Truth and Freedom in Economic Analysis and Economic Policy Making
Higgs, Robert, Independent Review
For thousands of years, philosophers have told us that if we are to live our lives at their best, we should seek truth, beauty, and goodness. Of course, each of these qualities has raised thorny issues and provoked ongoing arguments. That people have carried on such arguments, rather than surrendering themselves to their raw appetites and animal instincts, may be counted a valuable thing in itself, but a final resolution of such deep questions may lie beyond human capacities.
In regard to goodness and beauty, I have nothing worthwhile to add to the discussion. For guidance in seeking goodness, we may look to saints, theologians, moral philosophers, and moral exemplars of our own acquaintance. For demonstrations of beauty, we may turn to nature and to artists, great and small, who have adorned our lives with the grace of music, poetry, and the visual arts. My own professional qualifications, as an economist and an economic historian, do not equip me to contribute anything of value in these areas.
I do feel qualified, however, to speak in regard to truth, because the search for truth has always served as the foundation of my intellectual endeavors. Moreover, my study, research, and reflection within my own professional domains have brought home to me a relationship that others might do well to ponder and respect--a relationship, indeed, an array of relationships, between truth and freedom, such that anyone who seeks the triumph of truth must also seek to establish freedom in human affairs.
When I began my academic career in 1968, my research specialty was the economic history of the United States. I was expected to publish the findings of my research in reputable professional journals. For a young man just beginning to master his field, carrying out publishable research was a daunting task. Thousands of other writers had already contributed to building up the literature in my field, so adding something of enough importance to merit its publication in a good journal was hardly an easy task.
I discovered, however, that one way to proceed was by identifying significant mistakes in the existing literature and correcting them. Moreover, I soon found that many such mistakes had been made. To put this statement in another way, I found that the existing sources often failed to tell the truth about one thing or another, and in some cases the falsehoods propounded by one writer led later writers, who relied on those false statements, to make additional errors of their own. We often think of the scientific or scholarly enterprise as a cooperative process in which the establishment of one truth facilitates the establishment of another, but, unfortunately, the process often works in an adverse way, too, as the establishment of one falsehood fosters the establishment of another.
The errors in my fields of study and research take two main forms: factual and interpretive.
Factual errors arise on a few occasions from deliberate falsification, but they arise far more often from sloppiness in the observation, measurement, transcription, and processing of data. In checking quotations, for example, I often found discrepancies between the words quoted by a writer and the words appearing in the source from which the quotation was taken: some words or punctuation marks were omitted, or other words or punctuation marks were inserted, without any indication being given of such changes. Many writers are simply not careful and therefore make false statements of fact.
For example, I found that in a well-regarded article the increase in U.S. cotton production in the United States between 1850 and 1860, compared to that between 1860 and 1880--an essential fact for the argument being made--had been measured with a large error in part because the original researchers had assumed that a bale of cotton contained the same amount of lint at each of these three dates, whereas the amount of lint per bale had actually increased from 400 pounds in 1850, to 445 pounds in 1860, to 453 pounds in 1880. …