Errol Morris, A Wilderness of Error *** 1/2

Years ago, I was involved in a murder case in Dallas that I was able to resolve. This case, often I don't know what to make of it. – Errol Morris

A Wilderness of Error revisits the case of Jeffrey MacDonald, the Green Beret doctor convicted of murdering his wife and daughters in 1970. Morris has two major arguments: that MacDonald was convicted because the prosecution told a better story than the defense, and that the full set of evidence supports at least reasonable doubt about MacDonald's guilt.

I read Fatal Vision back when it was published – indeed, I'm a huge fan of the book. Fatal Vision is infamous for the way author Joe McGinniss started out supporting MacDonald but ended up portraying him as guilty. That is the strength of his book: you start out thinking there's no way MacDonald could have done it, and your doubts build slowly until the shocking denouement. McGinniss makes a convincing case, but I'm not at all surprised that it's somewhat fictional... or let's say speculative. The narrative works too well in literary terms, and MacDonald's protestations of innocence have always given me pause.

Morris makes a convincing case for reasonable doubt and the unfairness of MacDonald's trial. His analysis of the weakness of the physical evidence – the "impossible" coffee table, the holes in the pajama top, and so on – is especially compelling, because it is the physical evidence that provides the clinching arguments in Fatal Vision and ultimately doomed MacDonald.

Morris spends many, many pages on Helena Stoeckley, the woman who repeatedly confessed to being the "woman in the floppy hat." The judge ruled her testimony inadmissible, which seems prima facie prejudicial. However, the sheer number of pages dedicated to Stoeckley's confessions and retractions makes me sympathize with the judge's ruling that the "probative value is substantially outweighed by the danger of unfair prejudice, ... waste of time, or needless presentation of cumulative evidence" (Rule 403).

Finally, Morris ridicules the idea that MacDonald had a psychotic break. Both versions of the story agree that MacDonald's prickly personality is a factor in people's conclusions, but MacDonald has never shown any sign of the type of personality disorder necessary to support the prosecution's version of the crime. I wish Morris had delved deeper into this aspect of the story.

The organization is a bit of a mess: Morris creates a fog of doubt rather than a reasoned argument. That's all a defense needs to do. I still wonder, though: Why was there so little evidence of four people trampling around the house on a rainy night? That's the question that started it all, the reason the prosecution never seriously investigated who the four people might be.