Forensic Linguistics : An Overview of the Intersection and Interaction of Language and Law

Forensic Linguistics (FL) is a relatively new subfield within applied linguistics that studies the different intersections between language and the legal field, which is heavily linguistic by nature. In order to have a fair, legal and effective procedure, anyone involved in a legal process (lawyers, judges, police officers, members of a jury, etc.) benefits from possessing a certain awareness of linguistic principles. With this purpose, the expert testimony of a linguist could contribute to the understanding or recognition of possible interpretations or points of view that might have gone otherwise unnoticed. This article provides the general linguist with an overview of the broad field of FL and highlights the different ways the discipline can contribute to the criminal justice system. It presents a summary of some of the most well-known and discussed legal cases and outlines the intersections between applied linguistics (mainly pragmatics, discourse analysis, and sociolinguistics) and this emerging field in three interrelated areas: (1) language as the medium of communication between law enforcement authorities and suspects/witnesses or as the medium of legal argumentation in the courtroom, (2) language of the law (issues of intelligibility, interpretation and construction of legal language), and (3) crimes of language and linguistic evidence (use, validity, and reliability in the courtroom). Challenges and limitations of the field are also discussed.


Introduction
Law is codified in, and later mediated through, language.This means that without language, there is no law.However, the language of the law is very different from everyday language, which often results in disadvantages for the ordinary user.For this reason, the shades of meaning of legal language often have to be meticulously reviewed by forensic linguists, i.e., linguists who apply their knowledge of linguistic theory to the forensic context of the law: [J]ust as physicians are trained to see things in an X-ray that the average person with excellent vision cannot see.So linguists are trained to see and hear structures that are invisible to lay persons (Shuy, 1993, p. xvii).
The aim of this article is to provide an overview of the Forensic Linguistics (FL) field by highlighting the different ways this discipline has made an important contribution to the criminal justice system.The next sections present some of the most well-known and discussed legal cases and outline the intersections between FL and applied linguistics in three interrelated areas: (1) language as the medium of communication between law enforcement authorities and suspects/witnesses, (2) language of the law (issues of intelligibility, interpretation and construction of legal language), and (3) crimes of language and linguistic evidence (use, validity and reliability in the courtroom).

Linguistic Evidence
Forensic linguistic evidence is any type of text (spoken, signed or written) that can be used in a criminal investigation or as evidence in court.These texts include emergency calls, ransom notes, anonymous letters/calls, suicide letters, text messages, police records, confession statements, etc.Although the most well-known task that forensic linguists undertake as expert witnesses might be author identification, they also deal with other crimes of language, such as threats, bribes, conspiracy, or perjury, among others.

Author Identification
Forensic stylistics (or stylometry) is a technique that utilizes the linguistic analysis of writing style for the purpose of authorship identification (McMenamin, 2010, p. 487).
Based on the premise that there is individual variation in the use of language and that much of this variation is unconscious (and thus difficult to disguise), in order establish the linguistic fingerprint of a specific text, several methods of rigorous quantitative and qualitative analysis can be used.
As a general rule, the forensic linguist compares the text presented as evidence (questioned text) to other texts written/spoken by the presumed author (known writings) and determines the likelihood that the same (author identification) or different (authorship exclusion) person produced the questioned text.One famous example of forensic speaker identification is the Prinzivalli case (Labov & Harris, 1994).Prinzivalli was an employee of Pan American Airlines suspected of making telephone bomb threats to his employer in Los Angeles because: a) he was known to be an unhappy employee, and b) he was a New Yorker (the caller making the threat was believed to have a New York accent).Labov was given a tape with the original threat and another one with samples produced by the suspect.Based on the distribution of certain vowels, he determined that the person who issued the threat was actually from Eastern New England and not from New York City (Prinzivalli was acquitted).
Another occasion where dialectic variation gave information about the suspect's identity was the case of a ransom note analyzed by Shuy (2001).Although the suspect included misspellings of words such as dautter for daughter or kops for cops, his correct spelling of more difficult words such as precious, diaper or watching led Shuy to believe that the author of the note was trying to appear less educated than he was.However, what really helped determine the writer of the note was the uncommon use of devil strip, a term denoting the strip of grass between the sidewalk and the curb that is only used in the area surrounding Akron, Ohio.As there was only one educated man from Akron in the suspect list, the police did not take long to find other clues that also incriminated him.
Telephone text messages (SMSs) are another type of linguistic evidence that has increasingly been used in court.For example, in the case of Danielle Jones, a girl who disappeared in 2001, two messages that were sent from her phone to her uncle after her disappearance were crucial in identifying her possible abductor and killer.Coulthard (2008) was asked to compare the 65 texts that the girl had sent in the three days previous to her disappearance with those two last texts in question.Based on a series of linguistic choices absent from, or rare in, the Danielle corpus, he determined that "it was fairly likely" that she did not write them, which meant that someone else pretending to be her probably wrote them (her uncle, in this case) (Coulthard & Johnson, 2007, p. 202).
Another occasion in which a linguist could have been called to provide expert testimony was the case of four men convicted for having killed a paperboy.One of the four men, Patrick Molloy, made a confession admitting to killing the boy, but he later said that the police forced him to confess.The police claimed that they had a recording of an interview prior to the confession in which he also admitted to the same crime.Molloy denied that this confession ever took place and alleged that the police had fabricated it after the fact.Coulthard concluded that, based on the uniqueness principle, which states that the same person telling the same story at two different points in time would select "an overlapping but different set of lexicogrammatical choices" (2005, p. 49), both texts were too similar in vocabulary and phrasing to have been produced independently in two separate occasions: one was derived from the other or both were derived from a third one.This led to the conclusion that the written confession was indeed fabricated.

Crimes of Language
There are a variety of crimes that are committed through language alone, such as solicitation, conspiracy, bribery, perjury, defamation, threatening, and plagiarism, among others (Coulthard & Johnson, 2007;Fraser, 1998;Shuy, 1993).The main difference between these and other types of crimes where linguistic evidence is involved is that: [o]ne does not actually need to do harm to the person threatened, give the bribe, have the wife killed, or engage in sex with the prostitute.The language threat, offer, or solicitation is enough to constitute a crime (Shuy, 1993, p. 1).
In these cases, the issue generally is not determining authorship, but rather in identifying whether these crimes happened or not.As these crimes are, in essence, speech acts, it must be taken into account not only what has been said (locutionary act), but also what is meant (illocutionary act) and the effect it has on the listener (perlocutionary act).For example, although a threat, a promise or a warning might have the same effect on the listener (getting him to do something) it could only be considered a threat if there is intimidation involved (Fraser, 1998): As the information in Table 1 shows, when uttering a threat, the speaker is in full control: it is generated from his/her perspective, it is for his/her benefit and he/she is the one in control of the outcome.Solan and Tiersma (2005) mention the case of Hoffman, a young man who sent Ronald Reagan a letter that said "Ronnie, Listen Chump!Resign or You'll Get Your Brains Blown Out".The use of the passive voice (vs. the active: "I will blow your brains out") leaves the agent unspecified, which opens the possibility that this is a prediction or warning and that someone else could kill him (the same way a father tells his daughter "Do not play with the pan or you will get burned").In that case, if the outcome is not in the hands of the speaker, it could not be considered as a threat.However, the jury thought the opposite and Hoffman was convicted to four years in prison for threatening the president (U.S. v. Hoffman).
The concept of conversational contamination, developed by Shuy (1993), is also paramount in determining whether a person committed a crime of language or not.In a conversation where, for example, conspiracy or solicitation takes place, it is very important to look at who introduced the topic and how the other person responded.For example, agreeing with a plan by giving more details on how it is to be carried out is very different from responding "uh-huh" or remaining silent.Shuy presents the example of a Japanese engineer who was accused of consenting to buy internal product secrets from an undercover FBI agent.When analyzing the conversation, it can be seen that he responded "uh-huh" several times, but he never elaborated on the plan.Shuy concludes that "uh-huh" cannot mean agreement to the new information presented by the FBI agent, but instead it is a marker to signal that the addressee is listening or understanding (consistent with the Japanese culture of politeness).
Another example in which conversation analysis shed some light on the nature of the interaction was the case of an athlete accused of conspiracy to sell drugs (Solan & Tiersma, 2005).The linguist in charge of analyzing the evidence-a taped conversation between the dealer and the athlete-concluded that the use of the singular pronoun I by the dealer (vs. the plural we) was an indicator that he planned to act in an individual capacity and not with the athlete.
Finally, in another case mentioned by Solan and Tiersma, a man called Lawrence Gerenstein was accused of conspiring and soliciting to kill his wife.Even in the absence of a direct request for the other man to kill his wife, he incriminated himself by discussing different types of weapons that could be used to perpetrate the crime.

What can the (forensic) linguist do?
The duty of a forensic linguist, as in any case of forensic investigation, is to see what might not be evident to the naked eye: [L]inguists know what to listen for in a conversation.They listen for topic initiations, topic recycling, response strategies, interruption patterns, intonation markers, pause lengths, speech event structure, speech acts, inferencing, ambiguity resolution, transcript accuracy, and many other things.Scientific training enables linguists to categorize structures that are alike and to compare or contrast structures that are not.Linguists understand the significance of context in the search for meaning in a conversation and are unwilling to agree with interpretations wrenched from context by either the prosecution or the defense (Shuy, 1993, p. xviii).
Forensic linguists called to provide information on the authorship of a text must have a strong background in several areas of linguistic analysis: sociolinguistic variation, stylistics, phonetics, syntax, dialectology, discourse analysis, etc.In the same way, those who deal with crimes of language need to possess strong training in pragmatics, among other areas, in order to identify whether a crime or speech act was committed or not.After they reach their conclusions, these have to be transmitted in a simple, nontechnical manner to their audience.

Limitations
Although regular methods of forensic identification-such as DNA or fingerprinting-have accuracy rates approaching 95%, the many variables that play a role in forensic linguistic author identification can considerably affect its reliability.For example, in the cases analyzed for this section, there was often sufficient linguistic evidence to support an identification or exclusion hypothesis.However, the linguist will not always be fortunate enough to find a word or other piece of evidence that leads to a solid conclusion.For this reason, many linguistic analyses may be inconclusive, and consequently, not used in court (and not reported in the literature).
The use of pragmatics in forensic investigation is not without its limitations, either.For example, the difference between locutionary, illocutionary and perlocutionary acts is not always clear, which makes attribution of intentionality highly problematic.Additionally, relying on audiotapes without taking into account body language or trusting witnesses' accounts of what they think was said (vs. what was actually said) might render evidence inaccurate or inadmissible in court.

Language of the Law: Intelligibility and Interpretation of Legal Language
The ordinary person has to deal with a number of legal documents on a regular basis (real estate, end-user license agreements, Medicare forms, military disability forms, lease agreements, contracts, statutes, deeds, wills, and all kind of policies, among others).In order to be explicit and avoid the ambiguities present in everyday speech, all of these legal documents need to make use of: […] a register marked by redundancy (e.g.repetition of full noun phrases instead of pronouns), technical terms with precise definitions (to reduce ambiguity), complex sentence structures (through which information otherwise available from non-linguistic context is given expression) and formulaic expressions (to assure consistency across cases) (Hall, Smith & Wicaksono, 2011, p. 279).
The downside of this degree of explicitness is that, more often than not, legal language ends up being "extremely user-unfriendly for its non-expert consumers" (p.279).As a result, forensic linguists have carried out a considerable amount of research that shows, contrary to popular belief, legal language used in everyday transactions is only accessible to a reduced percentage of the population.
In his analyses of pension plan documents and credit card notices, Stygall (2010) noted that these lengthy, complex texts "present excellent examples of legal language unintelligible to most people" (pp.51-52).The problem, according to that study, is that although the literacy rate of the US population is 99%, the level of literacy required to understand the nuances of this type of texts is only attained by 3-4% of adults nationwide (p.59), which leaves the rest in a disturbingly disadvantaged position.
Although judges often rely on dictionaries to provide them with the official definition of words found in legislation, forensic linguists prefer to base their definitions on the observation of actual usage of those words (Goddard, 1996).The reason is the number of common, everyday words that, when used as legal jargon, have very different meanings (legal homonyms).Stratman and Dahl (1996, p. 212), for example, mentioned the case of a man who had a restraining order lodged against him (State v. Hardy).When he slipped an apology letter under his partner's door, he was accused of having "molest[ed], interfere[ed] or menac[ed]" her.Clearly, the issue here is that the defendant was unaware of the legal definition of those terms and did not think that an innocent apology might actually be considered by the law as an act of molestation or menacing.This case led the researchers to conduct a study in which they provided ordinary readers with a restraining order, four different scenarios, and asked them what scenarios violated the order.Researchers concluded that, after reading the restraining orders, most readers still failed to determine whether there had been a violation or not (they said that the restraining orders had not been violated when, in fact, it was).
Several pieces of research have analyzed the complexity of language in jury instructions and how faulty comprehension can result in fatal consequences.Forensic linguist Levi reported a death penalty case (U.S. ex rel.Free v. McGinnis) in which Levi herself was called to provide expert testimony and demonstrate that the majority of the members of a jury could have misunderstood central points of law that were "essential" to apply in this case (1993, p. 23).On the same note, Saxton (1998) found that, although 97% of a group of jurors claimed to have understood the instructions given to them before the trial, 40% of them still believed (after the trial) that having a defendant charged with a crime was strong evidence that they had committed the crime (p.96).Additionally, and quite ironically, looking up confusing words in the dictionary is strictly prohibited for jurors.As a consequence, those who have tried have been accused of misconduct for consulting an outside source (Tiersma, 1999), which strongly highlights the need to revisit accessibility of legal language for the average citizen.

What can the (forensic) linguist do?
There is no doubt that each discipline needs its own jargon to facilitate communication within the profession.However, it is also undeniable that people have the right to understand the laws that pertain to them.If comprehension of legal language is often impaired by "linguistic features that are not specifically legal" (Tiersma, 1999, p. 203), there is no reason why that language cannot indeed be modified in order to be made more accessible for its users (as the Plain Language Association International (PLAIN) has demanded since 1993).
As a starting point, linguists can work with document designers and attorneys to conduct usability testing on a representative (in terms of literacy level) sample of the target audience (Stygall, 2010, p. 64).Then, the linguist could suggest different phrasings while the attorneys make sure that the meaning is still the intended one.As suggested by Tiersma (1999), special attention should be paid to the following: In most of these cases, a solution that does not compromise the legal ramifications of the language can be found (as has been done with relative success in California (Tiersma, 2010).The explanation or defining of unusual or confusing terms, the use of verb forms instead of their nominalizations, or the avoidance of the passive, multiple negatives or excessive clause-embedding can improve substantially the understandability index of many legal texts.

Limitations
While it may seem straightforward to prove that a text is dense, complex or difficult to understand for the average layperson, it is more difficult to demonstrate whether a particular person (mis)understood it or not.For this reason, although forensic linguists may feel limited when it comes to using this type of evidence in court, they can indeed work on making these documents more understandable and accessible.

Language During Legal Procedures and Courtroom Discourse
In addition to written laws and legal documents, language is also the medium of communication between law enforcement authorities and suspects/witnesses, and also the medium of legal argumentation in the courtroom.There has been a great deal of research on: a) interactions between police officers and suspects (before, during and after arrest); b) vulnerable populations as witnesses or defendants; and c) faulty court interpretation.

Interactions Between Police Officers and Suspects and Courtroom Discourse
In order to understand the intricacies of any interaction between interlocutors, Paul Grice's Cooperative Principle (1989) and the theory of speech acts is helpful (Austin, 1975).
According to the Cooperative Principle (Grice, 1989), there are four maxims that need to be respected for efficient communication: 1. Quality: do not say what you believe to be false or that for which you lack adequate evidence.
2. Quantity: make your contribution neither more nor less than is required.

4.
Manner: avoid obscurity of expression and ambiguity; be brief and orderly.
When the speaker flouts a maxim (without the apparent intention to mislead), the hearer then tries to reconcile what the speaker said with the assumption that the speaker is cooperating.It is precisely this situation that leads the hearer to infer what the speaker means (conversational implicatures): He has said that p; there is no reason to suppose that he is not observing the maxims, or at least the Cooperative Principle; he could not be doing this unless he thought that q; he knows (and knows that I know that he knows) that I can see the supposition that he thinks that q is required; he has done nothing to stop me thinking that q; he intends me to think, or at least willing to allow me to think, that q; and so he has implicated that q (Grice, 1989, p. 31).
For example, in the following interaction: According to the theory of speech acts, although the police officer is only asking about the trunk working properly (locutionary act), the driver of the car interprets the speaker's intent as an order (illocutionary act) and he opens the trunk (perlocutionary act).Since illegally procured evidence (without a warrant or consent) cannot be admitted in court, it is important to determine whether the speech act was indeed a question (to which a response like "yes, it does open" would have sufficed), or a command by a person in a power position.In this particular case, the Supreme Court found that the suspect had given voluntary consent to the search and the stolen checks found in the trunk were admitted as evidence.
Another issue that has received a great deal of attention by forensic linguists is the administration of the Miranda warning, which is a set of rights that are read to a suspect upon their arrest and before interrogation.Although its utterance is assumed to constitute a performative speech act at the locutionary level (the warning cannot happen unless it is actually said), the perlocutionary effect it has on the person being arrested (whether they understand their rights or not and whether they say they do) is of critical importance.In other words, if the reading of these rights informs and/or reminds the arrestee that they have the right to remain silent, why do approximately 80% of them still answer police questions before they hire a lawyer (Ainsworth, 2010, p. 111)?One of the possible answers to this question might be that, as we have seen with other types of legal language, the Miranda warning: violates most of the norms of spoken English and would be challenging to parse even in formal written English and it would be a difficult utterance to understand fully even in the best of circumstances (p.115).
Although this warning informs the arrestees of their right to "the presence of an attorney during any questioning", this right has to be exercised by the suspect through another performative speech act (requesting a lawyer).Ainsworth (2010) noted that the use of an interrogative such as "Could I call my lawyer?" or "Do you mind if I have my lawyer with me?" can prevent the arrestee from getting a lawyer, as an interrogative does not necessarily have to be interpreted as a request.Again, this is paradoxical, since we have already seen that interrogatives can and are indeed interpreted as commands ("Does the trunk open?").The only difference here is that, in this interaction, the one asking the question is not the person in power.
In the same way that illegally procured evidence cannot be used in court, a confession that is found to be involuntary or coerced cannot be admitted into evidence either.In his analysis of police-suspect interactions, Shuy (1997) stressed the coercive nature of questions to suspects and the difficulty they have remaining silent in the interrogation room: The difference between the examples above is that, while (a) is a yes/no question, (b) is a question that presupposes the answer is yes, and (c) does not even question whether there was intercourse because it is already taken for granted (for a complete categorization of questions in court, see Harris, 1984).

Vulnerable Populations as Witnesses, Suspects or Defendants
Children, juveniles and people with communication, cognitive or other mental disorders are especially susceptible to waiving their rights, changing their statements, making false confessions, or accepting plea agreements (Cloud et al., 2002;Drizin & Leo, 2004;Redlich, 2007): The words of the Miranda warnings themselves are 'meaningless' to mentally retarded suspects, who simply do not understand them [….]Disabled suspects' waivers of the rights described in the Miranda warnings are 'voluntary, knowing, and intelligent' only if we are willing to manipulate and distort the very meaning of these terms.
[….]When subjected to the pressures of custodial interrogation, mentally retarded people are more likely than others to confess to crimes they did not commit (Cloud et al., 2002, p. 591).
In their analysis of 125 cases of proven interrogationinduced false confessions, Drizin and Leo (2004) found that interrogators' manipulative methods of psychological interrogation often lead an innocent person in a vulnerable group to make a false confession.Redlich (2007) went further and mentioned that even though 70% to 100% of juveniles involved in judicial proceedings have "diagnosable disorders", many are still questioned as healthy adults, which makes the likelihood of false confessions even more alarming.
Minors and people with mental disorders are not only vulnerable populations when they are suspects or defendants in a case, but also when they are witnesses or, more importantly, victims of abuse.Aldridge (2010), Eades (2010), and Ellison (2002), for example, each explored the interaction between suspects/defendants and their interviewers and concluded that, even though some measures to protect them are usually put in place, cross-examination is littered with linguistic devices and interrogative techniques that disadvantage those with language capacity limited by immaturity or disability (Ellison, 2002, p. 10).
Lastly, in the group of vulnerable populations are found people whose cultural or linguistic background prevents them from being accurately protected or represented in the adversarial system.Eades (1994) examined the communicative disadvantages that Aboriginal Australians experienced in police interviews and courtroom interactions due to cultural and linguistic differences between them and other non-Aboriginal English speakers.This population, who values indirectness, for example, is unaccustomed to direct questioning (either-or or whquestions) and eye-contact.Additionally, they often display the "conversational pattern of agreeing with whatever is being asked, even if the speaker does not understand the question" (p.244) and value silence positively, which is often interpreted by Western societies as "evasion, ignorance, confusion, or even guilt" (p.243).As a consequence, and as in the other cases seen in this section, they are more vulnerable to self-incrimination and their testimony has been judged to be insufficient, inadequate, unreliable, or simply invalid (with all the ramifications that this entails).

Court Interpretation
Depending on the case, hard-of-hearing persons and speakers with limited English proficiency (LEP) sitting on the accused bench have the right to the assistance of an interpreter (in some states, witnesses and jurors do as well).What is essential to note in these cases is that, when the services of a court interpreter are used, it is their interpreted words that are transcribed and, thus, become the official record.Likewise, members of the jury are instructed to take only the interpreted words into consideration, even when or if they understand the original language.Although these interpreters have to uphold high standards of professionalism and ethical conduct, it is not uncommon to find interpretation issues that might, in the long run, invalidate a statement.
In her ethnographic observation of interpreted judicial proceedings, Berk-Seligson (2002) revealed not only that interpreting is an inherently highly complicated process, but also that the mismatch between an interpreter's role and what others (court personnel and clients) perceive it to be might result in catastrophic consequences.Eades (2010) pointed out five common arguments against the use of interpreters in the courtroom: 1) Will the interpreter modify an answer (for or against the defendant's case)?
2) Will the interpreter help the witness (or does the defendant expect the interpreter to help them)?
3) Will the use of an interpreter give extra time for the witness to prepare an answer (if the defendant understands English partially)?
4) Will it be harder to gauge the credibility of the witness?5) Will the interpreter provide a 'buffer' between lawyer and witness?
One of the major obstacles, according to Berk-Seligson (2002), is that the pragmatic and syntactic content of interpreted testimony does not usually receive much attention: [A]ll interpreters tended to omit those seemingly unimportant features of speech style that can impinge on the evaluation of witnesses' speech by those judging them.The interpreters' stylistically inaccurate renditions can therefore potentially alter the outcome of the case.The results of this study show that court interpreters are not consistently interpreting 'truly and faithfully' and that linguistic training is required for interpreters to become aware of the importance of style in the courtroom (Hale, 2002, p. 44).
For example, while Laster (1990) observed that by adding sir and madam markers of politeness, jurors tended to consider the witness more competent and convincing than when they were not.McMenamin (2002) found that making the English version more vulgar than the original (by adding a word like fucking to a phrase that does not contain it) "reflects negatively on the speaker (defendant)" (p.249).
However, interpretation does not exclusively take place in the courtroom.The reading of the Miranda rights and the interviews that take place between the police and the suspects are all necessarily mediated through language.As Eades (2010) or Berk-Seligson (2002) noted, those interviews, which can also become part of the evidence, more often than not, take place without a (certified) interpreter.Moreover, in the cases where there is someone willing to mediate between the suspect and the police, whether a friend or the police officer him or herself end up taking this interpreter role, which brings with it a whole array of implications (conflict of interests and faulty interpretation, among others).Deaf suspects are especially vulnerable in this situation, since many times their condition is not immediately recognized and, as a consequence, their silence can be interpreted as resistance or lack of cooperation by the police (McKee, 2001).

What can the (forensic) linguist do?
Although judges and jurors are the ones that ultimately have to decide whether a speech act has or has not taken place, linguists can be "extremely helpful in analyzing the discursive structure and linguistic content of interrogations" (Ainsworth, 2010, p. 122).As has already been seen, linguists "are trained to see and hear structures that are invisible to lay persons" (Shuy, 1993, p. xvii) and their expert testimony could be crucial in a trial.For this purpose, it has been repeatedly suggested that, as a precautionary measure, interviews between law enforcement agents and suspects be recorded at all stages of the process ( Another way linguists can help is by suggesting ways in which the police can make use of less coercive language during interactions (Eades, 1994;Shuy, 1997).In this way, the police could avoid the invalidation of statements or confessions due to improper questioning (exclusionary rule).In the case of children and people with communication disabilities, the linguist's task would be to provide specialized training for interviewers (detectives, lawyers…) so they are able to evaluate and recognize the linguistic characteristics of a suspect and act accordingly.
When deciding whether a defendant needs an interpreter or not, the presiding judicial officer (the judge or magistrate) makes use of questions such as "How long have you been in this country?",or "Where did you learn English?".However, all linguists would agree that these questions have nothing to do with the complexity or type of language that will be used in the courtroom (Eades, 2010, p. 66).Linguists, then, can assess the linguistic competence and performance of non-natives and decide whether they have basic interpersonal communication skills, or if they can, in fact, detect "subtle differences in word choice, tricky manipulation of presuppositions [or] three questions in one" (p.67).In order to give the defendants the same treatment as a native-speaker, the linguist can also ensure that the interpreter is not giving the defendant an advantage (or disadvantage) due to unprofessional practices.Again, taping the actual conversation (that includes the original language) would allow linguists (and certified interpreters) to make sure that the services provided by an interpreter are fair for everyone involved.In these cases, even if the original interaction is not part of the official record, it could be presented as evidence later in court.

Limitations
Grice's Cooperative Principle presents two main problems: it assumes an ideal speech situation, and is limited to the locutionary speech act.As a consequence, it also presents some limitations: Why should it be assumed that anyone involved in a crime or criminal investigation would want to cooperate?What happens when both parties in a conversation have conflicting goals?Human beings lie, exaggerate, and make use of politeness, hedges, irony, sarcasm, figures of speech, etc., which all makes the assumption that they adhere to the cooperation principle more unrealistic than it seems.
In the case of court interpretation, the limitations are very clear: as long as the original testimony is not recorded in order to be able to compare it with the interpreted version, linguists cannot do much more than speculate on the quality of the interpretation.Additionally, an important limitation on the results regarding the effect of (faulty) interpretation on jury decisions is that these studies used mock trials that were especially manipulated for the research at hand, which means that the effects of faulty interpretation on real trials has not, or perhaps cannot, be tested.

What the Forensic Linguist Doesn't Do
Contrary to popular belief, a forensic linguist's duty is not to perform text analysis with the objective of discovering the writer's intent or describing his/her psychological profile or state (McMenamin, 2002;Solan, 1998).For example, one book that has received a great deal of criticism is Author Unknown: On the Trail of Anonymous (Foster, 2000) because the analysis presented in it is "purely speculative" (Chaski, 2001, p. 3) Other areas that fall outside of forensic linguistics are, among others, graphology, handwriting analysis or document examination.Graphology has been repeatedly questioned for its pseudo-scientific nature.Handwriting analysis and document examination, on the other hand, although helpful in shedding light on criminal cases, bases their research on scientific theory other than linguistics, such as chemistry, computer science, or physics, among others.
Lastly, it is vital to note that what the forensic linguist analyzes is language, not guilt or innocence (Shuy, 1993, p. xxi).In other words, although the expert testimony of a linguist might be helpful in a case, it is the prosecution's burden to prove the accused's guilt beyond a reasonable doubt.Legal decisions are for judges to make and the forensic linguist's testimony is just one piece in the puzzle.

Conclusion
Law is inconceivable without language: without language there would be no laws, no trials, and in some cases, no evidence.Although the field of forensic linguistics is still in its infancy, its contributions to the criminal justice system are nonetheless significant.
This article has provided the reader with an overview of the intersections between forensic linguistics and other areas of applied linguistics (mainly sociolinguistics, pragmatics, and discourse analysis) in three interrelated areas: linguistic evidence, language and the law, and language during legal procedures and courtroom discourse.It has shown how applied linguistics can contribute, not only to a more understandable codification of the law, but also to the maintenance of the rights of linguistically vulnerable populations.
Like any other emerging discipline, forensic linguistics presents numerous limitations that should not be overlooked.First, linguistic evidence alone is often not enough to convict or exonerate a person, although it may contribute to a larger body of evidence.Second, while linguistic analysis is becoming increasingly accurate with the aid of technology, it is still not 100% infallible and it is still subject to interpretation.Finally, the impossibility of experimental manipulation in the courtroom makes some assumptions about what happens there difficult to demonstrate.While this may be the case, what needs to be clear is that when linguists serve as expert witnesses, their aim is mainly to assist the jury in understanding the evidence by shedding light on issues that might not be obvious otherwise.
a) technical vocabulary: legal homonyms, unfamiliar legal terms…; b) archaic, formal and unusual words; c) impersonal constructions; d) nominalizations and passives; e) modal verbs; f) multiple negation; g) long and complex sentences.