH indeed, however the SPM is of course only a summary, meaning there might be some more details in the body of the report, for instance the error bars.
Dikran Marsupial,
My understanding is that *if* sources of internal variability and natural forcings conspire to cause decadal trends which are lower than a longer term rate or expectation, an anthropogenic forcing is still detectable over a sufficiently large interval, which according to Santer is *at least* 17 years. IOW, attribution is trend-agnostic so long as we can suss out the individual signals. Obviously, this can only reasonably be done post hoc.
That’s a long way of saying I agree with you; failure to predict internal variability, major volcanic eruptions, solar output fluctuations, actual emissions, etc. does not necessarily entail “falsification” of teh modulz. (We already know they’re wrong anyway ….)
Turbulent Eddie said on May 31, 2016 at 3:48 pm,
“Warming at a rate of 2C/century, for any trend of 17 years or more through 2015 has not occurred, so theories which postulated such warming are false.”
Irrespective of whether this claim “warming at a rate of 2C/century, for any trend of 17 years or more through 2015 has not occurred” is true (it seems that it’s not), let’s examine first the larger claim that this prior claim implies that “theories which postulated such warming are false”:
This larger claim is an example of what I said and amplified symbolically in my comment on May 30, 2016 at 12:51 pm, which is that those who claim such and such has been falsified in climate science commit the fallacy of not taking into consideration and properly dealing with the set of background assumptions or auxiliary hypotheses, which exists *outside* the core hypothesis or test hypothesis.
A general version of this larger claim (irrespective of whether the prior, antecedent claim is true, which it seems is not) is that “if p has not occurred up to now, then p cannot occur in the future”, and it seems to be an example of the fallacy of hasty generalization.
This type of reasoning especially shows itself to be wrong if we translate it to a mathematical context. For instance, without knowing the formula for a real valued function on a given interval, it’s wrong to take a proper subset of the graph of that function and proclaim that since no part of that subset has a slope of at least x, no part of the whole function could possibly have a slope of at least x.
How is this last point at least possibly very relevant? Look at a graph of the 60 year running mean of the global temperature record since the late 1800s up to recently. Note two things: (1) It filters out any broad multidecadal cyclic behavior, and (2) it tracks not a straight line but a positively accelerated curve. Consider the possible implications of it tracking a positively accelerated curve in terms of possible future increase rates.
In my comment
under the post “Hmmm, entering a cooling phase?”, I presented a graph of a 60 running mean as well as some other graphs to point out that it seems that the underlying long term global warming is accelerating, and this comment includes a link to a post – which includes some comments – by Tamino at Open Mind that argue that the underlying long term global warming is accelerating.
”
…failure to predict internal variability, major volcanic eruptions, solar output fluctuations, actual emissions, etc. does not necessarily entail “falsification” of teh modulz. (We already know they’re wrong anyway ….)
”
Even if we accept that the teh modulz fail, and do not account for the temperature metrics that everyone likes to talk about, we have a bunch of other robust metrics that would have to be similarly ‘falsified’ or explained away…
You know, stuff like:
Historical and archaeological evidence
Glaciers
Arctic sea ice loss
Vegetation
Pollen analysis
Cloud cover and precipitation
Dendroclimatology
Ice cores
Animals
Sea level change
William Whewell called this confluence of evidence ‘consilience’, and it would be one hell of a feat to ‘falsify’ all that.
http://plato.stanford.edu/entries/whewell/
Now, other things have been falsified within the range of our available observations.
Warming at a rate of 2C/century, for any trend of 17 years or more through 2015 has not occurred, so theories which postulated such warming are false. …
The 17-year trend to 2006, according to WFT, using GISS L&O, is:
#Selected last 204 samples
#Least squares trend line; slope = 0.0214378 per year
So 100 X .0214378 is 2.14378 ℃ per century, right?
And when 2016 is done, the 17-year trend to 2016 is going to be higher. So by the end of 2020, is .2 ℃ per decade achievable? I think it is. The PDO remains positive; regardless of La Nina in 2017-2017, this heatwave is not done.
Very Reverend, we agree. I perhaps should clarify that I was attacking “teh modulz fail, so AGW is bunk” argument which is based on the notion that if a theory can’t make reliable predictions, it isn’t a very good theory. Which is a silly argument because it implies geologists’ perennial failure to predict the precise location, timing and magnitude of earthquakes “falsifies” the theory of plate tectonics.
The Very Reverend Jebediah Hypotenuse says: “…robust metrics that would have to be similarly ‘falsified’ or explained away…”
We seem to have a different idea what falsification *means*. It doesn’t mean “explain away”. Rather, it is the means by which a theory can be tested.
Theory: The fuel for a burning candle is the wax.
Falsification: Burn the candle. If the wax does not diminish, then it wasn’t the fuel for the candle.
Weakness of confirmation: Even if wax is consumed, it might not be the *only* fuel. In fact, the wick is also fuel, but a minor component.
That is why merely confirming your theory isn’t good enough. You must identify possible confounders or other causes of the observation. It is common, but erroneous, to assume only one cause can possibly exist for any particular observation. It may be *likely* but consider an example: Dropping a sandstone rock and an iron rock, on Earth, in a vacuum. The iron is attracted not only by gravity but by the Earth’s magnetic field and thus I expect it to accelerate slightly faster downward as compared to the sandstone.
To falsify the theory I have just proposed, you conduct an experiment. If they accelerate at exactly the same speed, then my theory is falsified. But the measurement will have to be extremely precise since the difference will be small.
Historical and archaeological evidence: Asserts much, falsifies some. For instance, Christianity used to assert without exception that all current forms of life were created ex-nihilo in their current forms. Finding fossils falsifies those claims, but also falsifies all other claims that depend solely on the first claim falsified.
long term global warming is accelerating
I don’t believe so.
Here are the trends through 2015 beginning with the year indicated for each point:
There are upward spikes for trends beginning around the El Chichon eruption and the Pinatubo eruption, and a dip for trends beginning around the 1998 El Nino.
But otherwise, trends are bouncing around the 1.6 to 1.7 C per century range.
> “teh modulz fail, so AGW is bunk” argument which is based on the notion that if a theory can’t make reliable predictions, it isn’t a very good theory. Which is a silly argument because it implies geologists’ perennial failure to predict the precise location, timing and magnitude of earthquakes “falsifies” the theory of plate tectonics.
It also implies we have adamant data, which is a bit strange considering Willard Tony’s quest to show otherwise.
If we don’t have adamant data, then the problem might not be with the theory at all.
Yabbut, we have very adamant data, Willard:
I note that TE still hasn’t disambiguated his claim.
ATTP May 31, 2016 at 5:22 pm “A “theory” isn’t the same as a prediction/projection.”
true but
Chris says: May 31, 2016 at 12:38 pm
“If you could acknowledge that for Popper we falsify hypotheses, not directly theories, let alone “core” theories, that would be great.Yes that’s a distinction that should be made (theory/hypothesis). Is it the case that Popper’s falsification referred explicitly to hypotheses rather than theories?”
Chasing chimeras [thanks Willard, better than Mosher’s Unicorns as more functional].
Chris,hypotheses and theories are the same thing, more or less, give or take.
See ATTP above for something different.
Word games to try to get around the problem of falsification of part of a hypothesis[or theory].
You should not do it.
If a hypothesis[or theory] has multiple components which are interlinked and dependent then falsifying one is capable of falsifying all.
As Willard and Duhem point out there is falsification in isolation which may* not falsify the whole hypothesis[or theory].
But the existence of the falsification demands more than sweeping it under the carpet [we falsify hypotheses, not directly theories] or pretending it is a help.
The hypothesis[or theory], not dying by minor flesh wounds, can only grow stronger,
Nietzsche and KeefeAndAmanda is a sweet and noble concept but fails Occam’s razor and the KISS test.
If** a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.
TE,
Any chance you could add uncertainty intervals to your figure?
angech,
Where does this come from? This isn’t obvious to me. What would it mean, for example, to falsify AGW as a whole?
Really angech? That sentence is potentially interesting but it’s also meaningless. In what way are they “the same thing” and what are the qualifiers “more or less”/give or take” meant to signify? Why not get of your perch, stop avoiding Popper and say something falsifiable? Give us some examples.
Have a read http://strangebeautiful.com/other-texts/popper-logic-scientific-discovery.pdf
angech said on June 1, 2016 at 8:03 am
“Nietzsche and KeefeAndAmanda is a sweet and noble concept but fails Occam’s razor and the KISS test.
If** a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.”
Actually, Occam’s razor and the KISS test don’t apply when one denies fact, such as the fact of the existence of auxiliary hypotheses that are distinct from the core or test hypothesis. See the below for more on this.
Also, note that although the negation of one of the conjuncts of a conjunction implies the negation of that conjunction, it’s false that the negation of a conjunction implies the negation of each of its conjuncts.
Study the article at the University of California, Berkeley site on bundling hypotheses I shared on May 29, 2016 at 5:36 pm, and then study the symbolic logic I gave in my comment on May 30, 2016 at 12:51 pm. This logic shows in symbolic form what is said in that Berkeley article and it shows in symbolic form the main mistake made by those who claim falsification.
This main mistake can manifest in at least a couple of ways. Referring to the symbolism I used in that comment, one way it can manifest is the denial of the existence of A (which can be a conjunction of statements) as distinct from T (which can also be a conjunction of statements). Another way it can manifest is the false claim that ~(A & T) implies ~T, this false claim being a partial example of the mistake you might be making above in which you might be claiming that the negation of a conjunction implies the negation of each of its conjuncts.
Note 1: The correct logic is that the negation of a conjunction is equivalent to (both implies and is implied by) a disjunction in which each of its disjuncts is negated – in general symbolic form, this correct logic is ~(p & q) <=> ~p v ~q
Note 2: Here I use “<=>” instead of “<->” to signify that the biconditional is a tautology (a statement true in all its substitution instances) rather than merely material (truth functional).
Final point to address the false use of Occam’s razor and the KISS test: Again using the symbolism I used in my aforementioned comment: In science, we can’t use T -> P instead of (A & T) -> P as a premise in a claimed proof (a claimed falsification is a claimed proof) of ~T without contradiction, since there are known instances in the history of science in which we had negation of this conditional T -> P temporarily hold, this negation being the conjunction T & ~P, where T was true even through ~P was measured. There’s a law in logic such that we can’t use a statement as a premise or axiom in a proof if we can’t define that statement to be true axiomatically, that is, if we can’t define it to be a tautology for the purposes of the proof. We can’t define conditional T -> P as a tautology so that it would be an axiom in a proof because of these known instances in which we temporary had T with a measurement ~P. But we can define (A & T) -> P tautologically so that it would be an axiom in a proof via an appropriate definition of A.
One way to have such a definition of A would simply be to (mathematically) define A to be the smallest conjunction of conditions such that the disjunction ~A v ~T v P (which is one of many forms equivalent to the conditional (A & T) -> P) is tautological. Note: This mathematical approach to defining A means that for the purposes of the definition, we don’t have to explicitly identify the conjuncts of A ahead of time. This last point neatly entails what scientists do when they look for a condition that would qualify as a false conjunct in A if they measure ~P. That is, if they find a condition such that if they rerun the test with the negation (which is simply the logical opposite) of that condition holding and this rerun measures P instead of the prior ~P, then for the purposes of these tests, that condition qualifies as a false conjunct in A they were looking for. (Note that this condition could itself be a conjunction or disjunction of conditions, in which case its negation would be either a disjunction or conjunction of conditions, respectively.)
(A recent notorious example in which we temporarily had T with a measurement that said ~P:
https://en.wikipedia.org/wiki/Faster-than-light_neutrino_anomaly
Some measurements said that some neutrinos were traveling faster than light. That is, we had measurement ~P. I recall bloggers and commenters all over the world proclaiming the end of Einstein. They evidently wrongly thought that it was OK to use T -> P as a premise rather that the required (A & T) -> P. In the end, conjunction A was found to be false – one of its conjuncts was found to be false. So we had ~A – we did not have ~T, the end of Einstein.)
Popper:
To begin with, I shall exclude from our discussion the application of the term ‘simplicity’ to anything like a presentation or an exposition. It is sometimes said of two expositions of one and the same mathematical proof that the one is simpler or more elegant than the other. This is a
distinction which has little interest from the point of view of the theory of knowledge; it does not fall within the province of logic, but merely indicates a preference of an aesthetic or pragmatic character. The situation is similar when people say that one task may be ‘carried out
by simpler means’ than another, meaning that it can be done more easily or that, in order to do it, less training or less knowledge is needed. In all such cases the word ‘simple’ can be easily eliminated; its use is extra-logical.
op cit:
Schrödinger’s theory, for instance, is of great simplicity in a methodological sense, but in another sense it might well be called ‘complex’. We can say of a problem that its solution is not simple but difficult, or of a presentation or an exposition that it is not simple but intricate
Backs up KeefeAndAmanda
To falsify AGW as a whole would entail turning science on its head.
It would deny the logic of CO2 being a GHG.
My concerns are and will be degree, danger, disinformation, alarmism and fairness.
–
Your other comment re falsification of part of a theory or hypothesis making the whole theory false not being obvious is disingenuous surely.
It is a basic statement of maths and science when they are not counterintuitive.
–
AAK and Willard are pretty high powered. It’s late. If I can get up early and adjust I will reply to AAK’s provocative statement putting up the exception that proves the rule argument, which is of course wrong, using that long tail of infinity that recurses around to the opposite argument.
In short what is suggested, in parts is individually correct but the summation of the parts gives a wrong answer.
Obvious to those of us who do Sudoku.
AK has got two nines in the same line through twisting the joining paths.
Paradox is not always obvious to those who do not want a paradox.
angech,
degree, danger, disinformation, alarmism and fairness.
Only one of these has the remotest chance of being included in a falsification framework. The others are about risk, which is something quite different.
Which one and how? (this is the core of what I’m working on, so it is a moot question, because I know the answer)
So
AAK introduces auxiliary hypotheses that are distinct from the core hypothesis.
To claim that disproving the auxiliary hypotheses does not disprove the core hypothesis.
?
That is stating the obvious is it not?
Except he took 50 lines (too tired to count), to state said obvious which has nothing to do with my argument. One could mention disjunction at this point .
This is no rebuff to my comment
–
“If a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.”
–
Nowhere is there any mention of disproving an auxiliary hypothesis.
Nowhere is there room to replace part of a destroyed theory with
“Oh, it was disproved so it was only an auxiliary theory.”
Which is the line being run by some to evade Popper.
–
If a branch of the theory dictates certain observations should occur and they do not occur,
The pause.
The hot spot.
The Antarctic ice increasing.
–
Calling them auxiliary theories is one way out.
Another is denying that they are happening.
And the third, as Mosher said, is to reassess the hypothesis.
angech you are again marching firmly in opposition to Popper who would never have considered that theories should be considered flawed by pretending that things that probably do exist (“hot spot”) don’t, or that things that are not integral elements of a theory are asserted to be so. Why the anti-Popper stance?!
The hot spot likely does conform to predictions from modeling – we aren’t completely sure since radiosonde and satellite MSU data are problematic, but recent analysis indicates that the upper troposphere is warming as expected:
http://iopscience.iop.org/article/10.1088/1748-9326/10/5/054007/meta;jsessionid=BEFE30FD77C410F9F3A11AD1BFE1E6A3.c5.iopscience.cld.iop.org
Click to access jtech.pochedley.2015.pdf
This is very interesting from a philosophy of science perspective, wouldn’t you say? While it’s very easy to attempt to trash models this is an excellent example of their value. We might tend to assume a model is wrong (all models are wrong!) when there is a disagreement with experimental/observations, but it may well be the observations (and their analyses) that are wrong, which has clearly been the case so far with the very difficult tropospheric temperature measurements (see my earlier post for earlier cock-ups with the observations/analyses:
This leads to the question – when do we decide our hypothesis/theory/model is “wrong” in the face of apparent observational conflict?
Note btw, that although some anti-Popperians love to beat up on models when that think they can winkle an apparent conflict with observations, scientists love these apparent incompatibilities since they provide an arena for a focussed research effort.
n.b. “The pause” and “Antarctic (sea?) ice increasing” are both compatible with expectations from earth response to enhanced greenhouse warming.
You need to sort out your theories from your hypotheses. To be fair you’re not the only one who’s been a little lax with this (me included)…
@-angech
“The pause.
The hot spot.
The Antarctic ice increasing.
–
Calling them auxiliary theories is one way out.
Another is denying that they are happening.
And the third, as Mosher said, is to reassess the hypothesis.”
A fourth might be to observe that the pause is a statistical artifact that is generated by cherry-picking dates.
The hot spot is below our ability to detect given the high noise level of tropospheric measurements and the dependence on satellite measurements that rely on models of atmopsheric energy absorption that have proved extremely error prone, requiring numerous corrections and amendments.
The Antarctic multi-millennia land ice is decreasing as shown by in situ and satellite measurement. AS is the area of grounded ice sheets. Only the seasonal sea surface ice is increasing for a few weeks each year, for reasons that are well understood. Judith Curry was an author on a paper that discussed this some years ago I believe.
Asserting something is a falsifiable hypothesis that has been refuted by observation when the issue you raise is neither a hypothesis, central or auxiliary, and has NOT been contradicted by any observation is unpersuasive.
Presumably in reply to my comments on May 30, 2016 at 12:51 pm and June 1, 2016 at 11:17 am, angech said on June 1, 2016 at 1:06 pm,
“So AAK introduces auxiliary hypotheses that are distinct from the core hypothesis.
To claim that disproving the auxiliary hypotheses does not disprove the core hypothesis.
?”
Why the “?”? I did not invent the idea of auxiliary hypotheses that are distinct from the core or test hypothesis. It’s what everyone should know. And these auxiliary hypotheses or background assumptions always exist in the background. Note that the form A -> (T -> P) makes explicit this background function and is equivalent to the form we usually see, (A & T) -> P. See such articles as
Duhem-Quine thesis
https://en.wikipedia.org/wiki/Duhem%E2%80%93Quine_thesis
and
Duhem
https://thephilosophyofscience.wordpress.com/2011/06/15/duhem/
for more.
The latter article is one example of the many times people have written the form using variables, and that author puts it this way:
“1. If T and K1 and K1 … and Kn is true, then O is true.
2. Not-O.
3. Either T or K1 or K2 … or Kn is false.”
Recall that I put it this way, using only symbolism:
(A & T) -> P
~P
~(A & T)
~A v ~T
The former puts it this way, in words, where it goes the extra step of including the inductive step of taking A to be true – don’t forget that A is equivalent to ~~A, by which we then use disjunctive syllogism to infer ~T:
“Although a bundle of hypotheses (i.e. a hypothesis and its background assumptions) as a whole can be tested against the empirical world and be falsified if it fails the test, the Duhem-Quine thesis says it is impossible to isolate a single hypothesis in the bundle. One solution to the dilemma thus facing scientists is that when we have rational reasons to accept the background assumptions as true (e.g. explanatory scientific theories together with their respective supporting evidence) we will have rational – albeit nonconclusive – reasons for thinking that the theory under test probably is wrong in at least one respect if the empirical test fails.”
Angech wrote:
“This is no rebuff to my comment
–
‘If a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.’
–
Nowhere is there any mention of disproving an auxiliary hypothesis.”
Yes it is a rebuff. Your mistake is to not address the existence of the conjunction of auxiliary hypotheses A. Why must A be addressed and properly dealt with? We can’t even begin to talk about “destroying” (your term below) a test theory or hypothesis T without giving good reasons why the conjunction of all auxiliary hypotheses A holds, so that you can take the inductive step of affirming A to infer ~T.
“Nowhere is there room to replace part of a destroyed theory with
“Oh, it was disproved so it was only an auxiliary theory.”
Which is the line being run by some to evade Popper.”
First, there is no evasion of Popper – he accepted auxiliary hypotheses that are distinct from the test hypotheses. Second, no part of the “destroyed” test hypotheses is being replaced by something. You’re getting it all wrong.
“If a branch of the theory dictates certain observations should occur and they do not occur,
The pause.
The hot spot.
The Antarctic ice increasing.
–
Calling them auxiliary theories is one way out.”
I’ll deal with this last statement first: It’s incorrect that we call them auxiliary hypotheses. (Replace “pause” with “statistically insignificant slowdown”, by the way.) In an attempted falsification of ~T (the test hypotheses), one would have to try to make them as instantiations of ~P. They would not be instantiations of A, and they would not be instantiations of ~A. In each of those three instances, many scientists have published many papers that show solid reasons that ~A holds, which means we can’t infer ~T even if we tried to take these three as instantiations of ~P. Via adjustments to the assumptions, these three properly stated have been shown to be instantiations of P, consistent with T. Consider that even Judith Curry says you’re wrong on the last of those three. Don’t forget that she coauthored a paper in 2010 that essentially argues that global warming actually is causing an increase in Antarctic sea ice that may continue for decades until it finally starts to reverse itself – that is, they argue that increasing Antarctic sea ice would actually be an instantiation of P, meaning that it is consistent with T. See
Resolving the paradox of the Antarctic sea ice
https://www.sciencedaily.com/releases/2010/08/100816154958.htm
for more.
Finally, you have yet to tell us what you mean by “in whole” in your above claim. You could be committing the mistake of thinking that the conjunction of auxiliary hypotheses or background assumptions A is part of the test hypotheses T. It is not.
angech wrote ““If a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.””
No, the hypothesis may be that if A or B then C, in which case demonstrating that A is false does not falsify C.
angech wrote “If a branch of the theory dictates certain observations should occur and they do not occur,
The pause.
The hot spot.
The Antarctic ice increasing.”
Funny that Arctic sea ice declining faster than is predicted by the models is never raised as an example of falsification. ;o)
Also this is not the best representation of falsificationism, for falsification you need a model that specifies what cannot happen, not what should occur. The pause is not a good example, as the model runs also contain “pauses”, it is just that the timing is unpredictable, so the models tell us that we should expect to see them (apparent pauses), and indeed we do.
> Is it the case that Popper’s falsification referred explicitly to hypotheses rather than theories?
Popper explicitly referred to statements as falsifiable, but also all kinds of things, according to his whim. Popper also talks about falsifying a model (e.g. Rutherforld falsified Thompson’s atomic model). There’s also the idea that one could test an explanation, for instance when he says that the most satisfying explanation is the one that is the most severely testable or tested (Induction, 15, p. 145).
When discussing his logical (or rather meta-logical) argument against induction and confirmation, Popper handwaved to simplistic theories that could be expressed in a single, basic statement. If one could produce a falsifier of a basic statement that represents a theory as a whole, then of course we could say that the theory is falsified as a whole. To go from basic statement to theory he needs to maintain what he calls the principle of empiricism. There’s a difficulty here in which I can’t delve for the moment. Suffice to say that all the things that can be falsified for Popper are conjectural, which means they can become an hypothesis that we can test.
By his own admission, Popper wasn’t very precise in his wording, and prefers clarity over precision and commonsensical usages over exact definitions. Yet he got to define “falsifiability” in his Realism and the Aim of Science to correct popular misconceptions in the sequel to his Logic. In his 1982 introduction, he distinguishes a first sense of falsifiability (p. xxi):
This quote shows very well how Popper, with his swan example, shifts from theories to singular statements of them – “All swans are white” is a falsifier of which theory, except the one according to which swans are white?
Also, note how the “All swans are white” can be falsified: by producing a swan that is not white. A true observational statement would refute it, e.g. “here’s a black swan.” Which means that the overall color theory of swans requires very little formal apparatus. There’s no maths, no quantified relationships, no measurement – a universal quantifier, a natural kind, and a property.
Finally, note that the overall testing procedure to produce the falsifier is abstracted away. There’s no background like we read in papers. In this case it could be something like this:
As he lay out a bit below, we need to specify the conditions under which we would be prepared to regard a theory as falsified. Only when we specify such conditions do we operationalize the whole gamut and work by way of hypothetical testing. Only then models get some ice time.
When we consider scientific theories as a complex comprising non-observable constructs and belonging to network of other related theories, the Popperian caricature falters. It made sense to make an epistemological point. That point, however, has little impact for philosophers of science.
Popper’s conception of falsifiability has little scientific grit because it was meant to solve the demarcation problem, which for Popper is a logical (or rather a metalogical) problem. Hence Popper comes up with a second sense for falsifiability (p. xxii):
So here we have Popper himself saying that there’s no falsification (of theories) that is usually comprised by contrarians.
PS: That note is a bit rough, but I have dinner to prepare. I may revisit it later.
Let’s add this TL;DR because I need to go.
For Popper:
(1) Science is a bunch of conjectures you can in principle falsify, but never confirm.
(2) Scientific theories are conjectures should be amenable to universal statements.
(3) Scientists should be able to specify counterexamples to their theories.
(4) Scientists are never compelled by logic alone to reject their theories.
From this we can see that Popper:
(5) sometimes conflates scientific theories with testing hypotheses;
(6) only considers scientific theories that contain natural laws;
(7) acknowledges that his epistemology doesn’t replace psycho-sociological explanations;
(8) does not go as far as contrarians usually go in his name.
***
Popper cranked up his argumentation with at least two interesting moves. First, he’ll appeal to rationalism and criticism, with hints of moralism that has been immortalized in ClimateBall ™ with “but Feynman.” Second, he’ll generalize human activity to problem solving.
To be fair, that was pretty much my point.
“Let’s add this TL;DR because I need to go.”
Thanks Willard, good synopsis
Alfred Wegener and his theory of Continental Drift:
While all of this is essentially true, Wegener’s theory is not considered correct. He lacked the correct mechanism for how continents moved. Neither of his two proposed possible mechanisms (subordinate (hypotheses) were correct. Today we recognize that Pangaea did exist and that the continents have moved to their present positions due to plate tectonics. Falsifying the subordinate hypotheses did not falsify the main theory.
Elegantly argued, O’Niell, thanks.
Dikran Marsupial says: June 1, 2016 at 3:49 pm
“”angech wrote ““If a hypothesis[or theory] is falsified in part it suggests strongly that it must* be falsified in whole.””
No, the hypothesis may* [for Brandon] be that if A or B then C, in which case demonstrating that A is false does not falsify C.”
–
No let up is there in misinterpreting the statement.
ie reinterpreting it to say something you can then be happy with.
The statement makes no provision for “if A or B then C”.
So do not do it.
Same problem as KeefeAndAmanda above, Ignore the quote and chuck in an auxiliary hypothesis or theory [C in your case].
It is “if A then B”. that is the problem.
–
A to B has a number of interdependent steps, If any one is falsified the theory is then falsified.
It may be resuscitated by finding a way to save that step but it will involve changing the theory at that point.
You are better than that Gavin
Dikran Marsupial says: June 1, 2016 at 3:53 pm
angech wrote “If a branch of the theory dictates certain observations should occur and they do not occur,
“”Funny that Arctic sea ice declining faster than is predicted by the models is never raised as an example of falsification. “”
The only thing funny is that is an example of verification.
Your idea of humor?
> A to B has a number of interdependent steps, If any one is falsified the theory is then falsified. It may be resuscitated by finding a way to save that step but it will involve changing the theory at that point.
Indeed, if by “interdependent” you mean that B follows logically from A. All you need, besides love, is to apply modus tollens. That B follows from A is not always the case. It wasn’t in O’Neill’s example.
More importantly, we need to be clear about what “changing the theory” means here. It could be to replace it with another one. It could also mean we patch it to bypass the falsifier. This patch could be found in the theory, but it could also be tweaked in another theory that is somehow connected with the falsifier. In fact, you can go as far as to say that theories don’t get evaluated in isolation. This would bring us back to Duhem’s thesis.
It should be clear by now that the concept of theory is far from being clear in our (and Popper’s) discussion.
angech,
I think the problem here is understanding how that maps to AGW.
Great, I’m crossing Willard now.
“I think the problem here is understanding how that maps to AGW.”
Theory
Increasing CO2 causes an increase in temperature.
Man is causing an increase in CO2 .
Hence it should warm. AGW.
–
Some exists.
–
Theory then has updates.
-Positive Feedbacks
-100% AGW, 110% AGW, etc
-CAGW Alarmists, Lukewarmers, Deniers
–
The essence of these arguments is about how much AGW is due to man and how bad it will be.
CAGW goes further and has positive feedbacks which enhance the basic scientific arguments.
– The theory then has to fit the observations, AKA the facts.
Depending on which AGW theory you adhere to certain facts must derive.
When these do not occur, ie are falsified, the particular theory needs revising if possible or discarding.
–
Grubby concepts like clouds and negative feedbacks must be considered.
Positive benefits of warming and sea level rise need to be mentioned, not dismissed and derided as never gong to happen.
–
A fair playing field where flaws in the theory are considered on their merits and positives are acknowledged by both sides.
angech.
go read this https://www.amazon.com/Search-Planet-Vulcan-Clockwork-Universe/dp/0738208892/188-0492106-1603928?ie=UTF8&*Version*=1&*entries*=0
or watch
angech wrote “The only thing funny is that is an example of verification.
Your idea of humor?”
Have the models been historically “verified” with respect to the rate of Arctic sea ice decline, or has the rate of decline been generally more than can be explained by the models?
@angech wrote “No let up is there in misinterpreting the statement.”
No, I tried to answer your question, I am not interested in this kind of rhetoric. In real science, hypotheses often do involve “or”s as well as “and”s, this is a minor point, and there was no reason not to accept the minor qualification rather than going on the offensive.
“You are better than that Gavin”
using someones real name rather than their pseudonym is bad manners (especially if you remain anonymous yourself); if someone uses a pseudonym then you should assume there is a good reason for it and honour their wishes (even when their real name is not a secret). This is application of the “golden rule” and is not rocket science.
angech,
Ok, works for me.
Positive feedbacks have been there since Arrhenius.
Yes, I thought their placement in that list was odd but skip it.
As I mentioned previously, positive feedbacks were from the beginning. The “C” doesn’t strictly require feedbacks at all, it just requires unmitigated warming to be catastrophic … which is ironically fitting because the C was prepended as an attempt to make AGW sound ridiculous. I don’t personally mind so much, catastrophe is what I’d like to avoid and there’s good literature support suggesting catastrophe is what we’ll get *if* we don’t act to reduce emissions.
I’m not about to advocate for a position which demands that CAGW fits the facts before attempting to do something about it.
As they are.
Positive effects of SLR warrants derision. Benefits of warming? Sure, some climates at higher latitudes become milder.
A better place to talk about fair and balanced is weighing positive and negative effects and attempting to derive the net. Real Science™ does not require medical researchers to give the time of day to homoeopaths — it’s up to the homoeopaths to go through the process of hypothesis formation, experimentation and refereed publication just like all the rest.
angech,
I agree with Dikran. If you’re choosing to be pseudonymous you can respect other people’s pseudonymity.
“Positive benefits of warming and sea level rise need to be mentioned, not dismissed and derided as never gong to happen.”
There are benefits to sea level rise? I’d be genuinely interested to hear what they are, it hadn’t occurred to me that there were any. I’d also be rather surprised if the IPCC reports do not mention the benefits of warming (e.g. expansion of boreal forrest), and take them into account in assessing the likely impacts of different courses of action.
In reply to my comment on June 1, 2016 at 3:48 pm, angech said on June 2, 2016 at 4:23 am,
“The statement makes no provision for “if A or B then C”.
So do not do it.
Same problem as KeefeAndAmanda above, Ignore the quote and chuck in an auxiliary hypothesis or theory [C in your case].”
To address this point on C: Again you make the same mistake of saying that an auxiliary hypothesis in the setup occupies the position of the consequent of a conditional. It does not. Condition C is in the position of the consequent of a conditional. I pointed out this mistake to you already in my comment on June 1, 2016 at 3:48 pm. Could you please correct yourself on this? That is, given that T is the hypotheses to be tested, the phrase “auxiliary hypothesis” or “background assumption” is *defined* such that the conjunction A of all the auxiliary hypotheses or background assumptions is the antecedent of the conditional A -> (T -> P), which is the setup for the falsification test, where P is obviously that which is implied by T given A. We can call P a prediction or projection of T. By a rule of the propositional calculus called exportation, we have the (tautological) equivalence
(A -> (T -> P)) <=> ((A & T) -> P).
This latter form (A & T) -> P is the more common form of the setup. I use the symbol “&” to denote the conjunction operator.
(I use the symbols “<=>” and “=>” to denote tautological equivalence and implication, respectively, and the symbols “<->” and “->” to denote material equivalence and implication, respectively.)
Again: The conjunction A of auxiliary hypotheses in the setup of the falsification test does not occupy the position of the consequent of a conditional. That position is ultimately occupied by the prediction or projection P. Again: Could you please correct yourself on this?
On June 2, 2016 at 6:15 am, angech said,
“When these do not occur, ie are falsified, the particular theory needs revising if possible or discarding.”
False. The condition ~P in and of itself does not imply ~T. The condition ~A can completely explain “when these do not occur”, which would mean the theory could need no revising or discarding at all. This happens many times in science, and either directly or indirectly via links I’ve given a number of examples in my comments in this thread of this happening. You are again doing what I said everyone who claims falsification of a climate science hypothesis do, which is not addressing A and properly dealing with it. I’d like to see you write out in symbolic logical form how to derive ~T from ~P. You will use T -> P as a setup for the falsification test, right? Recall my comment on June 1, 2016 at 11:17 am, in which I explained that you *cannot* use the setup T -> P for a falsification test. Doing so is the denial of A, committed by all who claim the said falsification. You must include A per the above.
And the degeneration of this blog to Russell and Whitehead continues
There are harsher criticisms than to degenerate to Russel and Whitehead ;o)
Dirkran, I apologise, I was not aware that you only wished to go by this name and did not intentionally try to upset you. I am naive at times, this was one of them, will not happen again.
Thanks to ATTP for explaining it as well.
–
Positive effects of sea level rise?
–
Serious ones. Yes.
-more free water (not ice) in the world means more moisture in the air , more rainfall , more crops less deserts, more greening of the world hence a more liveable world overall.
-more shallow oceans (as in more land close to sea just underwire) means more coral reefs, more fish , hence more food and more life in the world. Not great for some of the air breathing minority but great for the planet as a whole.
–
Less serious ones,
Beach front property will go up in price (less of it). Buy now!
–
humorous ones
The English Channel will keep the French out even better.
–
Historic ones
The refilling of the Mediterranean Sea ,which turned a giant salt pan into the vibrant centre of human life it is today.
–
I am sure that there are many more. The sea has covered the land many times in the past and it is hubris to say only this level suits us and what we want is what we should get and that sea level can only be beneficial at a fixed level.
Or that any and all change is bad, which seems the only way to interpret such negative comments by Brandon etc
> False. The condition ~P in and of itself does not imply ~T.
We’ve already been there more than a week already in another thread, K&A. Your T is not Angech’s T, nor is it Pop’s. (I could redefine T & A as U and say that by “theory” I was referring to U.) There’s an equivocation on that T, and because of it both you and Angech are talking past one another. It’s time to give it a rest.
Popper’s point wasn’t to logicize science, but to “solve” induction. His solution is to use modus tollens as a model (gasp!) of scientific inference. We are powered by contradiction. When we see one, it creates a problem. We like to solve problems. Etc.
angech,
This is not obviously true. In fact, I suspect it isn’t really true at all since you’re ignoring a number of things. It will be warmer, so there will be more evaporation and some regions will get drier. You’re also ignoring that even if we could adapt to whatever changes may take place, if the pace of the changes is sufficiently fast, then doing so will be difficult and could lead to all sorts of other problems; we don’t live in a world where we can simply move somewhere new when it becomes difficult to live where we do currently.
Huh? We’ve just had what might have been one of the worst coral bleaching events and you think they will benefit from a warmer world?
angech,
Part of this doesn’t make any sense, but you’re strawmanning. This isn’t about current levels being the only ones that suit us, it’s about them being the ones to which we’ve currently adapted. Sure, we can move and changes things, but that may not be easy.
Again, noone is saying that all change is bad. However, we are producing changes that may be faster than at any time in human history. That could lead to all sorts of complications that we may find difficult to deal with.
> The essence of these arguments is about how much AGW is due to man and how bad it will be.
This “and” connects two very different beasts, i.e. level 0:
and level 2:
angtech wrote “Serious ones. Yes.
-more free water (not ice) in the world means more moisture in the air , more rainfall , more crops less deserts, more greening of the world hence a more liveable world overall.”
No, this is not correct AFAICS. The amount of water in the atmosphere depends on air temperature (Clausius-Clapeyron relationship), not the surface area of the oceans. Perhaps you would like to defend this first point before moving on to the others.
BTW I don’t think it would mean less deserts either as most deserts lie at latitudes where Hadley cells descend (with their dry air).
Good point, I missed that. As I understand it, warmer air can hold more water vapour and the expectation is that relative humidity will remain almost constant meaning more water vapour in the air and more evaporation from the surface.
“His solution is to use modus tollens as a model (gasp!) of scientific inference.”
Though I think he did recognize the obvious point that one person’s modus tollens is another’s modus ponens. And so on …
ATTP yes, that is my understanding, and as water vapour is a GHG it means we have known about at least one source of positive feedback in the climate system for a very long time. IIRC a warming climate is likely to result in an intensification of the hydrological cycle, so where it is rainy at the moment it is likely to become more rainy; where it is dry, more dry. However I don’t recall ever reading that an increase in ocean surface area being a factor, probably because it isn’t, but as usual I am amenable to evidence.
Incidentally, the idea that more rainfall is a good thing is not obviously true.
> Though I think he did recognize the obvious point that one person’s modus tollens is another’s modus ponens.
I don’t think he ever did, Slizard. Just to please Eli, here’s Jean-Yves:
Click to access TS2.pdf
Layer 4 is the deontic layer:
Op. Cit.
Jean-Yves is wrong in saying that Popper was explaining truth (he was a Tarksian and did not believe it was possible), but his point shows that at the logical level, something’s amiss with falsificationism. Breaking symmetry is a steep price to pay if ultimately we only get to counter induction with deduction + liberty.
Willard said June 2, 2016 at 2:07 pm
“> False. The condition ~P in and of itself does not imply ~T.
>
I could redefine T & A as U and say that by “theory” I was referring to is U.”
If you wish to define U = A & T, and in so doing you’re referring to my conjunction antecedent A & T, which includes referring to my definition of A that I gave in one of my prior comments, the one on June 1, 2016 at 11:17 am, then sure, no problem.
But are you really doing that? I ask this question because you used “theory” to name U = A & T. First, note that my definition of A is not a fixed set for all possible instantiations of T. Second, I defined A so that (referring now to the set of conjuncts that comprise conjunction A) it could be a large set of statements to draw from (a subset of which would be a sufficient subset of the set of all of Popper’s background knowledge for a given instantiation of T [this sufficient subset could be of course equal to the set of all Popper’s background knowledge for a given instantiation of T], even though A as defined is of course a conjunction that contains a much smaller set than the set of all possible statements. This all means that it could – and very probably would – be *quite* a stretch to *reasonably* call U “theory”.
> This all means that it could – and very probably would – be *quite* a stretch to *reasonably* call U “theory”.
As far as I’m concerned, it’s even more of a stretch to separate a theory from all the other theorical constructs without which it can’t be articulated, from all the instrumentation apparatus by which its “active” components are tested, from the obvervations made in this setting, and from the background knowledge that makes all this relevant to us.
No semantic sleight of hand can counterbalance Pop’s elementary modus ponens. When you expect an event E to happen because your knowledge base B implies E and the contrary (not-E) occurs, there’s something wrong with that B. If you get a knowledge domain that can’t abide by this kind of inference, it can’t be science, at least according to Pop.
Pussyfooting about the true definitions of B’s components can only make Popperians smugger than they already are.
angech,
This recent paper suggests a significant decrease in future rainfall for Iberia. The critical point is the projected decrease in overall rainfall as well as rainfall in spring and autumn as this results in a longer dry season therefore increasing the risk of drought.
http://link.springer.com/article/10.1007%2Fs10584-015-1575-z
I’m not sure how you can conclude that sea level is a good thing give 1.3million people in the US alone are projected to live below the high tide level by 2100 for the lowest sea level rise projection . This increases to 7.8million for the highest projection. Managed retreat for parts of Florida and New Orleans would likely be the only option.
http://www.climatecentral.org/gallery/graphics/sea-level-rise-and-population-impact
According to the world bank almost 5% of the world population live below 5m elevation and this is likely to increase as population increases and more people migrate to coastal cities.
angech wrote “Positive benefits of warming and sea level rise need to be mentioned, not dismissed and derided as never gong to happen.”
It ssems that the reason that these are dismissed (note neither I, nor hyperactive hydrologist nor ATTP did any derision AFAICS) as never going to happen is that there are good scientific reasons why they will not happen (e.g. Clausius-Clapeyron). As I said earlier, I would be very surprised if the IPCC had not properly assess the benefits of warming as well as the costs, so I suspect this is a bit of a straw man anyway. If angech can provide some evidence to address the counter-arguments that have been presented above, and that there are non-negligible benefits that have not been taken into account by e.g. the IPCC, then his argument may have some merit.
I was going to leave angech’s second argument until later, but as HH has addressed it, the worlds continental shelf amounts to about 8% of the surface area of the worlds oceans, which is about 30million square kilometers, which is about three times the surface area of the USA. I don’t think the increase in useful shallow sea is going to be significant, even just looking at the numbers. Of course the value of the shallow sea created is unlikely to compare with the value of the land that would be submerged, which is (a) likely to be heavily populated, and hence invested in and (b) of high agricultural value (river deltas, like the Nile etc have lots of sediment deposition, which is likely to make them good arable land (when not flooded). Now if angtech could provide some evidence that (i) the area of additional continental shelf was in itself significant, and if so (ii) that the benefit of the additional fishery was non-negligible compared to the current value of the land that would be submerged and if so (iii) that the IPCC don’t already take this into consideration, then the argument might have some merit, but we need evidence, not just assertion.
“Popper’s point wasn’t to logicize science, but to “solve” induction. His solution is to use modus tollens as a model (gasp!) of scientific inference. We are powered by contradiction. When we see one, it creates a problem. We like to solve problems. ”
My sense is that if folks actually looked at how scientists handle anomalies ‘over the board’
one could build a book of tactics.
some nice history different strokes…
http://www.pbs.org/wgbh/nova/physics/solar-neutrinos.html
https://en.wikipedia.org/wiki/Tests_of_general_relativity
View at Medium.com
Willard said on June 2, 2016 at 5:32 pm in reply to my comment on June 2, 2016 at 5:10 pm,
“> This all means that it could – and very probably would – be *quite* a stretch to *reasonably* call U “theory”.
As far as I’m concerned, it’s even more of a stretch to separate a theory from all the other theorical constructs without which it can’t be articulated, from all the instrumentation apparatus by which your “active” components are tested, from the obvervations made in this setting, and from the background knowledge that makes all this relevant to us.”
We can separate some of it. I again recall that notorious event a few years ago when many bloggers all over the world were proclaiming the end of Einstein due to the measurement of neutrinos traveling faster than the speed of light. I think that it should be obvious that we can separate Einstein’s theory from the condition of some wires in the measuring equipment not being hooked up properly. And I think it should be obvious that we can find many examples throughout history where we can separate the theory to be tested from much of this other stuff.
Also: Nothing of what I have written implies that we must be able to “separate” such, anyway. There’s nothing in what I wrote that implies that the pair-wise intersections of the domain sets of the variables A, T, and P in my A -> (T -> P) (A & T) -> P must be empty. And there’s nothing of what I wrote that implies that we must be able to identify every element of every domain set before we can acknowledge the existence of these domain sets or their logical relations. Also, please recall that mathematically, we can talk about the existence of subsets of sets and many of their logical relations among each other without ever having to identity which element is in which subset. And note that these conditionals in these falsification setups I define are mathematical objects, mathematical abstractions.
“No semantic sleight of hand can counterbalance Pop’s elementary modus ponens.”
Actually, there is. Here’s how:
“When you expect an event E to happen because your knowledge base B implies E and the opposite (not-E) occurs, there’s something wrong with that B.”
Not necessarily, since for all you know there could be something *outside* that knowledge base B than explains why event E occurs. Please know that the knowledge base of a human is seriously limited. And so this possibility of something *outside* that knowledge base B is always there, logically inescapable. Hence my background A (it is defined to have a sufficient set of information “outside” your B that is limited by the human capacity for knowledge) as I defined it, which, again, is a conjunction that contains a certain subset of the set of all possible statements. Hence my form using your variables B and E of A -> (B -> E), which is equivalent to (A & B) -> E. Note that I said that we don’t (and, by the way, can’t) know what A is ahead of time as I defined it. We don’t need to know what it is – it’s sufficient to simply know that it exists (see existence definitions and existence theorems in mathematics).
That is, things don’t imply things in vacuums. Even in my field of mathematics, even though every theorem is or can be put into the broad form of a conditional p -> q, there is always a set of background conditions that function as the antecedent for this proved conditional. That is, when we prove p -> q, we really prove A -> (p -> q), which means that we simply draw from this background set A whenever we need to via such as definitions, axioms, postulates, and the intuitive motivations for definitions, and so on. It’s no different outside of mathematics when addressing a conditional p -> q, including using it a setup for modus tollens to investigate whether p is true – A is always in the background, its existence inescapable.
Dikran Marsupial,
You’re Doing it Wrong; we’re arguing hypotheticals not specifics here. Given that *many* warmists say disparaging things about opposing points of view because they can’t tolerate dissent, let me extend to a generalized hypothetical scenario: *IF* one’s interlocutor makes an unsubstantiated assertion which flies in the face of large bodies of extant literature, all the tone trolling in the world won’t change the fact that the argument is unsupported and *likely* ridiculous.
I leave it for otters to decide whether derision is warranted in specific cases. As for me; I’ve tried the “polite” route and found that it doesn’t make a damn bit of difference *if* folk don’t like the message I’m bearing. Way I figger, why then bother pussyfooting around what I really think? Does not Integrity demand I be honest about what’s on my mind?
thanks Willard (re: “Is it the case that Popper’s falsification referred explicitly to hypotheses rather than theories?”)
The falsifiability of statements seems like a good starting point. A statement is scientific if it is falsifiable.
Obviously a concept doesn’t become falsifiable by being formulated as a statement (“God exists”). And predictions are problematic. In fact the contemporary and rather urgent application of scientific knowledge/theory in a predictive sense (that is, predicting the progression of phenomena into the future) may actually not fall within Popper’s remit.
For example statements like – “we’ve received the biopsy result and we’re sorry to say you’ve got less than 6 months to live” or “given expected progression of greenhouse gas emissions the Earth surface will be between 0.4-0.6 oC warmer in 2040 than now”…are both unfalsifiable (there’s no set of experiments and observations that could be made now that could falsify these statements), and also perfectly falsifiable (are you still alive after 6 months – yes or no?….wait until 2040 – is the surface temperature in 2040 in the predicted range – yes or no?) And if it isn’t in the predicted range, what exactly has been falsified – nothing more than the statement. The theory that informs the predictions in both cases is likely to be secure.
Which is why we address predictions into the future in terms of probabilities. And we can’t really apply Popperian falsifiability to these I think. Can we?
Hypotheses, properly formulated, are eminently falsifiable. What about the falsifiability of theories? One problem is that it isn’t totally obvious what a theory is. For example if we consider these 4 things that are called theories (Theory of Evolution; Plate Tectonics Theory; Chemiosmotic Theory; String Theory), at least two of them (Plate Tectonics and Chemiosmosis) seem to be so secure as to be beyond the possibility of falsification even if these might be refined in the future – the same possibly applies to Evolution, although here one can more easily imagine some astonishing revelation that overturns this paradigm. Perhaps Plate Tectonics and Chemiosmosis shouldn’t be called theories any more. And maybe String Theory doesn’t deserve to be called a theory yet.
Brandon,
I guess it depends what your goals are and on who else might be reading the exchange. I’ve not always managed to remain calm during discussions that I’ve found frustrating, but I’ve rarely not regretted losing my cool, even if I thought I was justified in doing so and only saying what I really thought.
> Not necessarily, since for all you know there could be something *outside* that knowledge base B than explains why event E occurs.
That’s just another way to appeal to ignorance. I see at least three problems with that kind of argument against Pop’s thesis.
First, there’s no such thing as necessity in empirical sciences. Falsificationism remains at its core a fallibilist doctrine. Once you see a problem with your conjectures, you can deal with them however you fancy, as long they preserve their explanatory power. (Besides, there are people who will never grasp modus tollens.) The “refutation” part is simply there to specify a necessary condition: if can’t imagine falsifiers to your theory, it can’t be science.
Second, once that something *outside* B helps explain E, then B changes, since B represents everything you know that is relevant to infer, predict or deduce E. If you want to keep B the same, then your knowledge of B changes, which means you’ll have to replace B with whatever letter in whatever alphabet. Either way, Pop’s covered – there’s some kind of deductive engine to criticize and reorganize conjectural knowledge and not an inductive one.
Third, we must ensure that the thing *outside* B should preserve explanatory power, otherwise the explanation becomes ad hoc or worse circular. What needs to be explained must remain independent from the explanation, and must carry enough information that it comes with testable consequences.
> Which is why we address predictions into the future in terms of probabilities. And we can’t really apply Popperian falsifiability to these I think. Can we?
Unless you buy Pop’s propensity stuff, Chris, you can’t. If you bought into induction, like just about any contemporary scientist who deals with statistical hypothesis testing, you can dispense yourself from the stringency of translating everything you do into popperese. As long as we’re clear about what Popper did and why, no harm done.
Perhaps it would be better to understand Popper if we simply present his modified essentialism (RAS, 15. The Aim of Science, p. 137), which is the crux of his Realism-with-a-big-R:
Popper will later on clarify what he means by depth and structure. Interestingly, he’ll note in parenthesis that structures, like laws, explain things because “laws impose a certain kind of structure upon the world, and that may be interpreted, alternatively, as descriptions of that structure.
By “explanatory theory”, Popper has in mind a very simple deduction setting with two parts, traditionally called an explicans, and an explicandum:
[Explicans] Natural laws and initial conditions
====================================
[Explicandum] State of affairs, predictions, consequences, etc.
Popper satisfies this model by formulating theories by way of one single, basic sentence. This basic sentence works as an explanation. When he states Galileo’s and Newton’s theories about the behavior of projectiles, he uses one sentence in each case. But look how “basic” they appear to be (p. 140-141):
[Galileo] A thrown stone moves in parabola, except in the case of a free vertical fall when it moves, with constant acceleration, in a straight line.
[Newton] A projectile on earth moves along an ellipse with finite excentricity (rather than a parabola).
The whole distinction between explicans and explicandum has disappeared. Why? The explicans, i.e. the theory that serves to explain projectile movement, is implicit and taken for granted.
Anders,
We agree.
Sure. I’ve been known to completely melt down online, and yes I definitely regret doing that. Part of what I’m disputing here is the notion of saying “harsh” things about an argument I (or anyone) thinks is crappy constitutes a loss of control, or means that the rebuttal is irrational, unfounded, etc. “Tone” is so subjective that it’s easy for anyone to dismiss a rebuttal simply because they’re able to give the “advice” that more people might be convinced of it had it been written in a more “polite” manner. As well, let’s not forget that opinions on behaviour split fairly reliably along partisan trench lines. Ye olde zero-sum game the tone wars are.
Finally, *if* I think an argument is ill-formed, unfounded and/or stupidly dangerous, I really do feel somewhat obligated to let otters know that in no uncertain terms. *Maybe* I could get that point across without mockery and derision … but then again, maybe not. I’ve done it a lot of different ways.
It isn’t even clear to me that what I write has any real effect outside of satisfying my own desire to voice opinions.
Willard said on June 2, 2016 at 9:25 pm in reply to my comment on June 2, 2016 at 7:52 pm,
“> Not necessarily, since for all you know there could be something *outside* that knowledge base B than explains why event E occurs.
That’s just another way to appeal to ignorance. I see at least three problems with that kind of argument against Pop’s thesis.”
I committed a typo, and I apologize for not being more diligent in preventing it. It should have read as follows:
“Not necessarily, since for all you know there could be something *outside* that knowledge base B than explains why event ~E occurs.”
And so, because it’s about the negation of E, appeal to ignorance does not apply, and the rest of what you said also does not apply.
Willard, do you argue against the existence of a background set that could have an impact on this modus tollens? I would have thought we should all know it’s there, even if just implicitly. All I’m doing is this: I’m explicitly pointing out that if you claim you have obtained ~B via modus tollens on your B -> E via ~E, you are at least tacitly claiming that there’s no background set that would change this result of ~B from ~E, and then I’m asking for a justification for this at least tacit claim.
If you’re not arguing against this, then what are you arguing against, since, again, that’s essentially all I’m saying.
> [B]ecause it’s about the negation of E, appeal to ignorance does not apply, and the rest of what you said also does not apply.
Appealing to our ignorance of all the possible background assumptions has nothing to do with any conclusion that the knowledge base B is supposed to imply, or not. As soon as you appeal to something that lies outside what you know, you appeal to ignorance. I have no idea why you think that replacing E with ~E changes that fact.
***
> Do you argue against the existence of a background set that could have an impact on this modus tollens?
On the contrary, I would argue that this argument could be used to destroy any inference based on incomplete and fallible knowledge. It’s bread and butter for contrarians. Of course deductive inferences can be wrong from time to time. Popper never argued otherwise. We still use them, even if we don’t live in a closed world.
I’d rather accept that modus tollens can sometimes fail (say because we’re not omniscient) than to accept that because we don’t know everything we can’t deduce anything.
Eli,
I’m having fun watching Willard singlehandedly write the post-post-modern play, Waiting for Gödel.
I think I am basically in support of Willard’s arguments.
“”As far as I’m concerned, it’s even more of a stretch to separate a theory from all the other theoretical constructs without which it can’t be articulated, from all the instrumentation apparatus by which its “active” components are tested, from the observations made in this setting, and from the background knowledge that makes all this relevant to us.”
“”Your T is not Angech’s T, nor is it Pop’s. (I could redefine T & A as U and say that by “theory” I was referring to U.) There’s an equivocation on that T, and because of it both you and Angech are talking past one another.”
particularly ” It’s time to give it a rest.”
Dikran Marsupial says: June 2, 2016 at 2:34 pm
“angtech wrote -more free water (not ice) in the world means more moisture in the air
No, this is not correct AFAICS. The amount of water in the atmosphere depends on air temperature (Clausius-Clapeyron relationship), not the surface area of the oceans.
Perhaps you would like to defend this first point before moving on to the others.”
–
Sure.
*The amount of surface area of the oceans also depends on the temperature.
Hence if the surface area of the oceans is increased the average temperature has also increased.
Hence there is more water in the atmosphere [C-C relationship.
–
“A warmer atmosphere can hold more moisture, and globally water vapour increases by 7% for every degree centigrade of warming. How this will translate into changes in global precipitation is less clear cut but the total volume of precipitation is likely to increase by 1-2% per degree of warming.”
Chris says: June 2, 2016 at 9:19 pm
The falsifiability of statements seems like a good starting point. A statement is scientific if it is falsifiable.Obviously a concept doesn’t become falsifiable by being formulated as a statement (“God exists”).
–
Hmm? And a scientific concept as a statement “The sun comes up in the morning”
Or are all concepts not scientific or are all scientific statements are not concepts.
–
“And predictions are problematic.”
As the great Yogi Bera also said “It’s tough to make predictions, especially about the future.”
–
“In fact the contemporary and rather urgent application of scientific knowledge/theory in a predictive sense may actually not fall within Popper’s remit.”
I wouldn’t go there if I was you.
–
“For example statements like – “we’ve received the biopsy result and we’re sorry to say you’ve got less than 6 months to live”…are both unfalsifiable (there’s no set of experiments and observations that could be made now that could falsify these statements),”
–
Justifying unfalsibility as at only one time like this is not correct.
The simple fact is that experiments and observations have been done thousands of times on similar cases in the past just so such a prognosis could now be given with a high certainty of correctness on said biopsy result.
That one would not tell the patient this for other humane and medico-legal reasons is beside the point, sort of, but I will belabor it.
Doctors rarely give anyone a timeline of life such as 6 months, it takes hope away from the patient and is seen as cruel. Instead they speak of options and possibilities though they are fully aware of said prognosis being blunt and tactless is not helpful].
–
and also perfectly falsifiable (are you still alive after 6 months – yes or no?]
That is the medico-legal and social problem.
The odd patient that does survive after being told they were going to die is surprisingly unthankful, not to mention the relatives who have to undivide the will.
> I think I am basically in support of Willard’s arguments.
I think we could get K&A’s support, Angech, if we rewrite falsificationnism as the doctrine according to which a theory and its auxiliary hypotheses is deemed scientific if and only if we could in principle produce a falsifier of that theory or of its auxiliary hypotheses (and that these preserve the explanatory power of that theory).
Maybe it’s a vocabulary thing.
***
> The simple fact is that experiments and observations have been done thousands of times on similar cases in the past just so such a prognosis could now be given with a high certainty of correctness on said biopsy result.
It’s important to bear in mind that a prognosis is not exactly a prediction, i.e. the logical consequence of a theory (and its auxiliary hypotheses).
angech,
The fluid having the highest heat capacity is prolly gonna be the one setting the temperature, not vice versa. See also, >90% of heat retained in the system is going into the water.
Willard said June 3, 2016 at 2:10 am in reply to my comment on June 3, 2016 at 12:08 am,
“> [B]ecause it’s about the negation of E, appeal to ignorance does not apply, and the rest of what you said also does not apply.
Appealing to our ignorance of all the possible background assumptions has nothing to do with any conclusion that the knowledge base B is supposed to imply, or not. As soon as you appeal to something that lies outside what you know, you appeal to ignorance.”
It seems to me clear now that you’re arguing past me. I think you’re saying that I’m saying things that I’m not saying – you’re not interpreting my definitions as I intend them to be interpreted.
For instance, take my A. You call that the set of all background assumptions. I did not define it as any such thing. You say that I’m appealing to A. I don’t see how I’m doing any such thing, since simply knowing that A exists and simply pointing out its existence is not appealing to A in some argument. Do you think that I’m appealing to A to rescue T from being falsified? If you think that, then that is a confirmation that you’re missing my definitions. I’ve defined A so that when scientists go looking for and then successfully find conditions that successfully rescue their ideas from being falsified, they’re getting these conditions from a set that already exists in the abstract mathematical universe, where this set is A, which is a certain subset of this universe. See further below about this universe – it’ll blow your mind.
I see your position as rejecting every one of the articles I cited in this thread (I gave links to these articles) that speaks to the existence of conditions outside the test condition, and yes, they include unknowns. That is, I gave these links to show that I’m doing what others are doing. I’m just expanding on the same exact idea of auxiliary hypotheses that exist outside the test hypothesis.
For example, one of the articles (that Berkeley article) gave an example in the history of science of an idea that was wrongly falsified because they didn’t know about the condition that was later found that would have saved that idea from being falsified at that time. This condition exists in A, which is a subset of this abstract mathematical universe, just as are all conditions.
Now, about this universe:
Do you know what a disjunctive real number is? (How is this relevant? Wait for it.) Check this out:
https://en.wikipedia.org/wiki/Disjunctive_sequence
https://en.wikipedia.org/wiki/Disjunctive_sequence#Rich_numbers
http://euler.math.uga.edu/wiki/index.php?title=Normal_number
Quote:
“A rich number in base b is one whose expansion in base b is disjunctive:[8] one that is disjunctive to every base is called absolutely disjunctive or is said to be a lexicon. A lexicon contains all writings, which have been or will be ever written, in any possible language.”
So am I committing appeal to ignorance by simply pointing out that a lexicon exists, that it contains the set of all possible writings in all possible languages, that this set exists in the abstract mathematical universe, that this set is a superset not only of the set of all possible theories but also is a superset of the set of all possible auxiliary conditions that scientists could find to save their theories from being falsified, which in turn is a superset of A?
The set A exists. That’s a mathematical fact, since by the above, it’s a subset of the set of all possible auxiliary conditions that scientists could find to save their theories from being falsified, which is a subset of the set of all possible writings in all possible languages, which is a subset of a lexicon. Therefore, no appeal to ignorance is being committed since I’m not appealing to these facts to *make an argument* to try to prove or disprove T.
Yes, the definitions I gave are abstract, but so what? Don’t you find this the least bit interesting?
ATTP wrote “I guess it depends what your goals”
Indeed, in my case I am keen on ClimateBall minimisation. Sometimes I think an argument is ridiculous, but that doesn’t mean that I have to point out that it is ridiculous, especially if that means I invite evasion via “tone of the discussion” discussion.
angech wrote “sure.
*The amount of surface area of the oceans also depends on the temperature. Hence if the surface area of the oceans is increased the average temperature has also increased.
Hence there is more water in the atmosphere ”
This is sophistry, and I suspect than angech knows that perfectly well. It is the temperature of the atmosphere that determines its capacity for water vapour, the fact that warming means the oceans will expand as well means the two things are correlated, but there is no causal relationship from ocean expansion to atmospheric humidity. It is a shame that this sort of contorted logic is used, rather than simply acknowledging the argument was incorrect.
Indeed, not only – as Dikran says – it the temperature of the atmosphere that matters, the change in surface area of the oceans is probably going to be negligible.
Dikran Marsupial,
Underscores the point I was making about subjectivity — I see little tonal difference between calling an argument ridiculous and calling it sophistry. The latter term is arguably more fancy, but no less derisive in my book.
And let me be abundantly clear: I’m only making this argument to underscore my point that tone trolling is a zero-sum game. To each their own as they will, according to what makes them most comfortable.
brandon, there is no derision involved in pointing out the use of sophistry. Sophistry is widely used in e.g. political debate, and in some cases actively appreciated; the point is that it is unacceptable in scientific discussion and angtech should avoid it if discussion of the science is his intention. Very few of us are mind readers, so it is generally best to stick to what is actually written, rather than assuming that the author meant something that is only (at most) implied by what was written (at least without asking). Unfortunately this is just the sort of “tone of the discussion” discussion that I would like to avoid, so if you don’t mind I’ll leave it at that.
KeefeAndAmanda says: June 3, 2016 at 5:35 am
“So am I committing appeal to ignorance by simply pointing out that a lexicon exists, that it contains the set of all possible writings in all possible languages, that this set exists in the abstract mathematical universe, that this set is a superset not only of the set of all possible theories but also is a superset of the set of all possible auxiliary conditions that scientists could find to save their theories from being falsified, which in turn is a superset of A?”
I suspect you are pulling a rabbit out of a hat.
Nowhere in your “lexicon” exists that set that absolutely forbids pulling rabbits out of the hat or as you call it auxiliary conditions existing in the parallel universe that conveniently save scientists’ theories from being falsified.
If you wish to pursue this line every time someone puts up an argument then It’s time to give it a rest.
Your argument in effect means that no arguments can ever be one or lost.
> It seems to me clear now that you’re arguing past me. I think you’re saying that I’m saying things that I’m not saying – you’re not interpreting my definitions as I intend them to be interpreted.
My argument is mostly general and independent from your interpretation of A, K&A: either you know it, or your don’t; if you do, in principle you can test it, or it’s ad hoc; if you don’t know it, you appeal to ignorance. See?
I insist on the “in principle” because falsificationism doesn’t compel you to reject a theory if you get unexpected results from a test. It’s only a doctrine according to which you could, at least in principle, produce a test that would falsify it. When that happens, your rejection of that test follows modus tollens. You don’t reject that theory because you always abide by modus tollens – applying modus tollens only describes what you do when you reject that theory. Basic learning by trial and error. Falsificationism could be reduced to the claim: no trial and error, no science.
There’s no amount of definitions, citations, or vulgar display of logical power that can counter that. Well there are, but not by playing any rational game. Here could be one way:
https://www.xkcd.com/169/
***
> I see your position as rejecting every one of the articles I cited in this thread (I gave links to these articles) that speaks to the existence of conditions outside the test condition, and yes, they include unknowns. That is, I gave these links to show that I’m doing what others are doing. I’m just expanding on the same exact idea of auxiliary hypotheses that exist outside the test hypothesis.
From unknowns one can infer just about anything.
Considering that I already cited “bundle up your theories” more than 1 1/2 year ago, K&A, you might need to curb my enthusiasm regarding which universes would flabbergast me. You’re just rediscovering holism, and holism is not incompatible with falsificationnism, but with the myth of the experimentum crucis. Your logical argument rests on an equivocation over the concept of theory, but I can’t bother to go there – you’ll just repeat the same thing over and over again, like a defensive math professor would to a mesmerized class.
Dikran Marsupial says:
*The amount of surface area of the oceans also depends on the temperature. Hence if the surface area of the oceans is increased the average temperature has also increased.
Hence there is more water in the atmosphere ”
This is sophistry, and I suspect than angech knows that perfectly well.”
Sophistry is all very fine but this is a fact.
Climate change involves increasing C02 increasing heat hence melting ice to give the sea level rise. The heat comes first, the ice melts second. The humidity goes up because of the temperature rise and the larger amount of water at the air sea interface.
I don’t get it. The consequences of increasing CO2 are of increased rainfall overall*.
It seems you are all denying one point you normally believe in to protect another because you are not happy with a seemingly logical outcome
angech wrote “Sophistry is all very fine but this is a fact.”
more sophistry, that rising global temperatures mean sea level rises and increasing atmospheric humidity are indeed both facts, but that does not mean that sea level rise is a cause of increasing atmospheric humidity, which is what you initially claimed.
Sophistry is not fine. In a scientific discussion it matters whether the arguments you provide are logically sound. Rising sea levels does not cause additional rainfall, and hence this is not a benefit of sea level rise that you claim is being dismissed.
Hyperactive Hydrologist says: June 2, 2016 at 5:39 pm
“According to the world bank almost 5% of the world population live below 5m elevation and this is likely to increase as population increases and more people migrate to coastal cities.”
Is this a bad thing?
Why do they live below 5 m elevation.
Well in the east it is out of necessity for fishing
But also for tourism [Bali, Phuket, Vietnam coast]
and Beauty. Lots of beach-houses in Australia
Commerce. Many big cities are trading ports London Venice etc etc. It is very hard to have shipping and commerce in a port above 5 meters elevation methinks.
But the main reason is mathematical. If you look at the percentage of the world that is under 5 meters and look at the population density I would guess that 5% of the world is below 5 meters elevation.
It is pretty hard to live at elevation as well by the way. Something to do with rarefied air and lack of [sea] city ambience.
In fact I am amazed that it is only 5%
Now what happens when the sea level goes up or down?
The people move back or forwards to stay within comfortable reach of the sea.
They don’t just sit there for a hundred years waiting for the sea level to go up 2 feet or 6 feet. They adjust.
Look at Miami, they want to build and live next to the water not even 5 meters up, people love being by the sea whatever level it is.
No Brandon it is not sophistry.
angech,
Yes, people can indeed move, but the idea that this is somehow obviously simple, cheap, and not a reason to consider reducing our emissions is what I find somewhat bizarre.
angech wrote “Look at Miami”
I can see why angtech might want to look there, rather than say Dhaka.
The first world have the economic resources to ameliorate the worst affects of climate change, but that doesn’t mean that the rest of the world is similarly capable of dealing with the climate change that by and large (i.e. per capita) isn’t substantially caused by them.
Tell us why you think so angech. I would have thought it was an entirely appropriate statement (at least the way I stated it). In fact pondering this allows us to approach the core of what this thread is about (Popper and his use as a bludgeon in “debates” about climate). It also helps us to think about these things in a wider context which is nice.
Think of a theory that has been essentially falsified (not easy since most theories aren’t falsified – they are mostly superceded and are left like rusting hulks on the shoreline). For example, take the theory that everything is composed of earth, air, fire and water. This really has been falsified (even though it’s not such a bad theory). We could notice that analysis by spectroscopy, chemistry, X-ray diffraction, electron microscopy, atomic force microscopy etc. shows that matter is actually composed of atoms and elements, and that many things (plastic, rubber) contradict the theory.
Now imagine someone stating “I am exploring the theory that all matter is composed of earth, air, fire and water”. We might say, “that’s interesting – we should be able to falsify that theory e.g. by doing such-and-such spectroscopic analysis on stuff, or finding things that aren’t composed of these classical elements.”
Note that there is no temporal element to this. We can do the experiments right now (“Get up angech, we’re going to do an experiment!”). We wouldn’t say, “In 6 year’s time we will know whether or not the statement about the composition of matter is true or false.
This is quite different from the non-falsifiable predictive statements. Even though we will know perfectly at the appropriate time whether our predictive statement was proven or falsified, we can’t address the falsifiability of the statement at the time it was made (or any time before the point at which the issue is resolved) by any set of experiments or observations we might make.
I think you’re some way towards getting this yourself as judged from your next response which has an admirable element of explanatory detail; viz:
Well yes. As we agreed earlier in the thread, we don’t make statements of certainty (I suspect you really meant “high confidence of correctness”) about the future. We assign probabilities to our predictions/projections based on the strength of our theoretical and observational knowledge. [n.b. careful with the use of “prognosis” which itself contains the element of uncertainty].
What do we need in order to do this usefully and (hopefully) productively? At the very least we need to have an honest and informed relationship with our knowledge base. And even though we don’t expect our prediction statement to be falsifiable (in the Popperian sense), we do expect it to be consistent with our knowledge.
And when we do reach the point in time where we can come to some conclusion about the success of otherwise of our prediction/projection, we again should do this in the context of an honest and informed relationship with our knowledge. So we don’t say, for example, that we’re unhappy because we haven’t found a putative “hot spot” that we consider is expected from our theory, when in fact that “hot spot” is quite likely actually to have been found…etc.
Why not angech? Give your reasons please. In any case we have “gone there” (it’s kind of the point!).
Dikran Marsupial says: June 2, 2016 at 2:34 pm
“angtech wrote -more free water (not ice) in the world means more moisture in the air
No, this is not correct AFAICS.
Dikran Marsupial says: June 3, 2016 at 1:11 pm
that rising global temperatures mean sea level rises and increasing atmospheric humidity are indeed both facts.that does not mean that sea level rise is a cause of increasing atmospheric humidity, which is what you initially claimed.
– I think I see where you felt I was using sophistry as I had said prior to the quoted bit
Positive effects of sea level rise? – Serious ones. Yes.
“” Rising sea levels do not cause additional rainfall, and hence this is not a benefit of sea level rise that you claim is being dismissed.
–
Again then, while ” It is the temperature of the atmosphere that determines its capacity for water vapour,” the amount of water available at a temperature able to get into the atmosphere does depend on the surface area of the water source available.
If you had a land bound planet of earths characteristics [rotation solar input etc] the air would never be able to get as much moisture in as a sea bound planet despite all the CC’s in the world.
It is obvious that the amount of moisture in the atmosphere depends on the availability of the water source and as Neil King said on another matter it is all dependent on the surface area.
The effect may not be large for a 2 foot rise in sea level but If you consider the 100 foot plus rise from the ice ages vast swathes of land area are now under the sea.
The second consequence of sea level rise you ignore is ice loss on land including the Antarctic and the areas around the Arctic [you cannot have one without the other].
This decreases albedo and raises the temperature and humidity.
I do not think this point is sophistry, sea level rise means land ice loss.
More water to go into the atmosphere and more heat in the atmosphere.
More rain and more arable land.
Chris, I don’t think falsifiability does require the plausibility of immediate falsification, for instance Eddington’s observation of gravitational bending of light required a solar eclipse, and hence the test wasn’t immediately possible at an arbitrary point in time, but that doesn’t mean that the theories that predicted it were not scientific theories.
angech wrote “Again then, while ” It is the temperature of the atmosphere that determines its capacity for water vapour,” the amount of water available at a temperature able to get into the atmosphere does depend on the surface area of the water source available.”
The amount of water available is not the rate limiting factor, if it were, there would be no rain (hint if the air doesn’t have the capacity for storing moisture, it precipitates out). The limiting factor is air temperature. Consider this, if a parcel of air has travelled all the way across the Atlantic to the U.K., how much effect is the last kilometer going to have on the humidity of the air that arrives?
Dikran,
I was actually going to qualify my point by explaining that “immediate” is actually used to represent “timelessness” or “independent of time”. Conceptually, statements/hypotheses are falsifiable solely in relation to experiments/observations and if, for example, these were all at hand a statement/hypothesis could be falsified “right now”.
The reason for stating it like this is to contrast with predictions which are unfalsifiable (in the Popper sense), but also perfectly falsifiable since one just needs to wait until the prediction is realized or not (i.e. by waiting). But these can never be falsified ahead of time (i.e. “right now”).
I’m really trying to address the point of whether predictions lie within the remit of Popper, since that’s how the chap is often used as a bludgeon to beat on science!
I should have used your example Dikran – here’s another go at conveying what I mean:
Einstein’s theory makes a prediction that massive objects have sufficient gravitational attraction to bend light. That prediction can be tested in the way that you describe (observe the bending of the path of light by e.g. the sun). This test is timeless in the sense that whenever we had the results of the experiment we could apply it as a test of the theory. It turns out that Eddington had to wait for an eclipse; however he might well have had the results of the experiment at hand. it makes no difference in relation to theory testing/falsifiability etc.
Consider a prediction of the future of the type that we might consider important. “Biopsy results indicate that you have less than 6 months to live”…or the Earth surface temperature in 2040 will be 0.3 – 0.5 oC warmer than now”.
These statements are not falsifiable in the Popper sense. There is no set of observations we could make right now that would falsify these. Of course they are also perfectly falsifiable since we just have to wait around (6 months or til 2040, respectively). if they’re falsified by observation this may well have little effect on the confidence of our knowledge base (unless of course the predictions described progression of future events that were necessary consequents of the theory, which neither of the above are). The only thing that’s falsified is the specific prediction itself.
In fact we generally don’t make these predictive statements about the future. Predictions of how things progress in the future are expressed in terms of probabilities/likelihoods. These are also not falsifiable (in the Popper sense) – however we expect that they are based on honest and informed understanding of our knowledge base.
Tell me if I’m being confusing or you think I’m talking nonsense. Using the word “prediction” can be confusing. I’m contrasting Eddington-style predictions arising from theories (a v. good example btw) which are “timeless” in the sense I’m trying to convey…with the prediction of how specified things will progress (morbidity/mortality/sea level rise/surface temperature) in the future. Increasingly, it is the latter that we would really like to get a handle on.
Chris wrote “Consider a prediction of the future of the type that we might consider important. “Biopsy results indicate that you have less than 6 months to live”…or the Earth surface temperature in 2040 will be 0.3 – 0.5 oC warmer than now”. ”
I think both are falsifiable in the Popperian sense. In both cases there are experiments that we can run, that may take time, but will potentially falsify the theory. The 1705 prediction of the reappearance of Halley’s comet in 1758 was a scientific one, there was an underlying physical theory, there was some observational data, and there was a prediction. However the theory (that the previous observations were all of the same comet) were not falsifiable until the next appearance half a century later. I should think this was an excellent example of the sort of thing Popper had in mind.
angech,
Given that we don’t agree on the core argument, we’re not likely to agree on how to label it either.
Nomadic lifestyle choices are easier when one’s total assets fit into a few covered waggons — or in the case of livestock are self-motive. One does not simply relocate one skyscraper, let alone an entire farm of them. And as I’ve mentioned before, the dirt itself has value, a good portion of which is based on expectation of future appreciation.
Market correction to *present* reality predicated on physical stasis will not be kind once the literal fluidity of the situation becomes apparent to more than just real estate developers and flood insurers. Caveat emptor.
Willard said on June 3, 2016 at 12:54 pm in reply to my comment on June 3, 2016 at 5:35 am,
“My argument is mostly general and independent from your interpretation of A, K&A: either you know it, or your don’t; if you do, in principle you can test it, or it’s ad hoc; if you don’t know it, you appeal to ignorance. See?”
Two points:
(1) By your application of appeal to ignorance, since you say your application of appeal to ignorance is mostly general and independent of how A is defined, if I’m committing this fallacy by *merely* putting forth the form A & T -> P or any of its equivalent forms such as A -> (T -> P) as a setup for falsification, then so is everyone else who *merely* puts forth the form A & T -> P or any of its equivalent forms such as A -> (T -> P) as a setup for falsification. That is, to what degree is your claim that one is committing this fallacy merely by using this formula as setup dependent on how A is defined? And if you say I commit this fallacy not by merely using this formula as a setup but by how I use it otherwise, then what is this otherwise use? The *only* otherwise use for A is that it must be affirmed before one can claim that T is false given the falsity of P. And I show in (2) that *everyone* does this, and so if I commit this fallacy by this otherwise use, then so does everyone. See (2) below for more on this.
(2) By your application of appeal to ignorance, the part of “don’t know it”, since you say your application of appeal to ignorance is general and independent of how A is defined, if I’m committing this fallacy, then so are you and everyone else whenever you or they claim to have successfully falsified some test hypotheses. When, after looking for and not finding a condition that would have saved T from being falsified, one claims that T is false, then one is, at least for the time being, saying *at least tacitly* that there does not exist a condition that would have saved T from being falsified. That’s an application of “don’t know it” – one is appealing to something that one doesn’t know.
You might claim that it’s not an application of “don’t know it”, but here’s how you would be wrong: When one says even just tacitly that one expects that no one will ever find in the future a condition that would have saved T from being falsified, then one is *at least tacitly* affirming set A to be true. This is because “I expect that no one will ever find in the future a condition that would have saved T from being falsified” is equivalent to “I affirm A to be true”. Here’s how this is true:
I explained how it’s a plain mathematical fact that set A as I defined it exists – it’s a subset of each of uncountably infinitely many real numbers called lexicons, and real numbers exist. And so it’s a plain mathematical fact that the formula A -> (T -> P) exists and is true (the definition of A mathematically guarantees this formula to be a tautology). Therefore, since it’s a plain mathematical fact that this formula exists and is true, every time one claims to have obtained ~T via ~P, one *must* affirm *at least tacitly* that A is true to obtain ~T – that is, it’s logically inescapable that we have to *at least tacitly* use disjunctive syllogism to obtain ~T. (We have ~(A & T) => ~A v ~T, and the only way to get to ~T is to use disjunctive syllogism to affirm ~~A which is equivalent to A, to obtain ~T.)
Please respond in an academic way to what I wrote above and to the definition of A I gave in
on June 1, 2016 at 11:17 am, given that I now make a modification to this definition: Replace “the smallest conjunction of conditions” with “the smallest conjunction of conditions that are not contradictions” (a contradiction is defined to be a statement that is false in all its substitution instances).
“I insist on the “in principle” because falsificationism doesn’t compel you to reject a theory if you get unexpected results from a test. It’s only a doctrine according to which you could, at least in principle, produce a test that would falsify it. When that happens, your rejection of that test is based on modus ponens.”
Don’t you mean modus tollens? Yes, I know that modus tollens can be translated into modus ponens. Is that what you’re doing here?
To address “in principle”: There’s nothing in my definition of A that implies that there exists a conjunct of A that can’t in principle be itself falsified. Again, look much more carefully at the definition.
“From unknowns one can infer just about anything.”
But, as I showed above in (1) and in (2), you and everyone else *at least tacitly* use unknowns when you claim ~T. If you disagree, then, as I requested before, please respond in an academic way.
Finally, Willard, with respect to what you said on June 3, 2016 at 5:09 am,
“I think we could get K&A’s support, Angech, if we rewrite falsificationnism as the doctrine according to which a theory and its auxiliary hypotheses is deemed scientific if and only if we could in principle produce a falsifier of that theory or of its auxiliary hypotheses (and that these preserve the explanatory power of that theory).
Maybe it’s a vocabulary thing.”
As I understand what I say, I’m already meeting all these criteria. There’s nothing in the given definition of A or the setup A -> (T -> P) that implies the negation of any of these criteria. (Please don’t talk about unknowns being a problem to reply to this, since again, I already showed above that everyone inescapably *at least tacitly* appeals to unknowns whenever they finally go all the way to affirming ~T. On these unknowns: All I’m doing in the end with the given definition of A is explicitly talking about what everyone else at least implicitly does. What? It’s OK to do it as long as we don’t talk about it?)
angech said on June 3, 2016 at 12:51 pm in replt to me comment on June 3, 2016 at 5:35 am.
“I suspect you are pulling a rabbit out of a hat.
Nowhere in your “lexicon” exists that set that absolutely forbids pulling rabbits out of the hat or as you call it auxiliary conditions existing in the parallel universe that conveniently save scientists’ theories from being falsified.”
If I’m pulling rabbits out of a hat, then so are you and everyone else, since, by the definition of a real number called a lexicon, everything that you and everyone else writes is a subset of each of those uncountably infinitely many lexicons.
“Your argument in effect means that no arguments can ever be one or lost.”
Not so. My argument and your claim that what I argue, “pulling rabbits out of the hat or as you call it auxiliary conditions existing in the parallel universe that conveniently save scientists’ theories from being falsified”, are of course subsets of a lexicon, but they are mutually disjoint subsets. That is, one has nothing to do with the other.
Huh? !! Clearly I don’t belong here, so with apologies, please return to earth for a moment and indulge me. It’s nice having clever arguments amongst experts (and argumentative obsessives), but the problem at hand is to stop people using “falsification” as a phony argument. I promise this will be short.
A good friend sent me a good link on the problem of Popper and falsification and unskeptical “skeptic” distractions. So here’s a link he sent me from one Dr. Richard Lawson with the comment: “It’s a neat summary of the position from the point of view of philosophy and worth a careful read.”
https://philosophynow.org/issues/104/Climate_Science_and_Falsifiability
Basically, the contrarian hypothesis has been falsified already – the AGW one has not. Th-th-th-th-that’s all folks.
We have a much bigger problem than arguing minutiae using relevant academic language. We need to convince a bunch of people who don’t want to know that “earth, we have a problem.”
Now I have to apologize, not sure how obnoxious I’ve just been, but that’s my tuppence.
Susan,
Interesting article, thanks. Your definition of being obnoxious is clearly different to mine 😉
> one thing is certain: the hypothesis that the effect of carbon dioxide and other greenhouse gases is trivial and warrants no action simply does not hold up.
Just between you and us (we’re onto the 223rd comment), I count at least two things.
Neither of those specific statements of future progression of morbidity-mortality/surface warming is a theory. They’re simply statements about future outcomes (that are made within a framework of theory). I can’t see how these statements could be falsifiable in the Popperian sense even if they can be absolutely falsified or confirmed at the appropriate future time. Can you think of any of the experiments you are proposing (other than waiting and seeing what happens)?
Your Halley’s comet example is a good one, but I don’t think it’s quite what it seems in this particular context. It’s different to mine I think, by considering the temporal element. Conceptually, in terms of falsification, we might as well have the experiment/observational data at hand now (e.g. in Halley’s case, in 1705 when he made his prediction), since the hypothesis concerns a phenomenon embedded as part of the natural world (the orbiting of a comet). The fact that his followers had to wait til 1758 was not an inherent element of the assessment of the hypothesis (that the previous regularly-spaced sightings were all of the same comet which would therefore have a predictable reappearance). It just happened that they didn’t have other means to assess the hypothesis except by waiting.
However there are other observations/analyses one might have made to assess the hypothesis (e.g. if they were able to follow the comet from its arrival in 1705 and determine a significant part of its orbit, they could see whether this was following a calculated orbit and determine that therefore it must return to near-Earth in 1758). 3 years of observation, for example, might have been sufficient to assess whether the comet was following the calculated orbit and therefore test the hypothesis. They couldn’t do this of course but this is a nice Popper-style test of the hypothesis that is conceptually possible.
So the fact that Halley’s hypothesis required a long wait was simply because that was the only “experimental/observational” means at their hands. The wait is a bit of a red-herring I think, and not an inherent part of any possible test of the theory.
On the other hand our prediction of the value of a parameter (sea level, surface temperature) at some specified time in the future, whether given as an absolute value, a likely range, or a probability has no possible experimental or observational tests other than to determine, at the relevant point in the future, what actually happened. If one says such and such will be the case in 2040, then how can one possibly test this other than by waiting until 2040? And then all that’s been been falsified or confirmed is the specific prediction. Of course this may give us some insight into the validity of the knowledge framework within which the prediction was made.
I’m really trying to get a handle on the status of these sort of specific predictions that we consider nowadays to be very important (how things will progress in the future), in the context of Popper. Not that it matters whether or not one can do this since there’s no reason why Popper should have any necessary influence on how we think about this…
Susan Anderson,
I second Anders’ observation that your definition of “obnoxious” is curious. 🙂 You write with a passion that I share:
By way of response, let me quote Dr. Lawson’s lede paragraph:
This is one of my stump speeches. The people who most need to be convinced are the policymakers. We mock the “I’m not a scientist” talking point, but for once in their public life, they ain’t lying — they don’t speak scientist. It literally is not their currency. Find a better way to appeal to their chances of reelection, and we may find some better forward progress on our own agendas. [1]
That doesn’t mean that I think messaging for the general public is necessarily invaluable. Part of what I’m suggesting is that there’s at least an off-chance that some bi-partisan dealmaking could also improve public opinion. My “evidence” for this is anecdotal and dubious, nevertheless it is my (provisional and tenuous) belief.
You quote Dr. Lawson:
As we have already seen on this thread, that assertion will simply bounce off those already arguing that “AGW theory” isn’t falsifiable. Getting journalists to carry that message more frequently might sway those in the undecided middle, but I can say with absolute certainty I’m going to hate trying to defend an argument which takes that form. First thing they’re gonna ask me is, “Where’s your evidence which ‘disproves’ my argument?”
I apologize if my weariness and feelings of impotence and futility come across as sour or pedantic.
———————
[1] If that sounds facile to you, I agree. For an attempt at brevity, I leave out particulars. Besides, I’m not all that convinced any of my ideas are particularly novel, or even if they have any merit.
Brandon , conviction is the hub of the problem for me.
When people try to convince me of anything I immediately think of snake oil.
The harder they try the more my hackles get up.
You obviously feel this even more strongly than I do but on the opposite side of the argument.
As do Sue and Dickran etc.
When they move away and give me space I look at the arguments on my own and reassess. A good logical argument does not need to be sold by conviction, it can only be hurt.
The second point is that we often get stuck at one level of our argument/ conviction and do not move on to the other points or the long term outcome.
Hence we argue past each other.
Susan’s points are moot.
The argument is not whether the world is warming,
Or whether CO2 as a GHG is partly responsible .
These are settled facts.
Move on to the real areas of debate.
Benefits of a warming world. Benefits of sea level rise.
Benefits of more people living with more food,warmth, clothing,science and mobility than ever before with mass communication and 200 Einstein’s and Edison’s out there able for the first time in our history to communicate their discoveries to help humanity solve our problems.
You can solve global warming by living in a cave or by going boldly into the bright new world.
angech – you have repeatedly been asked to provide some of the benefits of sea level rise. Instead of providing the requested examples you just repeat your previous unwarranted assertion.
The US is cutting science funding. More science? Heat stress kills plants. Clothing? What? Mobility? Fleeing from natural disasters usually isn’t considered increased mobility or a net benefit. Have you factored in the spread of disease? How about chronic kidney failure?
Will a warmer wetter earth increase mosquito populations? Will it lead to an increase in mosquito borne diseases?
You throw out words with no analysis or even real thought behind them. Just a Gish Gallop of ifs, buts, maybes and fable.
Thanks for the responses. Yes, It’s depressing, but I’m glad some of you don’t think I was overdoing it. I was feeling a little snarky and was afraid of irritating people with my complaint about the academic discussion.
Willard: chortle, but you’re splitting a hair/hairs (one or two).
Angech: Try what is going on in Texas, Fort McMurray environs (fire still not out), France and Germany, the drought in India, Arctic melt, there are shifts in the problems, but they are increasing. Migration of disease vectors (mosquitos, tree killing critters). The refugee crisis is exacerbated by a compromised food supply in Africa and the Middle East. War/conflict is one normal reaction to desperation. I could have dug up one article, but here’s one possible search:
https://www.google.com/search?q=refugee+crisis+and+climate+change
I’ve had enough experience with people like you who constantly shift ground that I know it’s counterproductive to try, but since you reacted I did.
There is no road to progress with people who, for whatever reason, don’t want to hear. If nothing is said, the claim is there’s nothing to say. If something is said, it’s “hype”. In the case of climate, big fossil has helped fund an mirror universe carefully constructed to look just like the real thing. People like Hotwhopper, especially, and others at times, go down the rabbit hole and point out the failures of logic and inconsistencies and outright lies, along with snark and personal attacks, that make WattsUpWithThat, BishopHill, Judith Curry, and many others less than useful.
What is really unfortunate is that in most of the English speaking world, this intellectual dishonesty is exploited, often for profit, by our political masters. Rep. Lamar Smith and Sen. Cruz are at the head of “science” in our Congress, and are doing their best to suppress, discredit, and defund information they don’t like. There is also Judicial Watch, lesser known legal attack dog for the Koch wing of the far right.
In my little role in all this gallimaufry, I encounter the “falsifiability” and Popper from time to time, and am glad to have learned a bit more about it. My instincts are good, but sometimes one needs to know a tiny bit more than instinct about what one is saying.
oneillsinwisconsin says: June 4, 2016 at 11:05 pm
“angech – you have repeatedly been asked to provide some of the benefits of sea level rise.”
See my comments June 2, 2016 at 1:56 pm
-more free water (not ice) in the world means more moisture in the air , more rainfall , more crops less deserts, more greening of the world hence a more liveable world overall.
-more shallow oceans (as in more land close to sea just underwire) means more coral reefs, more fish , hence more food and more life in the world. Not great for some of the air breathing minority but great for the planet as a whole.
Plus, since you did ask, more jobs for people involved in reclamation of subsided land hence millions of extra jobs .
For an example various land reclamation works have helped expanded the land area of Singapore by about 20%.
More work for marine scientists and entrepreneurs. in Fish farming and sea mining
In the 1960s, Marine Diamond Corp. recovered nearly 1 million carats from the coast of Namibia. Today, de Beers obtains a significant portion of its total diamond production from the continental shelf of southern Africa, in water shallower than 300 meters (1,000 feet)
More livable land in the semi frozen north Siberia Greenland and England would become more habitable and productive.
Desalination plants closer to areas of need.
Tidal power renewable energy easier to achieve.
Mind you, I understand there are very compelling negatives as well, I just do not see black and white only
-.
Sue, I live in Australia
A land of sweeping plains, Of ragged mountain ranges, Of droughts and flooding rain.
I find comments about “the drought in India,” an unfortunate bargaining chip. Nature and natural disasters are very cruel and sad and will continue with or without AGW climate change.
It just makes me look bad and you feel good
You could send money but I fear I am skeptical of most relief organizations as well. They seem to have big offices in Insurance company buildings next to big oil.
Australia has the largest Carbon footprint in the world next to Saudi Arabia and the other pure oil producing companies, 500 million tonnes of coal was mined Australia was the world’s fifth-largest coal producer, after China, the United States, India, and Indonesia.
Per head of population that means I am “responsible” for 25 tonnes of coal a year , not to mention our oil and gas exports.
Do I care about the world, sure.
Will Australians compromise their standard of living, no.
By the way the coal from Queensland will go directly to India for power and transport and jobs and medication and food that they would not get anyway.
Perhaps that will help your concerns about drought.
angech,
No, it does not.
Yes, of course. However, if we continue to add CO2 to the atmosphere, we will continue to warm and we will change the hydrological cycle (precipitation). The changes could both be substantial and could occur faster than at any time in human history. Simply dismissing this – as you seem to be doing – seems to be ignoring these possibilities. We (the global human race) are doing this and we (the global human race) could choose to do something different, or could make decisions that would potentially change what we do in future. Suggesting that somehow disasters will happen anyway ignores that we could impact – in future – the scale of these disasters and we should, IMO, be willing to discuss both this possibility and whether or not we should be doing something to influence what might happen in future.
Angech,
Just curious. Do you think at all before you post your verbal stream of diarrhea?
Angech: “more free water (not ice) in the world means more moisture in the air , more rainfall , more crops less deserts, more greening of the world hence a more liveable world overall.”
Ever hear of the Clausius-Clapeyron equation or Hadley cells? Ever hear of rain shadows?
And your comments about more shallow oceans… Uh, dude, you do realize that we are talking about flooded coast line here. Anything that can be mined there will already be gone. And do you really want to eat fish raised on top of what used to be polluted asphalt, etc.?
As to your precious coal, did it ever occur to you that by burning that coal we are recreating conditions akin to those that existed before the coal formed–only with a hotter sun…
If you aren’t courageous enough to take action yourself, then do us all a favor and just get the hell out of the way.
As we’re back on the Popper thread, I thought I’d post something I’ve been thinking about. I think that the shortcoming in Popper mirrors a lot of what is wrong with the “climate debate”.
Because all models are wrong (as we know from G. Box), it is easy to attack the models. There will always be shortcomings in a model. What this ignores, however, is the fact that most successful models–and climate models surely qualify–will almost assuredly have some things right. It is not enough to simply falsify a theory by pointing out discrepancies in its performance. You have to also sort out what is wrong with the theory. If you haven’t, you really aren’t doing science, because science is at its nature model-driven empirical inquiry.
Earnest Rutherford said that, “All science is either physics or stamp collecting.” It is the existence of a well developed model that keeps a science from being reduced to stamp collecting–indeed, the model even informs the taxonomies of the stamps. So, really, the best way to falsify a theory is to develop a better theory.
OT alert: on reality in 2011:
In 2009 at Copenhagen, President Nasheed:
.”
youth ambassador Mohamed Axam Maumoon, http://www.democracynow.org/2009/12/8/would_you_commit_murder_15_year_old
Obama 2015, climate bit starts about minute 2:
Bali, 2007, Lima 2014, and elsewhere: “Lead, follow, or get out of the way”
You know, one of the the added bonuses of reading the comments is the predictable hilarity of the responses when denialiti visitors are called out to put up or shut up. And then start a bizarre tap dance.
A current example:
oneillsinwisconsin says: “angech – you have repeatedly been asked to provide some of the benefits of sea level rise.”
angech replies: More work for marine scientists and entrepreneurs in… sea mining
In the 1960s, Marine Diamond Corp. recovered nearly 1 million carats from the coast of Namibia. Today, de Beers obtains a significant portion of its total diamond production from the continental shelf of southern Africa, in water shallower than 300 meters (1,000 feet).
I mean, c’mon. Seriously, that’s a *benefit*? Now they will have to mine underwater for resources they could previously accessed conventionally? Tell it to the petroleum industry.
Or this:
angech replies: more shallow oceans (as in more land close to sea just underwire (sic. “underwater”?) )
Maybe the geometry/geography on bizarro world could work like this, but not on Earth, as far as I can tell. What, are new islands going to appear to compensate for the submerged ones, etc.
Wow.
> Earnest Rutherford said that, “All science is either physics or stamp collecting.”
Indeed, physicists are sometimes jerks.
***
> I think that the shortcoming in Popper mirrors a lot of what is wrong with the “climate debate”. […] It is not enough to simply falsify a theory by pointing out discrepancies in its performance. You have to also sort out what is wrong with the theory. If you haven’t, you really aren’t doing science, because science is at its nature model-driven empirical inquiry.
Popper doesn’t say that falsification is sufficient, but that it’s necessary.
Learn the difference, or else you’ll have stamp collectors kick your ass over things we learn in college.
> Willard: chortle, but you’re splitting a hair/hairs (one or two).
Distinguishing AGW from its consequences is the opposite of hairsplitting, Susan.
Drive by for more Waiting for Godot editorials all you want, but if you could stop suggesting things that burden otters with more ClimateBall, that’d be great.
Thanks Willard, you’re right. I stopped in and then went too far (that’s who I am, all too typical). I had long since concluded that I have little to offer here and certainly can’t weigh in on the more on-topic discussions you all have pursued, but no amount of complaint will keep me from pointing out what I think, which is that the only way to reach enough people to make a difference is to point out the connections between AGW and its consequences. This is as you say not the right place to do so.
I’m not driveby, I’m a lurker with other things that take up my time, inexpert at a level that makes me an intruder in many of the more technical arguments. I was irritated enough to get snarky, and you more than aTTP and brandonrgates perceived what I was apologizing for. So ultimately, I do mean thanks for the good advice. I certainly don’t want to make the ClimateBall worse.
> [T]he only way to reach enough people to make a difference is to point out the connections between AGW and its consequences. This is as you say not the right place to do so.
There can’t be connections between AGW and its consequences if we conflate the two, Susan. Conflating the two is the best way to get lukewarm concerns raised about their pet CAGW meme. There’s even a third thing we should distinguish: the need to do something about the consequences of AGW.
There are many reasons to rearrange our energy portfolio. AGW is only one. My favorites are risk management, security and public health. They’re tried and true – the frames work. Heck, even economics could work – there’s money to be saved by being smarter about the ways we dump CO2 in the atmosphere like there’s no tomorrow, and if the right doesn’t mobilize to negotiate anything with the powers that be, decisions will be made without them.
(The last point is from Scott Denning’s presentation to the Heartland Institute, BTW.)
This is a Popper thread. This is the place to discuss Popperian stuff, not to epilogue yet again about climate communication. Just about any other thread here could do for that – just take AT’s latest one on bullshit. Just like we’ve had for years at Keith’s, at MT’s, at BartV’s. As far as I am concerned, we might as well create a Samuell Beckett reading group than to talk about science communication, but to each his or her own.
Finally, we have barely touched encyclopedic entries. We’re light years away from what could be considered minutiae in philosophical circles. We’re not comparing various versions of a text, interpreting a personal letter, or delving into grey lichurchur. When most claims about Popper can be refuted by reading the very first page that should have been read, well, that should tell you about the limitations of paid lips service to (or dismissals of) philosophers in ClimateBall.
Okay. I started with Popper, and the next time I see an amateur making themselves look all wise and knowledgeable bloviating about Popper and falsifiability I won’t be so ill-equipped to respond.
I mostly understand your points but don’t agree; I wish you were correct that “if the right doesn’t mobilize to negotiate anything with the powers that be, decisions will be made without them.” So far what I’ve seen is the reverse: US Congress? Cameron/Osborne/Rudd? Australia CSIRO? Spain and solar? Or did you mean planetary dynamics will unquestionably provide consequences? Next stop geoengineering?
Most people don’t understand AGW but are open to thinking about consequences.
You make excellent points (in spite of my continued lack of self control in paragraph 2 above) and I’m still trespassing. I’ll go back to lurking.
Willard: “Learn the difference, or else you’ll have stamp collectors kick your ass over things we learn in college.”
Oh, Willard, you’re so cute when you go all Internet tough guy.
The point I’m making–and that Rutherford was trying to make in his rough Kiwi fashion–is that without a model, all you are doing is stamp collecting. A mediocre model leads to better science than no model–and so the way to falsify a theory is to develop a better one.
In the case of climate, the antipathy to models shown by the Glibertarians is silly, since the role of CO2 in raising the planet’s temperature is incontrovertibly supported by well established science that isn’t about to be falsified. In fact without a model, we are in a position of unlimited risk, since the models are among the more effective tools for limiting the high-end tail on Climate sensitivity to CO2. But the denialati excel at own goals.
> The point I’m making–and that Rutherford was trying to make in his rough Kiwi fashion–is that without a model, all you are doing is stamp collecting.
I’m not sure where you got anything about modulz with Rutherford’s All science is either physics or stamp collecting, Snarky, but in any case Rutherford’s point is a bit stronger than that:
(The quote is only there to show that the author doesn’t even bother to pay due diligence to reductionism – he’s just showing how silly it is when we analyze mental predicates.)
http://plato.stanford.edu/entries/physicalism/#RedNonRedPhy
Reductionnism was more in fashion a century ago. Many stamp collectors stamped this one to the ground. Rutherford doesn’t play home, and his epistemological remarks should be taken with a grain of salt. Most empirical sciences don’t need no stinkin’ physical model. For instance, try Chris’ examples in this thread, and report. In fact, most climate modulz only implement physics in a very loose fashion.
That said, physicalism still has good winning chances. Again, follow the stamp collectors and RTFSEP – or please do continue to show the relevance of AT’s latest entry.
angech,
I’ve noticed. You might be surprised to learn that we’re so dissimilar in that respect.
Nobody here controls what you do in your spare time, angech. I sincerely doubt that you would ever actually give someone else that kind of power over you.
I can run the same script on you in reverse: others will take you more seriously if you stopped pretending that there is some better way of convincing you to change your position.
Willard,
This theme disturbs me as it tends to torpedo many of my past arguments attempting to defend the crappiness of teh modulz. It upsets me so much that I’m already inclined to accept its veracity, even though “very loose fashion” is very loosely defined.
snarkrates says: June 5, 2016 at 12:52 pm “Angech,
Just curious. Do you think at all before you post your verbal stream of diarrhea?”
I did not want to. I was asked to. just following orders.
oneillsinwisconsin says: June 4, 2016 at 11:05 pm “angech – you have repeatedly been asked to provide some of the benefits of sea level rise. Instead of providing the requested examples”
rustneversleeps says: June 5, 2016 at 2:37 pm “And then start a bizarre tap dance.”
Countries like Australia have shallow coastal sea beds with steep drop offs [unusable/accessible sea floors] With a rise in sea level more livable agronomic accessible sea floor space is created which would increase fish food stocks for one. It would also increase coral reef viability and prevent a lot of bleaching which would be an environmental win. Also new islands form all the time, as for disappearing, I doubt the Maldives were ever 1000 feet above sea level, they will just grow new bigger reefs even quicker without humans there.
..and Then There’s Physics says: June 5, 2016 at 12:20 pm
“angech, more free water (not ice) in the world means more moisture in the air , No, it does not.” I feel this is something I might be able to come back with to argue with and will ask around.
It is a bit of a circular argument because I feel you cannot have more water in the world unless more ice melts and you cannot have more ice melt unless you have a warmer atmosphere in general and if you have a warmer atmosphere you can have more moisture in it.
Your rebuttal is that the temperature of the air is the sole determinate of moisture in the air
a, ignores this fact.
b. ignores the fact that there may not be enough available water to fully saturate the atmosphere at its temperature and that having an increased amount available through increased surface area will increase availability to reach that potential.
c. more water over shallow land results in increased evaporation due to the more rapid heat up of increased shallow waters, a bit like the bubbles around the edge of a saucepan of soup as it gets close to the boil, again putting more moisture into the air
Willard, without a model, how do you know what to measure?
Here, Snarky:
https://www.jstor.org/stable/2254959
Mellor has a good book on causation.
As for your question, start here:
Click to access hesse.pdf
Popper’s realism is a natural predator of instrumentalism.
@ Susan Anderson: “…and I’m still trespassing. I’ll go back to lurking.”
Please see my earlier comment: https://andthentheresphysics.wordpress.com/2016/05/27/popper/#comment-80160
angech wrote “Brandon , conviction is the hub of the problem for me. When people try to convince me of anything I immediately think of snake oil. The harder they try the more my hackles get up. You obviously feel this even more strongly than I do but on the opposite side of the argument. As do Sue and Dickran etc.”
You would be better off not assuming you know how other people think. As it happens I think a better way to approach things is to consider each argument on its own merits. The downside of “getting your hackles up” is that it makes it more difficult for you to accept your errors, e.g. that rising oceans do not lead to higher atmospheric humidity, and your doubling down/evasion just makes you look silly.
angech wrote “It is a bit of a circular argument because I feel you cannot have more water in the world unless more ice melts and you cannot have more ice melt unless you have a warmer atmosphere in general and if you have a warmer atmosphere you can have more moisture in it.
Your rebuttal is that the temperature of the air is the sole determinate of moisture in the air
a, ignores this fact.”
No, it doesn’t, as I pointed out it just isn’t the limiting factor. Again you are using faulty causal reasoning. A warmer world meaning more ice melt and more atmospheric moisture does not mean that melting ice is a cause of atmospheric moisture, correlation is not causation. BTW much of the sea level rise is not caused by melting ice, but by the thermal expansion of the oceans.
Phil Lately might be a good pseudonym for the person in general who believes any model is better than none, I notice the comment by Rutherford was for a mediocre model. Thanks Willard for that snippet of information.
Correlation might not be causation but it sure is a hint.
It takes a huge amount of thermal energy to cause expansion of the oceans ,so much so that there would have to have been a heck of a lot of ice melted in the same time.
One also wonders about a glass of water with a large ice block in.
As the ice melts does the temperature of the water get warmer?
Or does it stay the same until the ice has melted?
Then thermal expansion could be guaranteed with continuing heat input.
BTW the water level in the glass would not change as the ice melted (being in water and all).
I believe most people focusing on sea level rise stick to the land ice on Greenland, in glaciers and possibly from Antarctica (recent doubt as you know) as the main cause of sea level rise since the ice ages.
The sea temperature back then was possibly not much different to now, I doubt that thermal expansion has as much significance as you claim.
I am quite ready to be corrected on this latter point as I am arguing from logic not knowledge on ocean thermal expansion which is a real entity
angech,
This paper suggests that about 6cm of the 20cm we’ve experienced this century is thermal expansion. As a rough estimate, it’s 12cm per 1024J going into the oceans. Since about 1960, we’ve increase ocean heat content by about 2 x 1023J, so about 2.5cm of thermal expansion.
To be precise 0.38 metres per degree C of warming of the ocean.
20000 years ago the sea level was 125 meters lower.
Basic maths would suggest that this either meant, on a generous 25% equivalence to ice melt ,
41 meters at 2 1/2 degrees C a meter that the sea temp was 100 degrees Celsius lower than today.
Or conversely that say it was only 5 degrees colder (still far to large in my opinion) then 2 meters of the current 125 meter sea rise could be ascribed to thermal expansion.
Which is not much.
Happy to see a scientific or mathematical rebuttal of this theory by any Poppers or non poppers here.
angech perhaps you ought to ask yourself why it is that rather than dealing with the substantive issue (Clausius-Clapeyron being the limiting factor in atmospheric moisture) you have sidetracked the discussion of sea ice, which is of no relevance whatsoever, taking advantage of a minor clarification that I made?
angech wrote “Correlation might not be causation but it sure is a hint.”
This goes beyond mere sophistry IMHO; I have explained why there is a corellation without there being causation in this case, and angech’s response ignores that explanation again.
angech,
Okay, so you’re just saying that ultimately melting of land ice dominates. Fine, that’s probably true and is one reason there is so much uncertainty about future sea level rise, at least in terms of how quickly it will rise. However, about 1/3 of what we’ve experienced in the 20th century is thermal expansion, according to Gregory et al. (2013).
@Magma: thanks, that’s what my friend who sent me the Richard Lawson material said. I’m still happy to know there’s a big difference between the way pseudoskeptics use falsification and what it is. Unfortunately, people who don’t want to know use it as a conclusive argument against real science, and though I knew it was sideways, I didn’t know how or why. Despite being a fool who steps in where angels fear to tread, I like to have some idea what I’m talking about when I call people’s BS in amateurland. This is not about those who were trying to talk with each other about Popper: I plead “guilty” to Willard’s genuine complaint with the caveats I’ve already overstated.
[Dikran], I am happy to agree with you that there is a correlation between melting of land ice and sea level rise.
I respect your comment that CC is the limiting factor in the amount of humidity in the air.
I did not introduce the discussion of sea ice and thermal expansion sidetrack.
You have given a reason for correlation without causation for increased sea level rise and humidity increase and I pointed out extra reasons.
You do not like my reasons, fine.
But you did not rebut my argument, preferring to call it sophistry, and relying purely on your previous answer.
More water, more humidity possible up to the limit of CC still works for me, sorry.
Simple and obvious.
angech “, I am happy to agree with you that there is a correlation between melting of land ice and sea level rise.”
Evasion, the correllation under discussion was between ice melting and atmospheric moisture. Of course there is a corellation between melting land ice and sea level rise, there is also a direct causal link.
“More water, more humidity possible up to the limit of CC still works for me, sorry.”
but you provide no evidence whatsoever that CC is not the limiting factor (and ignore my arguments for why it is). If CC is the limiting factor then you can add as much as you like to sea levels and it will have no effect on atmospheric moisture. Hence no “benefit” to rising sea levels.
So, what evidence do you have that CC is not the limiting factor?
Dikran Marsupial says: June 2, 2016 at 2:34 pm
The amount of water in the atmosphere depends on air temperature (Clausius-Clapeyron relationship), not the surface area of the oceans.
I don’t recall ever reading that an increase in ocean surface area being a factor, probably because it isn’t, but as usual I am amenable to evidence.
-Dikran Marsupial says: June 3, 2016 at 6:56 am
It is the temperature of the atmosphere that determines its capacity for water vapor, the fact that warming means the oceans will expand as well means the two things are correlated, but there is no causal relationship from ocean expansion to atmospheric humidity.
–
Facts,
I have done your reading for you.
Evaporation of water from the surface of the Earth is the main process providing water
vapor transport to the atmosphere. Studies have shown that the oceans, seas, lakes, and rivers provide nearly 90 percent of the moisture in the atmosphere via evaporation. Yet only 70 percent of the Earth’s surface is covered by the oceans.
505 000 km3 , is evaporated from the surface of the World Ocean and only 72 000 km 3
from land.
Evaporation of water occurs at any absolute temperature of a water surface above 0 o C.
It is a process involves many factors, and it is affected by numerous factors.
For example, evaporation from a free water surface depends on the difference between the water vapor viscosity at the water surface and in the nearest air layers; it depends on the
air and water temperatures, on wind velocity, atmospheric pressure and water quality.
–
Having large areas of water surfaces, the oceans are gigantic evaporimeters of water which is transported to the atmosphere
-.
I have provided the facts for you.
The temperature of the water is at least if not more important than the temperature of the air.
Observe a boiling kettle .
The amount of water in the air is infinitely different to the capacity of the air for water vapor.
Surface area.
Temperature of the water
water vapor viscosity at the water surface etc.
–
“as usual I am amenable to evidence.”
angech wrote “Evaporation of water from the surface of the Earth is the main process providing water vapor transport to the atmosphere. ”
But that doesn’t mean that the area of the oceans is the limiting factor governing the amount of moisture in the air.
I note that yet again you have ducked the challenge to answer how much additional moisture a parcel of air that has traveled across the Atlantic picks up over the last kilometer.
“Observe a boiling kettle .
A kettle boiling actually makes my point about Clausius-Clapeyron. The stuff you see coming out of the spout is not water vapour, it isn’t steam, it is small droplets of water forming a cloud. Why does it condense out into a cloud rather than staying as steam?
The amount of water in the air is infinitely different to the capacity of the air for water vapor.”
Leaving the hyperbole aside for a moment (the difference is not infinite), I note the goalpost moving here from water vapour to clouds. Again, why do clouds form?
“I have done your reading for you.”
The hubris is unbecoming.
Pingback: Climate change deniers make a video about weather. La di da. | Red, Green, and Blue