Empirical Data and Unquestioned Assumptions


In an interesting twist, the early thinkers (in a modern sense) both within natural and economic history worked with a relative paucity of empirical data and a relative inability to process what data they had, yet their awareness of the fundamental issues at work is more acute than later researchers.  By contrast, the more recent researchers in both fields,  with far more empirical data to work with and better tools with which to analyze the data, have proven far more beholden to untested assumptions, and far more likely to allow those untested assumptions to overdetermne their results,  even to the point of manipulating the data they had to skew the results in favor of their preconceptions.   

Many of the best known 20th Century economists, such as Kuznets and Friedman, have tended to justify untested assumptions by an overfocus on data concerning systemic behavior during aberrant periods, those that reflect the impact of catastrophic events rather than data compiled during periods where events followed their more usual course, because data compiled during the more usual course of events tends strongly against their assumptions.  The opposite is true of the Neo-Darwinists, who have underemphasized the importance of systemic behavior and change during periods of catastrophe, precisely because that data undermines their preconceptions.

In both cases it’s demonstrable that empirical data, far from altering a given researcher’s preconceptions, is often manipulated specifically to justify those preconceptions.  It’s also demonstrable that the way in which data is manipulated is determined not by an a priori misunderstanding of the relative importance of various data, but by the method most conducive to maintaining assumptions that produce desired results.  Worse, from the perspective of empirical method, the very existence and ability to analyze data allows researchers greater freedom to manipulate data to suit their prejudgments, and that freedom has been exploited to its utmost extent, whereas in a situation of a paucity of data or an inability to process available data, researchers forced back onto their own ability to think and question tended to include a questioning of their own presumptions, largely because they had no manipulable data with which to justify them against anticipated objections.
In other words, the more empirical research has become, the less it feels the need to question its preconceptions, and simultaneously the greater the ability to manipulate data to confirm those preconceptions, the greater the tendency to do so.  The more available empirical data is, in other words, the greater the tendency to confirmation bias justified by manipulation of that data.
When you consider the further change over the past century and more from exact data to statistical data, the ability to manipulate data has obviously increased, but this has had precisely the opposite effect to what might have been anticipated.  Since statistical data is inherently more suspect, the change has not led to an increase in the manipulation of empirical data, but instead led to a renewed awareness of its relative inability to substantiate thoughtless assumptions, and a renewed tendency, at least in the top researchers, to anticipate objections to their work by spending more time questioning the assumptions underlying it.
The belief in empirical data as indisputable ‘hard facts’, a myth that had its greatest effect within actual science in the 19th century and the first 2/3’s of the 20th, is the underlying issue.  Although science itself is in the process of abandoning this particularly noxious superstition, for the majority of the general pubic, empirical data as indisputable fact is the predominant impression of the nature of science itself, and the justification of it’s privileged status.  However what is not maintainable within science itself will soon prove unmaintainable to the general public, and the result, since hte majority of that public will not immediately understand that this critique of a superstition about science is actually a positive correction to the notion of what science is and is not, will be at least initially a disillusion with science that will be more profound than any religious disillusion has been.  The extremism likely to accompany this disillusion arises from three factors – the suddenness with which it tends to occur,  the rate at which such understanding is shared, and the absoluteness of pronouncements made by scientists themselves and by laymen basing their pronouncements on those of scientists.
The most extreme of those who are anti-religious are almost invariably those who had the most belief in it, hence the extremism of atheism in former fundamentalists.  That extemism is liable to seem tame compared to the extremism when science as absolute truth is forced to admit not simply individual failures, but ageneral failure of scentific method made possible by science’s own faith that its methods were infallible.

 

Advertisements

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s