New Truth - as Only I See It

There are two infamous publications in the world of scholarly activities, "The Journal of Irreproducible Results" and "How to Lie with Statistics."  One is a spoof on science truth published regularly (http://www.jir.com/); the second is an actual book.  The journal is interesting and funny.  But the book is well known to all practitioners, and the best of them know how to use statistics as needed to make a point, a claim, or a theory.  

In a recent NY Times article by George Johnson (http://www.nytimes.com/2014/01/21/science/new-truths-that-only-one-can-see.html) the veracity of many publications are taken to the veracity task.  It is claimed that up to 80% of all publications are in error or just plain false. To quote from the article "It has been jarring to learn in recent years that a reproducible result may actually be the rarest of birds. Replication, the ability of another lab to reproduce a finding, is the gold standard of science, reassurance that you have discovered something true. But that is getting harder all the time. With the most accessible truths already discovered, what remains are often subtle effects, some so delicate that they can be conjured up only under ideal circumstances, using highly specialized techniques."

It really is not that practitioners are dishonest though some are to be sure.  It is that in the rush to publish, they take short cuts leading to insufficient, incomplete, and inaccurate conclusions.  Nothing here is new.  The system of peer review has been compromised by the massive number of journals needing food (new articles), the lack of scholarly scrutiny, the lack of scholarly review, and the lack of stringent principles in referee reporting. 

By this article, the gold standard of replication has been shattered, partly because there is no time and especially no inclination of others to validate or verify what others claims.   It is also expensive to duplicate complex studies.   In the explosion of knowledge over the past century, everyone is exploring their own research program, referencing only casually other works, and substantially not caring what they say except when in agreement.  In some areas, there is a "required" literature review section to each paper.  This is somehow considered the guarantor of honesty, but in reality this section is sometimes loaded with references to papers written by journal editors and suspected referees. 

There is an exception.  It concerns the vertical VS. the horizontal.  Too many fields these days are strictly horizontal meaning that knowledge is propagated on a horizontal plane with little reference except is type to previous knowledge.  Horizontal knowledge is something like an oil slick spread over thousands of square miles of microscopic depth.  Vertical knowledge is stacked, one result upon the next.  Veracity is essential; it is checked; it is validated. It is just not practical for any reason to proceed upon a false basis.  When knowledge is horizontal, there are no counter checks.  No one really cares about others in their push toward their own set of personal truths. 

There is an exception to the exception when vertical subject matter is at hand.  It concerns modeling.  The foundational lesson from the great Sir Issac Newton (1643-1727)  and his fabulous success with the law of gravitation, the proof of Kepler’s laws and the like, is that we should rigorously generate models of reality – of whatever flavor.   We are good.  We now generate new models by the score – all excelling in agreement with extant knowledge.  New dimensions are added as needed as we grab at the brass ring of experimental agreement.   Yet, in some ways we have transcended observability in favor of comprehensiveness.   This is a problem for practitioners yet to come.  Currently, there have been so many successes the shadow of doubt is not really allowed. 

It is singularly curious that in the world we live in politicians systematically distort the truth, tell us complete lies, and engineer opinion that the scholarly world should be different.  Corporate executive enhance the continuing success of their company.  Lawyers use every tool and trick of the law to argue, aka prove, their case. They all want the same thing: advancement, fame, power, influence, and acceptance. Their media is different, but in various combinations their goals are about the same. Some play in a larger sandbox than others.   The NY Times article has the tone of dismay when expressing this diminution of standard.  It is almost as though the author, while accepting on a daily basis lies and deceit from other venues of society is disappointed the plight of scholarly ethics and distressing lack of proper oversight.  Well, consistency has never been the strong suit of journalists - nor any of us.