The Annals of Improbable Research has announced the annual Ig Nobel prizes. (Their server appears to be overloaded at the moment, so you may have difficulty with the preceding link.) Most relevant to our work is the prize for literature awarded to Daniel Oppenheimer of Princeton University for his report "Consequences of Erudite Vernacular Utilized Irrespective of Necessity: Problems with Using Long Words Needlessly."
Too many judges and lawyers write with the apparent intent of impressing the reader with their vocabulary. See, e.g., United States v. Knights, 219 F.3d 1138 (2000), rev'd 534 U.S. 112 (2001). Oppenheimer's research quantifies what we have long believed -- such attempts backfire more than they succeed. In addition, most readers will not reach for the dictionary but rather skip over the obscure words, in which case the writing has failed in its central purpose -- to convey the writer's thoughts to the reader.
At the opposite end of the spectrum of judicial opinion writing are the opinions of Justice Clarence Thomas. They are not written to impress anyone. Their straightforward and unaffected style is crafted so that Joe Average can read and understand them. And that is impressive.