Authorship attribution under the rules of evidence: empirical approaches - a layperson's legal system

All documentary evidence admitted at trial must be authenticated, i.e., it must be demonstrated that the document is 'what its proponent claims' it to be. For authorship, attribution, this means that it can be demonstrated that the document in question has or has not been authored by a specific person. Among the different methodologies addressing this question of 'who wrote it,' those practiced bylaw enforcement specialists are often not scientifically reasoned or tested, nor are they empirically sound. Further problematic, the legal standards necessary to support a showing of authentication defers to the layperson's ability to evaluate the sufficiency of a given technique. In this paper, I will discuss, analyze, and pit a survey of current authorship attribution techniques against authentication and expert testimony standards. I will then discuss, through a recent case example, what can go wrong when law enforcement specialists are allowed to testify to the lay jury.