Monday, November 11, 2013

On the quality of research, or - are we doing any science?



On November 1, 2013, I spent half a day at the 5th Portuguese Conference on Polar Sciences, at the Algarve University in Faro. I was doing a social sciences experiment, making observations relevant to two science-related issues:

 

1. Do researchers know what science is?


I asked two of the researchers who were making their reports the following question: Taking into account that any progress of science is only possible if we formulate and question hypotheses, seeking experimental evidence allowing to reject some of these hypotheses, what were the hypotheses you have formulated, and how have your experimental evidence helped to reject some of them?

One of the presenters had no idea whatsoever what I was talking about. The other tried to excuse himself, saying that "we should not be so fundamentalist about this". Well, either we are doing science, or we are doing fact-finding or statistics (rather useless and wasteful, without a hypothesis that would define what to look for) - it is nice to be able to take photos of poorly accessible places, and do it on public money, but this is not what science is. His co-author, however, stated that they were discussing hypotheses in their poster - which I could not confirm; indeed, they were discussing issues related with human perception and theory of knowledge and learning, but I could not find any evidence of them trying to falsify some hypothesis that they have formulated.

One more researcher had a hypothesis formulated in a previous publication by a different group - and then was looking for evidence that conformed to this hypothesis. She went to the same place and collected some different species - which did behave in a similar way in similar conditions. So what? Was she looking for anything new, contradicting to her original hypothesis? Well, she was not. She does not understand what science is.

Still another (senior) researcher did have some hypotheses formulated in his slides, and mentioned it in the beginning of his talk, alerted by my questions to his colleagues. However, he mentioned no attempts to falsify any of these hypotheses - once more, they were searching for conforming evidence - and wasting public funding. In fact, no hypothesis may be definitely and unquestionably proven; those who think otherwise commit a logical error - and there is no sin in science worse than that. If your logic fails, how can you aspire to be a scientist?

There were more presentations, discussing new financing opportunities and collaborations that would lead to more public spending, but not science ...

2. Are scientific conferences discussing science?


Well, this one has not.

Firstly, there was not much of science to discuss - what was presented, apart from nice photos from Arctic or Antarctic, was some research, either without an idea (without a specific working hypothesis), or attempting to confirm an idea (a hypothesis). As already noted, no progress of science is possible in either of the two cases.

Secondly, there were no questions asked about science by the audience - apart from mine. However, I was not given an opportunity to question every presenter - the usual excuse was that they are behind the schedule, and have time for only one question - which was given to someone else. On at least one occasion, the lady who was presiding the session abused of her powers and asked the (only allowed) question herself, instead of giving an opportunity to the audience. The questions asked by other participants were very few, and usually touched either on minor details of the experimental procedures, or on the future plans of the respective research group ...

Obviously, the conference was supported by the taxpayer money, same as the rest of the "research effort".

_

Friday, June 7, 2013

Comment on science fraud submitted - and rejected

 

Pseudoscience Applied in Study of Pseudoscience 


Igor Khmelinskii1, Peter Stallinga2
1: University of The Algarve Faculty of Science and Technology Centro de Investigação em Química do Algarve e-mail: ikhmelin@ualg.pt
2: University of The Algarve Faculty of Science and Technology Center for Electronics Optoelectronics and Telecommunications e-mail: pjotr@ualg.pt

The recent paper by John Cook et al. (2013) analyses the scientific consensus over the anthropogenic global warming. We shall start by noting that the term 'consensus' belongs to politics and not to science. Science has never been done by consensus, it is done by facts. Also, science is not a parliamentary democracy, the scientific truth is not decided by vote; science may occasionally even amount to heresy, as happened, for instance, in the case of Galileo. Therefore, any attempts to divert from facts to anything else, including consensus, make us suspect pseudoscience.

We shall use the working definition of the scientific method as given by Feynman (1965): "... we look for a new law by the following process. First we guess it. Then we compute the consequences of the guess to see what would be implied if that law that we guessed is right. Then we compare the result of the computation with nature, with experiment, or experience, compare it directly with observation, to see if it works. If it disagrees with experiment it is wrong." It follows from this definition that we are unable to prove any hypothesis by providing conforming evidence; we are only able to reject a hypothesis by providing contradictory evidence. Reformulating, while we have several alternative hypotheses, each of them explaining the existing experimental evidence, we are unable to favor any of them. Effort should be spent on finding proof to reject some of the alternative hypotheses.
Note that all of the pro-AGW climate research is, by this definition, undeniably pseudoscientific and the respective researchers are pseudoscientists, for the simple reason that finding proof for any theory, including the AGW hypothesis, is not science. On the other hand, searching for failure of a theory is science. However, AGW has already been falsified many times, more than the single time required by the scientific method and no more research is needed in this field. Researchers are either not doing science (pro-AGW), or go about reinventing the wheel over and over again (anti-AGW). Indeed, Carter (2010) lists eight independent tests for the AGW hypothesis in his Ref. 228, and Khmelinskii et al. (2010) offer another two. The AGW hypothesis fails any and all of these tests, and must therefore be discarded.

Now, as follows from their conclusions, John Cook et al. (2013) assume that the level of AGW consensus, as derived from published peer-reviewed papers, accurately reflects the opinions of scientists. However, they ignored several (alternative or compatible) hypotheses that may also explain the observed prevalence of pro-AGW views in peer-reviewed literature. The first alternative is the 'verbal diarrhea' hypothesis. Indeed, pseudoscientists, in contradiction with common logics, try to prove the AGW hypothesis by publishing as much confirming data as they can, which causes a proliferation of pro-AGW publications. On the other hand, the scientific approach would require just one paper that falsifies the AGW hypothesis, whereupon the respective scientist could move to another topic of studies. The second alternative is the censorship by biased editors. Indeed, assuming a 1:1 proportion between pro-AGW pseudoscientists and anti-AGW scientists in climate-related research, and the same proportion between scientists (objectively minded) and pseudoscientists (biased) as editors, and also assuming that biased editors reject all of anti-AGW papers where anti-AGW scientists let pass 50% independent of type, we shall immediately obtain a 1:3 proportion between anti-AGW and pro-AGW publications. The third hypothesis is the grant-money corruption. Obviously, researchers need funds for their research, while a publication that demonstrates that the AGW problem does not exist will not look well on their curriculum if they apply for a grant to study climate change. Note that climate-related research and policies are being financed at US$10 billion a year (Carter, 2010, p. 149); this sum alone constitutes a strong temptation to embrace the climate pseudoscience. The fourth hypothesis is the noble-cause corruption (Carter, 2010, Ch. 7). Having ignored the alternative hypotheses, instead of falsifying them by providing contradictory evidence, John Cook et al. (2013) undertook a pseudoscientific statistical study, which does not enable to arrive to any valid conclusions about the real opinions of the scientific community on the AGW problem. As a more important conclusion, no further research that aims to prove or disprove the AGW hypothesis should be financed, as objectively it would be a waste of public funds – this hypothesis has already been falsified.

REFERENCES

[1] J. Cook, D. Nuccitelli, S. A Green, M. Richardson, B. Winkler, R. Painting, R. Way, P. Jacobs, and A. Skuce, "Quantifying the consensus on anthropogenic global warming in the scientific literature", Environ. Res. Lett. 8, 024024 (2013) doi:10.1088/1748-9326/8/2/024024
[2] R. Feynman "The Character of Physical Law", The M.I.T. Press. (1965), p. 156. ISBN 0-262-56003-8.
[3] R. M. Carter, "Climate: The counter consensus", Stacey International Publishers (2010). ISBN 978-1-906768-29-4.
[4] "Climate Change Policies for the XXIst Century: Mechanisms, Predictions and Recommendations", I. Khmelinskii and P. Stallinga, International Journal of Energy and Environment, 4 (2010) pp. 237-244. Online: http://www.naun.org/journals/ energyenvironment/19-660.pdf and http://arxiv.org/abs/1105.5845

Reply by the editorial office

Final decision on your article from Environ. Res. Lett. - ERL/476364/COM/310360
...
This is not a scientific comment. Comments on ERL articles must present a robust challenge the science or methods in a published article, as this article clearly does not contain this, we should reject the comment as unsuitable.
...