Expert evidence in handwriting identification - the claim to be able to identify an author from an examination of written documents - has, after a century of use, once again become highly controversial. As critics attack the so-called expertise as insufficiently proven and perhaps nonexistent, numerous judges have begun to rein in this form of expert testimony, and several have recently decided to exclude it altogether. This paper takes the current controversy over the admissibility of handwriting identification evidence as an opportunity to examine how handwriting identification evidence first became established as a legitimate and credible form of proof. Handwriting identification is an unusual form of expert evidence because it was the first kind of expertise that was primarily forensic, invented specifically for use in the legal arena. This paper is the first article-length examination of the history of this form of expertise, and offers a detailed examination of how handwriting experts convinced judges that their new form of knowledge was legitimate and worth permitting as legal evidence. It investigates the doctrinal and institutional context surrounding handwriting identification evidence, and examines both how expert evidence in document examination came to be admissible and how it came to be understood as persuasive. In addition, this historical study has two more general purposes. First, it seeks to shed new light on a dominant issue in the expert evidence literature, the tension between whether fact finders should defer to experts or be educated by them. Second, it challenges a bedrock assumption of most commentators on expert evidence: that reliability is exogenous to law, or, more specifically, that when judges make determinations about the validity and reliability of expert evidence, they are ratifying something that either exists or does not exist out in the world. This analysis suggests instead that, at least in some circumstances, judicial determinations can play a significant role in constructing broader cultural perceptions of reliability. At least in the case of handwriting identification evidence, judges did not simply import reliable evidence; rather, they helped to create it. Finally, this paper uses this history of handwriting evidence and the contemporary challenges to handwriting to offer some speculative musings on some of the problems implicit in Daubert's approach to expert evidence.