This measure allows the court to make an assessment of a person’s knowledge of a language other than English. The procedure can be used for virtually any language and can be applied by an examiner who speaks only English.

Measures 1.3.3 and 1.3.4 are unnecessary if courts already have a valid and reliable testing process for interpreters for the language in question, including freelance interpreters. If freelance interpreters are not tested prior to employment, the measure should be used.

Back translation is a technique in which a candidate interprets or translates English into the foreign language in question and, after the passage of time, interprets or translates her or his own foreign language version back into English. The interpreted or translated English version is then compared to the original English to determine how faithfully the original message has been preserved.

Planning/Preparation. Before undertaking this measure the court should acquire the textbook Fundamentals of Court Interpretation: Theory, Policy and Practice.²¹ The textbook includes a detailed description of the proper procedure for administering and scoring the back translation exercise, including 10 sample questions and statements with underlined scoring units. The measure also requires the use of two audiotape recorders, one for playing a recorded script and one into which the interpreter records her or his interpretation of the script.

A written script in English is then prepared in a form identical or similar to the script suggested in Fundamentals of Court Interpretation. The written script is read aloud into a tape recorder in the same way that an attorney would pose a question to a witness or a witness would answer a question. Between each prerecorded question or statement there must be a pause long enough for the interpreter to complete the interpretation.

To conduct the measure the court identifies all interpreters who work regularly in the court and plans a testing schedule. The schedule should require the interpreters to report to the testing room on two separate occasions. On the first occasion the interpreter listens to a tape-recorded passage in English and interprets it aloud in the foreign language, using a second tape recorder to record the foreign language rendition. On the second occasion the interpreter will listen to her or his own recorded foreign language rendition of the original script and interpret it back into English. The interval between the first occasion and the second occasion may be as little as one hour. However, separating the occasions by one or several days is not only acceptable but may result in a better test because the passage of time reduces the opportunity for the interpreter to rely on memory of the original English. Approximately 15 minutes should be allocated for each interpreter for each test session.

The final preparation step is to select one or two individuals to score the test results. These individuals should have highly developed language skills in English and be able to discern the difference between substitution of words and distortion of meaning.

Data Collection. Data collection consists of administering the test to the candidates as summarized above and as described in more detail in Fundamentals of Court Interpretation (pp. 196–199). Test raters listen to each interpreter’s back-translated English version of the script and compare it to the original. The script will contain approximately 40 underlined scoring units that are used to determine the individual’s score. The resulting data sources are scoring sheets prepared for each interpreter by the test rater showing the number of scoring units on the back translation that match the meaning of the original English script.

Data Analysis and Report Preparation. After all of the tests are scored, a listing of their scores should be prepared. The analysis should then report the summary results in terms of percentiles, as shown in the following table.