"‘9‘. .. —~. , I'II II'I III ”III § III IIIII IIIII IIIIIIIII III II III, IIIIIII' I I I IIIII 33'- I I EI‘II II' II I II I III IIII IIIIIII IIIII II $3.: I I] I II III-i HIIIIII IIII’I'.‘ ”II "III” II III :1"; " ‘III’II IIIIIIIII IIIIII III II III IIIIIIIII‘ I'IIIII IIIIIIII III: III;1 III'"’IIII1I“IIJ’I IIIIIII iIIiI I .‘ISI 83:” II‘IH‘II I I IIIII ”II III I IIIgIII-I, I!!!“ IIIIIIII II‘II II: . :4 ”a IIJIHI'III "“ 1' II “I; I I'IIIIIIIII II'HII'II' IIIIII'I III LIII, III II" III ‘IE I_ -‘ I3; 513' i: I‘ III II II {III IIIIIII I!” IIIIIIII IIIII III I ..II!.. IIIIIIIIII .I.'*_’, III IIIIII ”IIIIII IIIIIIIIIIII IIIIIIIIIIIIIII {II . IIIIIIIIIIIIIIIIIII IIII IIII 'I IIJI'II III III II‘II "III‘ IIII’INIIIII III 1 .' ‘4 MI II“III:1 IIIIIIII III] IIIIII ffififf‘gf‘ E; If? ‘II‘EI 1"; . [EFI7 film «.55 III‘ 45;: w I '4. f -m ‘ {‘J _ I . - ' L YIIV" 4..., A 5%??? .- "‘_‘ < . a-“ "q 3-“ -.. -‘...~_-, ~. g." u. w... —4.. MW“. .m'Ij: "' —..._fi Mm ma ~w. .0. .112: "“n. .3. v.2, n.“ v“ ~— ~. .._ ~— .- M‘, n c...» . -W w. . m m’“; : ' ‘ O W ..” .~ I A “-*-—'~'r-.:....,..~vrd.~—-"= .. n -.I.. . I .. . - a- A .. .- . '— ‘~ 0‘.“ M ‘ "8:521. ~0-a - ._,,‘. cm o ’1‘..-“ «~—- ”A... . M M "m '1‘1. , .IMI . I III .II I EI'IIIIIIIIIII‘I’I‘IIIIII ' ' II II..IIIII ‘II- ‘I'III'IIII'I' ‘ I ' IIIIII'II : IIIIII III II II {It IIIIIHAI‘EII I II'IIII‘I III? III III: III :III'I‘III III III IIIIII I,‘ I IIIIIIIIIIIIIIIIIIIII IIII‘III-‘IIIVI: 'I'III*“"’II" III III ' If: M II III II N. IIIIIIIIIIIII III III IIIII II III'IIIIII I"III‘I”IIIIII IIIIIIIII 'III'II “I” I II‘IIIIII IIIIIIII III II IIH'IIiIJ IIIIIIII :IIIIIII II II IIIIIIIIIII III III I},«I I III IIII III _ III ‘IIIIIII IIIIIIIIIIIIII“ IIIIIIIIII I III I I IIIII’II -_w_;_ m‘V‘MM ~.—: -w- _ III IIIIIIIIIIIIIIII..I- IIIIH i IIIIIII III IIIIII' II IIIIIIIIIIII MI I‘ II III IIIIIIIIIII. _~_______\H Li? I” 3 fi‘r i 1‘_‘ ‘Jili‘flelngit’r F ,1 —-— This is to certify that the dissertation entitled THE EFFECTS OF FREE AND DIRECTED ASSOCIATION CONDITIONS ON VERBAL PRODUCTIVITY AND THE SPONTANEOUS PRODUCTION OF FIGURATIVE LANGUAGE presented by Stuart Lee Doneson has been accepted towards fulfillment of the requirements for Ph.D. Psycho]ogy degree in fl/A’Lyéfl/M Major professor Date 11/4/82 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 iiiiiiiiiiii ,_ RETURNING MATERIALS: IVIESI_] Place in book drop to LIBRARJES remove this checkout from an your record. FINES will . be charged if book is returned after the date deh stamped below. A ~ _ 0 d .35, 23:: EE 3? of H 9 944 9.3030,. JUN122007 MM? ”7 THE EFFECTS OF FREE AND DIRECTED ASSOCIATION CONDITIONS ON VERBAL PRODUCTIVITY AND THE SPONTANEOUS PRODUCTION OF FIGURATIVE LANGUAGE By Stuart Lee Doneson A DISSERTATION Submitted to Michigan State University in partial fuifiiiment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychoiogy i982 bldUJJO ©1983 STUART LEE DONESON All Rights Reserved ABSTRACT THE EFFECTS OF FREE AND DIRECTED ASSOCIATION CONDITIONS ON VERBAL PRODUCTIVITY AND THE SPONTANEOUS PRODUCTION OF FIGURATIVE LANGUAGE By Stuart Lee Doneson The present study undertook to explore the relationships between contexts of association, framed by instructions for Free Associations and Directed Associations, on the fluency and figurativity of speech. It was hypothesized that Free Association conditions would reduce fluency (verbal productivity) and increase figurative productivity, (including the frequency and variety of categories of novel and living) figures). Differential effects of postural and eye closure instructions were also investigated. To test these hypotheses in a factorial design, sixty-four undergraduate subjects (32 male and 32 female) initially participated in one of four treatment groups: 1) Free Association/Reclining eyes closed; 2) Free Association/Upright eyes open; 3) Directed Association/ Reclining eyes closed; 4) Directed Association/Upright eyes open. A Barron's Ego Strength Scale was also administered to each subject. Verbal responses of all subjects were transcribed and scored for verbal productivity (number of words spoken) and for figurative productivity (number of novel and of living figures per 100 words of transcript, and the number of categories of novel and of living figurative language used per lOO words of transcript). Analyses of the results lends strong support to the hypotheses regarding the effects of Free Association on the fluency and figurativity of speech. Free Associating subjects were less verbally but more figuratively productive, i.e., although less fluent, they used a higher frequency of novel figures and a wider variety of novel figurative categories than Directed Associating subjects. Treatment groups did not differ with respect to frequency of living figurative use. While there were no significant differences for Reclining/eyes closed subjects vs. Upright/eyes open subjects, four of the five comparisons were in the predicted direction of less fluency and more figurativity for reclining/eyes closed subjects. Only for the frequency of living figures were the means in a direction opposite to the predictions, and this difference was not significant. Finally, higher Ego Strength was found to be correlated with higher frequencies of living figurative use. The results were discussed in the light of the scope and limits of psychoanalytic and speech act approaches to Free Association and figurative language. This dissertation is dedicated to my wife Susan who shares the hope who keeps the faith Amo Te ii ACKNOWLEDGMENTS If Socrates is right and the best life is the life of dialogue, then I have indeed been blessed with wonderful interlocutors: Professor Albert Rabin has been the fixed point throughout the twists and turns of my graduate career in psychology, the very model of personal decency and intellectual probity. As chairman of my thesis and dissertation committees, he has, over these eight years, kidded, coaxed and wrestled with my penchant for philosophizing. I have lost count of the many drafts of this dissertation that he labored over in helping me bring a clear and coherent piece of research to fruition. It has been an honor to have him as a teacher. It is a privilege to call him a friend. Professor Charles Hanley provided the statistical guidance for this research. Not only did he contribute bright ideas and inspired solutions to statistical dilemmas, but couched them all in good humor and tremendous erudition. Whether he was trying out novel analyses on his desk calculator or expanding my knowledge of the history of psychology, warfare, European manners or Freud's Vienna, Dr. Hanley's abilities as a psychologist and raconteur contributed memorably to my education. Professor William Mueller has been am inspiring presence throughout much of my graduate training in psychology. His fine ear for the nuances of language sharpened my own appreciation for the clinical power of figurative speech. The rare combination of excitement, thoroughness and intellectual penetration that he brought to teaching, psychotherapy supervision and dissertation committee membership is a lasting gift. iii Professor Ellen Strommen's participation on my committee was greatly appreciated. Her unfailing clarity and fairness won my appreciation of her as a teacher. Her keen critical intelligence earned my deep respect for her as a scholar and a researcher. My principal regret is that I didn't avail myself more of her sound judgment. I look forward to rectifying this oversight in the future. I wish to express my appreciation to Dr. Griffith Freed who actively, if inadvertently, contributed to the early conceptualization of the central problem this research undertook to explore. Over a period of years, our many conversations on the role of language in theory and psychotherapy helped focus my own interests in this area. To Dr. Mary Corcoran, for her research savvy and unstinting help with computer analyses, I owe special thanks. She has proven to be a good friend indeed. Dr. Howard Erman deserves special mention, as much for his capacity for friendship as for his intellectual contributions. My ongoing talks with Howard have helped me think through and keep in perspective the implications of this study. I have greatly valued these conversations and anticipate their continuation. I would like to thank Suzy Pavick for her typing expertise and personal calm. This was a maiden voyage for both of us. Her success speaks for itself. Over the years of my graduate education in psychology, I have received considerable support from the Department of Psychology at Michigan State University, the National Institute of Mental Health, and McLean Hospital, Belmont, Massachusetts. iv Most importantly, thanks go to my sons, Daniel (who is past master of ”the psychology game"), and David (whose very existence coincides with my graduate career in psychology), for their boundless joy and love which continues to delight and amaze me. It was left to Freud to discover how, in a scientific age, we still feel and think in figurative formations, and to create, what psychoanalysis is, a science of tropes, of metaphor and its variants. Lionel Trilling (1950, p. 53) vi TABLE OF CONTENTS Page LIST OF TABLES ........................... 1X Chapter I. INTRODUCTION ........................ 1 II. THEORETICAL BACKGROUND ................... 4 Deviant Symbolic Activity ................ 4 On Identifying Deviant Symbolic Activity ......... 5 On Modes of Regression and Indirect Representation. . . . l3 Psychoanalytic Views of Figurative Language ....... 18 The Paradox of Figurative Function ............ 22 The Paradox Developed .................. 25 Figurative Language, Speech Acts, and Free Association. . 29 Free Association: Classical and Ego Psychological Perspectives ...................... 32 Free Association: Language Perspectives ......... 34 Semantic Approaches .................. 35 Communications and Action Language Approaches ..... 36 Speech Act Approaches ................. 4O Toward Integrating the Approaches ............ 46 III. EMPIRICAL ORIENTATIONS IV. VI. VII. VIII. Figurative Language Studies ............... 48 Free Association, Regression and Body Posture Studies ........................ 50 Verbal Productivity and Silence Studies ......... 53 REVIEW AND STATEMENT OF THE PROBLEM ............. 56 HYPOTHESES ......................... 60 METHOD ........................... 64 RESULTS. . . . . . . . . .................. 7O Experimenters' Observations and General Characterizations of the Data ............. 84 Additional Findings ................... 88 Summary of Results .................... 94 DISCUSSION ......................... 98 Free Association: Regression in the Service of the Ego Model ..................... 99 vii Page Free Association: Contextual Model ........... 100 Free Association: Residual Paradoxes .......... 101 Figurative Language: The Novel-Living Distinction . . . 104 Figurative Language: Speech Act and Dynamic/ Motivational Models .................. 106 Postural and Eye Closure Effects ............ ll2 On the Use and Abuse of Figurative Competence ...... 113 IX. SUMMARY .......................... ll6 APPENDIX A .......................... 118 APPENDIX B ........................... 120 APPENDIX C ........................... 125 APPENDIX D ........................... 126 APPENDIX E ........................... 127 BIBLIOGRAPHY .......................... l32 viii LIST OF TABLES Table Page l. Analysis of Variance for Verbal Productivity (Number of Words) as a Function of Free Association vs. Directed Association Instructions ............. 72 2. Analysis of Covariance for Novel and Living Figurative Productivity as a Function of Association Conditions (with Ego Strength Controlled) ......... 73 3. Novel and Living Figurative Categories as a Function of Associative Condition .................. 75 4. Correlations Between Verbal Productivity and Figurative Productivity Corrected for Fluency ....... 76 5. Variety of Novel and Living Figurative Categories as a Function of Associative Condition (Uncorrected for Rate) ........................... 78 6. Correlations Between Verbal Productivity and Figurative Productivity Uncorrected for Fluency ............ 79 7. Analysis of Variance for Verbal Productivity as a Function of Postural and Eye Closure Conditions ...... 81 8. Analysis of Covariance for Novel and Living Figurative Productivity as a Function of Postural and Eye Closure Conditions ......................... 82 9. Novel and Living Figurative Categories as a Function of Posture and Eye Closure Condition ............. 83 lo. Variety of Novel and Living Figurative Categories as a Function of Posture and Eye Closure (Uncorrected for Rate) ......................... 85 ll. Novel and Living Figurative Productivity as a Function of Postural and Eye Closure Conditions (Uncorrected for Rate) ......................... 86 l2. Correlations Between Living Figurative Productivity and Living Figurative Category Scores (Corrected for Fluency) .......................... 89 13. Correlations Between Novel and Living Figurative Category Scores (Uncorrected for Fluency 9l ix Table 14.- 15. l6. l7. Comparisons of Means for Upright/Eyes Open Subjects and Reclining/Eyes Closed Subjects within the Free Association Condition .................. Living Figurative Productivity and Category Scores for First Fifty-Eight Words of all Subjects ....... Summary of Results ................... Summary of Results (Uncorrected for Rate) ........ Page 92 93 95 96 CHAPTER I INTRODUCTION It is easy to overlook the manifold ways in which humans use and abuse their capacities for meaningful communication; the extent to which the normal, ordinary, serious and literal uses of speech are haunted by (the possibility of) the abnormal, the extraordinary, the playful, and the figurative. That these and other "unhappiness conditions" (Austin, 1970) can and do repeatedly intrude into a great variety of speech situations is a fact often hastily passed over; due in part to our idealization of clear and distinct ideas and "secondary process" dominated conceptions of mature functioning. Yet the inevitable deviations from this ideal norm are not without theoretical and practical significance. If we are to understand fully realized speech acts--including the ways in which speakers are fully present in and authentically behind their words--we must also understand the way these performances do not quite come off or misfire altogether. As such, this undertaking becomes part of "the long term project of classifying and clarifying all possible ways and varieties of not exactly doingthings, (and I would add, 'saying things'), which has to be carried through if we are ever to understand properly what doing (and saying) things is" (Austin, l970, p. 27l. Parentheses added). Austin was neither the first nor the only theorist inviting us to notice and take interest in the ways people mean more or less or other than what they say. There is an uncanny, almost pre-established harmony between Austin's interest in the ambiguities, infelicities, l 2 paradoxes and pathologies which beset speech acts and language games and Freud's similar concerns. Freud, too, was attuned to the various "etiolations" of language, including pretense, hypocrisy, deceit, lying, play acting, duplicity, as well as the other forms of non-direct communication which he classified under the heading of defense mechanisms. Moreover, he developed the technique of psychoanalysis which was designed to highlight the functioning of these non-direct communicative forms, with their attendant erosions of meaning and personal responsibility. From this perspective, the Fundamental Rule of psychoanalysis - the instructions to free associate - disrupts the tacit integration of experience codified in ordinary language, therewith setting the stage for disintegrations of communicative forms as well as for new inte- grations. In the former case, speech is disfigured; in the latter, it is figured in original ways. Curiously, this activity of language as figuring, of figurative language, has rarely been isolated as a dependent variable in psychological experimentation, much less in psychoanalytically inspired research. Its critical role in the psychoanalytic process has remained underdeveloped. The present study will attempt to theoretically and empirically articulate the domain of figurative language as well as that of free association. Psychoanalytic conceptualizations of figurative language will be reviewed and supplemented with discussions drawn from inquiries in linguistics, the philosophy of language and literary criticism. Similarly, psychoanalytic, neo-analytic as well as speech act approaches to free association will be delineated. Finally, based on these con- siderations, which point to the unique impact of the Fundamental Rule on the quality and quantity of speech acts, the present study will 3 attempt to experimentally demonstrate that: l) Contexts of Free vs. Directed association have differential effects on the quality or figurativity of speech. 2) Contexts of Free vs. Directed association have differential effects on the quantity or fluency of speech. CHAPTER II THEORETICAL BACKGROUND Deviant Symbolic Activity Psychoanalysis has always been fascinated- and repelled-by deviance. The phenomena that captured Freud's attention were invari- ably those that failed to conform to norms of propriety and common sense. Dreams, jokes, neurotic symptoms, parapraxes, delusions, per- versions, psychopathology, are intrinsically anomalous, striking and perplexing by virtue of their very abnormality, their deviation from the rule. Moreover, it is not surprising that psychoanalytic practice has been centrally concerned with the specific communicative deviances that emerge within the special rule-governed context of psychoanalytic therapy. Yet, here as elsewhere, the rule, (the so called "Fundamental Rule" of free association), and its deviations (the resistances), con- stitute the conditions of intelligibility on which treatment may be meaningfully conducted: the resistances providing the objective clues for technical interventions. Freud, of course, did not restrict his analysis of deviance to overt psychopathology. In an ever-expanding orbit of explanation, jokes as well as religious and artistic genius could also be conceived as violations of customary expectation and practice. In this manner, the primary phenomena of psychoanalysis comprise distortions of symbolic structures in actions (e.g., repetitions and rigidities), in utterances (e.g., slips and delusions), and in bodily expression (e.g., conversions). From this perspective, the bulk of psychoanalytic theorizing represents 4 5 a sustained and multifaceted effort to explain deviant communication and symbolic activity (cf., Greenson, 1969; Habermas, l972; Kubie, l973; Rapaport, l959, l967; Rosen, l969; Schafer, l976, l978; Shave, T974). Hhile there is something incontestably appealing about a theory which purports to order phenomena as seemingly diverse as the delusions of psychotics and the artifacts of high culture, it is equally imperative that important distinctions not be blurred in the interest of a compre- hensive theory. Indeed, there has been no shortage of friendly critics outside the psychoanalytic fold who have cautioned about psychoanalysis' penchant for pejorative rhetoric (cf., Burke, l964). From inside as well, there have been analysts who have recognized the necessity of distin- guishing in theory between those deviances and misuses that are patho- logical and self defeating and those which prove to be the highest form of use (cf., Kris, 1952). Thus, while great novels and grandiose delusions may have elements in common, mountains, after all, are not molehills. The specific difficulties in accounting for the "normal aberrations" of symbolic activity has engendered on the part of psychoanalytic thinkers such apparent theoretical oxymorons as "adaptive regression" and "regression in the service of the ego". Curiously, although dreams, art and wit represent the non-pathological aberrations that most exercised Freud's ingenuity, he, as well as most of his followers, have been conspicuously mute regarding the most pervasive of all the "normal aberrations" of discourse, viz., figurative language.1 On Identifying Deviant Symbolic Activity Given the ubiquity of the jargon of abnormal psychology. not only in the technical discourse of experts but its seepage into the 6 quotidien life of plain people as well, it is especially noteworthy that the criteria for identifying specific instances or even classes of "abnormality" have not always been systematically elucidated. In one of the more ambitious efforts of this sort, Fenichel (1945) explicitly adopts a subjective or phenomenological criterion as definitive of neurosis. "In all neurotic symptoms something happens which the patient experiences as strange and unintelligible." (p. l8, my italics). The existence of experiences that appear as "outside the realm of conscious will", which cannot be voluntarily controlled, provide evidence for the proposition that "all neurotic phenomena are based on insuffi- ciencies of the normal control apparatus" (p. l9). While this may be clinically helpful as far as (self-referred) neurotic patients are concerned, it presents grave problems with respect to the so called "character disordered" patients for whom "symptoms" are ego syntonic. Typically such individuals run into trouble, not with themselves, but with others or with the social order. Consequently, in the discussion which follows, I will assume that a "norm deviation" model highlighting such features as contextual surprise, tension, contradiction, and anomaly has the potential for providing a more comprehensive criterion for identifying aberrant communicative behavior than does a "subjective disturbance" model. Moreover, Fenichel's discussion notwithstanding, I also assume that psychoanalysts, like the rest of us, rely on their implicit and explicit sensitivity to the relevant conventions that govern meaningful speech and action, for their diagnostic and therapeutic work. Suffice it to say that this is no more than a rudimentary sketch of an approach to the problem of providing criteria for identifying 7 psychopathology. Nevertheless, in what follows, I will offer reasons for thinking that approaching pathological symbolic deviance from the perspective of distorted communication provides a vital clue to the related problem of spelling out criteria for identifying those ggrmgl_ aberrations of symbolic activity termed figurative language. Just as there are deviant symbolic activities that are not pathological, there are "normal deviations" which are not figurative. In fact, figurative language belongs to a wider family of language procedures which involve departures from the common use of words, e.g., neologisms, use of foreign words, mistakes, malapropisms, word salads, etc. Moreover, there are several kinds of locutions which have been traditionally classified as figurative but which will not be considered here, largely because they involve non-semantic linguistic transformations. These involve the grammatical figures like ellipses and inverted word order; and the phonetic figures like syncope and elision (Paul, l970, p. 226). Each of these categories are linguistically non-standard with respect to either the normal grammatical forms of language or the normal phonetic characteristic of words. Instead, I shall focus mainly on those utterances which are semantically figurative, which don't mggg_what they say but use language in non-standard senses. Metaphor, simile, metonomy, synecdoche, oxymoron, paradox, hyperbole, litotes, irony, personification, and rhetorical question represent the figures especially relevant for present purposes, "hat is most conspicuous about the issue of identifying figurative language is the neglect of this problem in the psychological (and not only the psychological) literature on the subject. Not that those who discuss figurative language, or more commonly metaphor, evince any 8 difficulty in locating them or recognizing them. Writers continue to define, classify, experiment, and theorize about figures and metaphors, with barely a glance at this question. Rather than constituting an obvious oversight, the ubiquitous presumption that identifying metaphors is, as it were, self-evident, testifies to the confidence that native speakers of a natural language have in their intuitive knowledge of and ability to distinguish figurative usages from literal ones. Consequently, the present discussion is a preliminary attempt to clarify the criterial implications of this common, if naive, understanding. Such a capacity underscores what might be termed the "figurative competence" of a native speaker derived in turn from his involvement in and familiarity with the culture's relevant language games, and forms of life (cf., Wittgenstein, l953). More specifically, this capacity bears witness to an underlying consensus regarding participation in various communicative acts, or what Austin would call "total speech acts in total speech situations" (Austin, l970, p. 147). Consequently, in the background of every speech act is the assumption of being understood or making sense. (In this case, exceptions such as mistakes, insincerities, and nonsense serve to prove the rule.) Or, in the more formal language of Grice (1972), speech acts are regulated by a "Cooperative Principle" which states: "Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged" and the correlative maxims: "be informative", "be truthful", "be relevant", "be perspicuous" (Ibid.. p. 46). Thus, paraphrasing St. Augustine, we live and move and have our being within a matrix of meaning, of making sense. Yet few things are 9 as completely taken for granted--by layman and psychologist alike. Nevertheless, clinical work presupposes a reliable and intuitive grasp of the difference between sense and nonsense, meaning and its aberrations. For this reason, Grice's attempt to spell out (some of) the presupposi- tions of meaningful communication is especially helpful; as is Austin's account of speech acts as possessing locutionary, illocutionary and perlocutionary aspects. ("A locution is an act gf_saying something; an illocution is an act done jg_saying something; a perlocution is an act done by_saying something") (Austin, 1962).2 Taken together, they provide a framework for articulating and differentiating the conditions for normal and abnormal aberrations of communication, for successful and unsuccessful deviations within a total speech act. From this vantage point, direct speech acts are those in which what is said (the sentence) is consistent with what it is meant to convey (speaker's utterance meaning or intention). In a direct speech act the speaker's utterance meaning is reflected in the syntactic form used for its realization, as when the sentence “I promise" is used to make a promise. By contrast, nondirectgspeech acts are those in which sentence meaning and utterance meaning--what is said and what is meant--no longer coincide (Searle, 1979). The variety of ways that saying and meaning can be eccentric to each other mark off the categories of figurative language. Consequently, a necessary condition for figurative language to occur is the presence of an inconsistency within the speech act. The relevant context for identifying figurative language is not the gorg_(cf., Quintillian, quoted in Ricouer, 1977), nor even the sentence (Beardsley, 1958) but the speech situation in which the sentence is embedded. While the signal for a "metaphorical twist" is, 10 as often as not, some sentential aberration, an adequate statement of the necessary condition for identifying figures--viz., the oddly juxtaposed elements, the dissonance between saying and meaning--require that the sentence be considered along with its situation. This becomes especially apparent when the figures in question are irony, hyperbole, and litotes, rather than metaphor and simile. Although it is not hard to think of sentences whose metaphorical significance requires reference to the relevant context beyond the sentence, e.g., "Moscow is a cold city". Drawing the threads of this discussion together, the necessary warrants for the identification of figurative language involve reference to shared forms of life or background information, certain general principles of conversation, and a theory of speech acts. More specifically, figurative language appears to violate such super maxims as "try to make your contribution one that is true" (or perspicuous, or relevant, or informative) or "do not say what you believe to be false" (Grice, p. 98). Consequently, when an utterance flouts one or more of these maxims, as in the statement, "she is the cat's meow", or "Juliet is the sun", the hearer must either consider it pointless, or be challenged by the fact that the speaker may be trying to get across some meaning other than the one he is stating. Assuming a situation in which the principles of cooperative conversation are operative, the literal statement is something the speaker could not mean in that setting because it is contradictory, nonsensical or somehow odd, abnormal, or anomalous. Thus, one is led to construct a figurative meaning when confronted with obstacles which block access to literal meanings. Or, to be paradoxical: figures are ways of making sense with literal ll nonsense; with apparent violations of the conditions for making sense. Implicit in this analysis is the position that the more jolting the impediment to literal meaning the more tension there is in the figure. If we think of 'literal' as referring to relative frequency in ordinary usage, then figures may be classified according to their degree of tension or novelty with respect to conventional use. We would have to imagine something like a figurative spectrum stretching from dead figures through living figures to novel figures. A common practice is to discriminate two roughly drawn classes with somewhat permeable boundaries, viz., frozen figures and novel figures. Frozen figures are considered to be those "which are so commonly used that we have come to accept them as parts of the language" (Barlow, et al., 1970, p. 2). By contrast, novel figures are those that the hearer believes to constitute "an original contribution by the speaker to the content and context of his communication" (Ibid., p. 2). The problem with this venerable and seemingly omnipresent dichotomy of figurative language is that it fails to provide any criteria for making useful distinctions within the broad range of figures that run from stone dead to not quite brand new. A classifi- cation system that blurs the palpable distinction between such moribund metaphors as "he followed the argument" and such lively ones as "he's strung out on Vitamin C" is clearly falling asleep on the job. As a replacement, the present study proposes a tripartite classification system of figurative language into dead, living, and novel, based on what might be called an assessment of the figures' "vital signs", viz., semantic dissonance and tension within the speech act. Consequently, 12 ggag_figures will be described as those which are so commonplace as to have literal and dictionary usage within the language. Some paradigm cases are: "follgw_an argument", "the pgigt of a sentence", "the gggk_of a bottle", "the lgg_of a table". Ljyjflg (but not novel) figures are those which retain tension within the speech act, thus producing a "metaphorical twist". This category includes curses and cliches. Some paradigm cases are: "he's all strung out", "she's spgggyfl, "he's a fruitcake", "he's a 1199?, "he wolfgg_down his break- fast". Novgl_figures are those which possess semantic tension and are used for the first time by the speaker or heard for the first time by the listener in that context. Some paradigm cases are: "spiteful sun", "the bare ruined choirs where late the sweet birds sang", "man is a thinking reed", "life is but a walking shadow". While it may not be immediately apparent, psychoanalysis can be viewed as an effort to understand and explain (and when appropriate correct) deviant speech acts (cf., Schafer, 1976, 1978, for a somewhat related construction of psychoanalysis). Indeed, psychoanalytic inter- pretations appear to presuppose an enlarged view of figurative speech, in which anomalous communicative acts -- e.g., slips, parapraxes, symptoms -- are treated as intelligible with reference to the appropriate developmental context. Consequently, the same logic of identification may be operative here as in the case of successful figurative deviations. The communicative anomaly challenges the analyst to construct--or reconstruct--the appropriate figurative interpretations which render the act intelligible and meaningful. Regardless of the truth or utility of the view that psychoanalytic interpretations embody an implicit theory of metaphor--a view which has 13 only recently begun to receive sustained theoretical attention (cf., Benveniste, 1975; Doneson, 1981; Jacobson, 1955; Lacan, 1976; Rosen, 1968; Shave, 1974; Szasz, l960)--it remains a surprising fact that figurative language in its more prosaic as well as its poetic and scientific uses has seldom been explicitly treated in the psychoanalytic literature. And of these studies, virtually all have focused on the clinical utility of specific metaphors in the context of a case study (Aleksandrowicz, 1962; Ekstein, 1956; Sharpe, 1940; Voth, 1970). Indeed, I have only been able to discover two sustained, explicit theoretical treatments of figurative language in the psychoanalytic literature (Rohovit, 1960; Rubinstein, 1972), plus some rewarding obiter dicta in Jones (1916), Kris (1952), Kubie (1960), Rapaport (1952), Rosen (1967). Of these two major studies, Rubinstein's Chomsky-inspired treatment is by far the more ambitious. Unfortunately, by choosing to deal with only "simple" (i.e., nearly dead) metaphors, like "Mr. X is a fox", and treating them as the result of the ambiguity of the word 'fox' meaning both "a sly person" and "a particular sort of furry animal", Rubinstein begs the most interesting question, viz., how is it that the word 'fox' came to have the meaning "sly person"? This is precisely the work of metaphorical transposition that the theorist must explain, not assume. 0n Modes of Regression and Indirect Representation It is surprising to discover that for all its interest in deviant forms of communication, psychoanalysis has devoted little attention specifically to figurative language. This, in spite of the fact that figurative language in general--and metaphor in particular--constitutes a class of productive deviations formed by nondirect speech acts whose 14 special virtue it is to designate an indirect meaning through a direct meaning (cf., Klein, 1967). For this, if for no other reason, figurative language joins the family of symbolic activities--language games-~characterized by enigmatic indirectness and equivocity (over- determination), which includes dreams, jokes, art and symptoms. In this respect, Freud was a student of Hamlet: seeking, "by indirection to find direction out". From this vantage point, psychoanalytic theory--metapsychology-- takes its bearings from the attempt to provide an explanation for indirect forms of meaning. In contrast to the contextual-convention- alist account detailed above, the strategic conceptual move of Freud was to explain distortions in the speech ggt_by reference to the special psychological characteristics and intentions of the speaker thought to underly it; viz., the ongoing (unconscious) power of specific conflicts engendered in the course of the speaker's socialization history to influence and distort current communicative activity. To the limited extent that psychoanalysis has explicitly treated figurative language, the thrust has been to assimilate figures to the explanatory model developed for the interpretation of dreams. Chapter VII of the Interpretation of Dreams (Freud, 1900) remains the locus classicus of this metapsychological tradition. In it, the dream is established as the paradigm case, the model for understanding the other forms of deviant and distorted meaning. For dreams provide access to the phenomenon, and phenomenology, of regression. By exploiting the resources of this concept, Freud was able to define the scope of symbolic and communicative deviance and thereby establish a coherent range of phenomena where none had previously existed. 15 It is by now well known that in order to make sense of the perplexing mutilated text of the dream, Freud developed the concept of the "dream work" (Freud, 1900). The dream work makes intelligible the otherwise confusing distortions and contradictions for which dreams are famous. Although condensation and displacement are the principal "mechanisms" of the dream work, Freud regularly mentioned several allied processes, most notably "considerations of representability", or "plastic representations" and "indirect representation" which includes "faulty reasoning, absurdity,...representation by the opposite, representation by something connected, representation by something small, and various species of allusion" (Freud, 1905, p. 80). Briefly then, the "trans- positions" effected by dream work are carried out differently in the operation of condensation and displacement. The work of condensation is characterized by the formation of composite expressions out of several trains of thought. Freud also compares condensation to a "compression" and "synthesis" of many ideas into one (Freud, 1900, pp. 595-597). The work of displacement is produced by shifts or inversions of emphasis and value from one point to another. Displacement manifests itself in two ways: in the first a latent element is replaced not by a com- ponent part of itself but by something more re- mote--that is by an allusion; and in the second, the accent is shifted from an important element on to another which is unimportant, so that the dream appears differently centered and strange (Freud, 1916-1917. p. 174). Following Gill's (1967) lead, I intend to view the subsidiary mechanisms of dream work as derivative from these two principal mechanisms. 16 Taken together, these form the "mechanisms", the operationalization so to speak, of primary process thinking. Against this background, regression refers to the shift in "level of psychic functioning" (Schafer, 1967, p. 187) from the reality oriented secondary processes to the pleasure oriented primary processes (Freud, 1911). Moreover, the dimensions of regression are further specifiable by reference to the "metapsychological points of view“ (Rapaport and Gill, 1959; Gill, 1967). Freud distinguishes formal, temporal and topographic aspects (Freud, 1900, p. 48). Formal regression is characterized by transformation with "regard to representability" (Ibid., p. 344); including breakdown of syntax, temporal perspective and logical relations. Or to employ the allied terminology of Heinz Werner (1948), formal regression signifies a cognitive "dedifferentiation", a return to concrete, pictorial expression. In Freud's terms, regression is formal_"where primitive methods of expression and representation take place of the usual ones" (Freud, 1900, p. 548); regression is temporal, "in so far as what is in question is a harking back to older psychical structures" (Ibid., p. 548); regression is topographic when it involves a movement from conscious to preconscious and unconscious wishes, from motility to perception and hallucination. "All these three kinds of regression are, however, one at bottom and occur together as a rule; for what is older in time is more primitive in form and in psychical topography lies nearer to the perceptual end" (Ibid., p. 548). That condensation and displacement bear more than accidental resemblance to certain tropes of rhetoric has not gone completely unnoticed. In an important essay, the linguist Roman Jacobson (1956) has argued that condensation is tantamount to metaphor (which he equates 17 with relations based on similiarity). This view is also developed in the work of the French linguist Emile Benveniste (1971) and the analyst Jacque Lacan (1969) and his followers. While my own position is that the reduction of condensation and displacement to similiarity and con- tiguity relations fails in important ways to do justice to these concepts--or to metaphor and metonomy--these authors have noticed some- thing quite fundamental. By expanding on this notion, Freud's achievement may be described as the discovery of the implicit rhetoric of everyday life. Rather than persist in the traditional view of the tropes as essentially trivial attempts to decorate language, (but cf., Jones, 1915, pp. 90-92) Freud (re)discovered their pervasive influence in our diurnal and nocturnal lives. Indeed, in another place I will argue that the way metaphoric transformation is understood has decisive and far reaching implications for an adequate psychoanalytic view of creativity, sublimation and the so called “regression in the service of the ego". In order to get on with the argument, a major impediment to the full appreciation of the "metapsychological" potential of figurative language must be cleared away. I refer to the venerable dichotomy between the primary and secondary processes. As long as these processes are viewed as polar opposites, as mutually exclusive contradictories, attention is deflected from the myriad and subtle ways available to language for bridging this putative gap. Or, to put it in a way that is more faithful to the facts, strictly speaking there is no "gap" to bridge, but rather a diversity of "symbolic forms" which, for different purposes, mgy_be ranged on a number of overlapping continua, embracing abstract-concrete, literal-figurative, denotative-connotative, 18 3 In cognitive-emotive, conceptual organized-drive organized, etc. reality, the hard and fast version of the dichotomy was more honored in the breach than in the observance, trotted out for theoretical state- ments but disregarded in practice. Certainly, Freud's celebrated respect for the harder facts of mental life led to frequent inter- minglings of primary and secondary processes (1900, p. 599; 1911, p. 221; and especially 1905, pp. 171-172). Rapaport gave de jure recognition to this de facto practice in his discussion of shifting levels of psychic organization and functioning anchored at both ends by the correlative constructs of the pleasure and reality principles (Rapaport, 1951, 1953, 1957; cf., also Hartmann, 1939; Kris, 1952; Schafer, 1954). Psychoanalytic Views of Figurative Language Returning to the issue of the traditional psychoanalytic approach to figurative language, we find the basic conceptual move has been to treat figures, along with other modes of indirect representation, as instances of regression with respect to (mainly) linguistic modes of representation. The foundation for this line of approach is already laid in Freud's early writings. In his jokebook he writes, "Indirect representations of this kind (e.g., symbols and analogies) and allusions whose reference to the thing intended is easy to discover, are indeed permissible and much-used methods of expression in our conscious thinking as well...the dream work, however, exaggerates this method of indirect expression beyond all bounds (Freud, 1905, pp. 171-172). Although Ernest Jones went to great pains to distinguish what he calls "true symbolism" from other forms of "indirect figurative representation" (Jones, 1911, p. 90), when the dust finally settles, "true symbols" differ from other figures only topographically, i.e., with respect to 19 the "distance" from consciousness. Thus, "if regression proceeds only a certain distance, remaining conscious or at most preconscious, the result is metaphorical.... If, owing to the strength of the unconscious complex, it proceeds further--to the level of the unconscious--the result is symbolism in the strict sense" (Ibid., p. 144). Rapaport (1951, p. 210) is critical of Jones' "narrow" attempt to segregate true symbols from the other forms of indirect figurative re- presentation. He prefers Silberer's (1951) view of the psychoanalytic concept of symbolism as inclusive of "metaphors and related phenomena" (Ibid., p. 210). Yet even for Jones, once the topographic differences are acknowledged, all indirect figurative representations--among which he especially singles out "the various figures of speech and modes of thought such as simile, metaphor, apologue, metonomy, synechdoche, allegory, (&) parable" (Jones, p. 88)--constitute formal and genetic regressions to simpler, more primitive types of mental processes (Ibid., p. 137). In this view, he mentions reversions to simpler pictorial and concrete modes of representation. And in an intriguing parallel to the speech act theorists, who also suggest that "politeness is the chief motivation for (certain forms of) indirectness" (Searle, 1975, p. 64), Jones points to "the inhibiting influences (Freud's ethical censorship)" (Jones, 1911, p. 141), which prevent unconscious complexes "from coming to direct expression" (Ibid., p. 141). Jones quotes Rank and Sachs for a definition of symbolism and the other modes of indirect representation such as "simile, metaphor, allegory, allusion and other forms of pictorial representation of thought material (after the manner of a rebus), to all of which it is related. The symbol represents an almost ideal union of all these 20 means of expression: it is a substitutive perceptual replacement- expression for something hidden, with which it has evident character- istics in common or is coupled by internal associative connections. It§_ essence lies in its having two or more meanings" (Ibid., p. 96; cf., Szasz, 1964, pp.148-163). The basic feature of all indirect figurative representation is the fundamental tendency of the mind to "identification" (Jones, Ibid., p. 138), establishing equivalencies via substitutions and common elements. He goes on to consider metaphors as compressed (con- densed?) similes, in which one of the terms of comparison has been suppressed. For Jones, all such relations of comparison and substitution constitute regressions to simpler, earlier and more concrete modes of representation conditioned by a particular state of consciousness, (Silberer's "affective insufficiency", [1951]), and motivated so as to overcome inhibitions and censorship (Jones, Ibid., p. 194). Thus, aside from topographic considerations, all forms of indirect representation--from "true dream symbols", to jokes, to metaphors--rely on the same set of processes and lend themselves to identical treatments. Thus, Freud could as easily analyze jokes as he could his patients' metaphors, or "switchwords" (Freud, 1905b). Ella Freeman Sharpe (1940) extends the view that metaphor presents opportunities for the indirect expression of forgotten infantile wishes and tensions in speech. Like dreams, the "verbal imagery" of metaphors correspond to repressed ideas and emotions (Ibid., p. 204). In particular, "metaphor evolves alongside the control of bodily orifices. Emotions which originally accompanied bodily discharge find substitute channels and materials" (Ibid., p. 212). Based on clinical examples, Sharpe presents a content analysis of metaphors derived from oral, 21 anal, urethral, and oedipal conflicts. On this analysis, the metaphorical figure unconsciously trails clouds of forgotten glory--or baseness; viz., the forgotten or repressed infantile meanings. An interesting attempt to systematize the psychoanalytic view of metaphor can be found in a little known article by 0. Dean Rohovit entitled "Metaphor and Mind" (Rohovit, 1960). Adopting I.A. Richard's [1965 (1935)] analysis of the metaphor into tenor (or subject matter) and vehicle (or figurative attribution), Rohovit argues that metaphors, like dreams, contain a manifest content (viz., the conscious tenor or literal meaning), and a preconscious vehicle or figurative expressions, (e.g., "he §gjl§_into me"); linked to an unconscious or latent tenor discovered by free association to the vehicle. In this way, "the vehicle functions as an associative linkage between the conscious tenor and the unconscious tenor" (Ibid., p. 306). For Rohovit there are mechanisms of "metaphor work" (Ibid., p. 308) which, like dream work, involve condensation and displacement. Stated more rigorously, figures--and especially metaphors--project a set of relations, a system of implications,from one domain to another. The implication complex derived from the metaphorical ascription (the vehicle) functions as a model, a system of relations imputed to the primary subject (the tenor). Thus, "every metaphor is the tip of the submerged model" (Black, 1979, p. 31). For the psychoanalyst, the underlying models tend to be transference paradigms, i.e., sets of internalized roles and relationships rooted in childhood experience. While these discussions by no means exhaust the psychoanalytic treatment of metaphor, it appears that the bulk of the other studies simply assume that figurative language involves latent meanings derived 22 from primary process sources and make no further effort to theoretically support this assumption. As a result, most of these are case studies with various types of patients, in which the therapist employed and/or interpreted the patient's metaphors to beneficial effect (e.g., Aleksandrowicz, 1962; Cain and Maupin, 1961; Ekstein, 1956; Ekstein, 1966; Searle, 1962; Shave, 1974; Voth, 1970). The general drift of this approach is well summarized in Arlow's words: Metaphor constitutes an outcropping into conscious expressions of a fragment of an unconscious fan- tasy...The fact that the analysis of metaphorical expressions may lead associatively to repressed fantasy material comes as no surprise to the analyst, versed as he is in dream interpretation (Arlow, 1969, p. 7). The Paradox of Figurative Function Even such a compressed presentation of the treatment of figurative language in psychoanalytic theory cannot help but convey a certain lack of precision, a kind of equivocation that appears unresolved from study to study. I refer to what might be termed the paradox of function: is figurative language generated by the exigencies of censorship and defense, or is it an autonomous form of linguistic activity? (It is, I think, no accident that this paradox of figurative language is formally identical to Gill's paradox of primary process; Gill, 1967, pp. 279-280). In every case the theorists remained undecided on this issue, seeming to opt for both positions simultaneously, or at least from one paragraph to the next (cf., Freud, 1905, pp.169-170; Jones, 1950, pp.138- 141; Sharpe, 1940, p. 201; Rohovit, 1960, pp.306-307). I will try to show that this paradox stems from two unexamined assumptions that govern the psychoanalytic treatment of figurative language. The first of these may be labelled "the assumption of the primacy of content"; the second, "the assumption of the primacy of the word". 23 By the "assumption of the primacy of content", I refer to the view that figurative language provides a means of expressing indirectly--at various removes and with different degrees of distortions--some unacceptable and usually unconscious content, be it a thought, a wish, a fantasy, etc. By the "assumption of the primacy of the word", I refer to the view that language in general--and figurative language in particular--can be best understood in terms of its smallest unit of meaning, viz., the word taken in isolation. I.A. Richards 1965 (1935) characterizes this view as the "Proper Meaning Superstition. That is, the common belief...that a word has a meaning of its own (ideally, only one) independent of and controlling its use and the purpose for which it is uttered" (Ibid., p. 11). (Although this view may accurately characterize some aspects of the scientific use of language which establishes rules and distinctions for eliminating the ambiguity and polysemy of ordinary discourse, it fails to capture the necessary indeterminateness and variability of meaning that is the hallmark of common speech and creativity alike; cf. Austin, 1970; Heidegger, 1962; Wittgenstein, 1953). Both of these assumptions come together in the psychoanalytic treatment or metaphor--almost as though there existed a pre-established harmony between them. For psychoanalysis, as in traditional rhetoric, the focus is on the figurative wgrg_as a substitution for the common proper word, i.e., as no more than a truncated simile that is reducible to literal statement (cf., Beardsley, 1958, pp.136-137; Jones, 1911; Sharpe, 1940). Granted, that by construing the figurative vehicle as the bearer of preconscious and unconscious significance, psycho- analysis adds a new dimension of its own to this traditional view. In 24 effect, this 'topographic twist', turns the psychoanalytic position into an enlarged form of the connotation theory of meaning. The formal structure of this position is clearly laid out by Beardsley (1958, pp.115-147). Basic to this theory is the thesis that words of a language have a "primary" central or standard meaning and "secondary" marginal or accompanying meanings. This is the distinction between designative and connotative levels of meaning. For example, the word 'wolf' designates the characteristics of a certain class of animals. In addition, wolves are thought to have certain other traits, such as fierceness, ravenousness, clannishness. These latter characteristics comprise part of the range of connotations of the word. "What a word connotes, then, are the characteristics that it does not designate but that belong, or are widely thought or said to belong, to many of the things its denotes. This is the word's range of connotation" (Beardsley, Ibid., p. 125). It would appear that the contribution of psychoanalysis is to supplement the range of public connotations with an additional range of private, idiosyncratic connotations, which are brought to light by the procedure of free association. Thus, as we have seen, Sharpe endorses a view of metaphor as: "a transference of a wgrg to a sense different from its signification", adding that "the displacement is from physical to psychical and not visa versa" (Sharpe, p. 201). Similarly, Rohovit focuses on the hint provided by the figurative word (vehicle) as to the content of the unconscious subject (tenor) (Rohovit, p. 308). Not surprisingly we are back once again to Freud's insight into "switch- words" (Freud, 1905b). Against this background, the assumptions of the primacy of content and word result in a view of figurative language 25 functioning as a means of distorting-~concea1ing and revealing--some unacceptable latent content through an acceptable though ambiguous con- crete wgrQ-image (cf., Empson, 1950; Kris and Kaplan, 1952). Such a view supports the interpretation of figurative language as essentially defensive in nature, generated by the exigencies of censorship. Hence, the largely implicit psychoanalytic theory of figurative language may be explicitly elaborated as follows: a conflictual situation facili- tates a regression from the primary designative literal meaning of a word to its secondary connotative (preconscious and unconscious) meanings which are then given disguised expression in the form of a figure of speech. Such an interpretation has the advantage of being consonant with the psychoanalytic practice of employing the Fundamental Rule of free association to induce (an adaptive) regression and thereby uncover and call into play the latent transference meanings of words, images, objects and situations for the patient. Thus, like dreams, jokes, and symptoms, figures of speech, when properly understood, are "derivatives" providing privileges access to unconscious attributions of meaning.4 The Paradox Developed Given the related assumptions of the 'primacy of content' and the 'primacy of the word' in framing the psychoanalytic model of figurative language, the theoretical resolution of the 'paradox of function' appears to be straightforward: figurative language is unequivocally motivated by the defensive demand to evade censorship. Yet lest we become too complacent with this one-sided solution, or with its opposite by positing a faculty of metaphoric identification (Jones, 1915), we would do well to heed Freud's remark in another connection 26 that "the apparent contradiction is due to our having taken abstractions too rigidly and attended exclusively now to one side and now to the other of what is in fact a complicated state of affairs" (Freud, 1926, p. 97). Thus, it may turn out that the force of the paradox is kept alive by the blinding power of the model (based on the assumptions of 'word' and 'content') which rivets attention on the continuities and analogies uniting the entire range of indirect representations, thus obscuring equally decisive differences and disanalogies. To loosen the grip of this model of figurative language, I propose that the focus on the figure as content be complemented by a focus on the figure as fgrm, and that the focus on the figure as wgrg_be complemented by a focus on the figure as sentence. In certain respects, this terminological expansion parallels the relation of psychoanalytic instinct psychology to ego psychology. If approaching figurative language with the assumptions of 'word' and 'content' primacy results in a focus on the actualization of private and connotative meanings (derivatives), the approach via 'sentence' and 'form' results in highlighting the speaker's action leading to changes of meaning (cf., Ricouer, 1974, 1975, p. 138). The former represents a "substitution view of metaphor", "which treats the metaphorical expression as a substitute for some other literal expression which would have expressed the same meaning had it been used instead" (Black, 1960, p. 31). The latter leads to the "interaction view": "in the simplest formulation, when we use a metaphor we have two thoughts of different things active together and supported by a single word, or phrase whose meaning is a resultant of their interaction" (Richards, 1965, p. 93). 27 Instead of pursuing the marginal and idiosyncratic association of a key word, we are enjoined to study "the interanimation of words" (Richards, Ibid, p. 47) in particular contexts. Consequently, the construction of a figure of speech produces a verbal structure in a state of dynamic tension, built as it were out of a semantic clash or "verbal opposition". In this--but only in this--there is an analogy with the form of dreams and symptoms. I refer to the view of the dream and symptom as symbolic forms representing compromises between the conflicting forces of drive and defense (cf., Fenichel, 1945; Horowitz, 1970). And while it may be tempting to view dreams and symptoms as unconscious figures--metaphors and metonomies operating in different modes of representation--a temptation Lacan (1969) and his epigones find irresistible--I believe there is at least as much to be learned from studying the differences that separate these symbolic forms, as there is from the likenesses which unite them.5 From an interaction perspective, the metaphor represents a success- ful resolution--an integration--of conflicting semantic fields and intentions, just as symptoms, and to a lesser extent dreams, represent unsuccessful resolutions. If metaphors move in the direction of the new and creative, symptoms and dreams are in the grip of the old and ossified. Moreover, if dreams and symptoms conceal vital connections, metaphors--at least novel ones--also reveal and create connections for the first time (Black, 1961; Ricouer, 1974). This perspective sheds light on the so-called 'paradox of function'. We may note that Gill offers as a solution the view that "the mechanisms of the primary process are not created by censorship, but are called into play as the result of both the imposition of Censorship and the lifting of censorship" (Gill, 1967, p, 286). Regardless of whether 28 Gill's view does justice to the paradox of primary process, it fails to fully engage the issues involved in figurative language. It would be convenient to say that figurative language is not created by censorship, but used by it. And this is true--as far as it goes. Unfortunately, it does not go far enough. For as we have seen, figurative language embodies intrinsic tensions between semantic, intentional and con- ventional elements which may be exploited for defensive purposes, but can only be eliminated on pain of death. Only when a figure expires and is declared dead does the tension 90 out of it. And at that point it no longer functions figuratively. Not surprisingly, Gill's solution would be more relevant to the substitution-connotative aspects of metaphor than to the interactive ones. Since psychoanalysis has by and large thrown its cards in with a substitution view, it is no accident that Gill's solution has considerable appeal. Perhaps a more judicious response to the paradox would be to declare its terms illicit and thereby reject it as, at bottom, a false paradox. As we have come to understand, figurative language is neither autonomous (meaning conflict free), nor defensive in origin. On the contrary, it grows out of a conflict of meanings, a clash of semantic fields which it contains within a tensional symbolic form (cf., Berrgren, 1962; Wheelwright, 1962). As such, it constitutes a paradigm of (creative) symbolic integration in contrast with the (defensive) symbolic distortions and mutilations of dreams and symptoms. Polanyi captures this side of figurative language with his observation that in the metaphor "the subsidiary clues--consisting of all those inchoate experiences in our own lives that are related to the two parts of a metaphor--are integrated into the meaning of a tenor and a vehicle as 29 they are related to each other in a focal object (a metaphor)” (Polanyi and Prosch, 1975, pp. 78-79). As such, a metaphor discloses "man's basic imaginative capacity for integrating two or more disparate matters into a single novel meaning" (Ibid., p. 79). Figurative Language, Speech Acts, and Free Association Thus far, I have argued that there are two approaches to figurative languages within psychoanalysis. On the one hand, there is the more or less explicitly elaborated "substitution-connotation" view found in Freud, Jones, Sharpe and others. On the other hand, there is the "interactional" view that is implicit in treating symbolic structures as compromise formations. The former proceeds by focusing on the elements, on word and content. The latter proceeds by focusing on the larger unities of sentence and form. Moreover, the possibility was advanced that the paradox of the function of figurative language--as autonomously or defensively inspired--could be resolved by considering the implications of these unthematized assumptions. It appeared that the paradox was based on a false dichotomy, and could be "dissolved" by discriminating the proper scope and limits of each perspective. Consequently, a 'substitutive' perspective highlights aspects of figurative language which lend themselves to defensive uses and abuses, while an interactive perspective has more to contribute to the study of creativity and sublimation. In the former, the focus is on the regressive play of private, marginal connotations and typically more concrete, pictorial verbal representations. In the latter, the focus is on the integrative function of the figure as a symbolic form in which disparate elements are brought-- sometimes kicking and screaming--into play with one another, thereby enlarging both as they jointly bear on the creation of new meaning. Like 30 a funny joke, figures derive their interest and power by bringing together unexpected, seemingly unrelated, and deeply conflicting elements in ways which surprise, delight and illuminate us. They lose interest and power when the elements are too banal or too autistic; when the tensions between them are too weak or too great. In extreme cases, such as psychotic utterances, semantic clashes are embedded in symbolic verbal forms that have all but abandoned their integrative capacity and communicative intention to the play of private and autistic connotations. At such times, "things fly apart, the center cannot hold", and metaphor formation miscarries. Since the present study is not directly concerned with the pathological distortions of figurative language, it is not necessary to inquire into the fate of private meanings excommunicated from public language games. Nor is it possible to pursue the disruptions and breakdowns of the formal organization of words, imagery, action and emotion with the ensuing fragmentation of language games. Suffice it to say that the former resembles dreams and neurotic symptoms; the latter, psychotic disintegration. There is a schematic quality in this classification of figurative language which bespeaks its abstractness. Indeed, the source of the abstractness is not hard to pin down, viz., the focus on the figurative word or sentence independent of the total speech act in which it is embedded. Consequently, the discussions of the figures 'as word' and 'as sentence' will now be supplemented by discussing the communicative context in which the figure occurs. For it appears that taken by themselves, many (seemingly) figurative sentences may in fact be literal if situated in or referred to the appropriate context. Thus, "they really nailed him to the wall" may be an accurate report of a 31 crucifixion; and "Juliet is the sun" would be literally true in a language community which designated its sun by that name. Similarly, "John is a wolf (fox, dog, bear, pig, etc.)" may be literally true when used to name a particular mammal. And, Max Black notwithstanding, one can literally "plow through a meeting"--especially if the meeting is held in a cornfield. Yet, the fact that g9mg_metaphors can be literalized via the appropriate context, does not warrant the assumption that all lend themselves to like treatment, (just try, "virginity is the enamel of the soul"). The main thrust of these examples is to suggest that many sentences, perhaps a11--and I have not been able to think of a single good counterexample-~can function figuratively depending on the context, tone, and communicative intention. These considerations point to the generalization that rather than constituting a special class of speech acts, figurativeness can happen to any speech act depend- ing on the wgy_the sentence means given the proper "contextual twist". Thus, not only does the speech act supply the court of appeal in deciding whether a given sentence is figurative or not, but it also provides the place in discourse where the abstract interactional and substitutional dimensions of figurative language may be concretely embodied. If the substitutional axis signifies the (potential) pull of regressive connotations of words for the subject, and the interactional axis signifies the formal organization of semantic tensions into a sentence; the speech act (potentially) integrates word and sentence, subject and object, intention and convention by striving to construct a living figure to meet the public criteria of communicatiVe significance. Consequently, shifting focus from word and sentence to the actual speech event calls particular attention to the role of 32 context, conventions, and interpersonal communications in complementing the intrapsychic dimensions of regression to private connotations. Furthermore, the multiple perspectives on figurative language opened up by considering the speech act brings the characteristics of the speech situation into sharp relief. In the case of psychoanalysis, the emphasis falls on the Fundamental Rule and the characteristics of free association. Free Association: Classical and Ego Psychological Perspectives The leading view of free association within psychoanalysis stresses the regressive potential of the procedure (Bellak, 1967, p. 53). Free association is contrasted to directed association which involves "deliberately and consciously choosing a topic to talk about (which) results in a preconscious selection of related items with a concommitant rejection of irrelevant and intrusive thoughts. This has two conse- quences. It places the emphasis on logical rather than analogical relationships thus impeding any creative discovery of new relationships" (Kubie, 1973, p. 56). In contrast to this topical and logical organi- zation which, according to Kubie, primes preconscious selective scanning operations, free association facilitates the deactivation of the "automatic preconscious processes of selection and rejection from all possible elements in the stream of association" (Ibid., p. 57) that are 6 Thus, set in motion by the topical organization of directed thought. free associations provide the link between the rational and the irrational, the conscious and the unconscious. "Strictly speaking associations are the manifestations, and often the only manifestations (if action is excluded) of the hypothetical constructs we call id, ego, superego" (Bellak, 1967, p. 59). 33 No less an authority than Leopold Bellak (Ibid., p. 53) has called attention to the conceptual underdevelopment of the Fundamental Rule. He attributes this to be procedure's roots in the now superceded "tapographical model" of mental life. In that context, free associations were thought to gravitate to the traumatic events believed to be causative of neurosis. Moreover, "free" association was based on the assumption that a strict determinism ruled the processes of mental life. Consequently, associations provided the links in a causative chain that leads back to origins in affective and traumatic factors (Ibid., p. 53; cf., Rapaport, 1942). On these assumptions, the "freeness" of associations referred to the absence of the external control of the analyst or the internal, conscious self-editing of the patient. In a similar vein, Menninger and Holzman (1973, p. 47) state that free association makes possible the laying aside of the "facade of pretense and normality". For his part, Bellak proposes an ego psychological approach to free association. Structurally, he regards free association as a form of "regression in the service of the ego" (Kris, 1952), in which the ego oscillates between regressed and full functioning; i.e., "there is a swing from regression to vigilance of cognitive, adaptive, and synthetic functions. This produces emergent qualities which we know as the creative process" (Bellak, 1967, pp.55-56). Consequently, the relative reduction of certain secondary process functions leads to temporary dedifferentiation of the sharply conceived boundaries of spatial, temporal and conceptual distinctions, hierarchies, and relationships. This regressive phase is suceeded by a phase of synthetic ego functioning in which new boundaries may be drawn and new relationships may emerge. By facilitating new syntheses, regression 34 is "in the service of the ego" (Ibid., pp.56-57; cf., also Loewald, 1960). Turning from the formal characteristics of thought during free association to the conditions of their emergence, Bellak points out that following the Fundamental Rule requires: (1) a decrease of external cognitive functions, (2) a focus on internal 'stimuli', (3) and an exclusion of judgmental reality testing (Ibid., p. 56). Taken together, the result is a (partial) disruption of the cognitive and perceptual links which maintain the ordinary adaptation of ego to environment. When these processes are further supplemented by reclining posture and eye closure instructions, kinesthetic, perceptual, and motility ties maintaining the "general reality orientation" (Shor, 1970), are also loosened (Menninger and Holzman, 1973, p. 473; especially Brenman and Gill, 1959, p. 55). In summary, free association instructions specifically affect the topical scanning operations of organized thought. As a variant of sensory deprivation, the restriction of reality ties due to postural and eye closure instructions further amplifies the movement from logical to analogical thought engendered by the injunction to free associate. Free Association: Language Perspectives While the traditional and ego psychological perspectives shed light on the motivational and intrapsychic aspects of free associations, in important ways this has been at the expense of an adequate theoretical appreciation of the communicational aspects of the Fundamental Rule; or what might be called, following Austin, "the total speech act in the total speech situation" (Austin, 1970). Consequently, although metapsychologically oriented approaches--like those of Brenman and 35 Gill, Kubie, Menninger and Holzman, Bellak--have called attention to the regressive potential of free association, none, so far as I know, have explicitly drawn the connection between free association and in- direct representation. Certainly none have claimed a necessary connection between free association and indirect representation, much less between free association and figurative language. In what follows I shall argue for both of these propositions. It will be proposed that a speech act analysis will show how the Fundamental Rule produces communi- cation deviances, including that class of indirect representations termed figurative language. To develop this position, I will especially draw on insights of Hellmuth Kaiser (1965), Thomas Szasz (1964), and the recent writings of Roy Schafer (1978). 1. Semantichpproaches. Lest it appear that a speech act per- spective is totally alien to psychoanalytic thinking about these matters, I call the reader's attention to a seminal paper by Rudolph Loewenstein entitled "Some Remarks on the Role of Speech in Psychoanalytic Technique" (Loewenstein, 1959). There are sentences in that paper that are almost Austinian in their appreciation of the "performative" aspects of speech. But more germane to present purposes, in one notable paragraph Loewenstein observes how the Fundamental Rule results in the "patients' utterances (becoming) more obviously influenced by the primary process" (Ibid., p. 258). By this he means the altered relationship between word representation and object representation, be- tween the signifying and the signified, due to free association leads to transformations in the "usual vocabulary". Instead of "a definite set of meaningful relations between signs and ideas" (Ibid., p. 258), there is a shift to the processes and vocabulary common to "wit, jokes, 36 and in actual love life" (Ibid., p. 258). In a footnote, Loewenstein goes on to quote Benveniste's "attempts to describe these phenomena in terms of well known figures of speech" (Ibid., p. 258). Victor Rosen makes similar reference to the similes and metonymies provided by displacements and substitutions of the process of free association" (Rosen, 1969, p. 205). Interesting as they are, the contributions of Loewenstein and Rosen occupy a position midway between an intrapsychic and a speech act view of free association and figurative language. It is almost as if these authors are transitional objects between an intrapsychic and a communications perspective. In fact, with respect to the critical question of indirect communications and figurative language, both authors move within the orbit of Saussurian structural linguistics. Consequently, Rosen in particular, but Loewenstein on this issue, are far more interested in unconscious metaphoric and metonomic trans- positions from signs to signifiers than in analyzing the manifest forms of indirect speech acts.7 2. Communications and Action Language Approaches. By contrast, there is nothing half-hearted about Szasz's discussion of these issues. Explicitly rejecting metapsychology for a "semiotical analysis of behavior", Szasz attempts to assimilate psychoanalysis to a communication-rule-following-game playing model (Szasz, 1964). Indeed, what he lacks in detailed analysis, he more than makes up for in the boldness of his views. Suffice it to say that Szasz explicitly raises the issue of indirect communication and provides some useful hints about the impact of free association on figurative language. 37 Szasz distinguishes direct and indirect communication on the basis of the ambiguities attendant upon the latter's "multiplicity of meaning" (p. 148). The intentional use of expressions known to be interpretable in more than one way--i.e., indirect communication--"is called hinting, alluding, or talking in metaphor....In addition to the terms listed, double talk, innuendo, insinuation, implication, and punning are terms used to identify and describe indirect communication of different types" (p. 149). While there may be some problems with his stress on "intentional" and "deliberate"--after all, figures of speech are routinely employed by speakers who are not explicitly aware of using figures, not to mention the speaker's "obtuseness to his own duplicities" (Freed, l978)--Szasz does provide the rudiments of a communications approach to the relations between free association and figurative language. The crux of Szasz's view is contained in the statement: "Whenever the relationship between two people is uncertain--hence when either or both communicants feel threatened and inhibited--the stage is set for the exchange of relatively indirect messages". Szasz takes hinting as the paradigm case of "speaking in metaphor", and goes on to analyze hysteria "as a typical metaphorical communication". For present purposes, it is sufficient to mention that Szasz sees indirect communication as (1) serving a protective function with respect to "ego alien" or "socially alien" wishes, and (2) providing insurance against disappointment and object loss. Hence, indirect communications provide an escape hatch for the speaker, affording protection "by enabling him to communicate without committing himself to what he says". 38 When brought to bear on the issue of free association, Szasz's analysis entails that the implicit or explicit demand on the patient to verbalize the "unacceptable" will perforce increase the use of indirect forms of communication. By cashing in on the ambiguities inherent in the polysemy of language, metaphoric use of "indirect speech" (Freud, 1900, pp.141-142) has the singular advantage of maintaining "communicative contacts when, without them, the alternatives would be total inhibition, silence, and solitude on the one hand or communicative behavior that is direct and hence forbidden on the other hand" (Szasz, 1964, p. 150). Implicit in this analysis, though undeveloped by Szasz, is a view of the spectrum of indirectness which runs from total silence, through speech disruptions, to figurative language. In a recent essay on free association, Roy Schafer (1978) has developed a position, which, though more subtle, has certain affinities with that of Szasz. The main point of contact between them follows from Schafer's argument that the Fundamental Rule, by suggesting that associations are "free", serves to "free" the patient from responsibility for his speech (Ibid., p. 40). By virtue of this, the Fundamental Rule establishes a context in which associations come to be regarded as "disclaimed actions" (Ibid., p. 40). While this is not the place for a detailed consideration of Schafer's Ryle-inspired claims for a new "action language" for psychoanalysis, (Schafer, 1976), it is enough for present purposes to note that Schafer's "disclaimed actions" are essentially forms of indirectness imposed by (unconscious) "danger situations" in which directness would be ominously conflictual (Schafer, 1978, pp. 31-32, pp.43-46). 39 For Schafer, "the genius of the free association method lies not in its establishing suspension of rules but in its making plain the rules embodied in the analysand's associating" (Ibid., p. 54). "Word surprises"--the special class of disclaimed action that Schafer considers at length--are viewed as "disruptions of free associating" (Ibid., p. 55) which bring to light the (unconscious) rules and fantasies that serve to conflictually complicate seemingly straight- forward situations. Thus, for Schafer, the basic rule helps to establish a context which promotes the construction of a different mode of dis- course comprising disclaimed, indirect speech acts. It is to this constellation that Schafer refers when he talks about "the psycho- analytic method and the data it generates through the rule of free associations" (Ibid., p. 51). With unusual clarity, Schafer shows how the apparent freedom from the usual restraints of verbal decorum serves only to produce dis- claimed actions which reveal the internal constraints, the complex rules which the analysand imposes on situations. Yet neither Schafer nor Szasz, for all their keen appreciation of the deviant rules people impose on ambiguous settings, seem to be much interested in a fine grained analysis of the special contextual factors brought into being by a paradoxical rule which announces that there are no rules (cf. Kaiser, 1964). It was Hellmuth Kaiser who saw the radical implications of this state of affairs. Going beyond even such a minimal and paradoxical rule as the Fundamental Rule, Kaiser refused to offer any rules to the patient (or to the therapist for that matter). By doing so, Kaiser realized that his "decision not to structure in any way the patient's 40 behavior" (Ibid., p. 138), "did the only thing which could activate the universal conflict and in consequence enhance the patient's duplicity" (Ibid., p. 138). While I am inclined to agree that Kaiser's is the rock bottom position, it can be argued that free association is at least a distant relative, and, in any event, a vast improvement over such highly structured procedures as hypnosis and directive therapies. If duplicities are forms of indirect communication-- and I think there can be no question about this--then on Kaiser's principles free association should promote more indirect communication than the more highly structured directed association instructions. Thus, 'ambiguity' emerges as a critical contextual dimension invoked by the Fundamental Rule. Consequently, certain observations concerning the nature of the 'stimulus' and the 'situation' of the projective techniques have special relevance for this discussion: Earlier defintions of projective techniques have stressed the response and its interpretations, but not the stimulus or 'situation'. More recent attempts include an awareness of the nature and characteristics of the stimulus, its objective features which may evoke common response, as well as its 'unstructured' nature which allows for uni- queness and 'private world' responsiveness (Rabin, 1968. p. 11). This is precisely the position I wish to develop with respect to the Fundamental Rule. Where traditional interpretations have high- lighted its 'private world responsiveness" (cf., Fenichel, 1945. pp. 23- 24; Rapaport, 1967, pp. 196-197), I hope to detail its (public) situational characteristics and determinants. 3. Speech Act Approaches. The notion of a 'speech act' calls attention to language as action--as agent speaking to hearer in a specific setting in certain ways for certain purposes--rather than as 41 a set of abstract, self-contained relationships. Through his analysis of "performative utterances" and "illocutionary acts", Austin attempts to draw out what people do in speaking; i.e., that "speaking is a bit of conduct, a public performance or action"(Rosenberg and Travis, 1971, p. 557). Consider such diverse illocutionary acts as "promising", "begging", "christening", "accusing", "denouncing", "appraising", "accepting", "apologizing", "blaming", "praising", "wishing", "congratu- lating", "recommending", "warning", "exhorting", "approving", "blessing", "surrendering", "assuring", "volunteering", and the hundreds of other predicates (performative verbs) which indicate to the hearer the speaker's position with respect to the subject of the utterance. In every case, such illocutionary acts provide the conventional means whereby the speaker adjusts his utterance to the immediate communicative situation. They represent the language conventions which the speaker may call upon to actualize his communicative intention. Consequently the requirements of circumstances supply the conditions for the appropriateness and implication of the utterance (Grice, 1976, pp. 43- 50). Moreover, the sequence and exchange of utterances in conversation may be thought of as being regulated by the sort of "cooperative principles" and maxims such as Grice suggests (see Ibid, pp. 44-46). By virtue of these (or some such) principles, the speaker's intentions and the conventional linguistic and rhetorical forms of utterances are integrated into the exigencies of the immediate communicative occasion. In a more perspicuous way, I.A. Richards appears to be making a similar point when he states: Language uses an incredibly elaborate system of co-operative relations among its parts at all levels in order to deal as best it can with the 42 infinite variety of situations speakers and hearers may find themselves in. Every utterance has both a form and a meaning. The form has been developed to handle the meaning and the meaning must be explored both through the language and through examination of the situation the utterer is attempting to deal with. Both the utterance and the situation are the outcomes of systematic selections from among alternatives. For the speaker, the form is his endeavor to accord his language activity with the situation (as he sees it) (Richards, 1966, p. 22). Against this background, and bearing in mind the earlier discussion of figurative speech as indirect language forms for articulating certain kinds of multiple meanings, I will try to show that the free association instructions (in contrast to directed association instructions) create a situation in which the speaker is more likely to avail himself of the forms of figurative speech, or, failing that, to employ such privative forms of indirect communication as duplicity and silence (cf., Enelow, 1960). The special character of the ambiguity of the free association condition (and to some extent, of the projective technique condition as well) can be spelled out by reference to the effects of the Fundamental Rule on the "conversational principles", and in turn on the speaker's construction of the speech act. Specifically, the Fundamental Rule creates paradoxical tension within each of the four categories or "supermaxims" derived from the "conversational principle" (Grice, 1976, p. 45). Thus, with respect to the category of quantity, the Fundamental Rule clashes with the maxim, "make your contribution as informative as is required (for the current purposes of the exchange)" (Ibid., p. 45). The free association instructions to "let your mind wander and report 43 whatever comes to mind (no matter how trivial, shameful, impolite) without leaving out a thing" provides no criteria for the speaker as to what counts as a sufficiently informative account; no way of knowing what is underinformative or overinformative. With respect to the category of quality, the supermaxims "try to make your contri- bution one that is true" (as well as the submaxims, "do not say what you believe to be false", and "do not say that for which you lack adequate evidence") (Grice, Ibid., p. 46), clash with the Fundamental Rule's (only slightly implicit) demand that the speaker disregard the usual rules of evidence and the underlying distinctions between reality and fantasy. While it is in effect, the Fundamental Rule abrogates the concern for truth which provides one of the usual assumptions of cooperative conversation. With respect to the category of relation, the supermaxim "be relevant" clashes with the Fundamental Rule's explicit interest in the "irrelevant" and "trivial". Even more to the point, the Fundamental Rule explicitly repudiates the relevance of criteria of relevance, denying that anything is more relevant than anything else. Finally, with respect to the category of manner, the supermaxim "be perspicuous" (and the various submaxims: "avoid obscurity" of expression; "avoid ambiguity"; "be brief [avoid unnecessary prolixityl"; "be orderly") (Ibid., p. 46) clashes with the Fundamental Rule's expressed interest in the irrelevant and the disorderly. In addition to these conversa- tional maxims, other explicitly aesthetic, social, and moral maxims also clash with the Fundamental Rule's demand that the speaker disregard the claims of propriety, politeness, and moral constraints (at least so far as speech is concerned). 44 The clashes between the Fundamental Rule and the conversational (and other) maxims create a series of ambiguities, of contextual dissonances for the speaker. With its insistence that the speaker verbalize gll_thoughts, feelings, etc., the Fundamental Rule also defines a situation from which there is no escape. It insists that the speaker speak in spite of the tensions and paradoxes of the situation. Moreover, it implicitly insists that the speaker make sense, in spite of the fact that the usual conditions for straightforward making sense have been completely undermined (cf. Kaiser, 1965, p. 66). As a result, the speaker is faced with the paradoxical task of (somehow) transcending the contextual strains, dissonances and ambiguities via meaningful speech (cf. Greenson, 1967, p. 197). By way of contrast, it should be apparent that the directed association instructions--"choose one subject to talk about. Select any subject you care to and describe your thoughts and feelings about it"--are essentially consonant with the conversational maxims. Although it fails to specify the precise 'quantity' desired, it provides for the topical organization of speech regulated by the maxims of truthfulness, relevance, and clarity. As a result, the semantic paradoxes and dissonances are not major factors in this condition, at least once a topic is ghgsgn, Since no criteria are indicated as to an appropriate topic, it is conceivable that this ambiguity will create specific semantic stress points around the selection process (cf. Shapiro, 1970). To review the course of the argument so far: it was pr0posed that the ambiguity generated by the free association instructions could be further differentiated by adverting to the paradoxes set up when the Fundamental Rule clashed with the "cooperative principle" and its 45 "conversational maxims". More specifically, tensions arise out of conflicting conversational demands (1) to be informative and to be autistic; (2) to be truthful and to be fanciful; (3) to be relevant and to be irrelevant; and (4) to be perspicuous and to be desultory; as well as out of conflicts with conventional social and moral norms of polite- ness and decency. These strains on the speaker may result in the mis- firing of the speech act, in dysfuencies, disturbances, inarticulate- ness and silence (Murray, 1971). On the other hand, the speaker may hit upon or construct linguistic forms which are responsive to the tensions of the communicative context. Various defensive strategies and modes of indirect communication may be brought into play, including ambiguities, equivocations, and the speaker's lack of commitment to what he is saying. In short, the stage is set for what Kaiser calls "duplicity". Since the paradoxes of informativeness, relevance, truthfulness, and clarity appear to rule out--or at very least render problematic-- literal utterances, the speaker may have recourse to silence, equivo- cation and duplicity; but he may also avail himself of those forms of meaning which "make sense with nonsense, (which) transform a self- contradictory statement into a significant self-contradiction" (Ricouer, 1974, p. 63). Thus, one productive solution to the paradoxes raised by the contextual conflicts is the construction of figurative meanings. Indeed, only figurative language--the various forms of verbal p1ay--can contain the conflict of contextual demands by unifying differences into a structure of paradoxical likeness. Thus, instead of falling silent or duplicitous, the speaker may_become ironic, humorous, metaphorical, allusive...i.e., he may become figurative and thereby avail himself of the suggestive ambiguity of figurative speech 46 to safely violate the usual conversational maxims. By producing speech acts which don't mean what they say--which are open to different and even contradictory interpretations--the speaker provides the simultaneously paradoxical actions required by the Fundamental Rule. In the same act he may honor and violate the rules, obey and transgress conversational social and even moral taboos. Thus figures enable him to do consciously what symptoms and dreams are thought to do uncon- sciously. Consequently, at one pole are the gaps and lacunae of dis- course, the defenses and duplicities; at the other pole, the creative figure skating over thin semantic ice. From this vantage point, the living figure is language taking leave of its ordinary senses. Meta- phor is language leaping over its own shadow. Toward Integrating the Approaches Two approaches to the bearing of free association on the production of figurative language have been explored. In one, the focus was on the Fundamental Rule's induction of a "shift in psychic level, consist- ing in the fluctuation of functional regression and control. When regression goes too far, the symbols become private, perhaps unintelli- gible to the reflective self“ (Kris and Kaplan, 1952, p. 254). Accordingly, it was indicated that this regressive movement is tanta- mount to the movement from the designative to the connotative, from the primary to the marginal meanings of words. And following Ernest Jones and many others, it was contended that when regression proceeded only a little ways (i.e., was adapative), the result was figurative language. Free association could then be viewed as an invitation to move freely among the implicit meanings of words, drawing out their private systems of association and implication. Figurative language 47 provides the linguistic vehicles for the expression of these forms of meaning. The other approach focused on the contextual strains and paradoxes generated by the clash between the Fundamental Rule and the conversa- tional maxims. It argued that figurative language, in contrast to the privative modes, were forms of making sense nondirectly, therewith deviating from ordinary linguistic rules in the process. Although these two approaches to free association have been developed separately, it is apparent that they are by no means mutually exclusive. 0n the contrary, as in the case of figurative language, they may be thought of as representing complementary aspects of a complex concept. Consequently, a rounded statement of the relationships between free association and figurative language must do justice to both dimensions. For the present, suffice it to say that the Fundamental Rule (and its postural accompaniments) sets up a number of centripetal and centrifugal tensions which bear on the intentional (including intrapsychic) and conventional (including linguistic) aspects of the speech act. In figurative language, both of these dimensions are integrated into a tensive form (cf., Wheelwright, 1968; Wimsatt, 1954). When integration fails, the speech act may be distorted into an array of privative forms, most notably silence, dysfluency, and duplicity. CHAPTER III EMPIRICAL ORIENTATIONS Figurative Language Studies To my knowledge, there have been no quantitative studies of the relationship between free association and figurative language. Indeed, I could not discover relevant empirical studies of any kind. The psychoanalytic literature contains a number of case studies in which the patient's figurative language--particularly metaphors--(cf., Alek- sandrowicz, 1962; Sharpe, 1940; Voth, 1970), or the therapist's figurative language (Cain and Maupin, 1961; Caruth and Ekstein, 1966; Ekstein and Wallenstein, 1956; Reider, 1972; Sledge, 1977) have been used to trace unconscious conflicts or further the course of therapy. However, none of these studies brings out any relations between the Fundamental Rule and the production of figures. Similarly, most of the academic psychological literature on figurative language, such as it is, has employed forced choice formats between individual metaphorical and literal words for the purpose of completing sentences (Koen, 1965), or other highly structured contexts and predetermined stimulus words (cf., Asch and Nerlove, 1960; Billow, 1975; Ervin and Foster, 1960; Gardner, 1974; Klorman and Chapman, 1969; Pickens and Pollio, unpublished manuscript; Pollio, 1977; Pollio and Burns, 1977; Pollio and Pollio, unpublished manuscript; Richardson and Church, 1959; Werner and Kaplan, 1963). While these studies are not without interest--especially those bearing on the developmental differences in the capacity to understand figurative language--they 48 49 all suffer in various ways from what I have called "the assumption of the primacy of the word", or at best, "the assumption of the primacy of the sentence". Indeed, Pollio and Burns (1977) have provided some empirical support for the speech act position that "speaking and use in context gave words life and what was nonsense under one set of conditions did not have to be nonsense under another. By attending only to the dictionary meanings of words the natural flexibility of language was ignored and metaphor denied its proper place as a significant aspect of human communication" (Pollio, 1977). By contrast, the few studies of figurative language that have employed an unstructured free verbalization format have some bearing on the present inquiry. Thus, in one study of creative thinking, Mawardi (1961) found that instructions encouraging the use of "feeling words" and a feeling approach to problems facilitated the production of metaphors. In a study involving a line by line language analysis of a psychotherapy session, Pollio and Barlow (1975) found that novel, as opposed to frozen figures, cluster around problem setting and problem solving segments. In another study from Pollio's laboratory, Chapman (1971) found no correlation between intellectual level and the production of metaphor, although it was a factor in the preference for metaphor. Indeed, the few attempts to relate novel metaphor production to personality variables have not met with much success (Lockwood, 1974; Porter, 1969); nor has any correlation been found between metaphor production and cognitive style as measured on O.J. Harvey's This I Believe Test (Chapman, 1971), or on the Rorschach (Lockwood, 1974). These latter studies contain suggestions that low novel metaphor producers are more apt to use constrictive, repressive defenses 50 resulting in "rigid and ineffective use of inner resources" (quoted in Pollio, Barlow, Fine and Pollio, 1977, p. 90). Consequently, these authors suggest that high as opposed to low novel metaphor producers are more likely to exhibit cognitive flexibility and ego integration (Ibid., p. 91). This line of reasoning suggests that novel metaphor production will correlate positively with ego strength. Free Association, Regression and Body Posture Studies If quantitative--not to mention experimental--studies of the spontaneous production of figurative language have been few and far between, this is equally true for studies of free association. Over ten years ago, Bordin drew attention to the surprising fact that: Considering its central position in psycho- analysis, it is surprising that so little research on free association has been attempted. What work there is has been centered on word association. My search of the literature... uncovered only two studies involving free association, one an unpublished doctoral dissertation, and remarkably little effort to deal directly with the process (Bordin, 1966, p. 131). After my own search of the literature, 1 find essentially an identical state of affairs. Most of the studies claiming to use free association actually employ word association tasks (Bushnell, 1968; Galbraith and Kahn, 1968; Lieberman, 1968). As is often the case, the use of the same word to refer to an experimental task serves to blur important distinctions in the actual task requirements. Thus, the selection of particular stimulus words provides a topical set for organizing associations. In this sense, word association instructions represent a more structured task than either the Rorschach or Free Association proper (Rabin, 1968, p. 12). 51 The one remotely relevant word association study (Bushnell, 1968) attempted to relate levels of arousal (operationalized by upright posture/ goal oriented word association instructions and reclining posture/ reverie oriented word association instructions) with differences in primary process word association (defined as drive/neutral word choice, primary process level, importation and closeness/distance) to homonymous words (with one neutral and one drive related meaning). Failing to find the predicted relationships, the author cited anecdotal experimenter observations and subject reactions to make the point that goal oriented subjects tended to interpret the task as one of problem solving. In addition, contrary to predictions, subjects given the reclining/reverie instructions seemed more anxious and constricted rather than more relaxed and loose. Using the more ambiguous stimuli of the Rorschach, one study of induced regression (via hypnosis and fantasy instructions) did find an increase of primary process responses (measured by Holt's scoring system) under regressed conditions (Ackman, 1960). Further support for this study can be found in Fromme,et al's (1970) finding that hypnotized subjects produced more primary process indicators on their Rorschach responses, but that only "highly adjusted" subjects demon- strate "regression in the service of the ego" (again measured by the Holt scoring system). Turning from tasks with predetermined stimuli to those employing unrestricted verbalization, one study of conversational speech under conditions of induced regression (via alcohol ingestion) found an increase in the use of first person pronouns, neologisms and expletives and a decrease in the use of conjunctions and complicated 52 future tenses (Consever, 1960). With respect to free association proper, Bordin (1966) failed to find important correlations between free association ratings (for involvement, spontaneity, and freedom as well as for content categories) and a large number of personality variables. Summarizing his results, Bordin observed that a well balanced personality is more able to respond effectively to a free association task; and he suggests using an Ego Control Scale to further test this conclusion. In contrast to Bordin's focus on subject variables, Kroth and Forrest (1969) studied free association effectiveness (using Bordin's scale) by manipulating situational variables. The study found that low anxious subjects free associated more effectively when in the supine position than in the upright sitting position. The reverse was found for high anxious subjects. The effects of changing body position on other cognitive variables has also been investigated. Berdach (1965) studied the differential effect of upright and reclining postures on the free recall of memories. She found that although there were no differences in the overall number of memories recalled, subjects in the reclining position recalled more childhood memories (from birth to age 3) than did sitting subjects, and that the mean age of memories recalled was also earlier for the supine group. This finding was supported in a study by Weinstein (1966). In a related experimental format, Morgan and Bakan (1965) found that in a sensory deprivation experiment, lying down subjects produced more sensory deprivation hallucinations than did upright subjects. 53 Verbal Productivipy and Silence Studies Several studies have drawn attention to the effects of anxiety and situational, interpersonal and stimulus ambiguity in the production of speech dysfluencies and silences. Students of "paralinguistics" (Harper, Wiens, Matarazzo, 1978; Mehrabian, 1972; Trager, 1958) have been interested in those measures of verbal behaviors that are content free. Of special relevance to the present investigation is the research on such quantitative paralinguistic factors as productivity, silence, hesitations, pauses. Employing the usual method of counting number of words to measure verbal productivity, Siegman and Pope (1972) introduced two ambiguity variables; one affecting the stimulus questions, the other the interpersonal context. They found high ambiguous questions (corresponding to "Directed Association" instructions), in contrast to low ambiguous questions, produced an increase in the interviewees' verbal productivity. However, when the visibility between interviewer and interviewee was eliminated, verbal productivity decreased. Summarizing a literature review devoted to "talk, silence, and anxiety", Murray (1971) concludes: There is a strong tendency for verbal quantity to be positively related to dispositional and concurrent anxiety, but negatively to situational anxiety...silence tends to be related negatively to dispositional, but positively to situational and concurrent anxiety (Murray, 1971, p. 244). In one study, Siegman and Pope (1972) found patients have more silences on high anxiety than low anxiety days, a relationship which was unaffected by manipulating the degree of anxiety arousal of selected topics. However, only inconsistent relations have been found between composite measures of silences and personality variables. 54 In another study, this measure was successful in differentiating the speech samples of compulsive, impulsive, depressed and delusional patients (Aronson and Weintraub, 1972). However, in a study of depressives compared to controls, no such differences were found (Hinchliffe, Lancashire and Roberts, 1971). For present purposes, the most interesting finding has been that the semantic complexity of the task is associated with longer silences (Goldman-Eisler, 1968; Ramsay, 1966). Given the differences in procedure, variables studied and measures employed, the interpretation of these studies is by no means self- evident. Yet, in spite of the ambiguities, and at least one negative finding, taken as a whole these data tend to support the position that certain situational alterations promote both formal and content mani- festations of regression. The procedures facilitating these alterations were: reclining posture, eye closure, hypnosis, fantasy instructions, alcohol, Rorschach testing and free association. If anything, the bearing of those few studies on figurative language that were not limited to a forced choice format, was even harder to determine. Nevertheless, there were clear indications that rather than being a function of intelligence or personality traits, the use of novel figurative language was promoted by instructional (e.g., "use a 'feeling approach' to problems") and situational (problem centered) factors. Finally, studies of verbal productivity and silence point to a U-curve relationship between task ambiguity and complexity and speech production such that high and low ambiguity (lack of specificity) reduce productivity and moderate ambiguity increase it(Murray, 1971). 55 Considered together, these lines of research--on figurative language, on free association and regression, and on verbal productivity-- are quite consonant with the theoretical positions taken in the main body of this paper. Stated briefly, and much too abstractly, specific forms of contextual, instructional and stimulus ambiguity promote reductions in the productivity, and regression in both the form and content, of thought and language. The present study will attempt to investigate the effects of specific contexts of association (Free and Directed) on verbal fluency and on the production of figurative language. CHAPTER IV REVIEW AND STATEMENT OF THE PROBLEM In the foregoing sections the subject of figurative language was raised within the broader framework of a psychoanalytic perspective on communicative deviance. After attempting to distinguish pathological from normal deviations, figurative language was identified as that pervasive class of meaningfully deviant communicative activity marked by contextual anomaly and semantic indirectness. At that point in the discussion, objections were raised regarding the adequacy of the usual dichotomous classification of figures into novel or dead--a dichotomy which in itself is an improvement over a total disregard for such differences. It was argued that such a distinction failed to provide criteriafbr discriminating stone dead figures that have become part of conventional and literal usage--e.g., " ollowing an argument," or "conceiving an idea"--and still living but not brand new ones. A further differentiation of the spectrum of figurative language into novel, living and dead was introduced. Anchored at both ends by novel and (stone) dead figures, the class of living figures may be defined negatively as those figures neither novel nor dead, and positively as those still capable of producing some "metaphorical twist". Implicit in the problem of identifying figurative language was the problem of defining it. Consequently, the classical psychoanalytic theory of symbolic deviance as formal regression with respect to modes of representation was reviewed. From this perspective, figurative 56 57 language (at least manifest in contrast to unconscious figurative activity) was regarded as a special form of regression in the service of the ego affecting verbal modes of representation. Specifically, it is a regression from the primary, designative to the marginal, connotative meanings of ggrgs, An alternative approach to figurative language was developed by focusing not on the figurative word but on the figurative sentence. Accordingly, the structural aspects of figures as integrating images and affects into meaningful linguistic forms, came into view. Finally, the integration of these two perspectives into the ongoing communicative act--the speech act--highlighted the contextual-conventional aspects of figurative language. Reference to the contextual aspects of figurative usage drew attention to the figure producing potential of the psychoanalytic setting; and in particular, to the Fundamental Rule of free association. The Fundamental Rule was approached from two vantage points. From one perspective it was viewed as promoting a regressive shift from actively selective, topically organized, directed association to the more passive, unedited thought of the primary processes. Postural changes and eye closure instructions were viewed as amplifying this regressive movement by further loosening the speaker's perceptual, kinesthetic and motoric links to the “General Reality Orientation" (Shor, 1959). A concomitant of this process was a movement from the primary designative meanings of words to the (entertaining of) marginal connotative meanings. From another perspective, it was argued that, in contrast to directed thought instructions, free association in- structions created a systematically ambiguous communicative context. The ambiguities were the result of the absence of feedback and 58 number of specific contextual clashes which created unavoidable para- doxes for the speaker. By making use of the inevitable ambiguity de- rived from the suggestiveness of the figures' multiple meanings, figurative language represents a way of meaningfully meeting and inte- grating the contextual paradoxes. By contrast, failures of integration were seen to lead to silences, dysfluencies, duplicities and so on. These two perspectives on the relation between free association and figurative language were treated as complementary,not contradictory. Consequently, the tensions generated by the Fundamental Rule encourage the speaker to suspend conventional literal ways of making sense--and therewith the primary, designative, meanings of words. Thus, the stage is set for abandoning the ordinary conventions for making sense (including the conventional and literal relationships between word and object, subject and predicate, sense and reference, speaker and listener); and the correlative conventions governing meaningful communication (including the principles of cooperation, politeness and sincerity). In their stead the speaker may call into play a different kind of making sense, a different mode of thought and language more conducive to indirect forms of reference (including humor, re- verie, word play), in short, figurativity. The Fundamental Rule's insistence that the speaker continue speaking intelligibly increases the likelihood that the flexible speaker will then call upon his stock of public and private, marginal connotative meanings; which are then organized by the figurative forms of language into the ongoing communicative event. Stated in the more technical language of psychoanalysis, the contextual ambiguity sets up a regressive movement, the insistence on intelligible speech sets up an adaptive movement. 59 Adaptive regression occurs when the flexible speaker integrates the contextual paradoxes and marginal meanings into meaningful linguistic forms (cf. Kris, 1952). On the basis of these considerations, which point to the unique impact of the Fundamental Rule on the quality and quantity of speech acts, the present study will attempt to experimentally demonstrate that: l) Contexts of Free vs. Directed Association will have differential effects on the quality or figurativity of speech, with Free Association resulting in more and varied use of novel and living figures of speech. 2) Contexts of Free vs. Directed Association will have differential effects on the quantity or fluency of speech, with Free Association resulting in decreased verbal productivity. 3) Reclining/eye closure instructions vs. Upright/eyes open instructions will have differential effects on the quality or figurativity of speech, with Reclining/eye closed instructions resulting in more and varied use of novel and living figures of speech. 4) Reclining/eye closure instructions vs. Upright/eyes open instructions will have differential effects on the quantity or fluency of speech, with Reclining/eyes closed instructions resulting in decreased verbal productivity. CHAPTER V HYPOTHESES Hypothesis I: Subjects under Free Association conditions will be less verbally productive than subjects under the Directed Association conditions. Rationale: The Fundamental Rule of Free Association (in which the patient is requested to say everything that enters his mind, "with- out exception", [Fenichel, 1945]), in contrast to that of topically organized Directed Association, establishes a paradoxical speech situation (Kelman, 1960). As Fenichel and others have pointed out, "to tell everything is much more difficult than one imagines" (Fenichel, 1945, p. 24; cf. Bellak, 1961; Brenner, 1976). Other investigators have identified the effects on verbal productivity of moderate and high stress speech situations based on increasing instructional ambiguity (Siegman and Pope, 1972). On this dimension, the different speech situations created by the Free Association as compared to the Directed Association instructions correspond to high and moderate stress conditions respectively. Moreover, an inverted U-curve has been shown to describe the relationship between stress and performance in general (Fiske and Maddi, 1961), and between stress and verbal productivity in particular; high and low stress has been found to reduce verbal productivity and moderate stress to increase it (Murray, 1971). Consequently, subjects in the moderately stress producing Directed Association conditions should be more verbally productive than subjects in the high stress Free Association conditions. 60 61 Hypothesis 11: Subjects under Free Association conditions will produce more*(a) novel and more (6) living figurative language than subjects under Directed Association conditions. Rationale: The Fundamental Rule of Free Association involves a "relative reduction of certain adaptive functions, including a reduction of secondary process qualities" (Bellak, 1961). This voluntary and temporary controlled reduction of secondary process functioning--inc1uding vigilance, self-editing, external practical orientation, judgmental reality testing--permits the emergence of aspects of less well differentiated and organized primary process thought forms and contents into awareness. The subsequent in- crease of critical scrutiny and selection facilitates awareness of new or formerly unperceived relations. This is the "oscillating function of the ego" from dedifferentiation to heightened self- observation. At such times, when "the regression facilitates the new synthesis...we understand it to be in the service of the ego" (p. 57). In this manner the Fundamental Rule of Free Association suspends the usual conventions for making sense and calls into play different forms of expression and communication (Loewenstein, 1959)- This promotes a movement from straightforward, designative, and literal uses of language to more indirect, connotative and figurative uses. Consequently, invoking the Fundamental Rule decreases the role of active, selective, topically organized modes of thought and increases the role of analogical. imaginative and figurative modes. *The term 'more' will be defined both by the absolute numbers of figures produced by each subject, as well as by the ratio of figures per 100 words spoken by each subject. will ppen 62 Hypothesis III: Subjects under Recliningjeye closed conditions be less verbally productive than subjects under the Upright/eyes conditions. Rationale: Lying down and closing eyes alters the subject's access to the ordinary perceptual, kinesthetic and inter- personal cues that provide the background for common social expectations and consensual validation. The resulting reduction in these accustomed channels of feedback increases the situational ambiguity in those conditions. Greater situ- ational ambiguity has been found to be associated with in- creasing manifest levels of tension and decreasing verbal productivity (Murray, 1971; Siegman and Pope, 1972; Weinstein, 1966). Consequently, the absence of ordinary visual and postural feedback should decrease the verbal productivity of subjects in the Reclining/eyes closed condition as compared to subjects in the Upright/eyes open condition. Hypothesis IV: Subjects under Reclininglgyes closed conditions willyproduce more*(a) novel and more (b)livingfigurative language than subjects under the Upright/eyes qppn conditions. Rationale: By loosening the perceptual, kinesthetic and motoric links to the "generalized reality orientation" (Shor, 1959), instructions to lie down and close eyes increase subjects' access to inner sources of stimulation, including what William James (1950) termed "fringes of consciousness" and Freud called "the preconscious". For similar reasons, reclining on the couch 1"*The term 'more' will be defined both by the absolute numbers of figures produced by each subject, as well as by the ratio 0f figures per 100 words spoken by each subject. 63 in psychoanalysis as well as in progressive relaxation, many hypnosis inductions, and sleep, is generally regarded as a means of reducing muscular tensions and sense impressions that might distract from an increasingly internal focus of awareness (Hilgard, 1979). Based on similar considerations, Paul Bakan and his associates have reported a series of studies which have experimentally demonstrated the differential effects of the lying down and sitting up positions on several cognitive functions. In one study, Berdach (1965) found that reclining subjects reported significantly more childhood memories (from birth to age three) than sitting up subjects (cf. also Weinstein, 1966). In another study, (Morgan and Bakan, 1965), reclining subjects reported significantly more sensory deprivation hallucinations than subjects in the upright position. A corresponding alteration in cognitive functioning from the more precise and literal anchorage of language and meaning to the multiple meanings of analogical and figurative modes has also been noted (Kubie, 1973; Rapaport, 1951). Consequently, if subjects in the Reclining/eyes closed conditions have greater access to preconscious and analogical modes of thought, this should be reflected in the increased use of figurative language. CHAPTER VI METHOD Subjects Sixty-four (32 male and 32 female) subjects were selected randomly from a population of undergraduate volunteers who had signed up to participate in an experiment entitled "mental processes". Due to experimenter errors in administering the experimental conditions, the results of three subjects had to be eliminated from the final data base. Experimenters One male and one female experimenter were selected from a group of advanced undergraduate psychology majors. Experimenters were blind to the experimental hypotheses and to the more general fact that the experiment concerned figurative language. Materials and Experimental Settings Two identical small experimental rooms were used, except that one contained a large black reclining chair (used by reclining subjects) and one small straight back chair (for the experimenter), while the other contained two straight back chairs (for upright subjects and experimenter). In addition, each room contained a small table on which a tape recorder was placed. The Barron's Ego Strength Scale (Barron, 1953) was also included as a control measure, but no specific hypotheses concerning it were stated. Procedure After subjects were randomly assigned to one of four treatment groups (to be described below), with the provision that no more than 64 65 eight male and eight female subjects could be assigned to any one condition, they were then randomly assigned to the experimenters. Subjects were contacted by telephone by the assigned experimenter and an appointment was made. At the appointed time, the experimenter met with the subject and conducted him or her into the experimental room. Subjects were then given a consent form and also notified in writing that they could terminate the experiment at any time. At the same time, the experimenter briefly explained the experimental tasks as follows: I will give you instructions which I think will be self-explanatory. If any questions occur to you, I will be glad to answer them at the end of the experiment. I will also ask you to fill out a fairly brief questionnaire (The Barron's Ego Strength Scale from the MMPI). After requesting permission to tape record ("if you would be interested in participating, this research requires that the session be tape recorded"), the experimenter gave the subject the experimental directions. Once the experiment was under way, the experimenter spoke only to give instructions. If a subject asked a question concerning the instructions after they had been given, the experimental directions were repeated. If the subject asked a different question, the experi- menter said, "I'll try to answer any questions you might have after the experiment is over." If one minute elapsed during which the subject remained silent, the experimental instructions were repeated. Each subject attended one session in which he or she received one of the following four experimental conditions: 1) Reclining/eyes closed with Free Association instructions 2) Reclining/eyes closed with Directed Association instructions 66 3) Upright/eyes open with Free Association instructions 4) Upright/eyes open with Directed Association instructions In each condition, the experimenter's chair was located behind the subject, out of the subject's field of vision. Each condition lasted for five minutes. Sixteen subjects received each experimental treat- ment. Each experimenter ran sixteen subjects, eight male and eight female, in each of the four conditions. The following instructions were given to the subjects at the beginning of the two (Reclining/eyes closed; Upright/eyes open) Free Association conditions. Now, with your eyes closed (open), I would like you to describe whatever thoughts and feelings occur to you. These thoughts and feelings may be about anything. Just let your mind wander and tell me the thoughts and feelings that come to your attention without leaving out a thing. Now, go ahead. The following instructions were given to the subjects at the beginning of the two Directed Association conditions (Reclining/eyes closed; Upright/eyes open). Now, with your eyes closed (open), I would like you to choose one subject to talk about. I would like you to select any subject you care to and describe your thoughts and feelings about it. Now, go ahead. When five minutes elapsed, the experimenter stopped the subject and asked him to fill out the Barron's Ego Strength Scale. After completion of this questionnaire, subjects were asked if they had any further questions. They were also asked about their impressions of the experiment and their personal reactions. All subjects were told that their efforts have been exactly what the experimenter had been look- ing for, and were asked not to divulge the procedures of the experiment. 67 Each experimenter transcribed the data he or she collected. They were also asked for a total word count as well as for a breakdown into hundred word sections. The Ego Strength Scale was also scored. The transcript and Ego Strength Scale for each subject was given a code number. Scoring The Training Manual for Identifying Figurative Langyage (Barlow, et al., 1970) was used to provide a framework for scoring the data. This manual provides a programmed instruction technique for training raters to recognize 14 different figures of speech--ranging from metaphor and irony to metonomy and pun. Basically, the training comprises didactic materials on the selected figures followed by independent rating tasks for four different prose passages; two literary selections, one speech and one transcript of a psychotherapy interview. After each passage was rated independently, the judges met to discuss their ratings, to evaluate their differences and attempt to reach consensus. In addition to recognizing instances of figurative language, raters were also asked to decide whether a given instance was "novel", "living“ or "dead". For reasons already enumerated, the present study departed from the Training Manual's dichotomous division of figures into "frozen" and "novel". Instead, a threefold division of the con- tinuum of figurative language was introduced. As indicated, "dead" figures were described as those which have become part of ordinary usage. In practice, this is reflected in their being listed among the basic, acceptable dictionary meanings of a word. On this criterion, ”1gg_of a table", "neck of a bottle", "point of an argument", "conceive an idea", "follow a tune", and so on, would count as "dead" figures. 68 By contrast, "npyglf figures were described as those linguistic usages that the rater had never encountered previously in that context. "_jyjngf figures were defined negatively as those neither "novel" nor "dead"; or positively as including the deviant dictionary meanings labeled, "informal", "curses", "slang", "vulgar", "idioms", and "cliches". On this criterion, "he's a (pig, ass, mule, lion, fox, bird, dog, jive turkey...)", "she's a (tomato, meatball, ice cube...)", “they're (birdbrains, pinheads, potheads, smackheads, 'wasserkopfs', feather- heads...)" would count as "living" figures. Although an effort was made to score the transcripts for "dead" figures, it became apparent that because so much of ordinary language is a graveyard for dead metaphors, it was exceedingly difficult to develop scoring criteria that would meet the usual standards of reliability. Since none of the experimental hypotheses concerned "dead" figures, no further mention will be made of them in the present study. Consequently, two major categories of figures were employed: the npy§1_figures and the liying_figures. Under each of these categories the following ten subcategories were considered: 1) metaphor, 2) similie, 3) hyperbole, 4) litotes,5) rhetorical question, 6) irony, 7) paradox, 8) personification, 9) metonomy, and 10) oxymoron. Using this procedure each subject received scores for the total number of words spoken, for the number of novel figures (N), and for the number of living figures (L). Since some subjects restrict their use to the same types of figures which they employ repeatedly, while others employ a variety of different types of figures, each subject received two sets of scores for novel and living figures: 1) the total number of figures used, i.e., "figurative productivity"; 2) the total 69 number of different types of figures employed, i.e., number of "figurative categories"; e.g., a subject using no other figurative category than, say,'hyperbole' would get a "categories" score of one, regardless of the number of hyperboles produced; whereas a subject using a variety of different figurative categories, say 'hyperbole', '1itote', 'metaphor' and 'simile', would get a 'categories' score of four regard- less of the number of instances in any particular category. To equalize subject performance of unequal lengths, all N and L values were divided by the total number of words produced by a subject and then multiplied by 100. As a result, scores represent the figurative productivity and number of figurative categories per 100 words of transcript. The data was scored independently by two judges; one, a professional editor, the other a writer of children's stories, and both former teachers of English. The transcribed data gave no indication as to which experimental treatment a subject had been assigned, nor the sex of the subjects or identity of the experimenters unless this information was presented in the subject's transcribed remarks. CHAPTER VII RESULTS Reliability For the first variable, figurative productivity, interjudge agree- ment was defined as an instance in which both raters assigned the same category (novel or living) of figuration to the same speech sequence. A disagreement was indicated when one judge scored a sequence which the other did not, or when judges assigned the same sequence to different categories. The percentage of interrater agreement was represented by the total number of agreements divided by the total of the agreements and the disagreements. Raters agreed on 988 out of a total of 1290 instances of living figures, and on 18 of 21 instances of novel figures. The re- sulting quotient of .86 for novel figurative language and .77 for living figurative language indicates adequate reliability. For the second variable, number of figurative categories, an agree- ment consisted in the two raters independently assessing that a subject's transcript contained one or more instances of a particular figurative category (e.g., metaphor, simile, hyperbole...). A disagreement was re- gistered when one rater assigned a figurative category to a transcript while the other did not. The percentage of interrater agreement was represented by the total number of agreements divided by the total number of agreements and disagreements. Raters agreed on 221 out of a total of 249 instances of living figurative categories, and on 16 of 19 instances of novel figurative categories. The divided resulting percentages of agreements of .84 for novel figurative categories and .89 for living figurative categories indicates adequate reliability. 70 71 After each transcript was rated independently, the judges met to discuss their ratings and evaluate their differences. The scores used in this study are those upon which both judges agreed after discussion. Experimentalgflypotheses Effects of Free Association Conditions on Verbal Productivipy. Hypothesis I proposed that subjects under Free Association conditions would be less verbally productive than subjects under Directed Association conditions. This hypothesis received strong and unambiguous support. Table 1 reveals that verbal productivity (number of words spoken) was significantly lower for Free Association subjects (p < .001) in a one-way analysis of variance. Effects of Free Association Conditions on the Spontaneous Use of Novel and Living Figurative Languagp. Hypothesis II predicted that sub- jects under Free Association conditions would produce more 1) npygl_and more 2) ljyjng_figurative language than subjects under Directed Association conditions. Table 2 addresses the issues of the figure producing potential of Free Association conditions. This table presents the results of a one- way analysis of covariance in which productivity of novel and living figures were the dependent variables and Ego Strength the control vari- able. Turning first to the results for productivity of figurative use, Table 2 reveals that the productivity (in terms of figures per one hundred words of transcript) of ngygl_figures was significantly greater (p < .02) for Free Association subjects than for Directed Association subjects. However, Table 2 also shows that a significant difference was not obtained when Free Association and Directed Association subjects were compared for productivity of living figurative use. These data lead Table 1: Analysis of Variance for Verbal Productivity (Number of Words) as a Function of Free Association vs. Directed Association Instructions Free Directed Free Association Association Association vs. Treatment Treatment Directed Association (N = 31) (N = 30) M SD M SD df MS F Verbal 306 213.84 468 233.62 1 406.408 8.12* Productivity *p < .01 73 Table 2: Analysis of Covariance for Novel and Living Figurative Productivity as a Function of Association Conditions (with Ego Strength Controlled) Free Directed Free Association Productivity Association Association vs. (N = 31) (N = 30) Directed Association M M df MS F N°Ve‘ 1 17 029 1 311 6 09* Figures ‘ ' ' ' Living Figuresz 5.48 5.72 1 4.137 .505 *p < .02 1 Since Ego Strength was not significantly correlated with novel figures, it was dropped from the analysis. Reported results are from a one-way analysis of variance. Full analysis of covariance table appears in Appendix 0. 2E90 Strength was positively and significantly related to productivity of living figures (r = .34; p < .01). 74 to the conclusion that Free Association conditions significantly increase the production of novel figures, but not the production of living figures. A further detailed analysis of the data appears in the examination of the variety of figurative categories used in the different treatment conditions. Table 3 shows that the number of categories of ngy§l_ (p < .01) and the number of categories of liying_(p < .001) figures were significantly greater for Free Association subjects than for Directed Association subjects. Since these findings were based on a rate measure (figurative categories per 100 words of transcript) further analyses were performed to determine whether the rate measure accomplished its goal of equalizing transcripts of unequal length (number of words). With respect to the productivity scores, there was no significant correlation between novel or living figurative productivity and verbal productivity (see Table 4). This indicates that the rate measure was effective in eliminating differences due to fluency, with respect to figurative pro- ductivity. However, with respect to number of living figurative categories, a significant negative correlation was obtained for the rate measure and verbal productivity (see Table 4). Thus, the rate measure failed to eliminate the effects of differences in fluency. Instead, it overcorrected for fluency, penalizing highly fluent speakers with lower scores. One splution to this dilemma would be to get rid of the rate measure altogether and simply score the absolute number of figurative categories produced by each subject. Following this line of reasoning, Table 5 presents the results of a two-way analysis of variance in 75 S van. «ammo.m mm mum. moo.~ New“. pou.F m=w>w4 «om.m mm moo. mpo. mmnm. mowmp. pm>oz mapa>uu we am 2 cm : mmegommumu unaaogh umuumcwo cowumwuomm< owed m>wumszmwm cowuwucou m>wuawuomm< we cowuocaa a ma mmvgommumu m>_pmg:mwa mcw>P4 use ~m>oz "m m_nah 76 Table 4: Correlations Between Verbal Productivity and Figurative Productivity Corrected for Fluency Verbal Productivity: Free Association Verbal Productivity: Directed Association Conditions Conditions) r r ESSEie'Eiiu‘fiZs -°31 "03' Categories of -.O34 .052 Novel Figures *p < .002 77 which the dependent variables were the number of novel and the number of living figurative categories per subject, and the treatments were Associative and Postural conditions. Table 5 shows that Free Associating subjects produced significantly more ngygl figurative categories than did Directed Associating subjects (p <.01). However, Table 4 also shows that there was no difference for ljyjng_figurative categories. These findings for number of figurative categories parallel the findings for figurative productivity: subjects under Free Association conditions produce more novel figurative language--both in frequency and category--but not more living figurative language, than do subjects under Directed Association conditions. When considering these findings for number of figurative categories, it is worth noting that while the rate measure overcorrected for fluency, the use of absolute number of categories undercorrected for it. A correlation of .545 (p < .002) was found between the absolute number of living figurative categories and verbal productivity for Directed Association subjects, and a correlation of .452 (p :<.02) was found between the absolute number of living figurative categories and verbal productivity for Free Association subjects (see Table 6). Consequently, by using the absolute scores, the deck was stacked against lower fluency subjects. Since Free Associating subjects had significantly less verbal productivity than Directed Associating subjects, the fact that significant results were obtained at all becomes even more impressive (see "Additional Findings" section below, for an additional approach to this difficulty). Effects of Postural and Eye Closure Changes on Verbal Productivity. Hypothesis III predicted that subjects under Reclining/eyes closed 78 Table 5: Variety of Novel and Living Figurative Categories as a Function of Associative Condition (Uncorrected for Rate) Free Association vs. Directed Association (If SS MS F Novel Figurative * Categories 1 2-1507 2.1607 7.249 Living Figurative Categories 1 2 2.571 .8104 *p < .01 Table 6: 79 Correlations Between Verbal Productivity and Figurative Productivity Uncorrected for Fluency Verbal Productivity: Free Association Conditions Verbal Productivity: Directed Association Conditions Production of Living Figures Production of Novel Figures Categories of Living Figures Categories of Novel Figures .797*** .436* .452* .346 .671*** .148 .545** .113 * p < .02 ** p < .002 ***p < .001 80 conditions would be less verbally productive than subjects under Upright/eyes open conditions. This hypothesis received no statistical support. Table 7 shows that there were no significant main effects on an analysis of variance, but the mean differences between the groups were in the predicted direction. These means were respectively 352 words for reclining subjects and 423 words for upright subjects. Effects of Postural and Eye Closure Changes on the Spontaneous Use of Novel and Living Figurative Languagp, Hypothesis IV predicted that subjects under Reclining/eyes closed conditions would produce more 1) npygl_and more 2) ljying_figurative language than subjects under Upright/eyes open conditions. Table 8 shows that none of the main effects for productivity of novel and living figures was significant. Nor were the comparisons for novel and living figurative categories significant (see Table 9). However, the mean differences for the productivity of novel figures and the variety of novel and living figurative categories were all in the predicted direction. These means were respectively, .11 and .09 for frequency of novel figures, .11 and .06 for number of novel figurative categories and 1.50 and 1.26 for number of living figurative categories. 1heseresults were computed on the basis of the rate measure of figures per one hundred words. Since it was determined that there is a significant correlation between verbal productivity and variety of figurative categories, these results were reanalyzed using the uncorrected number of figurative categories as the new scores for the dependent variables. On this basis, Table 10 shows that there were no 81 Table 7: Analysis of Variance for Verbal Productivity as a Function of Postural and Eye Closure Conditions Reclining/Eyes Upright/Eyes Reclining/Eyes Closed Closed Open vs. (N = 29) (N = 32) Upright/Eyes Open M SD M SD df MS F Verbal Productivity 352 212.39 423 254.96 1 82.663 1.64 82 Table 8: Analysis of Covariance for Novel and Living Figurative Productivity as a Function of Postural and Eye Closure Conditions . Reclining/Eyes Upright/Eyes Reclining/Eyes Closed Elg::::::$ty Closed Open vs. (N = 29) (N = 32) Upright/Eyes Open M M df MS F Novel] .106 .093 1 .009 .185 Livingz 5.51 5.685 1 7.063 .863 1Since Ego Strength was not significantly correlated with novel figures, it was dropped from the analysis. Reported results are from a one-way analysis of variance. 2Ego Strength was positively and significantly related to productivity of living figures (r = .34; p < .01). 83 Table 9: Novel and Living Figurative Categories as a Function of Posture and Eye Closure Condition Reclining/Eyes Upright/Eyes Figurative Categories ($12533) (Nogegz) M SD M SD df t Novel .ll .28 .06 .13 59 .952 Living 1.50 .83 1.26 .64 59 1.27 84 significant differences between the production of novel or living figurative categories for either postural/eye closure treatment con- dition. Table 11 shows that when figurative productivity scores are uncorrected for rate, there were also no significant differences for novel or living figurative productivity between these two treatments. Experimenters' Observations and General Characterizations of the 9222: Repeatedly the experimenters were struck by the sense of dis- comfort on the part of many of the subjects as well as the various forms of instruction seeking and approval seeking behaviors. Moreover, the 'pressure' of this activity on the experimenters occasionally provoked them to deviate from the stance required by the instructions. In particular, they were hard pressed to maintain silence in the face of requests for feedback, approval, and direction. On at least one occasion, a transcript had to be eliminated because the experimenter responded to a request for directions by presenting the Free Association instructions in the context of an already established Directed Association condition. Moreover, the experimenters acknowledged strongly negative reactions to subjects who subjected them to this kind of interactional pressure to deviate from their own rule-governed position. Several subjects complained that the eye closure/reclining requirement produced more discomfort than relaxation. Several other subjects volunteered the information that they did not comply with the instructions "to say everything" that came to mind. Others sounded especially stilted or especially superficial. Elaborate attempts at trying to be organized made one experimenter comment that the subject "sounded like he was on a job interview". Another presented a 85 Table 10: Variety of Novel and Living Figurative Categories as a Function of Posture and Eye Closure (Uncorrected for Rate) _ . Reclining/Eyes Upright/Eyes F19urat1ve Closed Open Categor1es (N = 29) (N = 32) M SD M SD df t Novel .24 .51 .31 .59 59 -.5 Living 4.03 1.68 4.25 1.74 59 -.49 86 Table 11: Novel and Living Figurative Productivity as a Function of Postural and Eye Closure Conditions (Uncorrected for Rate) Eiggratgvgt Recéigggg/Eyes UpriggggEyes ’° ”C V‘ y (N = 29) (N = 32) M SD M SD df t Novel .24 .51 .44 .84 59 -1.088 Living 19.69 13.63 23.34 15.52 59 -9.72 87 caricature of Free Association which seemed like he was responding to a self-administered word association test. In this context, it was plain to the experimenters, and also evident upon reading the transcripts, that some subjects failed to conform to the experimental instructions. More specifically, some subjects responded to the Free Association instructions as if they were the Directed Association instructions and vice versa; thus, some macalled "free associators" seized on a single externalized topic and never deviated from it for the duration of the experiment. Similarly, some so called "directed associators" flitted from topic to topic to subjective state to subjective state without ever choosing one topic to talk about. Finally, some subjects seemed to combine both kinds of responses. On a more general level, perhaps the most striking aspect of these data is their banality; their comparative lack of real involvement aided and abetted by a pervasive ambiguity which rendered many of the speech acts impossibly unclear. This lack of directness, perspicuous- ness and involvement made those instances of engaged speech all the more striking by their comparative rarity. It is noteworthy in this regard that many of the novel figures produced in this study came in the context of subjects actively engaged in and struggling with their puzzlement about what the Free Association instructions seemed to demand of them. That so many speech acts were infelicitous, tentative, lacking directness, and 'punch' or misfiring into various forms of dysfluency and silence provided graphic evidence of the very special speech situations created by the experimental conditions. 88 Additional Findings Although specific hypotheses were not formally stated, the findings bearing upon 1) the moderating effects of Ego Strength on verbal productivity and figurative use; 2) correlations between figurative productivity and figurative category scores; 3) effects of postural changes on Free Association subjects; 4) methods of eliminating differences in verbal productivity; and 5) the effects of Sex of subjects on verbal and figurative productivity§are presented in this section. 1) Contrary to what might have been expected, Ego Strength1 was not found to be significantly correlated with verbal productivity or with the production of novel figures on an analysis of covariance. Nor were there significant correlations between Ego Strength and the variety of novel or living figurative categories on Pearson product- moment correlations. However, a significant relationship (p < .05) was obtained between Ego Strength and productivity of living figurative use on an analysis of covariance, with higher Ego Strength correlated with higher productivity of living figures (see Table 2, foot- note 2). Thus, subjects with higher Ego Strength used more living figurative language than did subjects with lower Ego Strength. 2) Significant correlations were also found between figurative productivity and number of categories of living figures used, such that subjects producing more living figures also used a wider variety of living figurative categories (see Table 12). Similarly, for Free Associating subjects, there was a significant correlation between the variety of novel and living figurative categories employed, i.e., Free Associating subjects using a wider range of categories of living 1 Means and ranges on Ego Strength Scale appear in Appendix D. 89 Table 12: Correlations Between Living Figurative Productivity and Living Figurative Category Scores (Corrected for Fluency) Free Association Directed Association Figurative Figurative Productivity Productivity r r Living Figurative .479* .656** Categories * p < .006 **p < .001 90 figures also employed a wider range of novel figurative categories (see Table 13). 3) It was also observed that within the Free Association condition changes in posture and eye closure did not significantly affect verbal or figurative productivity. However, a comparison of group means indicated that Free Associating subjects who were reclining with eyes closed tended to produce fewer words but produced more novel figures and used more categories of novel and living figurative speech than did Free Associating subjects who were upright with eyes open (see Table 14). 4) An additional approach to the solution of the problem of the biasing effects of using either the rate measure of figures per one hundred words or the absolute number of figures produced was also attempted. Since the problem revolved around the differential impact of unequal productivity (fluency) scores, it could be solved by finding a method of equalizing verbal productivity for all subjects. One radical way of accomplishing this was to find the least fluent subject, get that subject's verbal productivity score, (which turned out to be fifty-eight words), and then score only the first fifty-eight words of all other transcripts. In this fashion all of the statistical dilemmas could be solved, albeit at the expense of the verisimilitude of the study. Nevertheless, these data are included. Nhile none of these findings reached significance, an inspection of the results on Table 15 shows that when comparing Free Associating subjects with Directed Associating subjects, the means for living figurative frequency and living figurative categories are in the predicted direction. The reverse is the case for Reclining/eyes closed subjects versus Upright/eyes open subjects. 91 Table 13: Correlations Between Novel and Living Figurative Category Scores (Uncorrected for Fluency) Free Association Directed Association Living Figurative Living Figurative Categories Categories r r Novel Figurative .429* .194 Categories *p < .02 92 Table 14: Comparisons of Means for Upright/Eyes Open Subjects and Reclining/Eyes Closed Subjects within the Free Association Condition Upright/Eyes Open Reclining/Eyes Closed Verbal Productivity 343 269 Novel Figurative .14 .20 Productivity Living Figurative 6.1 4.86 Productivity Novel Figurative .1 .21 Categories Living Figurative 1.53 1.82 Categories 93 um.p mmwcommumu mn.F mmwgommpmu mommgm>< mo.~ xuw>wuu=uoca mo.m xuw>muuzuoga Ppmcm>o m~.p mmmgommumu o¢.F mmwcommumu mo.~ mmwcommumu Fo.m apw>wpu=uoza mp.N »»_>w66=6628 mo.m sow>866=6668 cage aezpwma mm»m\a;owza= maxp mmgzmwu.mcw>wd mmgampmlumw>P4 sm.P mmmcommuou mm._ mmwgommumu ne.P momgommuau ¢~.~ auw>muo=uoga cp.m auw>wuoacoga mm.~ auw>wuuauoga cmmopo mmxm\m:_cmpumm mmcamMMTmcw>wg mogamwm mcw>m4 mommcm>< ppmgm>o cowumwuomm< umaomcvo cowumwuomm< mag; cowumpuomm< mo wash muumwnam ppm mo mucoz pnmwmuzumwm “mew; cow mmgoum Acommumu use apw>wpuzcoga m>waae=awa u=v>84 um_ «Pack 94 The results for novel figures are not included since in the first fifty-eight words, for all subjects, only two novel figures were produced! A final approach to the potential biasing effect of fluency on living figurative categories was done as follows. To partial out verbal fluency so that a living-categories score uncorrelated with flu- ency could be calculated, the correlations between fluency and living categories within both association groups were pooled using the Fischer r-to-z transformation. Scores on fluency and on living categories then were put in standard score form. The pooled correlation was .547. The standard fluency score weighted by this coefficient were next subtracted from the corresponding living category scores to yield the desired scores, which were uncorrelated with fluency. Since the results obtained by using these transformed scores in an analysis of variance were not significant, they have not been included. 5) The effects of the sex of subjects on fluency and on novel and living figurative productivity were determined by an analysis of variance. There were no significant main effects and no significant interactions. Summary of Results The major findings represented in this section have been summarized in Tables 16 and 17. Free Association vs. Directed Association. As predicted, Free Associating subjects produced fewer words than did Directed Associating subjects. They also produced a higher frequency of novel figures and a wider range of novel figurative categories. Free Associating subjects also produced a wider range of living figurative categories, 95 mo pm>wp anmuamuom mcwcwwuum «o: canon» cowuumcwc nouuwumca mg» cw mapsmmg mmpmopucmm poo. v akxx Fo. v a «x No. v a « om: umm om: umm cum mom: 0mm new: mom: 0mm **«wumg=mwm mcw>wd auw>vuoauocm m>wumcampg mcp>w4 mopgommumu m>Pamgzmrm Fm>oz zuw>wuu=voga m>wumg=apm ~m>oz a68>_»u=uoea .anem> Aomzv cage mm»m\g;mwca= .m> Aumav ammopu mm»u\u=m=wpua¢ A A66663666 m=6>64 auw>6pu=uog6 06: u 666 wpmczmw6 m=w>64 mmpgomwpuu cu: u gum 88wum6=mv6 Pm>oz xuw>wuu=uo66 06: u 666 k6666=av6 Pm>oz 606:6 666° mm»6\626666= «<66 covumwuomm< 66666669 .m> .m> Rummy ammo—u mmzm\mcvcwpumm <6v comummuomm< «~66 Amumm so» umuumgcoocav mupzmmm we acmsszm "up «Pam» 97 but the groups did not differ with respect to frequency of living figurative use. When figurative scores are uncorrected for rate per one hundred words, the significant finding for living figurative categories drops out. All other results remain the same. Reclining/Eyes Closed vs. Upright/Eyes Open. "Reclining/eyes closed" subjects did not produce significantly fewer words than "Upright/eyes open" subjects although the mean differences between the groups were in the predicted direction of less verbal productivity for Reclining/eyes closed subjects. Similarly, “Reclining/eyes closed" subjects did not produce a significantly higher frequency of novel or living figures or a significantly greater variety of novel and living figurative categories, yet, with the exception of frequency of living figures, all of these mean differences were in the predicted direction of greater figurative productivity for Reclining/eyes closed subjects. When the scores for figurative use are uncorrected for rate per one hundred words, the mean differences are still not significant. However, with this change of scoring method, the mean differences are opposite the predicted direction. CHAPTER VIII DISCUSSION The gap between theory and practice has always been something of an embarrassment of psychoanalysis. The habit of mind that tends toward the construction of highly elaborated cognitive, motivational and developmental structures, often operating at very different levels of abstraction, has led to a predictable reaction on the part of certain leading psychoanalytic theorists. In recent works, George Klein (1976) has attempted to distinguish what he conceded to be the gold of "clinical theory" from the copper of "metapsychology". Heinz Kohut (1977) calls attention to theorizing that is "data near“ rather than “data distant". And Roy Schafer (1976) has argued that the "native tongue" of psychoanalysis is a kind of "action language" which can dispense with the epicycles of metapsychology altogether. Common to all of these reactions is a desire to look more closely at what actually happens in psychoanalytic sessions, to loosen the hold of predetermined categories in order to allow the data to 'speak' in previously unrecognized ways. It is probably no accident that two of these three were trained as psychologists. The approach of the present study also takes its bearing from the effort to describe the effects of certain basic features of the psychoanalytic situation--(the Fundamental Rule of Free Association, the reclining posture)--on speech activity (viz., fluency/productivity and indirectness/figurativity). A discussion of the results presented in the previous chapter will help articulate some of the complex 98 99 relationships holding between these variables, thereby helping to clarify the meaning and the use of Free Association and figurative language. Free Association: Regression in the Service‘of the EggUMoggl A major finding in this study was that consistent with what classi- cal psychoanalysis has held all along, invoking the Fundamental Rule of Free Association creates a context which is different in important respects from the context established by requests for Directed Associations. In particular, Free Associating subjects were signifi- cantly less fluent (and more silent) and significantly more figurative than their Directed Association counterparts. This coincidence of less speech but more figurativity presents a challenge to the traditional analytic theory of Free Association as "regression in the service of the ego". On that view, the relaxation of secondary process screening operations and interpersonal vigilance permits the emergence of primary process thought forms into the flow of conscious experience (cf., Bellak, 1961 and Kris, 1952). The general picture is one of a free flow of ideas and affects available for expression and represented in involved, spontaneous and free speech (Bordin, 1966). In contrast to this, our data points to a greater constriction in speech (reduced fluency) at the same time we find a greater creativity, novelty, and richness of expression (elevated figurativity). Prima facie, this must be counted as a paradoxical finding. If necessary, the "regression in the service of the ego" theory can account for the creativity but it would have to employ concepts of resistance and defense to account for the dysfluency. But resistance and defense-guardedness-are the very qualities which must be overcome 100 for creativity and novelty to appear. Their simultaneous appearance is disturbingly contradictory. On the basis of these considerations, it appears that the results of this experiment cannot be said to unequivocally support the theory of Free Association as regression in the service of the ego. While this theory could predict one piece of our data, viz., the increased use of creative language under Free Association conditions, it fails to offer a persuasive account of the other major findings, viz., decreased fluency. Free Association: Contextual Model The contextual model, as represented by the speech act theory of Free Association, was strongly supported by the experiment. The critical test for this position was whether the Free Association condition simultaneously produced more figurative language and less fluency (productivity). Of the five basic tests bearing on this issue--verbal productivity, productivity of novel figures, variety of DOVE] figurative categories, productivity of living figures, variety of living figurative categories-~all but the productivity of living figures were significant and in the predicted direction. By focusing on the effects of the framework created by the Fundamental Rule, attention was drawn to the clash between this rule and other rules (conventions) which ordinarily regulate speech occasions. In an earlier chapter it was shown that the Fundamental Rule—~if strictly adhered to--demands the violation of each of the conversational maxims (quantity, quality, relation, manner) which comprise the basic cooperative principle. (Grice, 1975). In addition, certain 'non-conversational' maxims which ordinarily help to regulate relations between people, including 101 social rules of etiquette, politeness and morality, are also compromised. It was argued that the conflicting demands of the conversational, social and Fundamental Rules create a unique speech situation in which the subject must speak in spite of the requirement of contradictory speech actions. Faced with the inevitable conflicts thus engendered, speakers have two principal Options: to speak ambiguously (indirectly, figuratively) or not to speak at all. Our results confirm that this is precisely what happens. Quantitatively, subjects are less fluent and more figurative. Qualitatively, they seem more disoriented, indirect, instruction-seeking and ambiguous. Free Association: Residual Paradoxes The problematic character of Free Association has not gone altogether unremarked--even within the psychoanalytic fold. Fenichel's comment that Free Association "is much more difficult than one imagines" (Fenichel, 1945, p. 24) just begins to hint at the dilemma. It would also appear that Bellak's (1961) thesis of Free Association as the "ego's oscillating function" from regression in the service of the ego to heightened self observation may be unduly optimistic, more honored in the breech than in the commission. Mark Kanzer (1972) faces the problem squarely when he notes that since in Free Association "opposing trends are to be expected, the stage is set for great inner and outer conflicts. Through a combination of relative freedom to relax with relative responsibility for curing himself of ills, apparently paradoxical and contradictory forms of behavior are required of the analysand" (Kanzer, 1972, p. 248, italics added). The paradox lying coiled at the heart of the Fundamental Rule is brought out with striking clarity in the observation that "when a 102 patient finally learns to free associatie it is a sign that the analysis is over" (cf., Loewenstein, 1963). This is clearly an awkward state of affairs for a theory and practice in which access to data as well as success and duration of treatment “depends on the conscientiousness with which (the patient) obeys this fundamental technical rule of psychoanalysis" (Freud, 1915- 1917, p. 287). Nor is the situation clarified by such comments from leading theorists as "to a large extent, however, analysis often involves 'controlled' rather than 'free association'" (Bellak, 1961, p. 54). (It is worth noting that Bellak's "controlled association" is for all practical purposes identical to our "directed association" condition.) Even Charles Brenner, a leading light of analytic orthodoxy, asserts that "'free association' is a bad term to apply to the psychoanalytic method" (Brenner, 1976, p. 190); a principal reason being "because it obscures the fact that an analytic patient is often asked to associate to a specific conscious stimulus" (Ibid., p. 190). Implicit in these and similar proposals for circumventing the Fundamental Rule is the recognition that there are serious difficulties in adhering to it. In part, the dilemma can be traced to the ambiguity of insisting upon and demanding "obedience" (Freud, 1915-1917, p. 289) to a 'rule' that the analyst knows in advance cannot be followed. The capacity of the "normal ego" to adhere to the Fundamental Rule is supposed to provide a baseline against which fluctuations and deviations in associations are to be measured. In an important sense, Free Association is simultaneously the methgg_and the goal_of psychoanalysis, embodying the values of spontaneity, freedom, and involvement (cf., 103 Bordin, 1966). It provides the model of (passionately) engaged speech by which to confront and affront polite society with its rules of decorum, etiquette and so forth. The opposite of premeditated, con- stricted, inhibited, detached speech, Free Association serves as the norm (or ideal) against which the analysand's speech is compared, and more often than not, found wanting. Deviations from this norm come into view as resistances, defenses, inhibitions, transferences--the "precipitates" of earlier object relationships. Against this background, "regression in the service of the ego" may more accurately describe the molar functioning of a well established "bipersonal field" (Langs, 1976) viewed over time-~or even within a "good hour"--than it does the conditions obtaining at its inception. Certainly, there is little in the data presented here that resembled the prototypical picture of Free Association in the popular mind or in the classic statements of the method. If anything, the miscarriages of this process are much more abundant and striking than the successes. Instruction seeking, banality and circumlocution substitute for spontaneity, freedom or involvement. From this vantage point, Free Associating comes into view as a significant personal achievement not a native endowment, the goal to be sought not the starting point to be assumed. As Spinoza [1955, (1674), p. 2711 observed, "all things excellent are as difficult as they are rare". And yet--and this is the paradox--even from the outset, Free Association foreshadows the goal. The first minutes of the first "session" contain in latent form the seeds to be cultivated over the course of therapy. As a result, even at the very outset, instructions to Free Associate successfully facilitate the emergence 104 of the private and connotative aspects of language, i.e., the figurative vehicles through which idiosyncratic elements of personal style, feeling and vision enter into ongoing communicative acts. Where the linguist, the language philosopher and even the psycho- linguist have been chiefly interested in the common and public denotative aspects of language, "the lunatic, the lover, and the poet" (Shakespeare, 1936), and we might add, the clinical psychoanalyst, have been enthralled by language's "private world responsiveness" (Forrest, 1973; Rabin, 1968). Thus, in spite of the obvious differences between the results of a laboratory experiment and a five hundred hour psycho- analysis, the experiment succeeded in demonstrating that the Fundamental Rule significantly affects the quantity and quality of speech acts. In particular, it highlights the figurative strategies adopted by subjects when conventional means of expression are suspended by a paradoxical injunction. That a novel order of figurative meaning can and does emerge out of the collapse of the conditions for literal meaning opens a new perspective on the genius of "the native tongue of psychoanalysis" (Schafer, 1976). figgrative Language: The Novel-Living Distinction The results of this study support the value of Wittgenstein's (1953) admonition, "don't ask for the meaning, ask for the use"; that in spite of £95981, structural similarities, novel and living figures can function quite differently. By examining the kinds of figurative activity in different speech settings, it was found that novel figures occur significantly more frequently and in a greater variety of categories during Free Association conditions than they do under Directed Association conditions. Thus, it appears that novel 105 figurative activity is much more likely to emerge from the collapse of the conditions for making ordinary kinds of sense brought about by the Fundamental Rule clashing with other language rules and social rules regulating speech. More specifically, novel figures are more likely to emerge out of the tensions created for a speaker attempting to carry out simultaneously contradictory speech acts, viz., to say everything that comes to mind while at the same time attempting to observe the rules of cooperation and politeness, not to mention self- esteem maintenance. Novel figures constitute a crucial and constructive solution to the problem posed by the paradoxical conditions, just as silence, ambiguity and avoidance (e.g., acting as if the instructions asked for Directed Associations) represent other, less successful solutions. Although the observation that novel figures, unlike living figures, much more frequently occurred in situations where speakers were actively engaged in and struggling to express their perplexity was not put to experimental test in this study, it is an exceedingly suggestive observation. It lends further support to Pollio and Barlow's (1975) finding that novel as opposed to frozen figures cluster around problem setting and problem solving segments of a psychotherapy session. In addition, it was found that there were no differences for productivity of ljngg_figures between Free Association and Directed Association conditions. Moreover, the productivity of living figures, but not of novel figures, positively correlates with Ego Strength. Taken together, these findings suggest that unlike the struggle from which novel figures emerge, living figures are part and parcel of a 106 speaker's (especially, a flexible, high Ego Strength speaker's) adaptive repertoire. Thus, living_figures emerge as a dimension of a flexible speaker's coping behavior, what might be called a speaker's 'figurative competence'. Such considerations support the proposition that in important respects novel figures function differently from living figures. For one thing, the frequency of living figurative use appears to be part of a speaker's adaptive repertoire or figurative competence; whereas. novel figures are not part of a speaker's established repertoire in the same sense. Rather, they are more likely to come into play when conventional repertoires and coping strategies prove inadequate. Free Association instruction can be seen as disrupting these conventional strategies with the result that subjects employ a greater frequency of novel figures and a greater variety of novel figurative categories. Consequently, coping with Free Association conditions call upon a wider range of figurative strategies (novel figurative productivity and categories) than does coping with Directed Association conditions. Figurative Language: Speech Act and Dynamic/Motivational Models The hypotheses of this experiment regarding figurative language were derived from two models: 1) the traditional psychoanalytic model of figures as revelations, as derivatives of underlying motivational systems and related notions of regression to primary process forms of thought; 2) the speech act model in which figures perform certain functions set by contextual and interactional con- siderations. The results of this study shed interesting light on the scope and limits of these two models. From a purely motivational point of view, the distinction between novel and living figures is 107 essentially irrelevant, because any and all figures, no matter how moribund, can serve as vehicles for compromise expressions of uncon- scious conflict and drives. In fact, less vital figures might be preferred: by virtue of their very banality they more easily elude the watchful gaze of censorship. Strictly speaking, from the motivational point of view, figurative activity is not confined to the figures of speech and thought, but expands to embrace feelings, actions, persons, and situations. Indeed, when anything can, and does, serve as a metaphor (transference) for anything else, new problems arise. The literal and the ordinary modes of language become suspect, subject to "deconstructions" which not only undermine the novel versus living distinction, but the figurative-literal and the figurative-ordinary distinctions as well (cf., Brenner, 1976; de Man, 1979; Derrida, 1977; Langs, 1976). A more moderate response from the motivational position has been the introduction of the category of blatancy of derivatives, with greater regression resulting in greater blatancy. On this view, Free Association instructions can be expected to produce more blatant depictions of drives and other primary process activity, including more novel figurative language (cf., Holt, 1956). Figures of speech have a special place in this process because they provide acceptable solutions to superego-id conflicts which might otherwise result in symptomatic formations. Nevertheless, it is hard to see how this model might account for the finding that there was no difference in the productivity or in the variety of categories of living figures produced under Free Association or under Directed Association conditions. 108 Either subjects were regressing, in which case more living figures; or they were not, in which case, no changes in novel figures. It's hard to have it both ways. By contrast, the speech act model provides the terminological resources to more easily account for the unexpected finding that the pro- ductivity of living figures is no greater for Free Associating than for Directed Associating subjects. For one thing, the focus of the speech act model is on language as a special form of institutional action, regulated by a variety of explicit and implicit conventions. Furthermore, it helps to elucidate the ways in which The Fundamental Rule systematically undermines these conventions, setting the stage for the appearance of the unconventional. However, our operational definition of living figures as those usages classified in the dictionary as cliches, idiom, archaic, slang, curse, etc., ipso facto acknowledges that these usages, far from being unconventional, are well on their way to being domesticated. Consequently, living figures represent a pool of not quite standard verbal devices potentially available to (native) speakers at any time, i.e., unlike novel figures, they don't have to be invented from scratch. As such, the capacity to employ living figures has more to do with resourcefulness and Ego Strength (see above) than with regression, disintegration and new integrations. The results of this study, as they bear on figurative language, call attention to the special and conflicting properties of the communicational field established by Free Association instructions. In particular, attention was drawn to the corresponding evocation of speech acts which are indirect, ambiguous, and inherently paradoxical, i.e., figurative speech which says one thing and means another. The 109 speech act model provided differentiated ways of focusing on the conventional/communicational aspects of language and the related problems of the wheg_and whe§e_of figures, i.e., the occasions of their use are clarified by attention to special features of the context. By contrast, the traditional model focuses on the why_and hgg_of figures, on the complex strategies by which a system of associations and implications related to one domain are employed to organize our view of another. In the former, concern is with con- flicting rules and demands (unconsciously) operating in an inter- personal context. In the latter, concern is with the conflicting rules and demands (unconsciously) operating in an intrapersonal context. That Free Associating subjects produce more novel figures (in both frequency and category) than do Directed Associating subjects supports the strong psychoanalytic claim that the Fundamental Rule potentiates the revelation of unconscious personal models implicit in the emergent novel figures. It also supports the expectation drawn from speech act theory that conflicting conversational rules will increase non-direct forms of communication, most notably figurative language. Yet contrasting these models, or even suggesting their comple- mentarity, fails to clarify the deeper unity of saying, meaning, and doing embodied in the figurative speech act. In addition to their cognitive meaning, figures have interactive force. They are not only ways of saying something but also ways of doing something by virtue of that saying. At the same time they indicate a non-direct meaning through a direct meaning, they also signal a change in the character of the relationship between speaker and listener. In the 110 context of psychoanalysis, this implies that the meaning of the unconscious model (idea, emotion, object relational paradigm or system of associations) that is made manifest by the figure is simultaneously reflected in the mode of relatedness re-enacted or sought with the analyst. Consequently, in a novel figure, cognitive content is isomorphic with interactive force. From this perspective, the interactive force of figures can be specified by indicating the kind of object relation recreated and/or sought. Thus, metaphors attempt to create involvement and identification between speaker and listener; irony attempts to create distance and seize control of the interaction; hyperboles serve exhibitionistic and mirroring fantasies; and litotes conjure weakness and inferiority, evoking care—taking responses. While oversimplified, this schema adds a new dimension to Ella Freeman Sharpe's (1940) classic example of the metaphor of the analysand who complained of constantly "losing the point". Sharpe notes that these and similar metaphors referred to repressed ideas and emotions concerning specific "suckling experiences" (Ibid., p. 205). The added dimension of the metaphor's 'interactive force' draws attention to the analysand's attempt, via communicating figuratively in metaphor, to induce identification with the analyst, thereby recreating in their interaction, the symbiotic mode of relatedness figuratively alluded to in the metaphor of "not getting the point". Against this background the finding that the Fundamental Rule of Free Association significantly increases novel figurative speech acts provides further support for the view that subjects not only speak but relate differently under such conditions. Or, more accurately, since speaking is relating, modes of relatedness become 111 more figurative, non-direct and transferential (regressed?) under these conditions. Not only as revelations of unconscious ideas and emotions, but as vehicles for altered (regressive) modes of relatedness, figurative speech brings us back to a fundamental insight of Freud's: the very efforts we make to protect and hide ourselves, reveal us most deeply and most uniquely. Such considerations clarify the features of occasions which serve to elicit figurative activity, and thereby help to provide an answer to linguist Jerry Morgan's (1979, p. 147) basic question, "why bother?" using metaphors anyway. Yet other questions still remain for future inquiry. It is by no means clear, for example, whether speakers spontaneously show strong preferences for one category of figures over another. Is there a figurative consistency which characterizes a speaker's style in the same or similar situations?, or across situations?, as part of the speaker's figurative competency? Perhaps even more interesting would be a study of the breakdown products of such a style. When stressed, do subjects characteris- tically resort to certain categories of (novel) figures and not others? And do such preferences correlate with standard character- ological, object relational and defensive typologies? Do figurative styles provide the soil out of which predilections for types of defensive and neurotic styles grow or vice versa? For example, manic and hysteric styles might be expected to be more involved with the vicissitudes of hyperbole and allusion; and schizophrenics more involved with the identifications made possible via metaphor and personification; while the self derogations and self negations of depressives might trade on the potentialities of litote and irony. 112 From this point of view, developments in psychoanalysis and psychotherapy might be marked by the appearance of new figurative categories and the expansion of figurative competencies as well as the gradual or rapid eroding of habitual and rigidified figurative strategies for pre-figuring persons and situations. Postural and Eye Closure Effects In popular as well as technical writings, the couch has become a metaphor (more specifically,a synechdoche) for psychoanalysis, still held by many to be a necessary condition for an authentic psycho- analysis to occur. The recumbent posture with the analyst outside of the analysand's visual field are considered essential elements in producing the therapeutic regression, the sine qua non of psycho- analysis. With this as its point of departure, the present study hypothesized that differences in body posture and eye closure would promote greater use of figurative language and less verbal productivity. While the differences obtained concerning body positions and eye closure were not significant, four of the five experimental tests bearing on this condition were in the predicted direction. This was the case for verbal productivity, novel figurative productivity and cate- gories, and the variety of living figurative categories. Only for the productivity of living figures were the differences in the means opposite of the predicted direction. On the face of it, it appears that the Free Association treatment is a more powerful one than the Postural-eye closure treatment. Yet the consistency of the direction of the results suggests that instruc- tions to recline and close eyes do help foster the predicted "regressive" movement. To account for the failure to attain 113 significance, it can be argued that initially the request to recline and close eyes might serve to increase rather than relax security operations--both experimentally and clinically!--which would seem directly counter to the desired effect of promoting access to inner sources of preconscious awareness. For the analyst, this complication provides an opportunity for resistance analysis. For the researcher, it's back to the drawing board. One alternative strategy for furture research might include the incorporation of an actual relaxation procedure in addition to the request to lie down and close eyes. Another would be to work with the same subjects over a sequence of sessions and note the alterations in fluency and figurativity. Finally, one could set aside analogue and laboratory research and study ongoing psychotherapy and psycho- analytic sessions with the present variables in mind. On the Use and Abuse of Figurative Competence The results of this study shifted attention from the traditional psychoanalytic approach to figurative language as derivative expressions of unconscious motives to the communicational/inter- actional aspects of the phenomenon. The special aptitude of figurative speech for navigating the cross currents set in motion by the Fundamental Rule was highlighted. As a means of combining simultaneous— ly conflicting speech acts, which say one thing and mean another, into a meaningfully ambiguous verbal form, figurative speech is uniquely suited to meet the challenge of the paradoxical requirements of the Fundamental Rule. When literal and direct speech runs the risk of violating conventions of politeness or morality or self-esteem maintenance; and silence or nonsense violates the expectation that 114 experimental subjects--not to mention analysands--will cooperate with the procedural requests, (in this case to gpegk in the required manner); non-direct speech with its intrinsic ambiguity that allows for communication by implication, subject to multiple interpretations, provides an elegant solution. Something like this capability was found in the oracles of antiquity, which, according to Heraclitus, "neither reveal nor hide in words but give manifest signs" (quoted in Arendt, 1959, p. 162). It is in the region of meaningful ambiguity which opens itself to the interpretive and projective needs of speaker and hearer, that the special interactional effects of figurative speech resides. The great task for future theory and research is to explicate the differential interactional effects induced by the various categories of figurative speech. As a preliminary example of such a project, consider the community creating properties of the metaphor (cf., Booth, 1979); the unique way in which the metaphor hearer is invited into the world of the metaphor user: the more novel the metaphor, the more effort the hearer must expend to accept this implicit invitation. Successfully carrying out this transaction constitutes the acknowledge- ment of a community between the two (Cohen, 1979). Just as understanding a poet's metaphors draws us into the poet's world—-at least for the duration of the poem--and understanding a scientist's or a philosopher's metaphors brings about a comparable revisioning of the world, so, too, does the metaphor of the analysand pull the analyst (often unwittingly) into a shared appreciation of experience. Such a process would go a long way towards clarifying the obscure but much talked about defense of 'projective identification'. 115 What the empirical test of such a hypothesis would look like remains a difficult question. Perhaps investigating the parallelism between analysandhsand analyst's metaphors would be a start. While it is commonly observed that patients adopt certain metaphors of their therapists, the reverse process may also occur. Moreover, it would be possible to also study the parallels in frequency and categories of figures over the course of treatment. Presumably, convergence would represent increased identification. Similar studies could be undertaken for the other major categories of figurativity. Another approach would be to observe the changes in predominant figurative categories in the course of successful and unsuccessful therapies. Are these changes in the direction of greater novelty in the successful cases? Or, are there changes in modal categories?, or within character types? If figurative speech acts embody changing modes of relatedness, they provide perhaps the most available forms of data for the scientific study of these most subtle processes. CHAPTER IX SUMMARY The present study undertook to explore the relationships between contexts of association, framed by instructions for Free Associations and Directed Associations, on the fluency and figurativity of speech. It was hypothesized that Free Association conditions would reduce fluency (verbal productivity) and increase figurative productivity, (including the frequency and variety of categories of novel and living figures). Differential effects of postural and eye closure instructions were also investigated. To test these hypotheses in a factorial design, sixty-four undergraduate subjects (32 male and 32 female) initially participated in one of four treatment groups: 1) Free Association/Reclining eyes closed; 2) Free Association/Upright eyes open; 3) Directed Association/ Reclining eyes closed; 4) Directed Association/Upright eyes open. A Barron's Ego Strength Scale was also administered to each subject. Verbal responses of all subjects were transcribed and scored for verbal productivity (number of words spoken) and for figurative productivity (number of novel and of living figures per 100 words of transcript, and the number of categories of novel and of living figurative language used per 100 wOrds of transcript). Analyses of the resultslends strong support to the hypotheses regarding the effects of Free Association on the fluency and figurativity of speech. Free Associating subjects were less verbally but more figuratively productive, i.e., although less fluent, they used a 116 117 higher frequency of novel figures and a wider variety of novel figurative categories than Directed Associating subjects. Treatment groups did not differ with respect to frequency of living figurative use. While there were no significant differences for Reclining/eyes closed subjects vs. Upright/eyes open subjects, four of the five comparisons were in the predicted direction of less fluency and more figurativity for reclining/eyes closed subjects. Only for the frequency of living figures were the means in a direction opposite to the predictions, and this difference was not significant. Finally, higher Ego Strength was found to be correlated with higher frequencies of living figurative use. The results were discussed in the light of the scope and limits of psychoanalytic and speech act approaches to Free Association and figurative language. APPENDICES APPENDIX A REFERENCE NOTES APPENDIX A REFERENCE NOTES 1In fact, it is arguable that Freud was more than coversant with these phenomena, but instead of treating rhetorical devices, he classified them as indexes of defensive operations and derivations of unconscious meanings. 2Austin's theory "defines three types of speech acts--acts performed when one uses language....A locutionary act is the speaker's act of saying whatever it is he says, and illocutionary act is the one the speaker performs in_saying something (such as the act of ordering); a perlocutionary act is one the speaker performs by_saying some- thing (such as annoying someone)" (J.L. Austin, 1962). 3Robert Holt, (citing R. Schafer, 1954) has also suggested the inadequacy of employing a single continuum to classify all thought products (R. Holt, 1967, p. 294). 4The extensions of the view into analysis of art and creativity leads to the reductive unmaskings for which psychoanalysis is (in)famous. At this point, I will suggest, but not prove, that even heroic effort to restore the rights and dignity of creativity via the construct of "regression in the service of the ego" fails to fully explain how genuine novelty can be derived from a re- established pool of latent connotations and associations. 5Consequently, the present study focuses on the problems of identifying and defining manifest verbal figures. I shall not consider here the problems connected with identifying unconscious figurative activity, other than to suggest that these latter are more like allegories than isolated figures (cf., J. Arlow, 1969; D. Beres, 1969; G. Klein, 1976; and especially 0. Shave, 1974 and J. Lacan, 1969). 61D the contemporary jargon, spilling over from the studies of the hemispheric laterializations of the brain, this difference is brought about by shifting from left hemisphere digital modes of information processing to right hemisphere analogical modes. Whether these distinctions add anything more than fancy neuropsychological meta hors--long on pseudo-scientific persuasiveness and short on real clarifying power--remains to be seen. 7Given a theory in which "la parole" and "la langue"--the speech event and the underlying objective structure of signs comprising a language--are segregated into two different categories, w1th only the latter having scientific dignity, it could hardly have been any other "a" 118 8 119 Borrowing for a moment the terminology of Ernest Kris and Abraham Kaplan (and through them, William Empson (1961)),these responses may be considered the "disjunctive" ambiguities (E. Kris, 1952, p. 245), the effects of full contradiction,"marking a division in the author's mind". (W. Empson, 1961. pp. vi and 192). In addition, there are the "conjunctive ambiguities" where "the separate meanings are jointly effective in the interpretation" (E. Kris, 1952, p. 247); i.e., where distinct and even opposed meanings are responded to conjointly (Ibid., p. 247). An example would be irony. Finally, there are the "integrative ambiguities" in which "manifold meanings evoke and support one another. They interact to produce a complex and shifting pattern; though multiple, the meaning is unified" (Ibid., p. 248). In spite of Kris and Kaplan's zeal to relabel them 'ambiguities'--a move which seems both gratuitous and motivated by a special theory of poetic language as necessarily ambiguous--these latter two categories contain the traditional figures of speech. The value of introducing these distinctions is that they point to the function of figurative language in integrating semantic clashes, of "bridging the gap between feelings and words to describe them" (Fine, Pollio, Simpkinson, 1973, p. 87). APPENDIX B TABLES OF DATA FOR ALL SUBJECTS APPENDIX B TABLES OF DATA FOR ALL SUBJECTS Table a: Verbal Productivity Free Association Directed Association Row Means Reclining/ Eyes Closed Upright/ Eyes Open Column Means N = 15 N = 14 269 434 348.72 N = 16 N = 16 343 502 423 307.10 470.37 120 121 6<.<~ 666:666 m=P>66 om.AF 8663666 mcw>66 mp. mmcamm6 Fm>oz mm. mmgamw6 Fm>oz memo: asapou mm.m~ 6663666 m=6>66 m_.mN maeamw6 m=6>66 m._~ macame6 mer>e4 <<. mmcamw6 pm>oz mp. mmgaaw6 pm>oz mo. magamm6 Fm>oz cmao mmzm \pgmwcaa 6P 1 2 mp u 2 68.6, mmeamw6 m=_>66 ee.mN 6663666 m=6>es 6o.

oz no. mug—5w...— _.o>oz ¢. mmgammn— _.w>oz e— u 2 mp u z uwmo—o mmxm \acecepumm memo: 3oz comumpuomm< vmuumgwo cowuawuomm< mmc6 Amocmap6 6o» umpumggoocav zuw>ruu=uoca w>pumgzmm6 "a wFaMF 122 mm.< 6663666 aceseu Am.m mac=e6 P. mmcaap6 ~m>oz m<. mwgzmm6 Po>oz mcmmz caspou mN.< mmcamw6 mcw>64 m<.¢ mmcamw6 u=w>wg oo.< megsmw6 mcv>64 Fm. magamw6 Pm>oz mp. mmcsmm6 Pm>oz m. mmczmw6 Fm>oz :mgo mmau \6566663 mp u z 64 N.< mm63m66 mce>64 mm.m mmczmw6 m=w>w4 em. mmcamw6 —m>oz so. mmcamw6 Pm>oz e. mmczmw6 _m>oz ummopu mmzm \m=6=w_666 e? n 2 mp u 2 menu: :om coppopuomm< umpumcwo copumwuomm< moc6 Axucm=~6 Low umuumgcoucav mmwgommumu m>wumcamm6 no open» 123 66.6 66e=66 om.m 6663666 aceses mo. megamw6 Pm>oz up. mmcamw6 ~m>oz mcmmz csspoo mo.m mugsmw6 mcp>wg um.m mmcamw6 mcw>wg P.o mogzmw6 mcw>wg mo. mmczmw6 Pm>oz mo. mmgamw6 ~o>oz up. mm6:a_6 pm>oz coco mmxm \6:6666= up u 2 mp u z mm.m mmcamw6 mcw>64 op.o mossmp6 m:_>wg ow.< mugsmw6 mcp>w4 Np. mmczmw6 ~w>oz Po. mwgauw6 pm>oz om. mwcamw6 Pm>oz vmmopu mmxm \mcecepumm «P u 2 m_ u z mcmmz 3oz scrumwuomm< vmuum6wo cowumwuomm< mm66 Axucmap6 6cm umuumcgouv xup>vuu=uo66 m>vpmcamw6 "u mpamh 124 ao.p 6663666 m=6>64 o~.P 6663666 m=6>e6 No. mwgauw6 pm>oz mp. mmgamr6 Pm>oz mcwmz seapou m~.p macamp6 mcw>m4 mm. mmgzmw6 mcm>64 mm.~ mmgamm6 mcw>64 mo. mmczmw6 Fw>oz mo. mmgamw6 pm>oz F. mmczmw6 Fm>oz ammo mmxm \ucmeeaa or u 2 mp u z om.P macamm6 mcw>w4 w~.p mmgzmw6 mcm>m4 mm.F mmcamw6 mcm>64 PP. magamw6 Pm>oz Po. mmcsmw6 —m>oz Fm. mmgzmw6 Pm>oz «P u 2 mp u z ammopu mmxm \aewcepuaa memo: 3oz cowporuomm< uweumgwo copumwuomm< mm66 Axucmzp6 Low umuomggouv mmwcommpmo m>wumgzmw6 um mpnmh APPENDIX C ANALYSIS OF COVARIANCE 0F EFFECTS OF ASSOCIATION AND POSTURAL/EYE CLOSURE CONDITIONS ON VERBAL AND FIGURATIVE PRODUCTIVITY WITH EGO STRENGTH AS CONTROL VARIABLE APPENDIX C STRENGTH AS CONTROL VARIABLE ANALYSIS OF COVARIANCE 0F EFFECTS OF ASSOCIATION AND POSTURAL/EYE CLOSURE CONDITIONS 0N VERBAL AND FIGURATIVE PRODUCTIVITY WITH EGO Free Association vs. Directed Association Reclining/Eyes Closed vs. Upright/Eyes Open Association X Posture/Eye Closure Productivity df MS F df MS F df MS F Novel Figures Living Figures Verbal Fluency 1 .311 1 4.137 .505 1 400.237 7.927** 1 6.090* 1 .009 .185 1 7.063 .863 82.663 1.64 1 .048 .943 1 19.358 2.364 1 225 .004 * p < .02 **p < .01 125 APPENDIX D MEASURES OF CENTRAL TENDENCY AND VARIABILITY 0F SCORES 0N BARRONS EGO STRENGTH SCALE FOR THE 61 SUBJECTS APPENDIX D MEASURES OF CENTRAL TENDENCY AND VARIABILITY 0F SCORES ON BARRONS EGO STRENGTH SCALE FOR THE 61 SUBJECTS Mean 82.02* Standard Deviation 5.03 Range 70-95 *Lower scores represent higher "ego strength", with 61 the lowest possible score and 122 the highest possible score on this version of the scale. 126 APPENDIX E EGO STRENGTH SCALE sooowcnm-s: 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. APPENDIX E EGO STRENGTH SCALE It does not bother me to see a woman smoke. My table manners at home are as good as when I eat out in a restaurant. I have had blank spells in which my activities were interrupted and I did not know what was going on around me. I can be friendly with people who do things which I consider wrong. I never attend a sexy show if I can avoid it. I am sometimes irritated by people who ask favors of me. I am always courteous, even to peOple who are disagreeable. I can remember "playing sick" to get out of something. I like to talk about sex. If I could get into a movie without paying and be sure I was not seen I would probably do it. Often I cross the street in order not to meet someone I see. There have been times when I was quite jealous of the good fortune of others. When I am with people, I am bothered by hearing very queer things. I am always careful about my manner of dress. I very much like horseback riding. At times I have fits of laughing or crying that I cannot control. There have been occasions when I took advantage of someone. There have been times when I felt like rebelling against the people in authority even though I knew they were right. I'm always willing to admit it when I make a mistake. I am attracted by members of the opposite sex. My hands have not become clumsy or awkward. 127 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 128 I infrequently find myself worrying about something. On a few occasions, I have given up doing something because I thought too little of my ability. I like science. When I get bored, I like to stir up some excitement. I sometimes try to get even rather than forgive and forget. On occasion I have had doubts about my ability to succeed in life. Parts of my body often have feelings like burning, tingling, crawling, or like "going to sleep." I am afraid of finding myself in a closet or small closed place. I have never had a fainting spell. I have a cough most of the time. I feel weak all over much of the time. I like to flirt. If I were an artist, I would like to draw flowers. I get mad easily and then get over it soon. I have had very peculiar and strange experiences. I like to cook. I feel tired a good deal of the time. I have diarrhea once a month or more. I believe my sins are unpardonable. I feel unsympathetic towards people who tend to hang on to their griefs and troubles. I have never felt that I was punished without cause. I do so many things which I regret afterwards (I regret things more or more often than others seem to). No matter who I'm talking to, I'm always a good listener. I have never been irked when people expressed ideas very different from my own. ‘ There have been occasions when I felt like smashing things. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 129 I sometimes think when people have a misfortune they only got what they deserved. I never make a long trip without checking the safety of my car. When I don't know something I don't at all mind admitting it. I am in just as good physical health as most of my friends. I never resent being asked to return a favor. When I leave home, I do not worry about whether the door is locked and the windows closed. I sometimes feel resentful when I don't get my way. My plans have frequently seemed so full of difficulties that I have had to give them up. I have never deliberately said something that hurt someone's feelings. One or more members of my family is very nervous. I would never think of letting someone else be punished for my wrong-doings. I have a good appetite. I find it easy to keep my mind on a task or job. I don't find it particularly difficult to get along with loud mouthed, abnoxious people. I never hesitate to go out of my way to help someone in trouble. Everything is turning out just like the prophets of the Bible said it would. I seldom worry about my health. I have often been frightened in the middle of the night. It is sometimes hard for me to go on with my work if I am not encouraged. I like collecting flowers or growing house plants. Before voting I thoroughly investigate the qualifications of the candidates. I have almost never felt the urge to tell someone off. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 130 At times I have really insisted on having things my own way. I dream infrequently about things that are best kept to myself. I have had some very unusual religious experiences. I am not afraid of fire. I do not have strange and peculiar thoughts. I have met problems so full of possibilities that I have been unable to make up my mind about them. I feel unable to tell anyone all about myself. The man who had most to do with me when I was a child (such as my father, stepfather, etc.) was very strict with me. I am made nervous by certain animals. I am not easily downed in an argument. During the past few years I have been well most of the time. I always try to practice what I preach. My way of doing things is usually understood by others. I brood a great deal. I like to gossip at times. When someone says silly or ignorant things about something I know, I try to set him right. My skin is not unusually sensitive to touch. I have had no difficulty in keeping my balance in walking. I would certainly enjoy beating a crook at his own game. I sometimes feel that I am about to go to pieces. I have never intensely disliked anyone. At times I hear so well it bothers me. Dirt frightens or disgusts me. I pray several times every week. 131 93. Sometimes some unimportant thought will run through my mind and bother me for days. 94. I think Lincoln was greater than Washington. BIBLIOGRAPHY BIBLIOGRAPHY Ackman, P. The effects of induced regression on thinking processes. Dissertation Abstracts, 1960, 21, 364. Aleksandrowicz, D.R. The meaning of metaphor. Bulletin of the Menninger Clinic, 1962, 26, 92-101. Anderson, C.C. The psychology of metaphor. Journal of Genetic Psychology, 1964, 195, 53-73. Arendt, H. The human condition. New York: Doubleday Anchor Books, 1959. Arlow, J.A. Unconscious fantasy and disturbances of conscious experience. Psychoanalytic Quarterly, 1969, 38, 1-27. Aronson, H., and Weintraub, W. Personal adaptation as reflected in verbal behavior. In A. W. Siegman and B. Pope (Eds.), Studies in dyadic communication. New York: Pergamon Press, 1972. Asch, S.E., and Nerlove, H. The development of double-function terms in children. In B. Kaplan and S. Wagner (Eds.), Perspectives in psychological theory. New York: International Universities Press, 1960. Atkin, S. Psychoanalytic consideration of language and thought: A comparative study. Psychoanalytic Quarterly, 1969, 38, 549-582. Austin, J.L. How to do things with words. Oxford: Clarendon Press, 1962. Austin, J.L. Philosophical papers. London: Oxford University Press, 1970. Barlow, J.M., Kerlin, J., and Pollio, H.R. Training manual for identifying fi urative language. TeChnical Report No. 1, University 0 Tennessee, Knoxville, 1971. Barron, F. An ego strength scale which predicts response to psychotherapy. Journal of Consulting_Psychology, 1953, 11, 327-333. Beardsley, M. Aesthetics: Problems in the philosophy of criticism. New York: Harcourt, Brace and Company, 1958. Beardsley, M. The metaphorical twist. Philosophy and Phenomenological Research, 1962, 22, 3, 292-307. ‘ 132 133 Bellak, L. Free association: Conceptual and clinical aspects. spurnal of the American Psychoanalytic Assgejatjgp. 1961, 42,‘9-20. Benveniste, E. Problems filgeneral ljn uistics. Coral Gables: University of Miami Press, 1 . Berdach, E. Bodyspositign and_memsry recall. Unpublished master's thesis, Michigan State University, 1965. Berggren, D. The use and abuse of metaphor. Review of Megaphysjss, 1962, 237-258; 1963, 450-472. Billow, R.M. A cognitive developmental study of metaphor comprehension. Developmental Psychology, 1975, 11, 415-423. Billow, R.M. Metaphor: A review of the psychological literature. _Esysnological_8u11etin, 1977 (Jan.), 88, 1, 81-92. Black, M. Models and metaphors. Ithaca: Cornell University Press, 1962. Booth, W.C. Metaphor is rhetoric: The problem of evaluation. In S. Sacks (Ed.), On metaphor. Chicago: University of Chicago Press, 1979. Bordin, E. Personality and free association. Journal sf ansulting Psychology, 1966, 88, 1, 30-38. Brenner, C. An elementary textbooksof,psyshoanalysis. Garden City: DouBleday Anchor BoOks, 1957. Brenner, C. Ps choanal tic techni ue and s chic conflict. New York: International Universities Press, 1975. Burke, K. Terms for order. Bloomington: Indiana University Press, 1964. Bushnell, D. O. Level of arousal and the primary process properties of associative responses. Dissersatjon_Abssrasts, 1968, 88, 1-8, 363-364. Cain, A., and Maupin, 8. Interpretation within the metaphor. Bulletin of ED? Menningersgjinjc, 1961, 88, 307-311. Cansever, G. Language structure of primary process thinking. Disserta- tion Abstracts, 1960, 81, 954. Caruther, E., and Ekstein, R. Interpretation within the metaphor: Further considerations. ,Journal.pf,Chj1gsEsychiatry, 1966, 5, 35-45. Chapman, J.W. The pergeption'and expression 9f a metaphgr_as a fus sjon ‘ of cognitive st 1e and intellectual 18vél. Unpfiblished doctoral dissertation, Georgia State University, 1971. 134 Cohen, T. Metaphor and the cultivation of intimacy. In S. Sacks (Ed.), On metaphor. Chicago: University of Chicago Press, 1979. de Man, P. The epistemology of metaphor. In S. Sacks (Ed.), On metaphor. Chicago: University of Chicago Press, 1979} Derrida, J. Limited inc. Glyph 2, 1977, 8, 162-254. Doneson, S. The figurative turn: Psychoanalysis and the illusions of meaning. Unpublished manuscript, 1981. Eagle, C.J. An investigation of individual consistencies in the manifestations of primary process. Dissertation Abstracts, 1964, 88, 2045. Ekstein, R., and Wallerstein, J. Choice of interpretation in the treatment of borderline and psychotic children. Bulletin of the Menninger Clinic, 1956, 8, 199-207. Empson, W. Seven types of ambiguity. Great Britain: Penguin Books, 1930. Enelow, A.J. The silent patient. Psyghiatry, 1960, 88, 153-158. Ervin, S.M., and Foster, G. The development of meaning in children's descriptive terms. Journal of Abnormal and Social Psychology, 1960, 2, 271-275. Evan, S. The creative work of free association. Psychoanalysis and Psychoanalytic Review, 1961, 88, 3, 89-96. Fenichel, O. The psychoanalytic theory of neurosis. New York: W.W. Norton and Company, Inc.,’l945. Fierman, L. Effective psychotherapy: The contribution of Hellmuth Kaiser. New York: The Free Press, 1965. Fine, H., Pollio, H., and Simpkinson, C. Figurative language, metaphor, and psychotherapy. Psychotherapy: Theory,,Research and Practice, 1973, 18, 1, 87-91. Fiske, D.W., and Maddi, 5. Functions of varied experience. Homewood, Illinois: Dorsey Press, 1961. Forrest, D.W. 0n ones own onymy, Esychiatry, 1973, 88, 266-290. Freed, 6.0. Personal communication. Freud, S. The interpretation of dreams. Standard Edition, 4 and 5, London: Hogarth Press, 1900. 135 Freud, S. The psychopathology of everyday life. Standard Edition, 6, London: Hogarth Press, 1901. Freud, S. Jokes and their relation to the unconscious. Standard Edition, 8, London: Hogarth Press, 1905. Freud, S. Fragment of an analysis of a case of hysteria. Standard Edition, 7:7-122, London: Hogarth Press, 1905. Freud, S. The unconscious. Standard Edition, 14:161-215, London: Hogarth Press, 1915. Freud, S. Introductory lectures on psychoanalysis. Standard Edition, 15, London: Hogarth Press, 1916-1917. Freud, S. Dora: An analysis of a case of hysteria. New York: Collier Books, 1963. Fromm, E., Oberlander, M., and Gruenewald, D. Perceptual and cognitive processes in different states of consciousness: The waking state and hypnosis. Journal of Projective Techniques and Personality Assessment, 1970, ss, 5, 375-387. Frost, R. The constant symbol. In H. Cox and E.C. Latham (Eds.), Selected Prose of Robert Frost, Collier Books, 1968. Galbraith, G.G., Kahn, K., and Lieberman, H. Personality correlates of free associative sex responses to double entendre words. Journal of Consulting and Clinical Psychology, 1968, 88, 193-197. Gamble, K.R., and Kellner, H. Creative functioning--Cognitive regression. Journal of Personality and Social Psychology, 1968, 2, 266-271. Gardner, H. Metaphors and modalities: How children project polar adjectives onto diverse domains. Child Development, 1974, 45. 84-91. Gill, M. The collectedspapers of David Rapaport. New York: Basic Books, Inc., 1967. Gill, M. The primary process. In R. Holt (Ed.), Motives and thought: Psychoanalytic essays in honor of David Rapaport. New York: International Universities Press, Inc., 1967. Goldman-Eisler, F. Psycholin uistics: Experiments in spontaneous speech. New York: cademic Press, 1968. Greenson, R. The technique and practice of psychoanalysis (Vol. 1) New York: International Universities Press, Inc., 1967. 136 Grice, H.R. Logic and conversation. In P. Cole and J. Morgan (Eds.), 8yntax and semantics (Vol. 3). New York: Academic Press, 1975. Gruenewald, D., Fromm, E., and Oberlander, M. Hypnosis and adaptive regression: An ego-psychological inquiry. In H nasis: Research developments and perspective. Chicago: gidine- Atherton, 1972. Habermas, J. Knowledge and human interest. Boston: Beacon Press, 1972. Harper, R.G., Wiens, N., and Matarazzo, J.D. Nonverbal communication: The state of the art. New York: Wiley, 1978. Hartmann, H. Egopsychology and the problem of adaptation. New York: International Un1ver§ities Press, 1939. Heidegger, M. Being and time. New York: Harper and Brothers, 1962. Hester, M. The meanjgg ofspoetic metaphor. The Hague: Mouton and Company,’l967. Hilgard, J. Personality and hypnosis. Chicago: University of ChicagoTPress, 1979. Hinchliffe, M.K., Lancaster, M., and Roberts, F.J. Depression: Defense mechanisms in speech. British Journal of Psychiatry, 1971, 118, 471-472. Holt, R. The development of the primary process: A structural view. In R. Holt (Ed.), Motives and thought: Psychoanalytic essays in honor of David Rapaport. New York: International Universities Press, Inc., 1967. Holt, R. Gauging primary and secondary process in Rorschach responses. Journal of Projective Techniques, 1956, 88, 14-25. Holt, R. Motives and thought. Psychological Issues, 1967, 8, 2-3. Holt, R., and Peterfreund, E. (Eds.). Psythoanalysis and contempOrary science. New York: The MacMillan Company, 1.71972. Horowitz, M. Modes of representation of thought. Journal of the American Psychoanalytic Association, 1972, 20, 793-819. Jacobson, R., and Halle, M. Fundamentals of language. The Hague: Mouton, 1956. James, W. Principles ofpsycholpgy, Dover Publications, Inc., 1950. Jones, E. Papers onspsychoanalysis. London: Bailliere, Tindael and Cox, 1950. 137 Kaiser, H. The universal symptom of the psychoneuroses: A search for the conditions of effective psychotherapy. In L.B. Fierman (Ed.), Effective psychotherapy: The contribution of Hellmuth Kaiser. New York: *The Free Press, 1965. Kanzer, M. Verbal and non-verbal aspects of free association. Psychoanalytic Quarterly, 1961, 88, 327-350. Kanzer, M. Superego aspects of free association and the fundamental rule. Journal of the American Psychoanalytic Association, 1972, 88, 246-267. Kelman, H. Free association: It's phenomenology and inherent paradoxes. American Journal of Psychoanalysis, 1962, 88, 176-200. Khatchadourian, H. Metaphor. British Journal of Aesthetics, 1968, 8, 227-243. Klein, G. Peremptory ideation: Structure and force in motivated ideas. In R. Holt (Ed.), Motives and thought: Psychoanalytic essays in honor of David Rapaport. International Universities Press, Inc.,71967. Klein, G. Psychoanalytic theory; An exploration of essentials. New York: International Universities Press, Inc., 1976. Koen, F. An intra-verbal explication of the nature of metaphor. Journal of Verbal Learning and Behavior, 1965, 8, 129-133. Kohut, H. The restoration of the self. New York: International Universities Press,*l977. Korner, I., and Allison, R.B., Jr. Comparative study of metaphoris Thinking. Journal of Psychology, 1965, 88, 67-70. Kris, E. Psychoanalytic explorations in art. New York: International Universities Press, Inc., 1952. Kris, E., and Kaplan, A. Aesthetic ambiguity. In E. Kris (Ed.), Psychoanalytic explorations in arts. New York: International Universities Press, Inc., 1952. Kroth, J.A., and Forrest, M.S. Effects of posture and anxiety level on effectiveness of free association. Psychological Reports, 1969, 88, 725-726. Kubie, L. Neurotic distortion of the creative_process. Kansas: University of Kansas Press, 1973. Lacan, J. The language of the self. New York: Dell Publishing Company, Inc., 1968. 138 Laffal, J. Freud's theory of language. Psychoanalytic Quarterly, 1964, 88, 157-175. Langs, R. The bipersonal field. New York: Jason Aronson, Inc., 1976. Lenrow, P. Uses of metaphor in facilitating constructive behavior change. Psychotherapy: Theory, Research and Practice, 1966, 3:4, 145-147. Levitt, M. (Ed). Readings in Psychoanalytic Psychology. New York: Appleton-Croft, Inc., 1959. Lockwood, B.R. Figurative language as a function of cognitive style. Unpublished doctoral dissertation, University ofTTennessee, 1979. Loewenstein, R. Some remarks on the role of speech in psychoanalytic technique. International Journal of Psychoanalysis, 1956, 37. Loewenstein, R. Some considerations on free association. Journal of American Psychoanalytic Association, 1963, 11, 45-473. Loewenstein, R., Newman, L., Schur, M. and Solnit, A. Psychoanalysis-- asgeneral psychology: Essays in honor of Heinz Hartmann. New York: International Universities Press, Inc., 1966. Margolis, J. The language of art and art criticism. Detroit: Wayne State University Press, 1965. Mehrabian, A. Nonverbal communication. Chicago: Aldine-Atherton, 1972. Morgan, J. Observations on the pragmatics of metaphor. In A. Ortony (Ed.), Metaphor and thought. New York: Cambridge University Press, 1979. Morgan, R.F., and Bakan, P. Sensorial deprivations and other sleep behaviors as a function of position, method of report, and anxiety. Perceptual and motor skills, 1965, 88, 19-25. Murray, D.L. Talk, silence and anxiety. Psychological Bulletin, 1971, 18, 4, 244-260. Noy, P. A revision of the psychoanalytic theory of the primary process. International Journal of Psychoanalysis, 1969, 50. 155-178. Paul, A. Figurative language. Philosophy and Rhetoric, 1970, 8, 234-241. Percy, W. The message in the bottle. New York: Farar, Strauss and Giroux, 1954. - 139 Piaget, J. Play, dreams and imitation in childhood. New York: W.W. Norton, 1951. Pickens, J.D., and Pollio, H.R. Patterns of figurative language use in adult speakers. Unpublished manuscript. Pollio, H.R., and Barlow, J.M. A behavioral analysis of figurative language in psychotherapy: One session in a single case study. Language and Speech, 1975, 18, 236-254. Pollio, H.R., and Burn, B.C. The anomaly of anomaly. Journal of Psychoanalytic Research, 1977, 8, 247-260. Pollio, H.R., Barlow, J.M., Fine, H.J., and Pollio, M. Psycholggy and thespoetics of growth. Hinsdale, New Jersey: Lawrence Erlbaum Aésociates, 1977. Pollio, M.R., and Pollio, H.R. A test of metaphoric comprehension and some_preliminary developmental data. Unpublished manuscript. Porter, C.M. Figures of speech, divergent thinking, and activation theory. Unpublished doctoral dissertation, North Texas State University, 1969. Rabin, A.I. (Ed.). Projective techniques in personality assessment. New York: springer PubliShing Company, Inc., 1968. Ramsey, R.W. Personality and speech. Journal of Personality and Social Psychology, 1966, 8, 116-118. Rapaport, D. (trans. and commentary) Organization and pathologysof thought: Selected sources. New York: Columbia Univeréity Press, 1951. Rapaport, D., and Gill, M.M. The points of view and assumptions of metapsychology. International Journal of Psychoanalysis, 1959. Q. 153-162. Rapaport, D. The collected papers of David Rapaport. (Ed., M.M. Gill). New York: Basic Books, 1967. Richards, I.A. The philosophy of rhetoric. New York: Oxford University Press, 1965 (1935). Richardson, C., and Church, J. A developmental analysis of proverb interpretations. Journal of Genetic Psychology, 1959, 94. 169-179. Ricouer, P. Freud and philosophy. New Haven: Yale University Press, 1970. Ricouer, P. Metaphor and the main problem of hermeneutics. Neg LiterarysHistory, 1974-1975, 8, 95-110. 140 Ricouer, P. The rule of metaphor. Toronto: University of Toronto Press,’1977. Rohovit, D. Metaphor and mind. American Image, 1960, 11, 3. Rosen, V. Sign phenomena and their relationship to unconscious meaning. International J0urnal of Psychoanalysis, 1969, 50. 197-207. Rosen, V. The nature of verbal intervention in psychoanalysis. In L. Goldberger and V. Rosen (Eds.), Psychoanalysis and contemporarysscience (Vol. III). Neinork: International Universities Press, 1974. Rosenberg, J.F., and Travis, C. (Eds.). Readings in the philosophy of language. New York: Prentice Hall, 1971. Rosner, S. "On the nature of free association. Journal of the American Psychoanalytic Association, 1973, 81, 3, 558-575. Rubinstein, B. On metaphor and related phenomena. In R. Holt and E. Peterfreund (Eds.), Psychoanalysis and contemporary science (Vol. I). New York: International Universities Press, 1972. Schafer, R. Psychoanalytic interpretation of Rorschach testing. New York: Grune and Stratton, 1954. Schafer, R. Projective testin and s choanal sis. New York: International Universities Press, I967. Schafer, R. A new language forspsychoanalysis. New York and London: Yale University Press, 1976. Schafer, R. Langua e and insight. New Haven: Yale University Press, 19 8. Searle, J.R. Indirect speech acts. In P. Cole and J. Morgan (Eds.), Syntax and semantics, (Vol. III). New York, 1975. Shakespeare, W. A midsummer night's dream. In The complete works of Shakespeare. Oxford University Press, 1936. Shapiro, 0. Motivation and action in psychoanalytic psychiatry. Psychiatry, 1970, 88, 3, 329-343. Sharpe, E. Psycho-physical problems revealed in language: An examination of metaphor. International Journal of Psychoanalysis, 1940, 201-213. Shave, 0. Communication breakdown. St. Louis: Warren H. Green, Inc., 1975. ' 141 Shor, R.E. Hypnosis and the concept of the generalized reality orientation. American Journal of Psychotherapy, 1959, 18, 582-602. Siegman, A.W., and Pope, B. (Eds.). Studies in dyadic communication. New York: Pergamon Press, 1972. Silberer, H. Report on a method of eliciting and observing certain symbolic hallucination-phenomena. Organization and PSYCDOIOQY 0f Thought,(trans. and commentary), 0. Rapaport. New York: Columbia University Press, 1951. Spence, 0. (Ed.). The broad scope of psychoanalysis: Selected papers of Leopoid Bellak. New York: Greene andiStratton, 1967. Spence, 0. (Ed.). Psychoanalysis and contemporary science. New York: The Macmillan and Company, 1975, 4. Spinoza, B. The ethics. In The chief works of Benedict de Spinoza. New York: Dover, 1955 (1674). Stern, G. Meaning and change of meaning, Bloomington: Indiana University Press, 1931. Szasz, T. The myth of mental illness. New York: Hoeber Medical Division, Harper and Row Publishers, Inc., 1964. Trager, G.L. Paralanguage: A first approximation. Studies in Linguistics, 1958, 18, 1-12. Trilling, L. The liberal imagination. New York: The Viking Press, 1950. Voth, H. The analysis of the metaphor. Journal of the American Psychoanalytic Association, 1970, 18, 599-621. Waelder, R. The principle of multiple function: Observations on over-determination. Psychoanalytic Quarterly, 1936, 8, 45-62. Waldhorn, H., and Fine, E. (Eds.). Monograph V: Trauma and s mbolism. New York: International Universities Press, Inc., 1974. Watzlawick, P., Beavin, J., and Jackson, 0. Pragmatics of human communication. New York: W. W. Norton and Company, Inc., 1967. Werner, H. Comparative psychology of mental development. New York: International Universities Press, 1940. Wheelwright, P. Metaphor and reality, Bloomington: Indiana University Press, 1962. 142 Wimsatt, W.K. The verbal icon. Lexington: University of Kentucky Press, 1954. Winnicott, D.W. Metapsychological and clinical aspects of regression within the psychoanalytic set up. International Journal of Psychoanalysis, 1955, 88, 16-26. Wittgenstein, L. Philosophical investigations. Oxford: Blackwell, 1953.