All of us who talk about the history of the 18th and 19th centuries can only make inferences based upon the data that we have available to us.  

No one disagrees that some of the WPA narratives contain favorable descriptions, by people who had been enslaved, of slavery.

The question before us is what to make of that data.  As many of us have tried to explain to you, at some length, the data must be taken with a grain of salt, for a host of reasons.  A persuasive interpretation of the past must not only adduce data to support its conclusions, it must also subject the data itself to critical analysis.  This is true, by the way, for any rational inquiry--historical or otherwise.  If you use bad data, you will get bad results.  In this case, the data is not so much "bad" as it is skewed.  It is skewed for a whole host of reasons, most of which I and various others on this list serv have discussed at some length.

I do not disagree that some slaves and former slaves wrote or said positive things about life under slavery.  The larger question before us is, given this fact, what can we conclude about the nature of slavery.  

Some of us here are well familiar with both the data (the WPA narratives) AND the analysis of their provenance and reliability. I conclude from reflection about this data that despite the existence of some favorable descriptions of slavery in the WPA narratives, slavery was a pretty wretched and awful institution, that slavery systematically degraded and dehumanized the people subjected to it, and that slavery brutalized both slave and slave-owner.  I conclude that the testimony common from apologists of the "lost cause"--that slavery was a mostly benign institution, that most slaves were well treated and lived happily, and that most masters were benevolent--is false, and certainly is not sustained by any reasonable and fair-minded assessment of what the data tell us.  

My guess is that most of the other people here who have read this data and have thought deeply about the circumstances under which it was collected will agree with me.  But if you do choose to disagree with me, it is incumbent on you to explain why you believe the data is NOT skewed the way that I have suggested that it is.
Kevin R. Hardwick, Ph.D.
Department of History
James Madison University