Interesting, scientists are human, but science is a collaboration which allows correction or errors made by individuals. And has over centuries made progress through the most difficult of problems.

Maybe it can take a decade or generation for a new idea to become accepted, but the testing process (of say, something like the theory of evolution) cancels out the individual biases of each scientist.

That said the 'history of science' is used as a teaching tool, to help students understand a complex process it helps to tell a step by step story.

They show the failings of previous theories with specific examples and then the simplest explaination which explains that experiment.

This helps students understand very complex ideas, and their justification, but not the real process of how scientific progress is made. It is a fabrication, and it is well known that 'science history' is just that.

If you ever learned any of the history of science in high school, it was likely a simplified story/model used as an aid to learning. And these simplified stories often result in misundersandings which i've seen used by creationists among others use to 'debunk' science. Most often a result of their own misunderstandings - sometimes coupled with over-simplification of the model - but made all the more convincing by how widely known the model is.

I just read this and some of the linked articles. It seems a lot of empty shadow-boxing to me, albeit this might not be the case for people that are less familiar with PoS/ToS/epistemology (or however you might call it). The article's pretty much are modern language rewriting of Kant's paragraphs on the role of "philosophers" (aka scientists in modern language) in the "Perpetual Peace." orathaic pretty much made the point of why and what HoS should tell us. I'd just like to point out where most critics or debunkers of science got PoS (and in particular Kuhn) wrong:

(1) science is a process, which does not rely so much on "testing" but on "reproducing." This basically means, if one can reproduce the discovered effect (or phenomenon) under the same conditions, it holds, if not, then something is fishy.

(2) Scientific knowledge is authoritative not because of the degree names or the tenure affiliations of their holders, but because of the possibility to access and evaluate the knowledge generating process. You can then change the process (eg. switching from thought experiments to actual experiments and then check the results again, or many, many more). Scientists are people trained at doing this kind of things.

(3) "Debunkers" (and this author here) make up some fictional type of scientist, then look for one or two examples where real world scientists don't fit the bill and derive from this the extreme conclusion (a) scientists don't know shit or the more moderate (b) outsiders are needed to fact-check untruthful scientists.

The claim scientists exaggerated everything might be right. In most cases though, I have the impression that the exaggeration is not the scientists doing, but the (non-scientific) audience's doing. This might stem from general unfamiliarity with scientific jargon (which differs across the fields), a lack in statistical knowledge, certain expectations, etc. That the author cites in particular biomedicine and economics as prime examples should not surprise here. Both concern us all in our daily lives, so we hope to improve our situation and to do so by acquiring more knowledge about it. We then are informed eg. that a hike in min wages costs us 1% of PIB growth. Most of the time, the information is something like 1% PIB growth ± some span. Most people read this as 3% PIB up this year, we had min wages so we could have had 4% FOR CERTAIN (or AT LEAST). It is very difficult to blame that exaggeration on the scientist. And if this wrong understanding didn't materialise (because it was never part of the claim anyways), you get yourself another argument against "all-knowing" scientists.

(4) The scientific community is not a court with authoritative judicial power or a legislature where you negotiate the validity of your claims. If a consensus builds up it works more like: "Hey you know, you drive on two wheels – No way, I tried and I fell, it is not possible after all. – well, you need of course build a bycicle to ride on them – AH, ofc then it works, but that is not impressive, because I wanted to ride on two wheels WITHOUT anything attached – you're right this doesn't work" This looks like a controversial discourse, but both of the "opponents" will agree, that you can ride on two wheels. Provided that you have a bycicle. Unfortunately, most debunkers (and many sociologists) don't get this part. They tend to make up a big pro and con story. For entertainment. And blame science for it.

Here is a great example of criticism of 'science' pointing to flaws in the communication that is going on: www.vox.com/2015/3/23/8264355/research-study-hype

It uses data from actual research to support its position, ie is itself scientific.

It uses data from actual research to support its position, ie is itself statistical. Not scientific. Which may make it more, or less, true, depending on the statistics. But statistics != science. (Although, as Adam Savage says "The difference between fooling around and doing science is writing your results down!" )

An interesting thread and one on which I have been meaning to comment.

1) '' But statistics != science''. Personally I find it useful to be clear whether the stats being employed are descriptive (e.g., means, standard deviation etc.) and when they are inferential (e.g., judgments of probabilities etc.). The former group are inherently potentially useful whereas the latter group may very well be useful, but also come with very big health warnings. For example, we all know that correlation does not necessary imply causation. To give an example from the scientific literature (Nature no less): Quinn, Graham E.; Shin, Chai H.; Maguire, Maureen G.; Stone, Richard A. (May 1999). "Myopia and ambient lighting at night". Nature 399 (6732): 113–4. doi:10.1038/20094. PMID 10335839.

2) I would make the further point that I do have a little sympathy with the sociologist quoted, at least on the grounds that scientists are often dedicated hyper-reductionists. Please don't misunderstand me reductionism is a great tool, but when it is employed to the exclusion of the ‘wider-world’ it too comes with a very big health warning. To finish on a positive note here is some very interesting news from yesterday which showed how cooperation between medieval scholars and microbiologists might well have resulted in a breakthrough around the very serious and growing problem of antibiotic-resistant infection. Inspiring to read and, without question, real science.

http://www.newscientist.com/article/dn27263-anglo-saxon-remedy-kills-hospital-superbug-mrsa.html#.VRvaz46rEuI

https://www.youtube.com/watch?v=mo4K51bQVs0