Keywords

Origin of Forensic Linguistics

Forensic Linguistics is a relatively new field, and the term was first coined by linguist Jan Svartvik when he wrote The Evans Statement in 1968. The book examined a murder case that took place in November 1949, in which Timothy Evans was arrested for the murder of his wife and infant daughter. His trial began in January 1950. The prosecution was able to obtain his written confession during the initial investigation. Based on his written confession as evidence, Evans received a death sentence and was put to death in March of the same year. Three years after Evans’s execution, John Christine was arrested for the murder of four women including his wife. During his trial, Christine confessed that he murdered Evans’s wife, which brought significant controversies and debates over Evans’s wrongful conviction and eventual execution.

Evans’s bereaved family requested to Svartivk for a linguistic analysis of Evan’s confession. Svartvik made a corpus analysis of the original written statement of Evans’ confession and found two distinctly contrasted grammatical styles: (1) an educated style, possibly coached by an investigating officer, and (2) a casual writing style reflected by the defendant himself. He concluded that the authenticity of Evans’ written confession was very questionable, suggesting that the content of the statement contained the sign of significant external influence, rather than his own.

Another pioneering analysis in forensic linguistics comes from the Bentley case involving the attempted burglary and murder of a police officer in 1953, for which nineteen-year-old Derek Bentley was convicted and later executed. Although the actual murder was carried out by sixteen-year-old Chris Craig, he was not given the death penalty because of his age at the time of arrest. It was stated that Bentley’s IQ was far below the average of his peers and he was also functionary illiterate. Recognizing that this case involved complicity in a burglary attempt, forensic linguist Malcolm Coulthard analyzed Bentley’s confession statement and argued that Bentley personally did not make a confession as noted in the statement to the police. Rather, using a corpus analysis of the term “then” in the confession statement, he found that large parts of Bentley’s writings reflected, and were composed of, words and language deliberately used by investigating officers assigned to the case.Footnote 1

This is how forensic linguistics, the application of principles and methods of linguistic analysis to the language of legal proceedings and documents, has become an established area in the interdisciplinary area of law and language in English speaking countries.

Forensic Linguistics in Japan

Forensic linguistics in Japan stayed dormant for nearly ten years after the first publication of the forensic linguistic paper “Hou-gengogaku no Taidou” (Embryonic Movements of Forensic Linguistics) in 1998.Footnote 2 However, with the preparation for the lay judge system beginning in 2005, there has been a growing interest among legal experts in making courtroom language clearer for lay judges. This has opened the way for recognition of forensic linguistic studies in general.

I will discuss a revised version of the first Japanese expert opinion of forensic linguistics, which was submitted to the Tokyo High Court in March of 2011. I presented an analysis of the testimony in a criminal case involving a charge of complicity.Footnote 3

Overview of the First Japanese Case

The following is an overview of the criminal case that was examined. A male F was found dead in a car that was submerged in an irrigation reservoir in Gunma Prefecture, Japan in July 2009. Five acquaintances of the victim (A, B, C, D and E) were arrested on charges of causing bodily harm resulting in death and disposing of a dead body. Three of them (A, B and C), who admitted to carrying out the crime, were given sentences of eight, nine and ten years, respectively. The other two defendants (D and E), however, denied any involvement in the crime. Defendant D had his indictment suspended, but Defendant E was charged as a joint accomplice in the conspiracy. Although Defendant E pleaded not guilty to the crime, she was sentenced to nine years of imprisonment by the district court in November 2010. She appealed to the Tokyo High Court, which dismissed the appeal in March 2011. The defendant then appealed to the Japanese Supreme Court but withdrew the appeal in October of 2011.

Before the Trial

The main issue in this case was whether or not Defendant E conspired with the three other defendants (A, B and C who were previously convicted of murder) to assault the victim. At the pre-trial conference, the defense lawyer made a statement that questioned the credibility of the three witnesses’ statements against the defendant. This provided the prosecutors with an opportunity to anticipate the defendant’s main trial strategy, thus prompting a series of visits to all three witnesses who were serving their prison sentences whereby each was interviewed ten times before the trial’s commencement. During the subsequent trial, all three witnesses A, B and C proceeded to give incriminating statements against Defendant E; yet, the content of their statements was different from that of the previous testimonies they gave in their own trials six months earlier.

Witness Preparation for the Prosecution

I focused on one of three witnesses (Witness B) and examined his testimony using linguistic analysis. This witness previously had an intimate relationship with Defendant E. During an interview with the witness, the prosecutor disclosed to him that Defendant E tried to intoxicate him with a stimulant drug in the kitchen with the intent to arouse him to attack Victim F. Prior to his testimony in court, then, it was clear that this witness had probable motive for testifying against the defendant. At the trial, Witness B clearly showed his anger at the defendant when he came into the courtroom to take the witness stand. But before analyzing the content of his testimony and examining the signs and traces of possible witness preparation by the prosecutor, I briefly review the method of forensic linguistics and how this investigative technique can be useful in the analysis of witness testimony.

Forensic Linguistic Analysis of Witness B’s Statement

Professional Language Features

Japanese police officers and prosecutors also import similar features of their professional language into the official records of suspects’ statements. They include the use of demonstrative pronouns (sono (its, the)) and the past progressive form, all of which aid in giving statements greater precision. First, I wish to show how these syntactic features are reflected in a suspect’s statement recorded by an investigating officer. We also cite examples from a handbook commonly used by investigating officers (Kajiki et al. (2006) Shin Sosa Shorui Zenshu Torishirabe (A New Complete Work of Investigating Documents: Interrogation) to facilitate my discussion and analysis. This is a standard textbook that teaches investigative officers about the techniques of suspect interrogation and the recording of verbal testimonies. I will then show that many traces of professional language used by the investigating officer appeared in Witness B’s testimony.Footnote 4

Interrogation Handbook Examples

1. SONO (THE): Demonstrative Pronoun

Constituents of a sentence are frequently omitted in the Japanese language, and such omissions are much more salient in spoken language, especially when the speaker believes that the hearer knows or can understand the context of a situation, as shown in the following examples.

Anata wa ashita eiga ni ikimasu ka? Anata wa ashita eiga ni ikimasu ka?

(Are you going to the movie tomorrow? Are you going to the movie tomorrow?)

The sentences below are taken from the handbook (Kajiki et al. 2006: 78). The words “my” of “my internet” and “her” are omitted because these demonstrative pronouns are easily recoverable from the context. On the contrary, the article “the” from “URL” or “picture” is not deleted because it clarifies “the URL” and “the picture” in question. This is how the handbook educates investigating officers not to omit the demonstrative pronouns relating to the key notions.

As I would make Mayu’s picture open to (my) internet homepage and send (her) the URL and cancel-key by mail, I was telling Mayu to delete the picture by herself…

2. –TE IMASHITA (WAS DOING): Past Progressive Form

The past progressive form frequently appears in a suspect’s recorded statement. This is because investigative officers or prosecutors are required to describe the crime scene vividly enough so that the judges can use the descriptions to recreate an accurate depiction of the crime and thus make factually correct decisions on the case.

I was telling lies.

Witness B’s Testimony

This section examines the different features of professional language and its usage that appeared in Witness B’s testimony. Witness B was originally convicted in the complicity case involving the same crime and was called to testify as a prosecution witness against Defendant E in her trial.

In Testimony (1) below, the prosecution witness’s statement contains many of the same linguistic and syntactic features used by professional investigative officers, including the demonstrative noun “sono (its)” and the pronoun “E,” that is, the defendant’s true name. If the witness had used ordinary spoken language, his testimony would cohere more with Example (2), in which both noun phrases (recoverable from the context) and formal expressions would be eliminated.

  1. (1)

    E got angry in regard to (the fact that) that son (her son) was beaten, called the other party’s parent and (his) son, and called out to E’s house to do the same to them.

  2. (2)

    E got angry in regard to (the fact that) that son (her son) was beaten, called the other party’s parent and (his) son, and called out to E’s house to do the same to them.

    Now look at the past progressive form ‘kuwaesasete-imashita’ (was causing or inflicting) in the sentence (3). This usage of the past progressive form by the witness describes the crime scene where Defendant E ordered A to physically assault F. These examples reflect formal linguistic phrases used by Japanese investigative officers.

  3. (3)

    E who got angry by it was using A to inflict violence on F.

Prosecutor’s Examples

Many instances of professional language from the interrogation handbook were found in the testimony given by Prosecution Witness B. Similar instances (sono, te-ita) can also be found in both the prosecutor’s opening and closing statements.

Examples (4) and (5) were taken from the prosecutor’s opening and closing statements, respectively. The word “Sono” is used in both instances in order to make a specific reference to the defendant’s daughter and the victim’s body.

  1. (4)

    the defendant’s daughter G, her boyfriend H

  2. (5)

    I have nothing to do with the disposition of the corpus.

    The past progressive form is also found in both (6) in the prosecutor’s opening statement and (7) in the final statement. Both examples refer to a description of on-going events.

  3. (6)

    A was watching the condition of Mr. F.

  4. (7)

    was talking with ~.

It is clear that these two features are usually found in the professional language of the prosecutors and/or investigative officers. Now I would like to demonstrate that they are in fact not a register of the witness himself but that of the prosecutors or investigating officers. I will show this by first tallying the number of occurrences of these features in five pieces of discourse: (1) a witness’s letter to the defendant’s daughter’s boyfriend; (2) the testimony of the prosecution witness in court; (3) eleven samples of the suspect’s statement taken from the handbook; (4) the prosecutor’s opening statement, and (5) the prosecutor’s closing statement.

None of these features (sono and te-imashita) were found in the witness’s personal letter. On the other hand, these linguistic features are found in the suspect’s testimony in court, as well as sample written statements from the handbook. The high frequency of these features in the suspect’s testimony and written statement suggests possible witness preparation or prosecution coaching prior to his testimony in court. The witness’s use of particular language patterns also parallels the language use of the prosecutor in his testimony (Table 1).

Table 1 Frequency comparison of statement and testimony
Full size table

Written Language Features

Written language is more complex than spoken language. Academic writing, which usually focuses on a specific theme contributing to the main line of argument without digressions, includes linguistic characteristics of noun-based phrases, subordinate clauses or embeddings, complement clauses, sequences of prepositional phrases, participles, passive verbs, lexical density, lexical complexity, nominalization, and attributive adjectives.Footnote 5 Among these characteristics, we discuss noun-based phrases below.

Location of Modifiers

One example that was found in the examination of written language is a modification of a noun phrase: a relative clause (noun + post modifier). A relative clause is used to provide additional information without the inclusion of another sentence. Nonetheless, unlike English, Japanese does not require the use of relative pronouns.

For example, the relative clause in witness testimony (8) directly modifies the noun phrase. The clause, (Sore ni hara wo tateta), comes before noun phrase (E) and is predominantly used in written language. In order to fully understand the meaning of this sentence, one must find the actor of the sentence (E), which comes after its modifier (“who got angry with it” (sore ni hara wo tateta)). The use of relative clause requires the process of reading back the whole sentence, which is suitable for written language, but not for spoken language. Thus the use of the relative clause in a normal conversation is extremely rare. In examining Witness B’s testimony, use of this relative clause in his speech was deemed very unusual and may imply the possibility of witness preparation conducted by a prosecutor during the ten pre-trial interviews in prison. In a normal spoken expression, it is more common and natural to express this with the use of a compound sentence as shown in (9).

  1. (8)

    E who got angry with it was using A to inflict serious violence on F.

  2. (9)

    E got angry with it, and he was using A to inflict serious violence on F.

Repetition

Coulthard suggested that it is rare for individuals to remember verbatim in its exact form or words in terms of what they themselves said, as well as what other people stated with respect to some past event. It is also a misconception that what people remember is the gist of what was in fact said and expressed.Footnote 6 This means that slightly different accounts are usually given at each retelling.

The witness recounted in court on November 10, 2010 about what had occurred from the Third to the Fourth of July in 2009. The witness, however, retold the same event using exactly same words and phrases, as shown in testimonies (10)–(11). Also, please note that Testimony (1) had two usages of “yobi” (call). This indicates that the witness retold the same event using the same word used by the prosecutor who also interviewed the witness in prison on repeated occasions prior to the trial.

  1. (10)

    It was because I was called out by E.

  2. (11)

    I was called out by telephone from E.

  3. (12)

    I was called out by Ms. E.

Characteristics of Witness B’s Testimony

Prosecution witness B gave his response to a direct question, using the prosecutors’ or investigating officers’ register, including the frequent use of sono, te-imashita, as well as written language features and repetitive expressions, all of which are not normally found in ordinary people’s verbal expression. The witness’s personal letter also had shown no indication of these characteristics or linguistic traits. Hence, it is possible that the prosecutor’s repeated contacts and detailed interviews with the witness influenced the way he responded to the question about the case.

The Japanese criminal justice system does not have a comparable process of discovery procedure like the one in the United States, and the prosecutors are not required to disclose the list of all of the evidence that they have collected. As a result, the defense lawyers must compile a specific list of documents or evidence needed to prepare for their defense strategies. During the course of a pre-trial conference, the defense lawyer makes a request for the disclosure of specific information, including material or forensic evidence, depositions, statements made during interrogation, or any other documents pertaining to the case. The defense’s specific request for materials or evidence often gives prosecutors a fairly good understanding of the defense’s likely strategy. The prosecution is then in a privileged position to formulate its own counter-defense plan prior to trial.

In the present complicity case, since the defense raised the question of the credibility of accomplices’ statements on Defendant E and requested relevant documents or evidence, the prosecutors then may have decided to conduct comprehensive interviews of the former accomplices in order to prepare them for their upcoming testimony in court. Indeed the prosecution conducted a total of ten interviews with all of the accomplices in a prison facility prior to the trial. If this was in fact the case, then the prosecutors’ trial strategy raises serious ethical questions regarding excessive witness preparation and even possible witness coaching.

I would like here to explain that witness coaching is illegal but preparation for examination of a witness is legal in Japan, as stipulated in Article 1913 of Rule of Criminal Procedure. This is because preparation for examination of a witness is nothing more than a method such as ascertaining the facts from the witness. However, unlike investigators’ interrogations of suspects, preparation for examination of a witness is not the target of audio and video recordings. This means that no one can verify the actual situation of preparation for examination of a witness as appropriate. It is therefore possible that preparation for examination of a witness is all but witness coaching behind closed doors, as reported in the article of January 5, 2014 in the Asahi Newspaper.

Conclusion

In this paper I have introduced an expert opinion of forensic linguistics, in which I identified the characteristics of prosecutor’s language which would appear in the prosecution witness’s answer during direct examination. By using qualitative and quantitative analysis, such as co-occurrence and concordance of words, I performed a linguistic comparison of the language that was used in a witness’s answer against that of five relevant documents. This included (1) a prosecutor’s opening statement, (2) a prosecutor’s final statement, (3) eleven samples of suspect’s statements from the handbook for investigating officers, and (4) two personal letters of the witness. The results of this analysis indicate that the witness’s responses had the features of the prosecutor’s written language. Therefore, I argued that the prosecutor’s ten meetings with the witness immediately before trial may possibly have influenced not only the witness’s language but also the content of the testimony itself.

The application of linguistic analysis to the language of legal proceedings and documents is now an established area in Anglo-American courtrooms, but not yet in institutions in Japan, Korea and Scandinavian countries. To enhance awareness among legal professionals, it is clearly essential for linguists to publish papers regarding the usefulness of such linguistic analysis. Mizuno is one of the Japanese linguists who actively publishes papers on court interpreting. In the next chapter Mizuno discusses the quality of court interpreting in lay judge trials using a linguistic analysis.