In the field of forensic linguistics, "linguistic evidence" is regarded as a type of evidence which is encoded in the form of language. Integrated within the rules of evidence in common law system, "linguistic evidence" is frequently interweaved with role or functions of linguists appearing in court: when the content or the form of language in a case is disputable , analyses and conclusions are usually provided by linguists. On one hand, in common law judicial practices, it is common that expert witness is employed by parties (plaintiff and defendant in civil lawsuits; prosecutor and defendant in criminal cases) of a case to settle technical or specialized disputes; on the other hand, a judge is not obliged to admit the expert testimony or to decide a case based on the expert testimony. For instance, in the United States, the rules of evidence require that expert testimony should rest on scientific or technical knowledge; therefore, it is controversial whether “expert testimony of linguistic evidence" is scientific or technical. On the contrary, in the People's Republic of China (hereinafter the P. R. C.), technical or specialized issues in judicial practices are in principle tackled by identification or examination reports rendered by judicial identifiers who are administrated by judicial authentication institutions. The institutions are set up after being accredited and registered by provincial judicial administrative authorities. Besides, categories of judicial identification and examination are refined by laws (except for handwriting, most "linguistic evidence" in theory does not fall into the categories). In 2012, the revisions of the Civil Procedure Law of the P. R. C. and the Criminal Procedure Law of the P. R. C. introduced a new role entitled "person with expertise" (there are scholars who refer to "person with expertise" as "expert assistant" or directly refer to "the person" as "expert witness") into judicial practices: parties may lodge applications to the people's courts to allow "the person with expertise" to appear in court and render opinions ("expert opinions") on identification/examination reports or on specialized issues. Grounded on the background information, this empirical study explores civil and criminal cases tried by the federal courts of the United States (expert witnesses proffered analytical analyses on "linguistic evidence" in all cases) from four perspectives: firstly, the manifestations of "linguistic evidence"; secondly, persons who were considered qualified to proffer testimonies; thirdly, analytical methods adopted by the expert witnesses in the selected cases; fourthly, standards and bases adopted by the judges of the cases when they conducted their discretions on the expert testimonies of different "linguistic evidence". There are four steps in carrying out this study: firstly, civil and criminal cases were searched and collected from LexisNexis and Leagle by entering keywords, e.g. linguistic evidence, linguist, etc.; secondly, the cases were classified and counted (with the assistance of Excel) according to different causes of the cases, distinct forms of linguistic evidence and diverse analytical methods on which the expert testimonies rested; thirdly, under what circumstances would the testimonies be admited or excluded were observed[whether the testimonies were adopted or rejected?]; fourthly, discussions and conclusions on the previous observations were presented. Though there is a disparity in systems of laws and rules of evidence between the P. R. C. and the United States, comparatively mature judicial experience of the United States is still able to enlighten the future orientation of the quasi-counterpart ("person with expertise") in judicial practices of the P. R. C. given that in the P. R. C. the role of "person with expertise" is yet to be clarified, and cases whose parties proffered "the person with expertise" to appear in court and render opinions are still limited.