Ling Midterm

theofordcomme's version from 2016-03-17 13:51

Section 1

Question Answer
What is the difference between the Daubert Standard for scientific evidence and the Frye Standard for scientific evidence? Which is used in US federal courts? State courts?Frye: court evaluating whether to admit lie detector evidence, still used in some state courts, when a scientific discovery crosses the line between the experimental and demonstrable stages is difficult to define so evidential force of principle must be recognized. the thing from which the deduction is made must be sufficiently established to have gained general acceptance in the particular field in which it belongs.
Daubert: replaced Frye in federal courts, trial courts need to ensure that expert testimony is admitted only when it is based on valid and reliable techniques (has theory been tested? has it been peer reviewed or published? known rate of error? generally accepted in the scientific community?) (comes from a supreme court ruling)
What are five characteristics of known suicide notes?apologies, pleas of forgiveness, personal names and references, expressions of love, explanation, instructions for the loved ones post death, etc.
What are the advantages and disadvantages of the aural-perception approach versus using computational methods in forensic phonetics?aural-perception eliminates mechanical error but allows for human error. it can catch things that computational methods wouldn't. but, computational methods can calculate rates of usage provide other mathematical and testable data
Explain what occurred in Pierson v. Post and explain the importance of the trial.Post is chasing the fox, Pierson kills his fox, Post is mad because since he was chasing it, it makes it his fox. He takes it up in court. the Court finds in favor of Post. Then later, it changes to find in favor of Pierson on the basis that there is no precedent for this type of case so it is hard to make a conclusion, but pursuit isn't enough, possession needs to somehow be proven. And no one can prove possession of a wild animal. The dissent argues that the rules of game hunting should have determined the case. The dissent identifies with the social status of Pierson while the majority opinion identifies with the social status of Post. Also this all quotes Latin frequently which only the legal folk can understand.
Name and explain a case which involves authorship identification? What are different ways we can determine individual authorship in a corpus text?(McMenamin) State v. Brown deals with authorship, it is the beginning of the stylistic approach, in Brown they are trying to determine if the typewritten diary found in the Brown household belongs to Mrs. Brown and, therefore, details the criminal activities of Mr. Brown. The experts developed a corpus of Mrs. Brown's known works, of which there were over 100. These included her personal letters to Mr. Brown and business letters to other recipients. All but 1 of the 125 letters in the corpus were typewritten and single spaced. This corpus was used to compare to the 53 entries in the diary. Format, spelling, organization, punctuation, sentence structure, and other factors were considered when comparing the questioned writings to the corpus. The topics used to compare were: format of page arrangement, punctuation, misspellings, profanity, expressions of time, expressions of location, expressions of cleanliness, attention to jewelry, characteristic words and phrases, expression of emotions, alternative forms, word-compounding, syntactic choices, exclamations, and "I" phrases.The formatting was almost identical, as was punctuation, misspellings, profanity, etc. The high level of stylistic similarity led the researcher to decree that the diary was in fact written by Mrs. Brown.
What are the different registers used in an individual’s writing styles and why are these registers relevant to forensic linguistics?People write in different manners depending on their intended audience. With close friends or family, slang and poorly constructed sentences and lack of punctuation may suffice, whereas business and professional emails are more organized, structured, and grammatically correct. The different registers used help to identify a corpus. When trying to identify the author of an email sent to a professor, it would be ineffective to go through all of the emails of a suspected group, rather than to focus on emails specifically to professors or to other superiors in a professional environment. These writings will likely be constructed in a similar manner. Everything I just said is for written stuff that has nothing to do with registers but regardless it is important to know. Registers are the tone your voice takes when talking to certain people and the way you say words. In a professional setting, you may pronounce words fully and correctly, while in a casual setting you might let the ends of words fade away. This is important because these differences could make the difference in identifying a person's voice. The same registers have to be compared.
Is there evidence that "expert" listeners can listen better than the average person?Expert listeners don't exist. We are all equally as likely to identify the voices of others. Many tests have been done and the inconsistent results show that there is no proven evidence that experts are any better than laypeople at identifying voices. Gender based studies have proven little, with some finding men better, some finding women better at women's voices, and some finding no distinct differences. Age did play a small role, with younger people doing better than older people. The longer a person is exposed to a voice, the better they are at recognizing or identifying it. This also is true with the quality of the sample. The most surefire way for someone to identify a voice known to them is to listen to a longer sample of the person's voice. Evidence shows that people who heard voices during stressful situations (high adrenaline) are better at recognizing those voices. The same is true with familiar voices than strange voices.
Describe three characteristics of intoxicated speech and why these were not sufficient to help get a conviction in the Exxon Valdez case.(Tanford) (Sidtis) Intoxicated speech is less consistent in pitch, the words tend to be unfinished there is typically audible slurring, the speed at which a person speaks changes, length of vowels, messy pronunciation, and certain consonant sounds are switched when speaking. Many of these traits were present with Cpt. Hazelwood in the Exxon Valdez case, but because intoxicated speech and stressed/emotional/fatigued speech sound the same and share similar traits, the experts could not say without a doubt that these traits were due to intoxication and intoxication alone. Therefore, the evidence was inadmissible. (Exxon Valdez- crash, oil everywhere, used radio transmissions and analyzed them, determined that his speech compared to speech at times before and after the accident was drastically different in ways that implied that it was drunk speech).
What is law? Is it possible for law to exist without writing? Is it possible for law to exist without language?Law is a set of guidelines (in the form of rules or consequences) that governs any given people. Tiersma determined that language was required in order to create laws and enforce them. Language is used to describe laws, to hold trials, and for deliberations by lawyers. Law can exist without writing, as long as there is language in existence. Language has phonetics, syntax, phonology, morphology, semantics, and pragmatics. Law has existed in many forms from the Tiv in Nigeria which involved local council, witnesses, and spoken pieces from each side where the council decides the outcome, Sumerian law (Ur-Nammu) which is the oldest surviving written law which was dictated by the king, Hitite Law which was situational and based on the outcomes of arbitration, civil law which is codified, which means that every thing is written down so each offense's punishment and outcome is dependent on the precedent set in that exact situation before and is constantly being updated, and common law which is not codified (there is precedent but that isn't all law). US has statutory and common law.
What are the 5 kinds of "register"?intimate (close family/friends), casual, consultative, friends, strangers, etc.
Explain how French came to be the legal language in England up until the 17th c.Latin was the legal language in England because of the lingering of Roman law. This stayed the same after the Norse invaded because their language was similar to old English so they just shared words and language and were able to communicate fine. When the Normans invaded though, they brought the French language with them. Because of this, French became the high language spoken by the higher classes, while English remained the low language (diglossic). French was then spoken in legal environments so it could be understood by the high class members of society. Latin words that could not be translated remained in the legal system.
What is a difference between the effects intoxication has on the voice and the effects stress/anxiety has on the voice?Both influences are very similar and have very similar effects.
Intoxication: consonant deletion, consonant substitution, slurs, deleting the ends of words, range inconsistency but often lower, speed inconsistency but often slower, etc.
Stress/Anxiety: faster speaking, higher pitch, similar inconsistencies but intoxication is more likely in line with the effects of fatigue
What effects does emotion have on the voice? Is this possible to study in a naturalistic environment? How accurate can a study be with these effects in a laboratory setting?The studies often done on the effects of emotion on voices use actors. Though they can portray what we expect to be signs of fear, excitement, etc., these are not genuine vocal patterns as they are being acted and not truly experienced.
sadness: slow, monotone, deep, quiet, downward inflection
fear: fast, high, abrupt inflection changes
anger: higher, louder, faster, changes in inflection
joy: louder, faster, higher, upward inflection
boredom: low, monotone, slow, no variability

Section 2

Question Answer
Li et. al.used forensic linguistics to identify which justice wrote the opinion. able to determine the three most likely justices to write any anonymous opinions. also ideologies ngrams and hfls
Johnson (plagiarism)compared 3 students suspected of plagiarizing and compared them up against controls from the same class who did not plagiarize. The three in question had higher similarities across the board, with less original word usage, and large portions that were close to matching. Lexical types and tokens were compared between each trio, these were significantly higher in the questioned works than in the control works. They determined that they did plagiarize and then one came forward and said one wrote it and "shared" it with the other two
Bailey (forensic linguistics)(Unabomber) He mailed letters to the people he would bomb. 35k word manifesto to the NYT, theorized from his writing that he was in his late fourties and from Chicago, there was a slight possibility that he was black. They focused on the number of times certain phrases or themes appeared per communication, the spelling and grammar rules practiced, proverbs metaphors and compound sentences, common grammatical errors, and textual influences and allusions. They found he had great grammar and was very smart, he alluded often to to a specific manifesto. "You can't eat your cake and have it too" was the damning evidence because Kaczynski said this often and most people say the reverse, which is actually incorrect. He used it correctly. He was identified by his brother based off the phrase "cool-headed logicians" and provided the FBI with a corpus of his prior works. It was determined that the works had the same author.
Solan, Tiersma (author id)admissibility of forensic linguistic practices in court: whether the theory offered has been tested, whether it has been subjected to peer review and publication, the known rate of error, whether the theory is generally accepted in the scientific community
Woolls (plagiarism)Plagiarism detection- words use and sequence usage play into this, a longer sequence excludes common phrases, if these show up in both works, there is likely plagiarism. If entire sentences or passages appear, if there is little original vocabulary between the two documents, if the subject matter is identical, etc. Electronic detection is limited because it won't recognize anything that undergoes a formatting change.
FL v ZimmermanZimmerman killed Trayvon Martin under circumstances that were only matter-of-factly known by Zimmerman and Martin, and only one lived to tell so we got Zimmerman's side of the story. There was screaming for help heard in the background of a 911 call made during the altercation. Experts were presented from both the prosecution and defense. Defense experts said that they could determine it was Zimmerman, and his family and friends claimed to recognize his voice. The same happened for the prosecution. It is unlikely that any friends or family would have ever heard Martin of Zimmerman screaming in fear for their life. There is no way to reenact that kind of vocalization so there is little scientific evidence of the effects of death fear on a person's voice. A person's voice also changes a lot and cannot be identified. Plus the quality of the recording made it very difficult to identify. This evidence was inadmissible.
Cornett v State3 years after a murder, defendants ex wife claimed that he was involved. She said she was there when the guy was killed and that they were there together for a burglary. Court held that spectrographic analysis was not accepted in the scientific community, the evidence of it was only used as supplemental. He was still convicted. She claimed that she did not make a phone call to an attorney previously stating that her then husband was not involved in the crime. Hubbell, the lawyer, said he recognized her voice and that it was she who said that. Experts used spectrographic evidence to prove that it wasn't Cornett who made the call. Spectrograms have positive elimination, positive identification, probable elimination, probably identification, or no conclusion. They determined that the spectrograms from Cornett positively eliminated her from being the caller.
Jessen (forensic phonetician)experts have specific training that helps them with acoustics

Section 3

Question Answer
Charley Ross
Bobby Franks