logo

Science and the Turkey’s Expertise

Albeit Christmas is rapidly coming to town, let’s imagine a turkey living in a farm. The turkey feels that everything is just fine because she is in safe, gets food on a daily basis; therefore she assumes that this trend will just continue in the future. However, when Christmas comes, people cut the turkey. All the turkey’s assumptions based on path-experiences and perceptions suddenly fell apart. It turned out that she did not know the reality, but her own perception of it.
We can say that by analysing and collecting statistical data and information, the observer does nothing, but synthetically manufacturing reality, which does not necessarily equal to the “real” reality. Along the course of this manufacturing, experts and economic practitioners are inclined to put more weight on numerical, statistically recorded data than subjective opinions across the board of various types of surveys and analysis.
The bias towards prioritising hard data is partly originated in our intention to be as scientifically rigorous as possible. In the 1930s, Karl Popper stated that the demarcation line between science and non-science lies in the issue whether the given theory can be falsified or not by using empirical analysis based on measurable factors. As a consequence, the bias towards using statistical data represents our claim to get closer to certainty which is considered as truths. People think that certainty is everything, certainty equals with truth and the ultimate goal of science is to find truth. This is why statistical data are widely held as fingerprints of what is really going on. To a great extent, this assumption is in order.
Experts are building on those recorded data in their analyses in which they use narratives to interpret the data. These analyses will influence the perceptions of other experts, market actors as well as that of policymakers over what is really going on in the economy. And, as experience shows, these perceptions can be rather wrong just like in case of a Christmas turkey, especially if the factor to be considered and interpreted is more featured with qualitative dimension, e.g. corruption.

Example: Stable or Lessening Corruption in Moldova

If one takes a mere glimpse into the World Bank’s Worldwide Governance Indicators, especially into its indicator entitled “control of corruption” on Moldova – which indicator reflects perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as the “capture” of the state by elites and private interests – the Moldovan reality was as follows: people seem to have perceived less and less corruption between 2009-2010 and 2012. What is more, the Moldovan corruption seemed to be relatively stable between 2005 and 2012 indicating the perception that the government could fend off increasing corruption via various anti-corruption policies.
However, this was just a manufactured and wrongly perceived reality. Up until 2012, control of corruption was improving, then, the government collapsed in early 2013 as a result of a huge corruption scandal which hit the media and in which senior members of the prosecutor general’s office and political elite were involved over years. In October 2015, the prestigious New York Times emphasised that Moldova “[…[ was rocked by the discovery that $1 billion had been fraudulently siphoned from Moldova’s banking system over a period of years, a huge amount for an impoverished country whose entire economic output is only about $8 billion a year.” Anti-government protest has become the new norm in Moldova by singing “Wake up Moldova, you have had enough!”
The new ‘control of corruption’ indicators were published on 25 September 2015 that were captured such a declining tendency as well.
Chart 1. Corruption perception index in Moldova (2002-2014)
falsificationSource: own compilation based on World Governance Indicators (2015).
To sum up, of course, manufacturing reality has always played a decisive role, because mankind pursues certainty by believing that measurable factors tell more about what is really going on. This approach seems to offer much more ammunition for scientists to test various theories in an empirically more vigorous way. Still, in social sciences, numerical data often hide just more numerical data by not adequately signalling what is really going on.
Interestingly, and recently, this guiding principle has been undergoing an observable softening even in natural sciences, like physics. Modern physicists have recently asked a fundamental question: Shall we insist on the Karl Popperian way of doing science (i.e. test empirically all potential theories) or can we make a shift towards doing science with a belief that even though we cannot test empirically our theories now (String theory, multiverse, M-theory etc.) due to practical limits, our current untestable theories cannot be discredited because maybe in the future we will be able to empirically test them through new technologies.
From our point of view, even if we can show empirically at a certain time of analysis that a social-economic variable (e.g. corruption) has been decreasing; we must always do it with some kind of moderation and with meticulous care without neglecting the dynamics of various processes behind the hard-data surface.
Olivér Kovács Ph.D is a Hungarian economist specialised in the research fields of sustainable development, fiscal sustainability, innovation and innovation policy.

Leave a Reply

*