Science Fictions - Excellent

I haven't enjoyed a new book quite this much in a long time.  It is Science Fictions: How Fraud, Bias, Negligence, and Hype Undermine the Search for Truth by Stuart Ritchie.  It directly addresses one of my constant themes on Thingfinder, the prevalence of cognitive pollution and the tendency of both academia and the mainstream media to misrepresent scientific research.  It goes to a broader issue - a rising incapacity in public discourse to make a reasoned, evidence-based argument.  

If our remarkable progress derived from the Age of Enlightenment values is to continue, we need to restore that capacity and discipline.  Even as I write this, NPR, New York Times, Twitter, Facebook and legions of other establishment institutions are trying to make the case that restricting information and stifling debate are ultimately communally good.  Kafka never woke to such a nightmare.  

Well, perhaps he did.  The lead character in Metamorphosis is Gregor Samsa.  He wakes one morning to find himself inexplicably transformed into a huge insect.  However, the German ungeheures Ungeziefer literally translates as "monstrous vermin".  And I cannot help but feel that transformation has indeed been visited on our mainstream media and academia.  Monstrous vermin of intolerance, hatred, coercion, and fearful of freedom, rule of law, equality before the law, human universalism, broad human dignity, etc.  

Ritchie does a magnificent job of framing the issue of science research fraud, bias, negligence and hype, thoroughly documenting the issues with both entertaining specific examples as well as providing a larger context and a sense of the magnitude of the problem.

By the end of the first seven chapters, there can be no doubt just how serious is the problem.  It is a problem of culture, of incentives, and systemic bias towards undesirable outcomes.  His closing eighth chapter is reasonable about multiple solutions which can be, and are being considered.  He resists the totalitarian mindset which prescribes a set solution and simply observes that some subset or combination of solutions being trialed or considered simply must work.  Otherwise, the engine of progress winds down.

Critically, Ritchie maintains a buoyant note.  Yes, it is a problem, but it is a problem for which there are viable solutions.  Too many books about an issue turn it into a apocalyptic harbinger of disaster.  Ritchie is content to demonstrate that it is a real problem, with real consequences and express confidence that we will overcome these shortcomings.  Hurray.

It won't be easy but we need to get serious about this.  Many reputations and practices will need to change.  

I highly recommend this book.

It is beyond the scope of his topic but in reading Ritchie's text three ideas kept pushing forwards.

1)  This a problem because reliable data is very expensive to create; 2) This is a TQM/Six Sigma problem subject to tried and tested quality improvement methodologies which have been around for decades; 3) This is a massive class issue.  

With regard to the latter, white and blue collar workers have been through the ringer in terms of learning and adhering to corporate and manufacturing Six Sigma practices at least since the 1980s.  Six Sigma thinking (ring out production failures by addressing root causes with data and decision-making) is hard work, wrenching, disruptive and challenging to adhere to.  

Much of the population in formal employment have been exposed to and subject to it in the private sector for decades.

But reading the examples, it comes across again and again that academia, media and other establishment bodies have excused themselves from the disciplines under which everyone else works.  No wonder there is rising distrust between citizens and the denizens of establishment institutions.  They live in different worlds, bearing dramatically different costs.  

Excerpts to follow.