r } 1-, I [it V:[hl.t'l| THESIS Mir inflimmmmmmmm ’71 “l" .. "a ‘ "’11-1 [Lazinm-vumun ,, n‘_ 7’.) g. . - 5“; ' ff: "' I ' 3 I \- r 4- - ,n r v - p. a -_’- ‘ 13' .1 J" ii. ‘5 v',‘ ' " ‘ )- 1 ';" ‘ r’ _ .. .13. 'Q 51 I- " ‘13: ugh v ..:. \. Jr 3 Q J This is to certify that the thesis entitled Attorney Strategies for Witness Interrogation in the Courtroom: A Sociolinguistic Study presented by Sue A. Cutler has been accepted towards fulfillment of the requirements for M-A' degree in Linguistics W Major professor '. * Date August 13, 1982 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU RETURNING MATERIALS: Piace in book drop to LIBRARIES remove this checkout from gusty—- your record. FINES wiH be charged if book is returned after the date stamped beiow. ATTORNEY STRATEGIES FOR WITNESS INTERROGATION IN THE COURTROOM: A SOCIOLINGUISTIC STUDY by Sue A. Cutler A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Linguistics and Germanic, Slavic, Oriental and African Languages 1982 ABSTRACT ATTORNEY STRATEGIES FOR WITNESS INTERROGATION IN THE COURTROOM: A SOCIOLINGUISTIC STUDY By Sue A. Cutler During a trial an attorney's goal is to convince the jury of the rightness of a client's position, or more simply stated the goal is to win the case, regardless whether the client be the state or an indivi- dual. Due to the nature of the courtroom situation, the most effective method that an attorney may employ to influence the jury is through the illicitation of testimony from witnesses. This paper consists of two parts. (1) A description is given of the linguistic features and the uses of two strategies of witness in- terrogation; these are labeled the Supportive Strategy (SS) and the Undermining Strategy (US). (2) A research project was conducted to test the following hypotheses: (A) linguistically naive members of a speech community (e.g. jurors) make consistent value judgments when an attorney uses the SS or US and the values are different for each stra- tegy and (B) the same speech community members also assign consistent values to a witness's speech choice. In brief, hypothesis A above was supported despite problems which arose from the construction of the research instrument. SUpportive evidence was not obtained for hypothesis B. ACKNOWLEDGMENTS First and foremost I would like to acknowledge the invaluable support, suggestions, knowledge and patience of Professor Carol Myers Scotton, the chairperson of my committee. During the years of working with her I have grown academically and personally. Special words of appreciation also go to Professor Barbara Abbott and Professor Ruth Brend for the valuable comments on this thesis, for serving on my committee and for providing both mental stimulation and frustration in the many courses I have taken with them. I would also like to thank Mr. James Wang, chairperson of the Depart- ment, who made the completion of this thesis and degree possible by providing financial support in the form of a teaching assistantship. Other people who assisted me in either the work for this thesis or in the maintenance of my sanity during the process truly are too numerous to name, but the following require mention: Blake Armstrong, for his assis- tance in the statistical portion of the work; the staff of the LGSOAL De- partment past and present, especially Jo—Anne and Cindy; also the staffs of the School of Journalism, Communication Department, the African Studies Center and the East Lansing District Courtr 54B; Diane Deutsch, and a myriad of other folk. This also requires a note to four other people: to Mom and Dad, who always believed this would be finished but were just never sure "when"; to A.P., who always wanted this to be finished (Now what?); and to my typist, editor and coffeemaker who resolves never again to type final copies of her own work, done at last. ii DEDICATION To Spinsters on the Journey who have the courage to Spark. And to my Great-Aunts Henrietta and Olive Cory. iii TABLE OF CONTENTS CHAPTER I - INTRODUCTION 1.1 Introduction 1.2 Overview of Study Organization 1.3 Concepts Central to the Study 1.3.1 Shared Norms 1.3.2 The Attorney as Entrepreneur 1.3.3 Strategy Choice and Codeswitching 1.4 Study Hypotheses 1.4.1 Attorney Strategy Choice 1.4.2 Witness Strategy Choice 1.5 Conclusion CHAPTER II — DATA COLLECTION AND ANALYSIS PROCEDURES 2.1 Introduction 2.2 Courtroom Language Data Collection 2.2.1 Delineation of Pretinent Features 2.2.1.1 Structural Features 2.2.1.1.1 Phonological Features 2.2.1.1.2 Lexical Choice 2.2.1.1.3 Syntactic Features 2.2.1.2 Interactional Features 2.2.1.3 Content Features iv 10 10 11 11 13 15 2.2.1.3.1 Confrontational Features 2.2.1.3.2 Editing Features 2.3 Interview Data Collection 2.3.1 Scripts 2.3.1.1 Recording of Scripts 2.3.2 Selection of Subjects and Use of Tapes 2.3.3 Instructions to Subjects at Interview _ 2.3.4 The Questionnaire 2.3.4.1 Design of the Questionnaire 2.3.4.2 Use of the Questionnaire 2.3.5 Analysis of the Data 2.4 Conclusion CHAPTER III - PRESENTATION OF FINDINGS 3.1 Introduction 3.2 Definition of Supportive and Undermining Strategies 3.2.1 Features of Strategies as Used by Attorneys 3.2.1.1 The Supportive Strategy (SS) 3.2.1.1.1 Structural Features 3.2.1.1.1.1 Phonological Features 3.2.1.1.1.2 Lexical ChoiceS' 3.2.1.1.1.3 Syntactic Features 3.2.1.1.2 Interactional Features 3.2.1.1.3 Content Features 3.2.1.2 The Undermining Strategy (US) 3.2.1.2.1 Structural Features 3.2.1.2.1.1 Phonological Features 3.2.1.2.1.2 Lexical Choices 15 15 16 16 17 18 18 19 19 19 20 22 23 23 23 23 25 25 25 25 26 27 27 28 28 28 28 vi 3.2.1.2.1.3 Syntactic Features 3.2.1.2.2 Interactional Features 3.2.1.2.3 Content Features 3.2.2 Effects of Strategy Usage on Witnesses 3.2.2.1 Effects of the Supportive Strategy 3.2.2.1.1 Structural Features 3.2.2.1.1.1 Phonological Features 3.2.2.1.1.2 Lexical Choices 3.2.2.1.l.3 Syntactic Features 3.2.2.1.2 Interactional Features 3.2.2.1.3 Content Features 3.2.2.2 Effects of the Undermining Strategy 3.2.2.2.1 Structural Features 3.2.2.2.1.l Phonological Features 3.2.2.2.l.2 Lexical Choices 3.2.2.2.l.3 Syntactic Features 3.2.2.2.2 Interactional Features 3.2.2.2.3 Content Features 3.3 Report of Interview Findings 3.3.1 Research Design 3.3.2 Predicted Results 3.3.3 Actual Findings 3.3.3.1 Tabulation of Raw Data. 3.3.3.2 Testing for Significance 3.4 Conclusion 30 31 31 31 32 32 32 32 34 34 34 34 35 35 35 35 36 36 37 37 37 37 39 4O 45 vii CHAPTER IV - DISCUSSION OF FINDINGS 4.1 Introduction 4.2 Discussion of Findings 4.2.1 Relation of Findings to Predicted Results 4.2.1.1 Tapes Inflected for Emotion 4.2.1.2 Tapes Uninflected for Emotion 4.2.2 Relation of Findings from Tapes Inflected for Emotion to Findings from Tapes Uninflected for Emotion 4.2.3 Relation of Findings to Study Hypothesis and Sub-hypothesis 4.3 Evaluation of the Study 4.3.1 Limitations of the Study 4.3.2 Suggestions for Future Research 4.4 Applications of the Study 4.5 Conclusion REFERENCES APPENDIX Instructions Read to Interview Subjects Questionnaire Script of Undermining Strategy Script for Supportive Strategy 48 48 48 48 49 51 53 55 57 57 58 59 61 62 64 64 65 67 73 3-10 3-11 LIST OF TABLES Features of Strategies Used by Attorneys. Features of Witness Speech--Effects of Strategy Use. Expected Responses to Questionnaire. Tabulation of Interview Results, Part I-Attorneys. Tabulation of Interview Results, Part II— Witnesses. Tabulation of Interview Results, Part III— Type of Examination. Z—values and F(z)-va1ues for Part I-Attorneys. Z-values and F(z)-values for Part II- Witnesses. Z-values and F(z)-va1ues for Part III- Z-values and F(z)-va1ues for Part III- Type of Examination. Actual Results of the Questionnaire--Set 1, Inflected for Emotion. Actual Results of the Questionnaire--Set 2 Uninflected for Emotion. viii 24 33 37 39 39 40 42 43 44 44 46 CHAPTER I INTRODUCTION 1.1 Introduction. During a trial the attorney's goal is to convince the jury members of the rightness of a client's position, regardless whether that client be the state or an individual. While opportunities exist for the attor- ney to directly address the jurors, i.e., opening and closing statements, there is a more effective means available to attorneys with which to shape the opinion of the jurors and, hence, to influence the outcome of the trial. The manner by which an attorney may most effectively influence jurors' opinions is the way an attorney illicits testi- mony from a witness. In this paper I claim that an attorney has two strategies available for use in the examination of a witness. An attorney's choice of stra- tegy can indirectly influence a juror's opinion of the case by contri- buting to or detracting from the credibility rating of a witness. So, a basic assumption of this study is that in the courtroom situation, there exist two strategies of language choice and these strategies are used by an attorney in the examination of a witness. A major purpose of this study is to define these strategies in terms of their charac- teristic linguistic features. This portion of the study is, of necessity, descriptive in nature. The strategies are described in part in terms of previous analyses of powerful/powerless speech, polite speech and studies of courtroom language. The manner in which an attorney chooses between the proposed strategies is related to the phenomenon of codeswitch- ing. Given that these strategies exist, the major hypothesis is that naive speakers of a speech community will assign consistent value judgments to the speakers using these strategies and that the values will be different for each strategy. A second premise of the study is that an attorney's strategy choice affects the linguistic features of a witness's presentation. The witness's variety choice provides the input for a juror's evaluation of a case. These Statements are taken as underlying the sub-hypothesis of the study: that linguistically naive members of a speech community (e.g. members of a jury) will assign consistent value judgments to a witness's variety choice. A juror's judgments of the witness are esential to the outcome of a trial be- cause they affect the juror's perception of the witneSs's credibility rating. 1.2 Overview of Study Organization. The remainder of this chapter is devoted to a discussion of (1) concepts basic to the study, (20 and a review of pertinent literature. Chapter 2 contains a description of the data collection process, including a descrip- tion of the features used to define the strategies. Chapter 3 contains a descrption of the findings and a discussion of how they support and question the hypotheses of the study. Chapter 4 includes a discussion of possible im- plications of the study as well as suggestions for further study and research. 1.3 Concepts Central to the Study. 1.3.1 Shared Norms. One of the central premises of sociolinguistics is that members of a speech community employ a set of shared norms to interpret their verbal interactions. Norms are assumed to govern the appropriate use of varieties composing the speech community's linguistic repertoire (Fishman, 1972; Gumperz, 1972; Hymes, 1968; Hymes, 1974). Different varieties within the repertoire are assigned consistent values by com— munity members; this assignment embodies the concept of shared norms. Due to these norms, the manner in which a statement or question is phrased is expected to affect the listener's perception of a speaker's character and motives in a prescribed manner (Crosby and Nyquist, 1977; Lakoff, 1973; Scotton and Ury, 1977). In the context of this study the concept of shared norms is appli- cable in two specific instances of languages usage: (1) the attorney chooses a strategy; its use is received and interpreted by the witness and the jurors and (2) the witness makes a speech choice-reaction to the at— torney; this linguistic output functions as input for the jurors. When a naive member of a speech community receives such input, that reception triggers the value judgments that the individual shares with the speech community of the speaker's background, character and intentions. Values assigned to the use of each variety are assumed to be unique. Recently studies have been conducted applying the concept of shared norms to the courtroom situation. Erickson et a1. (1977) dealt with the witness's use of different speech patterns and the effect of that usage on the evaluation of the witness by jurors. Danet et al.(1976), Loftus and Zanni (1975), O'Barr and Conley (1977) and Danet et a1. (1980) have all conducted investigations into lawyer/witness interactions. However, with the exception of Danet et a1. (1980), these studies have centered on the witness. 1.3.2 The Attorney as Entrepreneur. This study will focus on the role of the attorney as a questioner of witnesses and a manipulator of shared values of a speech community. By virtue of the very nature of his/her position, an attorney has a potential to direct a witness's testimony in the courtroom. In addition, it is common practice for an attorney to rehearse general courtroom procedures and expected order of testimony with her/his own witnesses prior to the actual trial. The purpose of the rehearsal is to insure that the mate- rial will be presented during the trial in the most positive and effective manner. On the other hand, an attorney rarely has the opportunity to interview the opposition's witnesses before a trial. It is in this latter case that the attorney's ability to utilize the existence of shared lin- guistic norms is especially tested. In such situations the attorney must quickly assess a witness and determine how to manipulate the linguistic varieties in her/his repertoire to call forth the desired linguistic reac- tions and variety choices from a witness--the attorney's intent being to minimize the negative effects of the opposition's witness. In summary, then, an attorney can utilize the assignment of societal norms and values of linguistic varieties to influence indirectly the jurors' evaluation of a witness's credibility. An outline of the pro- posed effects of this usage follows: (1) The attorney chooses one of the strategies from the available repertoire; the use of this strategy functions as a symbol. (2) The witness responds to the implications of the symbol inherent in the attorney's variety choice and selects an appropriate variety in which to respond. (3) Jurors note the varieties used by the attorney and the witness. They then assign values related to the speaker's background, character and intentions on the basis of these usages; each variety has its own unique set of values. Specifically, in relation to a witness, jurors assign a credibility rating to the witness's testi- mony based in part on the linguistic features of the witness's speech. A successful attorney is expert at the manipulation of existing sociolinguistic norms, and as such can be called an entrepreneur. A linguistic entrepreneur is a master at manipulating available linguistic varieties and obtains a living by the marketing of these linguistic talents (Minderhout, 1972). To be a successful linguistic entrepreneur, it is asserted that a lawyer must effectively choose between two main strategies of witness testimony manipulation. I label these strategies the Supportive Strategy (hereafter, SS) and the UnderminingStrategy (hereafter, US). It is further suggested that these strategies are characterizable with lin- guistic features and are recognizable to members of a defined speech com- mmunity. See Chapter 2 ( 2-2-1) for a complete discussion of the lin— guistic features of the strategies. It is hypothesized that the members of a speech community will not only assign consistent values based on the strategy or variety employed but also that the values assigned to one particular strategy would be different from those assigned to any other variety. 1.3.3 Strategy Choice and Codeswitching. The use of the strategies can be related to the phenomenon of code- switching. One might suppose that lawyer/witness interactions could take place in the transactional arena. In such a case, neither the con- cept of power nor solidarity would be salient (Scotton and Ury, 1977). If such were the case, the attorney would make no attempt to manipulate a witness's testimony. While such situations do occur, I suggest that they are rare, for in choosing the transactional arena, the attorney neglects the role of entrepreneur and leaves the trial outcome to chance; that is, such an attorney neglects her/his role of defending the client. Indeed, one may prefer to define attorneys who interact with witnesses in the transactional arena as novices, rather than as true entrepreneurs. Hence, in most trials an attorney will move the inter- action into (1) the identity or solidarity arena by using the SS or (2) the power arena by using the US. This model of attorney as 11H8Uistic entrepreneur is derived from the many hours of observation conducted.. In the identity arena, the participants share at least one characteristic which marks them as members of a common group. In terms of the courtroom the lawyer may attempt to convince the witness that they have a common goal relating to the outcome of the trial. In the power arena the attor- ney would emphasize the difference between the witness and her/him self by invoking the power position of legal professional (Scotton and Ury, 1977). 1.4 Study Hypotheses. 1.4.1 Attorney Strategy Choice. An assumption central to the study is: that there exist two stra- tegies of language choice for an attorney in the courtroom situation; these strategies are here called the Supportive Strategy (SS) and the Undermining Strategy (US); furthermore, these strategies are used by an attorney in the examination of a witness. The main hypothesis of the study is: (1) that members of a speech community will recognize an attorney's use of the US and SS; the recog- nition will be signaled by the consistent assignment of values to the attorney's use of a particular strategy; these values consist of judgments of personal background, character and intentions; and (2) that the value judgments that naive members make will be uniquely different for the use of each strategy. 1.4.2 Witness Strategy Choice. The other major assumption of the study is: that linguistic features of a witness's presentation is affected by the attorney's strategy choice. The sub-hypothesis related to this assumption is: that the members of the speech community will assign values to the witness's language choice. 1.5 Conclusion. In this chapter the sociolinguistic concepts and phenomena relevant to the study have been discussed; these include (1) shared norms, (2) linguistic entrepreneurs and (3) codeswitching. Literature pertinent to the study was cited. The major assumptions and their relation to the study hypotheses have been stated. CHAPTER II DATA COLLECTION AND ANALYSIS PROCEDURES 2.1 Introduction. The purpose of this chapter is to discuss the methods of data collec- tion used in the study of attorney speech in the courtroom. The collection was a two-step process. The first step consisted of working with natural data as they occurin the courtroom situation--a description of the method employed is provided. The second step involved an attempt to obtain in- dependent supportive evidence for the recognition of the strategies by naive speakers and, therefore, for the study hypotheses. In conjunction with this step I discuss the manner in which strategy scripts and tapes and a questionnaire were produced, how subjects were selected and the way the study interviews were conducted. The manner in which the question- naire results were analyzed is also discussed. 2.2 Courtroom Language Data Collection. I spent approximately 70 hours observing language use in the court- room environment. The observations were made in the East Lansing District Court 54-B, and in the Lansing Circuit Court. During that time 18 attor- neys and 30 witnesses were scrutinized. Both civil and criminal pro- ceedings were part of the original data corpus. Notes were made on general characteristics of attorney and witness speech. An additional 15 hours were spent transcribing language interactions from tapes made during proceedings. The East Lansing court system made these tapes available. 2.2.1 Delineation of Pretinent Features. The salient characteristics of the SS and the US were determined in the following manner: (1) I observed linguistic interaction in the courtroom and noted general impressions of the interaction--including the speech of the attorney and the witness. (2) Based on these initial im- pressions, an outline of strategy features was drawn up. (3) Previous studies dealing with the characteristics of courtroom language (Danet, 1980; O'Barr and Conley, 1976), powerful speech (Owsley and Scotton, 1982) and powerless speech (Lakoff, 1973) were consulted. (4) The original list was modified to reflect the terminology used in previous works; some of the original categories were further subdivided to correspond more closely to earlier research. The speech of attorneys and witnesses were categorized in terms of three major feature categories: (1) Structural Features, (2) Interactional Features and (3) Content Features. A general definition of each category follows. The applicability of the categories to the strategies and spe- cific examples of their application are given in Chapter 3 (Presentation of Findings), section 3.2.1. 2.2.1.1 Structural Features. The Structural Features category contains those features studied within the realm of traditional linguistics and is divided into three subcategories--(1) Phonological Features, (2) Lexical Choice and (3) Syntactic Features. lO 2.2.1.1.1 Phonological Features. This subcategory includes: aa. reduced pronunchtion or "sloppy speech"--characterized by con- sonant deletion and vowel reduction, e.g. the Speaker useslénl or [2‘1 for "and", [I m] for "him, them" and flu for "-ing"; the use of con- tractions is also very common, e.g. "do not" is realized asl:don{]; b. full forms or "careful speech"--a word is pronounced in its full phonemic forms, e.g. "do not" as [du#na€]; c. unmarked intonation--Q-l questions, Q-2 questions and YES/NO questions have a final high-low contour pitch and declaratives have an overall pitch downdrift or final falling pitch; d. marked intonation--Q—l questions, Q—2 questions and YES/NO questions end in a final falling pitch (which is usually characteristic of declaratives) and declaratives end in a high-low contour (which is caracteristic of unmarked questions); e. "friendly" vocal quality--the term "friendly" is difficult to define in precise linguistic terms 3however, it represents a consistent label used by member of the speech community to indicate a specific way of talking. It is considered to be a positive quality; and f. "nonfriendly vocal quality--this term is also taken from comments by the language community members and is indicative of a negative judgment. 2.2.1.1.2 Lexical Choice. This subcategory includes: a. informal word choice--vocabulary that is used in everyday speech ranging from slang expressions to informal style choice, e.g. "that guy, man"; 11 b. legalese word choice--vocabulary use draws heavily from the specific jargon used by legal professionals--it includes a formal style of English and a special vocabulary of legal terms and phrases, e.g. "the individual in question"; c. informal style of address--use of personal names or nicknames, e.g. "Johnathan," "John"; d. formal style of address--use of an appropriate title "Ms.," "Mrs.," "Dr.," etc. and an individual's last name or use of a term of respect,e.g. "ma'am, sir, your honor,‘ etc.; e. politeness markers--structures or lexical items which indicate deference as it is defined by the speech community, e.g. "could," "would," "please," or "thank you"; the use of models, e.g. "can," is also polite. 2.2.1.1.3 Syntactic Features. This subcategory contains: a. Q-l questions--questions containing the question words "who," "what," "where,' and "when"; b. Q52 questions--questions containing the question words "why" and "how"; c. YES/NO questions-—questions where the natural response possi- bilities are limited to beginning with "yes" or "no", e.g. Question: "Did you make the call?? Response: "Yes, I did."or"No, I didn't?; d. positive question types--questions that do not contain overt negatives, e.g. "What did you do after that?"; e. negative question types--questions that contain overt negatives, i.e.,"not," "never," "noone," etc., e.g. "Why didn't you do that?"; f. tag questions--questions ending in a tag, e.g. "You verified the number, didn't you?"; 12 g. referents overtlyimarked--nouns are reused in each sentence of a new turn rather than pronouns being used; h. referents poorly marked-~within a turn, pronouns are used rather than nouns and within the course of an examination, indefinite pronouns, e.g. "one," "that", are used; 1. unmarked word order-~the surface structure of a sentence is rela- tively simple, e.g. "The name on the box and the name on the list were the same";also characterized by an economy of words; j. marked word order—-the surface struCture of a sentence is rela- tively complex, e.g. "The name that was on the box was the same as the name that was on the list."; sentencescontain much excess verbage; k. indexical sentences--sentences contain indexical expressions which are referentially ambiguous, i.e., a pronoun may be interpreted as referring to more than one possible referent; e.g. "Where were you at that time?" or "And what happened then?" 2.2.1.2 Interactional Features. Interactional Features are used to manipulate the conversation. The user employs these features to control the sequencing of information, to regulate the content and to evaluate that content. Terminology and definitions for these features are taken primarily from Owsley and Scotton (1982). See Chapter 3, section 3.2.1 for the distribution of these features to the specific strategies. This category includes: a. interruptions--one speaker begins talking before the other speaker is finished; these may be accompanied by (1) an apology and/or an explanation or (2) no apology; b. discourse markers--use of short words or phrases to mark the beginning of a turn or to introduce a new topic; these words are l3 accompanied by a falling intonation and are stressed; they are also followed by a short pause; e.g. "80.," "Now,, etc.; c. minimal responses——short words used to indicate that previous speaker is to continue talking, e.g. "Yeah," "m-hm," "uh—huh," etc.; there are two types: (1) encouraging--signaled by a rising pitch and (2) intimidating signaled by a falling pitch; d. highlighting-—(l) c1arifying——speaker stops previous turn-taker to clarify facts of the testimony and (2) underlining-~speaker repeats a part of previous speaker's turn for emphasis; e. leading--trying to suggest possible answers to the other speaker. "Leading" questions used by attorneys are of a more restricted type than those which may occur in everyday speech. Courtroom procedures prohibit theuse of truly leading questions. An attorney is allowed to ask a question only if the witness has introduced the basic information during the examination. There are two types of leadings used: (1) single choice questions--offering only one choice, related to YES/NO questions and marked declaratives; e.g. Attorney: You gave him the money? Witness: Yes, I did. (2) multiple choice questions--offer more than one choice, at least one of the choices must have been previously mentioned by the witness or by a previous witness during the course of a trial; e.g. Attorney: What time did you go to work that night? Witness: Around 9 o'clock? Attorney: Was it 8:30, 9:00, 9:30? Precisely, what time was it? 14 f. length of total turn-taking time--(l) attorney speaks more than witness or (2) witness speaks more than attorney; and g. internal chronological consistencye-material is organized in chronological order for presentation; any deviation from the established time oriented ordering is accompanied by explicit instructions; e-g- Attorney: Let's go back to where you're standing by your truck with him five to six feet away from you. Did he have a weapon? 2.2.1.3 Content Features. The Content Features category has two basic subcategories: (1) con- frontational and (2) editing features. 2.2.1.3.1 Confrontational Features. This subcategory consists of: a. marked challenges--a marker used with a "confrontative" state- ment; markers include words like: yet," "however," "but, etc., e.g. Attorney: And yet! you just identified the defendant for the court. b. appeals to higher authorities quotes from previous testimony submitted by witness or by other witnesses; the quotes may be verbatum or paraphrased; c. set-ups or "dumb questions"-—the answer is known to all those involved in the interaction, is an obvious answer or is a "common sense" answer; and d. minimizing hedges--"Forms which reduce the force of an asser- tion by allowing for exceptions or avoiding rigid commitments." (Lakoff, 1973), e.g. "a little," "sort of," "kind of," etc. 15 2.2.1.3.2 Editing Features. This subcategory includes: a. summary--the speaker relates events and situations in an objec- tive or neutral manner; b. synthesis--the speaker intersperses personal opinion throughout the relating of the facts of a situation; there are three relevant types observable in courtroom language: (1) evaluative comments--speaker offers an opinion, e.g. "good," "okay," etc., (2) coloring-style shifting--speaker makes a codeswitch from a for- mal to an informal style or an informal to a formal style; the code- swith may occur either within the Speaker's own turn or from one speaker to the next, e.g. Witness: You mean by my truck? (informal) Attorney: Yes, by your vehicle. (formal) and (3) maximizers-intensifiers——words are used which request as explicit a response as possible; e.g. "precisely, exactly, absolutely," etc. 2.3 Interview Data Collection. After the language datanwereanalyzed and the salient features of the strategies outlined above were established, the second step of the study was initiated. Its purpose was to obtain supportive evi- dence for the study hypotheses, i.e., (1) that naive members of a speech community will assign consistent values to an attorney's use of the strategies and (2) that in turn they will also assign consistent values to the witness's speech choice, particularly in regards to the credibil— ity rating of the witness. The evidence was gathered through the use of recorded scripts and a questionnaire. 16 2.3.1 Scripts. A set of scripts was prepared for use in the study—-one tape repre- senting the $8, the other the US. The basic content of the scripts was the same. The purpose for maintaining content consistency was to control for the possibility that listeners might be reacting to information dif- ferences of tapes rather than linguistic differences. The scripts were based on a transcript of an actual case. At the same time the language of the original transcript was altered to maximize the linguistic differences of the tapes. Each script was "pumped full" of the linguistic features relevant to the particular strategy. The reason for maximizing the occurrence of the relevant features was to compensate for the shortening of the interaction time which was necessary to create a manageable study and to make the scripts as different from one another as possible on the basis of measurable criteria. Transcripts of the two scripts used in the study are in the Appendix . Both the attorney's and the witness's speech were manipulated. In future studies it might prove interesting to manipulate the speech of only one participant. While in such a situation one participant's speech wuld be "unnatural" in terms of the interactional influence proposed in this study, the expectation would be that the basic results would not differ markedly. The results would not be expected to differ because the underlying assumption, that naive speakers' reactions to speech variety choice is consistent, would remain constant. For further dis- cussion of possibilities of future study see Chapter 4. 17 2.3.1.1 Recording of Scripts. The scripts were then recorded. The actors chosen to record the scripts were seniors in the MSU Theater Department. The linguistic features of the strategies and their assumed intended effects were dis- cussed with the actors prior to recording. The use of experienced actors was deemed necessary in order to obtain as natural a rendition of the scripts as possible. The actors were also selected for sex--both were male. The assump— tion underlying this choice was that the reaction of naive members of a speech community will differ with the sex of the speaker. Past studies have overlooked the possible influence of this variable as well as those of education and race. In Danet et a1. (1980) native speakers were asked to compare the speech of witnesses with greatly divergent socio-economic backgrounds, i.e., that of an upper class, educated white male and that of a lower class, less educated black female. Two sets of tapes were recorded. Each set consisted of an example of the SS and the US. Each individual tape was about six minutes long. One set was inflected for emotion and one was not. The set inflected for emotion is defined as being closer to natural speech. However, the second set, which was uninflected for emotion, was prepared to test for thepossibility of subjects'reactions being based on emotive content rather than linguistic content. 2.3.2 Selection of Subjects and Use of Tapes. The tapes were played to 71 students enrolled in Communications 100-- a freshman-level communications course offered by the Department of Com- munication, MSU. Students may obtain extra credit in the course for 18 participating in approved studies. Subjects were required to be native speakers of English and to have had no special training in linguistics. The group size for a presentation varied from four to 16, but averaged 12. The order in which the tapes were played was changed with each pre- sentation, i.e., in one presentation the order was SS-US and in the next it was US-SS. This procedure was followed for both sets of tapes; the purpose of this alternation was to control for the possible influence of tape order on subjects' reactions. 2.3.3 Instructions to Subjects at Interview. Subjects were told that they would hear two tapes where an attorney was examining a witness concerning the facts of a case. They were in- structed to listen closely to the tapes and to imagine themselves in the role of "juror." They were also told that after listening to the tapes, they would be given a questionnaire to complete. I attempted to control for the possibility of unintentionally influencing the subjects by reading a set of instructions. The instructions to the subjects also included some background information on the trial. For a transcript of the exact instructions and background information, see Appendix . 2.3.4 The Questionnaire. 2.3.4.1 Design of the Questionnaire. The questionnaire consisted of three parts related to the subject's personal reactions to the tape content: (1) 22 items related to the attorneys; (2) 10 items related to the witness and (3) two items related to the type of examination the exchange exemplified. All test items were discrete; they required that a subject make a choice of A or B. In a 19 future study it would beinterestingto play one tape at a time and to have subjects rate each tape for each item on a continuum from 0% to 100% of a particular item. In Parts 1 and 2 the items were descriptive ad- jectives. The adjectives were selected to represent a wide range of positive and negative attributes of attorneys (in Part I) and of wit- nesses (in Part II). The adjectives were taken in part from comments and value judgments made about attorney style in articles and in part from my own reactions to the tapes. Prior to administering the questionnaire, predictions were made as to probable subject responses based on the study hypotheses. These predictions were then compared to the actual choices made by subjects. 2.3.4.2 Use of the Questionnaire. After the tapes had been played, the instructions from the ques- tonnaire were read aloud. The subjects were instructed to circle the number of the tape whichmore exemplified the particular item in each section of the questionnaire (attorney, witness and examination). "Tape 1" represented the first tape played and "Tape 2" the second. The subjects were further instructed to complete the questionnaire as quickly as possible. Emphasis was placed on recording initial impres— sions to the items. At this point, the questionnaire was given to the subjects. No questions were accepted from subjects until the questionnaires Inuibeen completed and collected. After the questionnaires were collected, subjects were provided with a short summary of the true nature of the study and questions were answered. Subjects were cautioned not to dis- cuss any aspect of the study with anyone not present at the sessions 20 until the study was completed. The entire process required approximately 25 minutes. 2.3.5 Analysis of the Data. The emotionally-inflected tapes were played for 46 subjects; the emotionally-uninflected tapes were played for 25 subjects. All results weretabulated and tested for significance. After the results of the two sets of interviews were tabulated, a z-test was used to test the results for significance. A significance level of .10 was assumed. The choice of a .10 significance level is less exacting than the more standard .05 level and was an arbitrary de- cision. However, upon reexamination of the data, it was determined that the choice did not make much difference in the results of the study. Tables of the raw data as well as the significance test results are presented in Chapter 3. 2.4 Conclusion. In this chapter the methods used in the collection and analysis of the data were discussed. A catalogue of the salient characteristics of the SS and US used by the attorney was presented. Information pertaining to the development of the scripts of the tapes used in the study and the study questionnaire and its use was given. The manner in which subjects were chosen was also delineated. CHAPTER III PRESENTATION OF FINDINGS 3.1 Introduction. The purpose of this chapter is to present the findings of the two steps of the study (i.e., natural data analysis and interview data ana- lysis). A definition of the features of the strategies (the Supportive Strategy and the Undermining Strategy) as used by attorneys is given. A description of the effects that the use of the strategies have on wtnesses is also included. The results of the interviews are reported —-both the predicted results and the actual findings. There is also a description of the use of a z-test which was employed to test the actual findings for significance. 3.2 Definition of the Supportive and Undermining Strategies. The Supportive Strategy (SS) and the Undermining Strategy (US) are characterized by a wide range of linguistic features as noted in Section 2.2. In this section the distribution of the pertinent features in the strategies is set forth and some additional examples of their usage are given. To reiterate, there are three major feature categories: (1) Structural Features, (2) Interactional Features and (3) Content Features. The Structural Features category is further divided into (1) Phonological Features, (2) Lexical Choices and (3) Syntactic Features. -It should be noted that the assignment of features to a particular strategy is based on frequency of occurrence rather than on exclusiveness, i.e., reduced pronunciation may occur intermittently 21 22 in the US but it occurs consistently and to a greater degree in the SS. 3.2.1 Features of Strategies as Used by Attorneys. This subsection contains a description of the salient character- istics of an attorney's use of (1) the SS and (2) the US. Table 3-1 lists the features of the strategies in table form for easy reference. 3.2.1.1 The Supportive Strategy (SS). The SS is usually used by an attorney who is direct-examining a wit- ness, i.e., the attorney has called the witness. It is supportive and by using it the attorney signals that he wishes to narrow the social dis- tance between him/herself and the witness. It uses an informal speech style. style. 3.2.1.1.1 Structural Features. 3.2.1.1.1.1 Phonological Features. The SS is marked by the use of reduced pronunciation, e.g. "I want you to speak louder." might be realized as [ay wan? ya ta spik lawdr]. Another example of reduced pronunciation is the use of contractions, using "don't" or "wouldn't've" instead of "do not" and "would not have," e.g. "The people didn't recognize the phone number?" Unmarked intona- tion patterns are characteristic of the SS, i.e., interrogatives end in a high-low contour pitch and declaratives have an overall downdrift or a final falling intonation. Another phonological feature relates to the overall tone or vocal quality of the speaker. In the SS one might describe the speaker as having a "friendly” vocal guality_or as being "supportive,' or "patient." This feature is measured by the sub- jective reaction of member of the speech community. Table 3-1. 23 Features of Strategies Used by Attorneys. FEATURES II. III. Structural Features A. Phonological Features B. Lexical Choices C. Syntactic Features Interactional Features Content Features -reduced pronuncia- tion: contractions -unmarked intonation -"friendly" vocal quality -informa1 word choices -informa1 style of address‘ -politeness markers -Q-l questions, very frequent -Q-2 questions, very frequent -YES/NO questions, less frequent -referents overtly marked -positive question types -unmarked word order -attorney speakes less than witness -interruptions, infre- quent, accompanied by apology -highlighting for clarifying - encouraging minimal responses -internal chronological -lead -minimizing hedges -set-ups -coloring synthesis (style shift to in- formal style) —full forms -marked intona- tion -"unfriendly" vocal.quality -legalese -formal style of address -YES/NO ques— tions, very frequent -Q-l questions, less frequent -tag questions -mu1tip1e choice questions -negative ques- tion types -marked word order -indexical sentences -attorney Speaks more than wit- ness -interruptions, no apology —underlining highlighting -intimidating minimal res- ponses -discourse mar- kers —leadings ~marked challenges -appeals to higher authorities -synthesis 1. coloring- style shift to formal style 2. maximizers- only intensifiers 24 3.2.1.1.1.2 Lexical Choices. The speaker of the SS uses very little "legalese." In place of common legal terms, the speaker makes informal word choices or uses words from "lay" speech, e.g. legalese "assailant, defendant" is realized as "person, guy." This choice of common words is explicable in terms of a codeswitch from the formal, power arena (where legalese is the norm) to the informal, solidarity arena. Further evidence supporting this explanation (of the use of the codeswitch) is the more extreme, occa- sional use of slang by SS-attorneys; slang has been listed as a feature of informal speech (Joos, 1967), e.g. "Let's jump back to where he told you to leave." The use of an informal style of address (e.g. "Bill" or "William" vs. "Mr. Jones") is a second type of lexical choice that ap- pears in the SS, e.g. "John, where were you employed on Nov. 6, 1980?" Politeness markers, such as "would,' "please,' and "thank you" are also lexical choices made by the SS-user, e.g. "Would you go over those three times again, please?" Modals are frequently used as politeness markers. 3.2.1.1.l.3 Syntactic Features. Only Q-l questions (using "what," "when, who," and "where"), Q-2 questions (using "how" and "why") and YES/NO questions are used by the SS-attorney. Of these three types of questions, Q-l questions and Q-2 questions are most frequently used. In questions referents are overtly marked in each new reference (stressing the use of nouns over pronouns), e.g. "The people living there said they didn't order a pizza." where "people living there" have been referred to previously. Generally, questions incorporate unmarked word order and are stated in positive terms, e.g. "Where did you park your truck?" In general, a question ap— pears in a non-indexical sentence form. 25 3.2.1.1.2 Interactional Features. In the SS, overall an attorney speaks less than the witness. A SS-user interrupts infrequently and apologizes for interruptions when they do occur, e.g. "Let's stop right there for a moment, please." Encouraging minimal responses are also characteristic of the 88; these ' etc., with a rising intonation. include words such as "uh-huh," "okay,' An attorney intersperses these throughout a witness's testimony as well as using them at the end of a response by a witness. An attorney may hightlight a statement made by a witness in order to clarify the res- ponse, e.g. (Witness has completed a lengthy description of monies given to a robber.) Attorney: So. At first you gave 'im $36; an' then you gave 'im another $19. The SS-user maintains an internal chronological consistency in the pre- sentation of questions, that is, opening questions deal with background information of the case and later questions relate to the incidents in the order in which they occurred during the committing of the crime. 3.2.1.1.3 Content Features. The SS has few salient content features. An attorney using the 88 can make use of minimizing hedges (e.g. "a little louder" in "You're gonna hafta slow down a little bit."), set-ups or "dumb questions" where either all those involved know the answer to the question or the answer is one of common sense (e.g. Witness: ...He asked me if I could identify him, ...and I jus' said no... 26 Attorney: Why'd you say that? Witness: Well, sure I'm gonna say I'm gonna identify 'im when he's standin' there holdin' a gun on me?), and a type of sypthesis called coloring which involves style shifting from a neutral-transactional arena to an informal, solidarity arena. Often the switch involves repeating vocabulary used by the witness, e.g. Witness: Yeah, but it was kinda hard. Attorney: Why was it hard? 3.2.1.2 The Undermining Strategy (US). The US is most likely to be used by an attorney during a cross- examination of a witness, i.e., the witness was called by the Opposition. It is non—supportive and its use signals that the attorney is emphasizing the power role existing between her/himself and the witness. It is generally characterized by the use of a formal style of Speech. 3.2.1.2.1 Structural Features. 3.2.1.2.1.1 Phonological Features. The US is marked by the use of full phonological forms, i.e., no contractions and no reduced pronunciations. Marked intonation patterns are characteristic of the US: (l) declarative statements often end in a high-low contour pitch (which is characteristic of unmarked interro- gatives) and (2) interrogatives statements commonly have a falling in- tonation in final position (which is characteristic of unmarked decla- ratives). The US is also marked by an overall "unfriendly" vocal quality, in opposition to the "friendly" tone of the SS-user. The US-speaker is likely to be characterized as "rude, unfriendly," or "condescending" by a member of the speech community. 27 3.2.1.2.1.2 Lexical Choices. The questions posed by the US-attorney involve the use of formal English (e.g. "individual" for "guy, person") and show a greater use of legalese than the SS-user. Legalese appears not only in single word choices but also in the choice of phrases, e.g. "then" may be realized in legalese as "at that point in time." The high level use of legalese is consistent with the hypothesis that the US-user is interacting with the witness in a formal, power arena. The US-speaker also uses a £23: mal style of address, e.g. "Mr. Wilson, you need to speak up so you can be heard." 3.2.1.2.1.3 Syntactic Features. In respect to the SS, a greater number of question types are available to the attorney who chooses to use the US. In addition, the usage frequency of the various question types differs greatly from that of the SS. In the US, YES/NO questions predominate; YES/NO questions have been determined to be most effective in controlling witness res- ponse (Danet et al., 1980). Q:l questions appear infrequently and Q-2 questions are extremely rare. In the SS, the latter two types, occur most frequently. The US-attorney may also use tag questions and mul- tiple choice questions. Questions are often phrased in negative terms, e.g. "Why didn't you come forward with this evidence sooner?" These negative question types contain overt negative words, e.g. not, "never," etc. and must be used carefully as the judge may rule that their use constitute "badgering" the witness. Tag questions, multiple choice questions and syntactic declaratives with question intonation have been listed as characteristics of powerful speech (Owsley and Scotton, 1982). The US is marked by an occurrence of indefinite pronouns 28 which have pgorlypmarked referents, e.g. "When did you go there?" where "there" could denote one of several places. Questions frequently con— tain marked word order. An example of the marked word order used in the US is "Was the address on the pizza box the same as the address on the house that you went to to deliver the pizza?"; in the SS the same sentence might appear as "Were the addresses on the pizza box and on the house the same?" Indexical sentences, e.g. "And did they?", also occur frequently in the Speech of the US-attorney. 3.2.1.1.2 Interactional Features. A general feature of the US is that the attorney speaks more than the witness. The controller (the attorney) holds the floor effectively. In fact, if the witness violates the "rules" by producing a longer than desired response, the attorney interrupts the response to reassert con- trol. The attorney also employs intimidating minimal responses (e.g. "m—hm," "okay," etc., accompanied by a falling intonation) to encourage short answers from the witness. Underlining highlighting (e.g. Attorney: Which was it, Mr. Wilson, nine or ten o'clock? Witness: Um. 9 o'clock. Attorney: Nine o'clock.), discourse markers (e.g. "Sp, You were looking at him for two minutes?"), and leadings (both single choice questions, e.g. "You gave him the money?"; and multiple choice questions, e.g. "Which was it...nine or ten o-clock?") are also used by a US-attorney. As previously mentioned, there are re- strictions against "leading" the witness and an attorney must use these items cautiously. Information used in multiple choice and single choice questions must have been previously introduced as relevant to the trial. 29 In the US, an attorney often violates the feature of internal chronological consistency. 3.2.1.1.3 Content Features. Owsley and Scotton (1982) have proposed a set of linguistic features to characterize "powerful language." I will use their terminology to describe the inventory of content features that the US-attorney has available for use: marked challenges (e.g. "And yet, you just identified the defendant for the court."), appeals to higher authorities (e.g. "In previous testimony you stated that you couldn't see the defendant clearly."), confrontative hedges and two types of synthesis: (1) coloring--especially style shifting from an informal to a formal style (e.g. Witness: You mean by my truck? (informal) Attorney: Yes, by your vehicle. (formal); and (2) maximizers--exclusively of the intensifier type (e.g. "precisely" or "absolutely" as in "Precisely, where was the bush located?"). 3.2.2 Effects of Strategy Usage on Witnesses. The sub-hypothesis of this study asSerts that an attorney's use of a particular Strategy will have a noticeable effect on a witness and, consequently, on the jury and that these effects will be manifested lin- guistically and assigned unique, consistent values by the jury members. Furthermore, these effects will be obvious to a linguistically naive, but communicatively competent member of the speech community. In this section the linguistic features of witness speech are dis— cussed. In many cases the relevant features of witness speech mirror the features of attorney Speech. However, the analysis of the descrip— tion of witness Speech features is not yet as advanced as the analysis 30 of attorney speech. It is possible that there are additional features which through future research may be found to further define witness speech choices. Some features outlined by Danet et a1. porated into the list of potential features. (1980) are incor- The same major categories used to describe attorney Speech are used in the description of witness speech, namely: Features, (2) Interactional Features and (3) Content Features. (1) Structural These features are presented in Table 3-2 for easy reference and are described. below. Table 3-2. Strategy Use. Features of Witness Speech--Effects of FEATURES SS US I. Structural Features A. Phonological Features B. Lexical Choices -reduced pro- nunciation -unmarked intonation -informa1 word choice -rare use of formal style of address -full forms -marked intonation -formal style of address -formal word choice -some legal- ese C. Syntactic Features -1ength-long -length- Short II. Interactional Features -W speaks -W speaks more than A less than A -W interrupts A III. Content Features —response -response -supplemen- only tary material -hesitations -hedges W = witness A = attorney 31 3.2.2.1 Effects of the Supportive Strategy. The attorney's use of the SS Signals solidarity or connectedness to the witness, i.e., that the attorney and the witness are working toward a common goal. In turn the witness appears relaxed and cooperates with te attorney; this type of presentation by the witness is expected to correlate with a high credibility rating. What follows is a description of the effect of the SS on witness speech. 3.2.2.1.1 Structural Features. 3.2.2.1.1.1 Phonological Features. The witness responds to the attorney's use of reducedypronunciation by also using reduced forms. A common reduction employed by witnesses but npp often used by attorneys is [gin] for [3151’ e.g. "coming" as Ecomin]. Phonologically reduced forms are a feature of an informal style. The witness's informality is a reflection of the informal style used by the attorney. The witness also uses unmarked intonation, i.e., falling intonation at the end of declaratives and a high-low contour pitch at the end of questions. 3.2.2.1.1.2 Lexical Choices. In keeping with the informal style being used, the witness most often makes informal word choices, e.g. "guy" instead of "defendant." The witness rarely uses "sir" or "ma'am" which would indicate the use of a formal style of address. 3.2.2.1.l.3 Syntactic Features. The only relevant syntactic feature for witness Speech is response length. If the SS is used, the witness will react with long responses. 32 The use of long responses indicates that the witness feels at ease, and does not feel limited to a "yes/no" response. 3.2.2.1.2 Interactional Features. Of those features delineated for the attorney, the only relevant feature for witness speech is length of turn, i.e., the witness speaks more than the attorney; this feature again illustrates the fact that flu witness feels comfortable speaking in the courtroom environment. 3.2.2.1.3 Content Features. For witness Speech, the most important content feature is "respon- siveness," i.e., whether a witness's response actually answers the question asked. When the SS is used, the witness is resppnsive, i.e., answers the question (e.g. Attorney: Could you see that night? Witness: Yes, I could see fine.) and also supplies supplementary material. Supplementary material is extra information, not directly asked for by the questioner, e.g. Attorney: Do you know anything about guns? Witness: Yeah, I've worked with guns since I was a kid. 3.2.2.2 Effects of the Undermining Strategy. To a witness, an attorney's use of the US Signals that the attorney and the witness are pp; equals and that the attorney does not intend to "come down to the level" of the witness. The witness reacts by being hesitant and uncooperative; I suggest that this type of witness will receive a low credibility rating. In this subsection the effects of the US on witness speech is presented. 33 3.2.2.2.1 Structural Features. 3.2.2.2.l.l Phonological Features. While a witness might employ reduced forms, in general she/he uses full forms. Full forms appear in witness speech in response to their use in attorney speech. The use of full forms is a general feature of the formal style, characteristic of the US. The witness may also employ marked intonation patterns, especially the ending of statements with a rising intonation. The use of a rising intonation on syntactic declara- tives has been noted in the Speech of women and other "less than power- ful" individuals (Lakoff, 1973). 3.2.2.2.l.2 Lexical Choice. In response to the US the lexical choices of the witness are made with an attempt to approximate the formal style of the attorney. There- fore, the witness uses formal word choices and may even include legal terms in her/his speech. A witness also uses a formal Style of address, e.g. "sir,' when addressing courtroom officials. 3.2.2.2.1.3 Syntactic Features. Syntactically, a witness reacts to the US by giving short responses, often of five words or less. Danet et al.(1980) have demonstrated that a correlation exists between the use of coercive question types (yes/no questions) and the witness's use of short answers. 3.2.2.2.2 Interactional Features. The witness speaks less than the attorney in terms of individual turn length, This feature relates to the witness's use of short responses. 34 The possibility of interrnppions exist as a feature of witness Speech. A witness may interrupt an attorney as a reaction to an attorney who consistently interrupts the witness. 'This action on the part of the witness maybe interpreted as an assertion of speaking rights. 3.2.2.2.3 Content Features. In reaction to the SS, a witness provides pnly a response, i.e., supplementary material is not given. Again, this brevity relates to the attorney's use of coercive question types and is indicative of the fact that the witness feels constrained regarding the permitted length of responses. The US witness also uses hedges when giving testimony, e.g. Attorney: Did you see the defendant clearly? Witness: Yes, but it was kind of hard." The hedges indicate to the jurors that the witness is uncertain of the content. A further in- dicator of uncertainity is the witness's use of hesitations or pauses. 3.3 Report of Interview Findings. The purpose of this section is to present the data gathered during theinterview process through the use of the questionnaire. First, in- formation about the research design is restated. Second , the predicted results of the study are indicated. And lastly, the actual collected data are presented. The latter are given in three forms: (1) raw numbers; (2) Z-values--as the result of applying a Z-test; and (3) F(z) values--which are used to determine the degree to which the results are significant. 3.3.1 Research Design. The subjects were drawn from a population of students enrolled in Communications lOO—-an introductory course offered by the Department of 35 Communication, MSU. The 71 subjects were self-selected, i.e., students enrolled in COM 100 obtained extra credit for the course by participating in the Study. All subjects were native speakers of American English and had no Special training in linguistics. Subjects were told they would hear two tapes, each of which con— tained an interaction between an attorney and a witness. They were given some background information as a lead-in to the tapes. They were instructed to imagine themselves in the role of "jurors" and that after listening to the tapes, they would be asked to fill out a questionnaire. (For a transcript of the instructions, tapes and background information ad for a copy of the questionnaire, see Appendix .) After listening to the set of tapes, one of which represented the SS, the other the US, the subjects were asked to record their personal value judgments to the tape content on the questionnaire. The ques- tionnaire consisted of three parts: (1) 22 items related to the attor— neys, (2) ten items related to the witnesses and (3) two items related to the type of examination that the exchange exemplified. Test items consisted of descriptive adjectives and the subjects were required to circle the number of the tape which most typified the characteristic indicated by the test item. 3.3.2 Predicted Results. The main hypothesis that linguistically naive but communicatively competent members of a speech community consistently associate specific values with the use of a particular strategy was operationalized. The operationalizing consisted of predicting that subjects would asso- ciate Specific items of the questionnaire with a unique strategy choice. 36 This prediction was based on the assumption that there exist two stra- tegies that attorneys use in the interrogation of witnesses and that the subjects were capable of recognizing their use. A second assumption was that the descriptive adjectives used as test items reflect the values that the speech community assigns to the use of the strategies. For Part I- Attorneys, it was predicted that the following items would be assigned to the SS-using attorney: friendlier (2), slower (5), more polite (7), more effective (9), more indirect (10), more supportive (11), more organized (12), more patient (15), more flexible (16), more wordy (20) and more experienced (22). The following items were expected to be assigned to the US—using attorney: more aggressive (l), more distant (3), more professional (4), more obnoxious (6), more efficient (8), more in control (13), more goal-oriented (14), more direct (17), more inti— midating (18), more confusing (l9) and better educated (21). For Part II-Witnesses, it was predicted that the assignment of the following items to the SS-affected witness would occur: more co- operative (4), more wandering (2), more at ease (6), more credible (7) and more clear (8). The following items were designated as likely to be assigned to the US-effected witness: more formal (3), more asser- tive (l), more confused (5), more defensive (9) and more uncomfortable (10). In Part III-Type of Examination, it was predicted that the SS would be interpreted as typical of a direct examination and the US would typify a cross examination. Table 3-3 is a list of the predictions that were made relative to expected subject choices. 37 Table 3-3. Expected Responses to Questionnaire. Section of Questionnaire SS US ATTORNEYS: friendlier (2) * more aggressive (1) slower (5) more distant (3) more polite (7) more professional (4) more effective (9) more obnoxious (6) more indirect (10) more efficient (8) more supportive (11) more in control (13) more organized (12) more goal-oriented (14) more patient (15) more direct (17) more flexible (16) more intimidating (18) more wordy (20) more confusing (19) more experienced (22) better educated (21) WITNESSES: more cooperative (4) more formal (3) more wandering (2) more assertive (l) more at ease (6) more confused (5) more credible (7) more defensive (9) more clear (8) more uncomfortable (10) EXAMINATION: direct examination (2) cross examination (1) *Numbers refer to the item number on the questionnaire. 38 3.3.3 Actual Findings. In this section I will present the actual findings of the interview portion of the study. The raw data of the subjects' choices are pre- sented in Section 3.3.3.1. In Section 3.3.3.2, these choices are tested for Significance by the use of a z-test and the results of the test are stated. In addition, the F(z) values, which are the likelihood of item choice on a scale from O to l, are also presented. 3.3.3.1 Tabulation of Raw Data. After completing the interview, the subjects' responses were individually tabulated for each item. Table 3-4 contains the raw data for Part I--Attorneys of the questionnaire in table form. "Set 1" was the set of tapes that were inflected for emotion and had an N of 46; "Set 2" was the set of tapes that were uninflected for emotion and had an N of 25. "SS" is the column heading under which the subjects' choices of the Supportive Strategy are tabulated. "US" is the heading under which the subjects' choices of the Undermining Strategy are listed. "NA" is the title of the column under which the number of "No Answer"'s are recorded. Table 3-5 contains the data for Part II- Witnesses and Table 3-6 contains the data for Part III-Type of Exami- nation. From a cursory examination of the raw data, it is apparent that on many of the items the subjects were able to make clear and consis- tent choices for both Set 1 (inflected for emotion) and Set 2 (unin- flected for emotion) tapes. One example of a clear, consistent choice is seen in Part I-Attorney (Table 3-4), Set 1 tapes where "more aggres- sive" (l) was chosen by subjects as representing the 88 only Six times but was chosen as belonging to the US 40 times. 39 Table 3-4. Tabulation of Interview Results, Part I--Attorneys. SET 1 SET 2 inflected for emotion uninflected for emotion ITEM (N = 46) (N = 25) says all says 1-more aggressive 6 4O 6 19 2-friendlier 41 4 18 7 3—more distant 10 35 1 9 l6 4-more professional 15 30 1 6 19 5-slower 37 99 19 6 6-more obnoxious 4 42 8 l7 7-more polite 42 4 17 8 8-more efficient 20 25 l 10 15 9-more effective 26 20 13 12 lObmore indirect 32 14 16 9 ll-more supportive 40 7 18 7 12-more organized 18 27 l 10 15 13-more in control 14 32 8 l7 14-more goal-oriented 10 35 l 6 19 15-more patient 44 2 19 6 l6-more flexible 43 1 2 20 5 l7-more direct 12 34 3 22 l8—more intimidating 3 43 7 18 l9-more confusing 13 32 l 11 14 20—more wordy 26 20 15 10 21-better educated 23 22 l 9 l6 22-more experienced 17 28 1 9 l6 Table 3-5. Tabulation of Interview Results, Part II--Witnesses. SET 1 SET 2 ITEM inflected for emotion uninflected for emotion (N = 46) (N = 25) says eels l-more assertive 19 27 ll 14 2-more wandering 31 15 15 10 3—more formal 18 28 6 19 4-more cooperative 39 7 14 ll 5-more confused 9 37 9 l6 6-more at ease 39 7 15 10 7-more credible 35 ll 14 ll 8-more clear 27 19 13 12 9-more defensive 7 39 14 ll lO-more uncomfortable 5 41 14 ll 40 Table 3-6. Tabulation of Interview Results, Part III—Type of Examination. 7 SET 1 SET 2 ITEM inflected for emotion uninflected for emotion (N = 46) (N = 25) eye as as l-cross examination 5 41 1 24 2—direct examination 40 6 24 l 3.3.2.2 Testing for Significance. A z-test was used to test the choices made by the subjects for significance. The test was applied to the data as it is presented in Tables 3-4, 3-5 and 3-6. The z-test is useful in a situation where there are only two possible choices for an item as was the case in this study. A significance level of .10 was assumed for the study for the reasons previously discussed; again, it is noted that the choice not to use the more common significance level of .05 was arbitrary and that the choice of either value does not drastically affect the results of the study. The formula for the z-test is: X - NP NPQ z: .5 is the number of observations or times that an item was chosen by the subjects of the Study. ‘N_is the number of total subjects who re- sponded to an item. ‘3 is the determined probability of an item choice. 9.13 the probability that the second item will be chosen; its value is determined by the value of P. 41 The null hypothesis was assumed in the testing and, therefore, P was determined to be .5 or 50%, i.e., it was assumed that if the use of the SS and US was np£_significant then the likelihood that a sub- ject would choose one item or the other (A or B) would be equal to chance or 50% (.50). The formulae for the determination of the values of P and Q (assuming the null hypothesis) are: Ho = p (TA) .5 = P Ho = p (TB) .5 Q As a result of assigning the value of .5 to P the value of Q is set at .5 also. Z-tests were only performed on the larger number of each item. The results of the tests for the raw data presented in Tables 3-4, 3—5 and 3-6 are given in Tables 3-7, 3-8 and 3-9, respectively. The z—values are the output from the z-test formula and are not directly interpretable; they must be converted into F(z)-values which are taken from Statistics textbooks tables of F(z) values. The only inference that can be drawn from z—values is that the higher the z-value the greater the likelihood that the item will be determined to be signi- ficant. The F(z)-values are also given in the above mentioned tables and represent the probability of the choice rated on a scale of O to 1. (To obtain the percentage of probability, one need only multiple the F(z) by 100.) The significance level of .10 that was assumed for the Study means that only those F(z)-values that are .90 (90%) or greater are accepted as indicative of a significant choice; in the tables, these values are marked by a (*). 42 Table 3-7. Z-values and F(z) values for Part I-- Attorneys. SET 1 SET 2 ITEM inflected for emotion uninflected for emotion (N = 46!) (N = 25) SS US SS US 2 F(275 z F(z) z F(z)? z F(E) l more aggressive 5.01 .99* 2.6 .99* 2 friendlier 5.31 .99* 2.2 .99* 3 more distant 3.73@ .99* 1.4 .92* 4 more professional 2.24@ .99* 2.6 .99* 5 slower 4.13 .99* 2.6 .99* 6 more obnoxious 5.60 .99* 1.8 .96* 7 more polite 5.60 .99* 1.8 .96* 8 more efficient .75@ .77 1.0 .84 9 more effective .88 .81 .2 .51 10 more indirect 2165 .99* 1.4 .92* 11 more supportive 5.01 .99* 2.2 .99* 12 more organized 1.34@ .91* 1.0 .84 13 more in control 2.65 .99* 1.8 .96* 14 more goal-oriented 3.73@ .99* 2.6 .99* 15 more patient 6.19 .99* 2.6 .99* 16 more flexible 6.33# .99* 3.0 .99* 17 more direct 3.24 .99* 3.8 .99* 18 more intimidating 5.90 .99* 2.2 .99* 19 more confusing 2.84@ .99* .6 .52 20 more wordy .88 .81 1 0 .84 21 better educated .15@ .56 1.4 .92* 22 more experienced 1.64@ .95* 1.4 .92* Significance level = .10 ! = For Set 1 N=46, unless otherwise noted. @ - For Set 1, N = 45. # - For Set 1, N = 44. * - F(z) is significant at the .10 level. On Part I-Attorneys (See Table 3-7.) of the questionnaire, results for the two sets of tapes were similar. of the 22 items were consistent to a significant degree in the tapes of Set l-1nflected for emotion; these items were: The subjects' assignment of 18 more aggressive (l), friendlier (2), more distant (3), more professional (4), slower (5), more obnoxious (6), more polite (7), more indirect (10), more supportive (11), more organized (12), more in control (13), more goal-oriented (l4), 43 more patient (15), more flexible (16), more direct (17), more intimi- dating (18), more confusing (l9) and more experienced (22). For Set 2-uninf1ected for emotion, the assignment of 17 of the 22 items were found to be at a significant level; these items were: more aggressive (l), friendlier (2), more distant (3), more professional (4), slower (5), more obnoxious (6), more polite (7), more indirect (10), more supportive (11), more in control (13), more goal-oriented (14) more patient (15), more flexible (16), more direct (17), more intimi- dating(18), better educated (21) and more experienced (22). The discussion of the findings is reserved for Chapter 4. On Part II-Witnesses (See Table 3-8.) of the questionnaire, the results for the two sets of tapes were very different. In the emotionally-inflected tapes of Set 1, the subjects' assign- ment of eight of the ten items was consistent to a significant degree Table 3-8. Z-values and F(z) values for Part II-- Witnesses. SET 1 SET 2 ITEM inflected for emotion uninflected for emotion (N = 46 ) ' (N = 25) SS US US 2 F(z) z F(z): z IF(z7* z F(z) l more assertive 1.18 .88 6 .52 2 more wandering 2.36 .99* 1.0 .84 3 more formal 1.47 .93* 2.6 .99* 4 more cooperative 4.72 .99* .6 .52 5 more confused 4.13 .99* 1.4 .92* 6 more at ease 4.72 .99* 1.0 .84 7 more credible 3.54 .99* .6 .52 8 more clear 1.18 .88 .2 .51 9 more defensive 4.72 .99* .6 .52 10 more uncomfortable 5.31 .99* .6 .52 Significance level = .10. * — F(z) is sienifican?’ at the .10 level. 44 The two items whose assignments were not found to be at a significant level were: more assertive (1) and more clear (8). In the tapes uninflected for emotion of Set 2, on the other hand, the assignment of pply two items was consistent to a significant degree; these two significant items were: more formal (3) and more confused (5). Table 3—9. Z-values and F(z) values for Part III-- Type of Examination. SET 1 SET 2 ITEM inflected for emotion uninflected for emotion (N = 46) (N = 25) SS US SS US 2 F(z) z F(z)' z F(z) z F(z) 1 cross examination 5.31 .99* 4.6 .99* 2 direct examination 5.01 .99* 4.6 .99* Significance level = .10. * - F(z) is significant at the .10 level. On Part III-Type of Examination, all items were found to be at a significant level in both Set 1 and Set 2 tapes. (See Table 3-9 for values.) A list of the actual results of the questionnaire for the emotionally inflected tapes, showing the choices that the subjects made, is presented in Table 3—10. It is interesting to compare these results with the expected results from Table 3—3. For the Attorney section (Part I) there are three major discrepancies. More organized (12) and more experienced (22), which were predicted to be SS items, were chosen as US items. And better educated (21), which had been designated as a US item, was chosen as a SS item. 45 Table 3-10. Actual Results of the Questionnaire--Set l, Inflected for Emotion. Section of Questionnaire SS US ATTORNEYS: friendlier (20 more aggressive (l) slower (5) more distant (3) more polite (7) more professional (U) more effective (9)* more obnoxious (6) more indirect (10) more efficient (8)* more supportive (11) more organized (12) more patient (15) more in control (13) more flexible (16) more goal-oriented (1H) more wordy (20)* more direct (17) better educated (21)* more intimidating (18) more confusing (19) more experienced (22) WITNESSES: more wandering (2) more assertive (l)* more cooperative (U) more formal (3) more at ease (6) more confused (5) more credible (7) more defensive (9) more clear (8)* more uncomfortable (10) EXAMINATION: direct examination (2) cross examination (I) * - indicates that the assignment of the item was not consistent to a significant degree at the .10 (90/) level. However, it is listed under the strategy for which it received the majority of choices. For the Witness and the Examination sections of the questionnaire, all the expected results were realized. A list of the actual results of the questionnaire, for the emotionally- uninflected tapes of Set 2, is given in Table 3-11. There are more dif— ferences between these results and the expected results from Table 3-3 than between those of Set 1 tapes and the expected results. In the Attorney section, more organized (12) and more experienced (22) were again assigned to the US rather than the SS. 46 Table 3-11. Actual Results of the Questionnaire-—Set 2, Uninflected for Emotion. Section off Questionnaire SS US ATTORNEYS: friendlier (2) more aggressive (l) slower (5) more distant (3) more polite (7) more professional (4) more effective (9)* more obnoxious (6) more indirect (10) more efficient (8)* more supportive (11) more organized (l2)* more patient (15) more in control (13) more flexible (16) more goal-oriented (1U) more wordy (20)* more direct (17) more intimidating (18) more confusing (l9)* better educated (21) more experienced (22) WITNESSES: more wandering (2)* more assertive (l)* more cooperative (4)* more formal (3) more at ease (6)* more confused (5) more credible (7)* more clear (8)* more defensive (9)* more ' EXAMINATION: direct examination (2) cross examination (1) * - indicates that the assignment of the item was not consistent to a significant degree at the .10 level. tegy that it received the majority of choices for. It is listed under the stra- In the Witness section, more defensive (1) and more uncomfortable (10) were selected as SS items rather than US items. In the Examination section, the expected results were realized. Further discussion of the results is reserved for Chapter 4. 3.4 Conclusion. In this chapter the findings of the study were presented. Defini- tions of the Supportive Strategy and the Undermining Strategy were given and related to the salient features that were outlined in 47 Chapter 2. A description of the effects of the use of the strategies on witnesses was also included. The results of the interviews were reported. The use of a z-test to test for significance of the results was justified and applied to the data to obtain z-values and F(z)— values. The differences existing between the predicted results and the actual results were noted. It was also stated that differences were found to exist between the actual results obtained with the two sets of tapes. Further discussion of the implications of the findings and the differences of predicted and actual results will be given in Chapter 4. CHAPTER IV DISCUSSION OF FINDINGS 4.1 Introduction. In Chapter 3 the findings were presented. In this chapter the findings ofthe study are discussed. The discussionfocuses on (1) the relation of predicted results to actual findings and (2) the relation of the findings to the study hypothesis and sub-hypothesis. The study is then evaluated in light of the obtained results. The evaluation consists of a discussion of the following points: (1) the limitations of the study, (2) suggestions for further research and (3) the theore- tical and practical applications of the study. 4.2 Discussion of Findings. 4.2.1 Relation of Findings to Predicted Results. In Section 3.3.3, some brief remarks were made about the similar- ities and differences between the predicted results and the actual findings. In this section, these remarks are expanded and explanations for discrepancies are offered. The reader is referred to Table 3-3 (p. 37) for the predicted results and to Tables 3-10 (p. 45) and 3-11 (p. 46) for the actual findings of the tapes inflected for emotion and uninflected for emotion, respectively. 'At this point a general comment can be made in relation to those questionnaire items which were not determined to be significant. Unless there is a comment to the contrary, it is assumed that the insignificance 48 49 of an item is due to the imprecise nature of an item, e.g. an ambiguity inherent in the item; for example, subjects may have interpreted "more wordy" (2) on (1) the sentence level, i.e., using long sentences when shorter ones would have sufficed--a possible feature of either the SS or the US or (2) the discourse level, i.e., the attorney speaks more than the witness--a US feature. 4.2.1.1 Tapes Inflected for Emotion. In Part I-Attorneys, there were interesting differences between the assignments made by the subjects for thestrategiesand the predict— ed results. First, regarding the SS, subjects consistently assigned seven of the 11 predicted SS features to the SS; and, therefore, described the SS-using attorney as friendly, slow, polite, indirect, supportive, patient and flexible. The relationship between these items and the concept of solidarity seems readily apparent. One possible exception is slow, Perhaps the connection is that one is willing to go slowly when discussing something unfamiliar with a "co-conspirator." Of the remainingfbur predicted SS-items, subjects selected two as SS-items but not to a significant degree; these were: more effective (9) and more wordy (20). The last two predicted SS-items, more experienced (22) and more organized (12), were consistently assigned to the US by the subjects. One unexpected result was that a predicted US-item, better educated (21), was assigned to the SS, but its level of assign- ment was not significant. Next the US, subjects consistently assigned nine of the 11 pre- dicted US-items to the US at a significant level. So, the defining 50 characteristics of the US-using attorney are: aggressive, distant, obnoxious, in control, intimidating, goal-oriented, professional, direct and confusing. Again, the relationship between these characteristics andthe salient concept, that of power, seems obvious. Of the other two predicted US-items, more efficient (8) was assigned to the US but not at a significant level and better educated (21) was assigned to the SS. Unexpectedly, two SS—predicted items were assigned to the US: more organized (12) and more experienced (22); both attained significant levels. The fact that a large number of test items were consistently assigned at a significant level to a particular strategy provides sup- portive evidence for the main hypothesis; that is, linguistically competent, but naive members of the speech community did, indeed, make differ- ential value judgments about the attorneys' characters based on their strategy choice. The similarities between the predicted and actual results provide supportive evidence for the validity of the test items. The differences between predicted and actual results are worthy of closer scrutiny. Only the differences will be discussed where an item was assigned to the opposite than predicted strategy at a significant level. More organized (12) and more experienced (22) were predicted SS- items but the subjects assigned them to the US at a significant level. If I were rewriting the questionnaire, I would reconsider the initial assignment of these items. It is easy to see the relationship between these two items; experienced people are often well organized. It is possible that subjects associated these values with being "in control" (13) and "goal-oriented" (l4). 51 In Part II-Witnesses there were no significant deviations from theexpected results. The only differences were in the level of signi- ficance of two items. In the SS, four of the five items predicted as SS—related were assigned to the SS at significant levels. The SS-affected witness was described as wandering, cooperative, at ease and credible. All of these words describe the average individual in a conversational set- ting among friends. More clear (8) was assigned to the SS category also, but not at a significant level. Of the five items predicted to be US-related, four were determined as significant. The US-affected witnesses were described as formal, confused, defensive and uncomfortable. All of these types of reactions describe how one might feel when facing a more powerful individual. While more assertive (1) was assigned to the US, its assignment level was not significant. In Part III-Types of Examination, the predicted results were realized. Subjects chose the SS as an example of a direct examination and the US as an example of a cross examination. In summary, it is possible that the differences between predicted and actual results are related to an initial misassignment of the items. 52 4.2.1.2 Tapes Uninflected for Emotion. In Part I-Attorneys, there are some discrepancies between the expected and actual results; they are described below. Regarding the SS, the same seven of the eleven predicted items,that were rated as significant for the emotionally-inflected tapes’were con- sistently assigned to the SS at a significant level. Once again the SS-using attorney was described as friendly, slow, polite, indirect, supportive, patient and flexible. Of the remaining four predicted SS— items, two were assigned to the SS, but not at a significant level: more effective (1) and more wordy (2). The other two were designated as US-related: more organized (12) and more experienced (22). Related to the US, all 11 of the predicted US-items were assigned to the US, but only nine of the 11 attained a significant level. The US-using attorney was, therefore, described as aggressive, distant, professional, obnoxious, in control, goal-oriented, direct, intimidating and better educated. The other two predicted US-items were not assigned a significant rating: more efficient (8) and more confusing (19). I suggest that more efficient (8) was not determined as significant partly because of the smaller N used to the emotionally-uninflected tapes. No explanation is offered for the lack of significance of more confusing (19) at this time. Two predicted SS-related items were assigned to the US, with more experienced (22) being determined as significant,and more organized (12) being defined as not significant. I again suggest that the lack of significance, in this case of more organized (12), is due in part to the study's small N. In Part II-Witnesses, there are a large number of discrepancies between the predicted results and the actual findings. 53 For the SS, all predicted five SS-related items were assigned to the SS, but none attained a significant level. It is possible that the insignificance of two of these items (more wandering (2) and more at ease (6)) may be due to the study's small N as their F(z)—values were both .84. The explanation for the insignificance of the other items is given below in Section 4.2.2. In addition, two items that were predicted as US-related were assigned to the SS: more defensive (l) and more uncomfortable (10); neither of these items was determined to be significant. For the US, only three of the predicted five items were assigned to the US by the subjects; and of these only two attained a sufficient level of significance: more formal (3) and more confused (5). More assertive (l) was not found to be significant. As noted above, sub- jects assigned the remaining two predicted US-items to the SS: more defensive (l) and more uncomfortable (10). In Part III-Type of Examination, the predicted results were realized. The use of the SS was determined to illustrate a direct examination and the US was taken as an example of a cross examination. In summary, a number of discrepancies were found between the predictedresults and the actual findings based on the tapes of Set 2, which were uninflected for emotion. Some of the discrepancies were related to an initial misassignment of an item. Others, e.g. the“ lack of significance of some items, were attributed to the study's small N. 4.2.2 Relation of Findings from Tapes Inflected for Emotion to Findings from Tapes Uninflected for Emotion. In two sections of the questionnaire, Part I-Attorneys and Part II-Witnesses, there are differences in the actual findings obtained 54 from the tapes of Set l--inflected for emotion and the tapes of Set 2-— uninflected for emotion. In this section, explanations are offered for these discrepancies. In Part I-Attorneys, there are three test items which were realized differently in each of the set of tapes; these were: more organized (12), more confusing (19) and better educated (21). More organized (12) was assigned to the US at a significant level (.91) when the emotionally-inflected tapes were used. When the unin— flected for emotion tapes were used, the item was still assigned to the US but the assignment level was not significant. This difference is attributed to the smaller N obtained for the Set 2 tapes. More confusing (19) was also assigned to the US at a highly signi— ficant level ().99) with the use of Set 1 tapes. The level of signifi— cance dropped to chance (.52) with the use of Set 2 tapes. It is possible that in order to judge an individual (e.g. a witness) as being confused that intonational or emotive content is important. When the emotive content was eliminated from the witness's speech in the Set 2 tapes, it is reasonable to suggest that it became impossible for the subjects to consistently determine whether the attorney was "confusing" the witness. Better educated (21) was assigned to the SS when subjects heard the Set 1 tapes; the level of the assignment was close to chance (.56). However, when the Set 2 tapes were played, the item was assigned to theUS at a significant level (.92). Again, it is likely that the presence and absence of emotive and intonational clues affected the assignment of this item. In Set 1, the subjects rated the attorneys 55 as equally well—educated. The "warm, friendly" tone of the SS-using attorney may have offset the use of informal English and slang. At the same time, the ' 'unfriendly" tone of the US-using attorney may have ne— gatively affected the listener's evaluation of the user's level of education despite the use of formal English and legalese. On the other hand, when the vocal quality was neutralized in Set 2 tapes, it is reasonable to suggest that the attorneys' actual lexical choices may have become more important to subjects. As a result of the Set 2 tapes, the US-using attorney was designated as significantly "better educated" (21). In Part II-Witnesses, the results from the two sets of data are different for six items: more wandering (2), more cooperative (4), more at ease (6), more credible (7), more defensive (9) and more un- comfortable (10). As indicated in 4.2.1.2 above, the insignificance of SS-items (more wandering (2) and more at ease (6)) in Set 2 tapes may be due to the smaller N obtained for that portion of the study. The other four items (more cooperative (4), more credible (7), more defen- sive (9) and more uncomfortable (10)) were all marked as highly signi- ficant in Set 1 tapes (F(z) =‘>.99) and as random choices in Set 2 tapes (F(z) = .52). Again, I suggest that the emotive content present in Set 1 tapes and absent from the Set 2 tapes accounts for the different evaluation of these items. In summary, discrepancies were found in the two sets of actual findings Most discrepancies were explained in one of two ways: (1) some items were found not to be significant for Set 2 tapes because of the smaller N used in that portion of the study; and (2) some items 56 l were not determined as significant for Set 2 tapes because of the absence of vocal quality or emotive content. 4.2.3 Relation of Findings to Study Hypothesis and Sub-Hypothesis. In the study, subjects made consistent choices with regard to the predicted results for the majority of items on Part I of the question— naire. Furthermore, most of these choices were determined to be signi— ficant by the use of a z-test at a .10 significance level. This information means that the study hypothesis has been supported; that is, subjects were able to make consistent value judgments related to the characteristics of the attorney based on the use of the strategies and the value judgments differed across the two strategies. However, the results relating to the sub-hypothesis were less conclusive; that is, the subjects were not consistent in their assign- ment of values to a majority of the items in Part II of the questionnaire. 4.3 Evaluation of the Study. In this section I will examine the merits of the study in relation to the obtained results. The evaluation of the study includes: (1) a discussion of the limitations of the study and (2) suggestions for further research. 4.3.1 Limitations of the Study. The major drawback to this study is the manner in which the scripts for the tapes were constructed. While the scripts were based on an actual trial transcript, they were reworked; as such,the scripts do not represent naturally occurring language from the courtroom environ- ment. However, it was deemed necessary to manipulate the speech 57 presented in the transcripts (l) to construct ideal examples of the strategies in use by including only the defined characteristics of the strategies and (2) furthermore, to include a large number of those features in each of the seven minute interactions to meet the time constraints imposed by the interview situation. For a future study, it would be valuable to obtain tapes of actual courtroom language and to play them for subjects. A second drawback to the study is that two variables were tested at the same time: (1) the attorney's use of the strategies and the manner in which their use affects the juror's opinion of the attorney and (2) the witness's reaction to the attorney's use of the strategies and the effect of that reaction on the juror's credibility rating of a witness. It would be interesting to attempt to hold one of the participant's speech constant, while manipulating the other. However, this suggestion also has a serious drawback-—it would not represent natural speech because, in fact, neither individual's speech remains constant when the speech partner's changes. A third problem was the choice of the items for the questionnaire, particularly in relation to Part II-Witnesses. In view of the results that were obtained from the emotionally-uninflected tapes, the items appear to have relied heavily on emotive content for their identifi- cation. It is possible that these items reflect the actual situation in the courtroom. However, it might be interesting to rewrite por- tons of the questionnaire and to choose new items carefully. It is aso possible that this section of the questionnaire was less precise because it contained fewer items. A future study should be constructed that contains a minimum of 20 items. Another possible explanation for 58 the failure of the sub-hypothesis to gain support may be that the characteristics used to define witness speech choice may not have been sufficiently delineated. More attention should be paid to constructing a more precise definition of witness speech features. Another minor difficulty of the study was the small N obtained for the questionnaire dealing with Set 2-the tapes uninflected for emotion. A larger N could have made a difference in the determination of some items' significance. Also a larger N for both parts of the study would have permitted the use of other statistical tests. The z-test, while adequate, is not a very sophisticated test. 4.3.2 Suggestions for Future Research. The limitations noted above provide some hints on where future research efforts may be concentrated. Studies should be conducted using natural language data from the courtroom, possibly even playing recordings of actual courtroom interactions. Also studies should be conducted whichtest only the variable of attorney strategy choice or the variable of witness speech features and their effects on jurors' perceptions of witnesses. The present study could be rerun—after the second part of tfluaquestionnaire had been reworked as suggested and the features of witness speech had been more clearly defined. In the present study, subjects were required to make a choice of item A or B while comparing two tapes. It would be interesting to see whether the results would differ significantly if a single tape were played and subjects were asked to either (1) assign a quality (e.g. "polite" or "not polite") to the strategy exemplified on the tape or (2) rank the linguistic interaction on a continuum (e.g. "polite 100% O 100% not polite"). It would also be revealing to 59 randomly select a sub-group of subjects for more indepth oral interviews. The oral interviews could consist of illiciting the "why's" of 3 subjects choices on the questionnaire. More effective was one of the items in the attorney section of the questionnaire. It was interesting that the subjects of this study did not assigned at a significant level to either the SS or the US. Therefore, it could be suggested that the subjects were unable to determine whether one strategy or the other was more effective in obtaining the desired outcome of a trial. An observational study could be made of a number of attorneys to determine whether a correlation exists between the use of one strategy and the winning of a case. Attorneys would have to be ranked on the percentage of features employed from a particular strategy. It is assumed here that the occurrence of a pure SS or a pure US is rare. The subjects of the present study determined that the US is used by a more experienced attorney. Again, an observational study could be made to see if a correlation exists between the use of the US and the amount of time spent defending or prosecuting cases in the courtroom. 4.4 Applications of the Study. Within the field of sociolinguistics there is a growing interest in the subject of linguistic interaction in the professional context, especial- ly in the medical and legal professions. For the most part the work that has been done on the legal profession has focused on the witness and how the witness's credibility can be effected by the type of language the witness uses (Erickson et al., 1977). Loftus and Zanni (1972) have dealt with the lawyer's linguistic ability to shape the witness's testimony but they limited their research to vocabulary choice. Danet et a1. (1980) .45 a v ti . .5 u; . u v u "- 60 have investigated courtroom linguistic interaction with an ear to the use of coersive and non-coersive question types and their effect on witness response. Danet et a1. and the present study go beyond the lexical and syntactic choices made by an attorney and attempt to describe the motiva- tion for the use of the linguistic forms. In the present study, I have attempted to show that the attorney's motives and roles (defender/friend or prosecutor/enemy) are reflected in linguistic choices (the US and the SS). Thus, one of the contributions which this study has striven to make is to supply data on a previously under researched area, namely, a lawyer's overall strategies in witness interrogation. The findings of the study may also be used to address the issue of juror interpretation of attorney speech. The subjects of the study iden- tified the user of the SS as a "nicer" person than the user of the US. It is possible that an attorney could be instructed on how to manipulate a jury's opinion of him/herself by varying the linguistic features of her/ his speech. Furthermore, it is possible that by manipulating a jury's opinion of him/her personally, the attorney may also be able to manipulate the jury's assessment of the facts of a case and improve the chances of "winning." As Danet et a1. (1980) have pointed out, the authors of legal textbooks agree that the goal of an attorney representing a client is not to make the whole truth known but rather to win the case. Continued research into the linguistic workings of the courtroom situation can provide attorneys with concrete information on how language choice can effect a jury's decision. 4.5 Conclusion. In this study, I have asserted the existence of two strategies available to attorneys in the interrogation of witnesses in the courtroom. 61 These strategies are called the Supportive Strategy (SS) and the Undermining Strategy (US). The strategies were defined in term of three major linguistic feature categories: structural features, interactional features and content features. Supportive evidence for the existence of the strategies was obtained through the use of a questionnaire/interview study. The study and its findings have implications in the field of professional language use and the training of attorneys for effective language use in the courtroom. REFERENCES REFERENCES Crosby, Fay; and Nyquist, Linda.The female register: an empirical study 1977 of Lakoff's hypotheses. Langugge in Society, 6, 313-322. Danet, Brenda; Kermish, Nicole; and Rafn, H. Jeffrey. A method for the 1976 study of courtroom questioning. Unpublished Paper. Danet, Brenda; Hoffman, Kenneth B.; Kermish, Nicole C.; Rafn, H. Jeffrey; 1980 and Stayman, Deborah G. An ethnography of questioning in the courtroom. IN Language Use and the Uses of Lagguage edited by Roger W. Shuy and Anna Shnukal. Washington, DC: Georgetown University Press, pp. 222-234. Erickson, Bonnie; Lind, E. Allan; Johnson, Bruce C.; and O'Barr, William. 1977 Speech style and impression formation in a court setting: the effects of "power" and "powerless" speech. Law and Language Project Research Report 13. Duke University: Law and Language Program. Fishman, J. A. The sociology of language. IN Language and Social 1972 Context edited by Pier Paolo Giglioli. London: Cox & wyman Ltd., pp. 21-44. Gumperz, John J. Introduction. IN Directions in Sociolinguistics: the 1972 ethnography of communication edited by John J. Gumperz and Dell Hymes. New York: Holt, Rinehart & Winston, Inc., pp. 1-250 Hymes, Dell H. The ethnography of speaking. IN Readings in the Sociology 1968 of Language edited by Joshua A. Fishman. The Hague: Mbuton & Co., pp. 99-137. Hymes, Dell H. Toward ethnography of communication: the analysis of 1972 communicative events. IN Language and Social Context edited by Pier Paolo. London: Cox & wyman Ltd., pp. 21-44. Joos, Martin. The Five Clocks. New York: Harcourt, Brace & WOrld, Inc. 1967 Lakoff, Robin. Language and woman's place. Language in Society, 2, 62 63 Loftus, E. R.; and Zanni, G. Eyewitness testimony: the influence of the 1972 wording of a question. Bulletin of the Psychonomic Society, 5, PP. 86-88. Minderhout, David. The entrepreneur's use of language. IN Sociolinguis- 1972 tics in Cross-cultural Analysis edited by David Smith and Roger W. Shuy. Washington, DC: Georgetown University Press, pp. 57-669 Owsley, Heidi; and Scotton, Carol Myers. What's my line? Conversational 1982 expression of power by tv interviewers. A paper. Scotton, Carol Myers; and Ury, William. Bilingual strategies: the social 1977 functions of codeswitching. International Journal of the Sociolggy of Language, 13, pp. 5-20. Terrace, Herbert; and Parker, Scott (eds.). Psychological Statistics, 1971 Vol. III and Vol. VI. San Rafel, CA: Individual Learning Systems, Inc. APPENDIX INSTRUCTIONS READ T0 INTERVIEW SUBJECTS You will hear two tapes where an attorney examines a witness regarding the facts of a case. You are a juror. Listen to the tapes closely. After listening to the tapes, I'll give you a questionnaire to fill out. Here's some background information: The witness is Mr. Jonathan Charles Wilson. He's an employee at Triominos Pizza Shap. He attempted to deliver a pizza at a house in North Lansing on November 6, 1980. The peeple living in the house told him that they didn't order a pizza. They then let him into their house to use the phone. He needed to verify the address on the pizza box. When he called, he got a doctor's office on West Michigan. We join the testimony at this point. (Turn on tape. After tape is over. Read instructions from ques- tionnaire. Instruct subjects to:) Fill out the questionnaire as quickly as you can. I'm interested in your first impressions. 64 65 ( ) QUESTIONNAIRE Interviewee No. You have just listened to two tapes, each of which contained an examination of a witness by an attorney regarding the facts of a robbery. ATTORNEYS The following list relates to your personal reactions to the attorneys. Circle the number of the tape on which you would describe the attorney as: TAPE TAPE 1. more aggressive l 2 2. friendlier l 2 3. more distant l 2 4. more professional 1 2 5. slower 1 2 6. more obnoxious 1 2 7. more polite 1 2 8. more efficient l 2 9. more effective 1 2 10. more indirect 1 2 11. more supportive l 2 12. more organized 1 2 13. more in control 1 2 14. more goal-oriented l 2 15. more patient 1 2 16. more flexible 1 2 17. more direct l 2 18. more intimidating 1 2 19. more confusing l 2 20. more wordy 1 2 21. better educated l 2 22. more experienced I 2 66 Questionnaire ( ) Page 2 Interviewee No. WITNESSES The following list relates to your personal reactions to the witnesses on the two tapes. Circle the number of the tape on which you would describe the witness as: TAPE TAPE 1. more assertive 1 2 2. more wandering l 2 3. more formal l 2 4. more cooperative l 2 5. more confused 1 2 6. more at ease 1 2 7. more credible 1 2 8. more clear 1 2 9. more defensive l 2 10. more uncomfortable 1 2 TYPE OF EXAMINATION Circle the number of the tape on which you would describe the examination as an example of: 1. cross examination (the witness l 2 and the attorney represent opposite sides of the case) 2. direct examination (the witness l 2 and the attorney represent the same side of the case) 67 SCRIPT OF UNDERMINING STRATEGY WITNESS: I got a doctor's office on West Michigan. LAWYER: Was the address on the pizza box the same as the address on the house that you went to to deliver the pizza? Yes, sir. You said the peOple did not recognize the phone number? Yes, sir. And you said they did not recognize the name? Yes, sir. And what did you do after that? (slight pause, hesitant start) Well, I took the whole thing as more or less a practical joke, just a bad delivery. And it was at that point that you left the premises? Yes, sir. And then what? I walked back to where my truck was parked. (Pause.) Near a bush on the side of the house. Stop there for a moment. Where had you parked your vehicle? Ah, in a vacant lot right beside the house. (Pause) In a driveway. And there was a bush near your vehicle? Yes, sir. Precisely, where was the bush located? Um. Right in front of my t-vehicle. Directly in front of your vehicle. Was there a street light in the vicinity? Yes, sir. 68 And exactly where was it located? Ah. Straight across the street from the house. Tell us, if you can recall, did you have any difficulty seeing that night when you returned to your vehicle? No, I could see fine. Would you describe it as being well lit at that time? Yes, sir! Continue with your description of what happened that night. Well. (Pause.) After I thanked the peeple I started back to my tru-, my vehicle. And as I was heading towards my vehicle, I saw some guy coming out between the vacant lot and a parked car over at the other end of the lot. The next time I saw him was when I had just started around that bush heading for my truck. Step there, Mr. Wilson. You have to speak slower for the record. And then what happened? OKay. (Pause.) Um. Just as I approached the driver's side door on my v-vehicle. This man was standing in back of the truck around the fender area. Was the individual in question carrying anything? Yes, sir. What was he carrying? A.gun. Did he say anything when you saw him? Yes, sir. What did he say ? He said, "Follow me to the backyard." 69 Was the individual in questoon standing on the same side of the vehicle as you were? Yes, sir. How many feet away was he from you when you first saw the gun? Approximately four to six feet. You're certain of that? Yes, about six feet. So he was about six feet away when you first saw the gun? Yes, sir. Is the individual under discussion present in the courtroom today? Yes, sir. Would you point to him? (No verbal response. Points.) Let the record show that the witness indicates Lloyd Jones, the defendant in this case. (Short pause.) Do you know the defendant by name? Yes, sir. Where did you learn his name? It was on my subpoena, and you just said it. Alright. What was he wearing when you first saw him? A light brown color of what appeared to be a leather jacket... (Interrupting.) It appeared to be a leather jacket? Yeah. It WAS a leather jacketaand he also had a nylon pulled down over his face. Mr. Wilson, describe for the court, if you can remember, the type of gun that the defendant had that night. 70 I guessed it to be either a .32 or else a .38 snub-nosed. In which hand did the assailant hold the gun? His right hand. So the defendant ordered you to go into the backyard with him? Yes, sir. Did you accompany him? Yes, sir. Did the assailant say anything else to you? Yes, sir. What did he say? Well. On the way back to the backyard, he asked me a couple of times if I could identify him. And what did you tell him? I just said, "No, no way, I don't know nothin'." And yet, you just identified the defendant for the court? Sure, I recognize him. Then why did you tell the defendant that you could not recognize him? I was afraid he'd shoot me. Did there come a time when you reached the backyard? Yes, sir. What transpired there? Excuse me? What happened in the backyard? Oh. We got back there and then he said, "Okay. Let me have your money." Did you give him any money? W: L: W: L: 71 Yes, sir. How much money did you give him? Thirty six dollars of company money and nineteen dollars of my own. So you gave him fifty-five dollars. (Pause.) Mhmm. What did he say to you after you gave him the money? He said, "Okay. Go on and get out of here." How long did it take for this conversation to take place? Not very long. A minute, maybe two. Two minutes? Yes. And during that time were you looking at the assailant? Yes, sir. So you were looking at him for two minutes. Yes, sir. When you FIRST saw the defendant, how long were you able to look at him? You mean by the truck? Yes, by your vehicle. How long were you able to look at him? Off and on, a couple of times for only a few seconds each time. What prevented you from seeing him for a longer time? Well, every time he looked at me and said something he kept trying to turn his back to me so I couldn't see his face straight on for more than a few seconds at a time. Okay. When he told you to leave the backyard, what were his exact words? 'He said, "All right, go on and get out of here." 72 And did you leave? Yes. Where did you go? I ran back to my truck and started it up. Was there ever another occasion when you saw the defendant? Oh. (Pause.) Yeah. Um, as I was starting to back out of the drive- way I remembered that I had a 250-watt spotlight in my truck that was hooked up to the... (Interrupting.) Mr. Wilson, (pause) when did you see the defendant again? When I turned the spotlight on him. And how long were you able to track him with the spotlight? About four minutes. Four minutes. The person you put the spotlighton was the same person who had confronted you with the gun? Yes, sir. So, would you say that there were three different occasions during which you had the opportunity to see the defendant facing you? Yes, sir. Would you summarize those three occasions for the court? Okay. (Pause.) When I was first at my driver's side door when he first approached me at the back of my vehicle; then in the backyard where the money transaction took place; and then again when I hit him with the spotlight. ' No further questions, Mr. Wilson. I have no further question, Your Honor. 73 SCRIPT FOR THE SUPPORTIVE STRATEGY WITNESS: I got a doctor's office on West Michigan. LAWYER: So, the address on the pizza box was the same as the address on the house where you went to deliver the pizza? Yes, it was. And the people didn't recognize the phone number? No, they didn't. And they didn't recognize the name? No, sir. Could you describe what you did after you made the phone call? (Begins slowly and speeds up.) Well, I took it more or less to be a practical joke or whatever, just a bad delivery. 80, then I thanked 'em for lettin' me use the phone an' I walked back out to the side of the house where my truck was parked, near a bush. Let's stop right there for a moment, please. You're going to have to slow down just a little bit, John, so we can make sure we get all the information you have. (Pause.) Where did you park your truck? In a vacant lot right beside the house, in a driveway. You said something about a bush? Yeah, there was a bush right in front of my truck, where I parked. Was there a street light in the area? Yes, there was. It was right across the street from the house. Did you have any trouble seeing that night when you went to your truck? 74 No, I didn't. Would you describe the area as being well lit at that time? (Emphasis.) Yes, sir! Now, let's go back to where you were walking back to your truck. What happened? Well, as I was headin' towards my truck, I saw this guy comin' out between the vacant building an' a parked car over at the other end of the lot. The next time I saw him was when I got to the driver's side door of my truck. I saw 'im standin' in back of the truck by the bumper. An' this is when he brought the gun up from his side. He had it in 'is hand. He brought it up an' says, "Follow me to the backyard." Okay, let's stop right there for a moment. Was he standing on the same side of the truck as you? Yes, he was. So, how many feet away from you was he when you first saw him with the gun? About four to six feet. Is that person that you're referring to in the courtroom today? Yes, he is. Would you point to him, please? That's him, over there. Let the record show that the witness indicates the defendant. (Pause.) Do you know his name? Jus' from the name bein' on my subpoena is all. Okay. What was he wearing when you first saw him? A light-brown color of what looked like a leather jacket, an' he also had a nylon pulled down over his face. 75 Do you know anything about guns? Yeah. Can you tell me what type of gun he had? I guessed it to be either a .32 or else a .38 snub-nosed. It was nickel-plated. In which hand did he hold the gun? In his right hand. Now, let's start at the point where you're standing by the truck with him four to six feet away, and tell me what happened then? Okay. He ordered me into the backyard an' I said that I wouldn't go but I'd give 'im the money or whatever he wanted. What was his reaction to your saying that? Well, (half laugh) he got a little more perturbed about it an' he come up an' grabbed ahold of the collar of my coat an' brought the gun up to the side of my head an' says, "Hey, listen, don't be a fool, I'll blow your brains out. Don't try an' run, jus' do what I say." So, after that I followed 'im into the backyard. (Pause.) Mhm, would you tell us what happened after that? Well, on the way back there he asked me a couple of times if I could identify 'im, if I knew what he looked like, an' I jus' said, no, no way, I don't know nothin'. Why'd you say that? Well, sure I'm gonna say I'm gonna identify 'im when he's standin' there holdin' a gun on me? (Pause.) Alright. Did you go into the backyard? Yes. What happened after you reached the backyard? 76 We stopped an' then he says, "Okay, let me have your money." I reached into my lef' coat pocket an' handed 'im the money. Then, he says... (Interrupting.) Let's stop there just a moment, please. How much money did you give him? About thirty-six bucks. Okay, what happened after you handed him the money? When I handed 'im the money, he says, "Okay, let's have the rest of it. I know you got to have more than this." I said, "I don't." He says, "I know you're lyin', you got to have more than this." I had about nineteen bucks of my own. So I got that out an' give it to 'im. So, at first you gave 'im thirty—six dollars and then you gave 'im another nineteen. Right. Then, he says, "Okay, go on an' get out of there." Alright. Let's stop right there now. How long did this conversation take? Not very long at all, a minute, maybe two. And during that time, were you looking at him? Yeah, I was. And could you see him clearly? Yes, I could. Now, going back to when you first saw him by the truck, your first meeting with him, how long were you able to look at him there? Off an' on, a couple of times for a few seconds at the start there. Did you get a good look at his face? Yeah, but it was kinda hard. 77 Why was it hard? Because every time he looked at me an' said somethin', at first he kept tryin' to turn his back to me, because of the light across the street, I don't know. But he kept tryin' to turn his back to me so as I couldn't see his face straight on. Mhm. Now, John, let's jump back to where he told you to leave the backyard. What were his exact words? He said, "Alright. Go on an' get out of here." Then I started to walk off an' go, may ten feet, an' he said, "I said get out of here, run!" So, then, I ran back to my truck an' started it up. Did you leave right away? No. As I was startin' to back out I remembered that I had a spot- light in my truck that was hooked up to the battery. I use it to read addresses on houses that sit back from the road. At that point he was still standin' there where I left 'im. Since I knew I could get out fast, I hit the spotlight on 'im. An' then he jus' stood there for a second an' then realized what was goin' on an' then he took off runnin'... (Interrupting.) Let's stop for a moment, John. I want you to talk a little slower. Okay? Mhm. Where did he take off runnin' to? Around the side of the garage an' I followed 'im with the spotlight through the backyard until I lost 'im. The person you put the spotlight on was the same person who had confronted you with the gun? Yes. 78 So, would you say that there were three different times that you saw this man facing you? Yes. Would you go over those three times again, please? When I was first at my driver's side door when he first approached me or whatever there at the back of the truck; then in the backyard when he took the money; an' then again when I hit 'im with the spot- light. Thank you, John. I have no further questions, Your Honor.