In uh: Nvfl-‘ra ~:\.v~‘n.'.0 ”Io Wan—3; sefimwgux _- .‘ ‘. f"; 1 ‘ ' gaudy-.wm: . :"LA '33; -“ikww ’:Va7«'x ;.‘fl- , r- . $27! ’ ’ It»: ‘ ‘ “,7 - y ‘ I q,‘ AA‘. {,5 r_ " "*4 'v ; 37: ’ 1‘ km!“ ‘ I I 9);; ct . , V i-‘. .‘ .- A‘K . ,' H " . Cl C fog—'03:. \ M“. ‘14 b.’ 1‘ ‘V .35.: m ' y“:,.1 ‘ ..'-u. "- ~ I t V7) .7 _ , my; . ‘. “‘..-'.'A;‘ V _ . 4. tc ' v ‘4' L? 21,?» r. w’w’ . .. “A? ’ A. i .V.‘ . «an: «7: A .. n. . __ ....— v - r. ' . a . . WI“: M. ...> ”425533,? 42; .‘._~. A :giffi‘é‘?‘ ' $4.7“? . "376“” 'nvhtfgw, “ 8;" ‘ . wig. - . . “,1 “5, ¢.. - ‘— (4‘1? A Six “r'v '.' ~\'T '1‘ f ".4? X .0 , Hint! J “.2 9y ,s'm “ ' v (no. 3:- 5". 2'? ‘7 IV" ”:8 f 4»- v t r'i 1m 9; ‘ ' J~ , ‘ J v " . i .3631, .g‘ww" 31.”?! “‘ " "HR“ 3 -‘ C 5:" ~;~.7‘..3:.-' " ' h 'm. .‘- x A’, Q; ‘03:. 7» h f" \1' \ K :2 , ‘ 5., ' ‘ . ‘ffiéw’? M? "" ~ ~iz"." if". 30 J! ‘\.’ "gm! “V55: “4.! . \1‘,'.—_‘,2-h:€,. , . . ‘Ol‘1"¥ “NF"; 5. - J. -t' V‘ ‘ .n y " . ‘ ; Law. I; ”\‘A. ‘ .':‘~":)' ‘54?» 'V'.~-}f\§ ~-;"~. '.‘W ' V ,, I 1 , 3 ~ 5".‘v Ii- '3’ "*3 . 7.‘ . a: ' n.‘ ."L “1‘“! 9M ‘.x U ) ff." 5) 7' $.01 Ila". ' V 3'.- .l ‘, ‘ ya "‘3: ,.: 2 Windy, .r A?! \'H' {WU ’7’: MIC CHIGAN STATE UNIVERSITY LII BRARIES IIII III III "II III III II IIIIII 3 1293 00604 3305 LIBR-kRY III Michigan State University J This is to certify that the dissertation entitled Integration of the Developmental Theory of Carol Gilligan with Object Relation Theories presented by Anne Cosgrove Cunningham has been accepted towards fulfillment of the requirements for Ph.D. degree in BSJQthQgL ,/ é/gil’g fitajézl‘“ Major professor Date October 25, 1989 MS U is an Affirmatiw Action/Equal Opportunity Institution 0- 12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or baton duo duo. DATE DUE DATE DUE DATE DUE II “()u‘ufs ICHB ‘ MSU Is An AfFLanglvo ActIoNEgnI Opportunity lnctIMIon INTEGRATION 0? TI! DEVELOPUEITAL THEORY OF CAROL GILLIGAN OT WOMEN'S DEVELOPHEIT WITH OBJECT RELATIOIS THEORIIS by Anne Coagrove Cunningham A DISSERTATION Subaittod to lichigan State University in partial fulfill-ant of the require-onto for the degree of DOCTOR OF PHILOSOPHY Depart-ant of Psychology 1989 l \yk/v ABSTRACT IlTEGRATIOH OF THE DEVELOPHENTAL THEORY OF CAROL GILLIGAN OF WOMEN'S DEVELOPMENT HITH OBJECT RELATIONS THEORIES BY Anne Cosgrove Cunningham This research investigated the potential relationship, denied by Carol Gilligan, that there is conceptual overlap between her ideas of advanced forss of developaent of self in relationships with advanced forns of self development in psychoanalytic object relations theories of developsent. Specifically, the research compared two operational neasures of the highest perspective in wosen's developsent fro- Gilligan's theory, ”self in self's terns and other in the other's terns”, developed by Jane Attanucci (1984), and ”self and other chosen freely,” a scale score of the Revised Relational Self Inventory (Stronsen et al., 1987; Reinhart et al., 1985; Pearson et al., 1985), with a seasure of ”quality of object relations in early sesories (Ryan, 1970, 1973, d 1974; Haynan & Ryan, 1973). The research sanple was conposed of 28 Diddle-aged wonen, who are students, exercisers, or both. The research results were reported for the couplets sasple, and for a restricted sasple excluding four wosen who were identified as psychologically distressed. The data show a positive relationship between Gilligan's highest developsental perspective and levels of object relations. They also indicate the possibility of different types of woaen who score highly on ”self in self's terns and others in others' terns,” and on "self and other care chosen freely.” In the forser case, there say be differences based on vhether wosen are in transition to a higher level. In the latter case. there say be differences in wosen's views of "care" depending on whether they choose nainly justice or care voices (Gilligan. 1987). The analysis was suppleaented by exaaination of case saterial. This reflects Gilligan's novenent toward a lore herneneutic perspective. It also includes a review of najor theses susaarized under issues of ”healthy” narcississ. These include: concern about authenticity of the self; view of the possibility of further developnent; significant transitions; re-assesssent of the leaning of tine; and coping with losses and death. Overall findings are discussed from both theoretical and neasurenent perspectives. ACKNOWLEDGEMENTS Thanks go to sy fasily, sy advisors, sy research assistants, sy friends, and sy acadesic and professional colleagues. Thanks go Ton, Ton D. and Katharine for patience for the sany tines I was unavailable. Hy husband gave se such encouragesent as well as a sodel of acadesic achievesent. Ry intelligent children are a source of delight to ac. Thanks also go to sy sother, sy sister Carolyn, and sy Aunt Rarian for years of love, and for understanding when I kept putting off sy visit because I was ”working on sy dissertation". Thanks also, Hon, for sharing your excitesent about the world of ideas with as when I was young. Thank you, Aunt Anne, for giving se a nodel of intellectual achievesent to isitate. Thanks to Dr. Ellen Stronnen for her inspiration and guidance around the work of Carol Gilligan and the RRSI, and for the opportunity to spend a year participating with the research study group and the developsental interest group that she and Dr. Donelson created. Thanks to Dr. Dert Karon for his cospany, support, sense of husor, and for providing office space and the use of his cosputer. Thanks to Dr. Elaine Donelson for introducing se to Gilligan's ideas through then- graduate student Ray Chin. Thanks to Dr. lorsan Abeles for saking tine for so when he really did not have the tine. Roy Gerard, R.D., Chair of Tasily Practice, generously extended sy adjunct appointsent over several years while I did sy cosprehensives and dissertation. Judy Conley introduced se to exercise in the siddle years of sy life, for which I will be forever grateful. I especially as grateful to Sharon Spryszak and Annie Hahle, sy student research assistants, vho thoughtfully and carefully coded the interviews based on Jane Attanucci's sanual. Thanks to Leslie Rolowitz for her invaluable assistance in coding the Early Resories data, to Drenda Hayne for statistical help, and to Dr. Edward Ryan, who provided se with an updated sanual. Without the expert secretarial support of Barbara Hallory, hynae Wozniak, and Vicki Conrad, I would never have finished in due tise. Departsental secretary Suzy Pavick guided se through the saze of requiresents for sy degree. Thanks to sy colleagues, especially those sesbers of the Michigan Psychoanalytic Council (UPC) who encouraged se in this project and in sy future studies in psychoanalysis, including Carol David, Ph.D., Edward Gibeau, Ed.D.and Teresa Bernardez, R.D. A very large debt of gratitude is owed to Dr. lax Bruck. The Rosen's Studies groups in the Richigan Society for Psychoanalytic Psychology (HSPP), under the leadership of Carolyn Paulter, Ph.D. and Patricia Zipper, H.S.I., helped se very such in the forsative stages of sy ideas. Thanks to Ellen Toronto, Ph.D., for her leadership in the EPC Rosen's Studies group, as well as to the other group sesbers. Thanks also to the sodel of achievesent given so by Sister nary Carita, S. C., who got her clinical psychology degree when she was age sixty. Thanks also to Peggy Iurphy who inspired se by getting her law degree at a sature age after a distinguished career in education. Phyllis Dunsay, you taught se to work hard at details, which is never fun, in sy opinion. There are nuserous persons who contributed to sy developsent when I will only sention by nase, although such could be said about the kindness of each person: Dave and Sue Bowen, Sally and Stan Busp, laureen Cronin, Eran De Tranzo, Lil Gallagher, Jo Ann Kelly, Carole and bob Laloreaux, Carole Walker Hurray, Bobbi and Nike Rice, Arleen and Bob Stapleton, nary Stapleton, and nary Steady. I also thank the following persons who assisted sy professional developsent both recently and in an earlier period of sy life: Drs. Hary Andrews, Bob Boger, Hargaret Dubolz, Jo Lynn Cunningham, John Hurley, Sidney Ratz, Jis O'Brien, Bea Paolucci, Al Rabin, Verda Schiefley, and Bob Zucker. I extend sy deep gratitude to the wosen in sy research study who allowed thesselves to be interviewed and who shared so such of their experiences and inner worlds. Finally, I thank sy father, no longer with us on this earth, but hopefully looking down with a chuckle at his daughter who finally took his advice about trying to understand the ”why” of things, not just the ”how to do it." Ry research shows the isportance of fathers in the subsequent lives of daughters, which I already knew anyway. vi Phyllis Dunsay, you taught se to work hard at details, which is never fun, in sy opinion. There are nuserous persons who contributed to sy developsent whom I will only sention by nase, although such could be said about the kindness of each person: Dave and Sue Bowen, Sally and Stan Busp, Eaureen Cronin, Tran De Pranzo, Lil Gallagher, Jo Ann Kelly, Carole and Bob Laloreaux, Carole Ralker Hurray, Bobbi and Bike Rice, Arleen and Bob Stapleton, Eary Stapleton, and nary Steady. I also thank the following persons who assisted sy professional developsent both recently and in an earlier period of sy life: Drs. Rary Andrews, Bob Boger, Hargaret Bubolz, Jo Lynn Cunningham, John Burley, Sidney Katz, Jin O'Brien, Bea Paolucci, Al Rabin, Verda Schiefley, and Bob Zucker. I extend sy deep gratitude to the wosen in sy research study who allowed thesselves to be interviewed and who shared so such of their experiences and inner worlds. Finally, I thank sy father, no longer with us on this earth, but hopefully looking down with a chuckle at his daughter who finally took his advice aboUt trying to understand the ”why” of things, not just the "how to do it.” By research shows the isportance of fathers in the subsequent lives of daughters, which I already knew anyway. vi TABLE OF CONTENTS Page LIST OT TABLES ................... .......... ......... ix LIST 0! PIGURES ...................... ............... 3111 INTRODUCTIOI AID REVIEW OP LITERATURE. .............. 1 Introduction ............................ ...... 1 Possible Overlap of Gilligan's Theories with Psychoanalytic Theories ...... ...... 4 Gilligan's Theory of floral Developsent .................... ...... ... 8 Object Relations Theories of Busan Developsent ....................... 16 Espirical Study of Gilligan's Developsental Perspectives: Attanucci's Studies ....... 20 Espirical Study of Gilligan's Developsental Perspectives: Based on Use of RRSI ..... 25 Espirical Study of Developsent in Object Relations Theory ................. 31 Research Question ............................. 39 RESEARCH EETBOD . .................................... 41 Overview .................... .................. 41 Subjects ...................................... 42 Selection of Subjects ..... . ....... ...... 42 Desographic Description ............ ..... 44 Instrusents ................................... 44 Brief Sysptos Inventory ...... ........... 44 Attanucci Interview Keasures ............ 49 Revised Relational Self Inventory ....... 56 Quality of Object Relations in Early Resories Scale ............... 57 Case Katerial ................................. 62 Content Theses ................................ 62 Process ....................................... 63 Recruitsent of Subjects ................. 63 Contact 1 ............................... 64 Contact 2 ............................... 65 Data Analysis ................................. 65 Additional Results ............................ 70 RESULTS ........................... ....... ........... 71 Bypothesis 1 ................................... 71 Hypothesis 2 ................................... 72 Other findings of Interest .......... ..... ...... 73 Descriptive Statistics ................ ..... .... 82 vii Case Studies .................................. Theses of the Riddle Years of Life ............. DISCUSSION Appendix A: Coding Eanual for Self in Relation to Others and Reliability of Raters ............ Appendix B: Coding Eanual for Quality of Object Relations in Early Resories ............. Appendix C: Posted Notice and Advertisesent ... Appendix D: Desographic Description of Restricted Sasple ....... ..... ................... Appendix E: General Inforsation Sheet .......... Appendix E: Schedules for First and Second Interviews ....... . ................... Appendix G: Consent Tors ....................... Appendix B: Telephone Screen ................... Appendix I: Revised Relational Self Inventory, Scales, and Scoring ........ .......... Appendix J: Additional Cluster Analyses ........ viii 90 103 113 132 137 137 158 172 173 177 180 185 186 189 199 Table 10 LIST OF TABLES Revised RRSI: Scale leans and Standard Deviations ...... . ...... . ............. Desographic Description of the Sasple (N I 28) ..................................... Cosparison of Scores of Pull and Restricted Sasples to Korss for Persons with an Average Age of 46 on the BSI ......................... Schesa for Data Obtained fros the Coding of Attanucci Categories for Self in Relationship for Research Uosen .............. RRSI Reliabilities and Scale Intercorrelations ............... .............. Pearson Product Kosent Correlations for Sussed Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Eesories Scores (Cosplete Sasple, l - 28) ...... .............. Pearson Product Kosent Correlations for Sussed Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Besories Scores (Restricted Sasple, K - 24) ......... .......... Cluster Analysis of Sussed Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Eesories Scores, Ward's Kethod, Squared Euclidean Distance Coefficient (Cosplete Sasple, l - 28) ..................... Cluster Analysis of Susned Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Besories Scores, Ward's Kethod, Squared Euclidean Distance Coefficient (Restricted Sasple, K I 24) ................... Cosparison of Sussed Category Scores for Attanucci Variables Between Cosplete Sasple (I - 28) and Sasple with Bigh Sysptosatic ix Page 29 45 50 55 58 74 76 78 80 11 12 13 14 15 16 17 A1 D1 II J1 Scores Excluded (Restricted Sasple, R - 24)... Cosparison of RRSI Scale Scores between Cosplete Sasple (R - 28) and Sasple with High Sysptosatic Scores Excluded (Restricted Sasple, I I 24) ................ ............... Cosparison of Quality of Object Relations in Early sesories Scores between Cosplete Sasple (N - 28) and Sasple with High Sysptosatic Scores Excluded (Restricted Sasple, l I 24) .......... Cluster Analysis of Sussed Attanucci Categories Vard's Hethod, Squared Euclidean Distance Coefficient, Cosplete Sasple (K I 28) ......... Cluster Analysis of Sussed Attanucci Categories Vard's Hethod, Squared Euclidean Distance Coefficient, Cosplete Sasple (l - 28) ......... Sussary of Scores for High-Scoring Vosen for the Attanucci Category IV (Sus IV 4) and for Quality of Object Relations in Early Hesories . ........ Sussary of Scores for Low-Scoring Vosen for the Attanucci Category IV (Sus IV 4) and for Quality of Object Relations in Early Hesories ......... Sussary of Scores for Low-Scoring Vosen for the Attanucci Category IV (Sus IV 1 or 2), for Quality of Object Relations in Early Hesories, and High Score on Brief Sysptos Inventory (BSI) ........ Reliability of Rater Codings for Attanucci Categories: Percent Agreesent before Di.cu..ion 0.0.......OOOOOOOOOOOOOOOOO......O00 Desographic Description of the Restricted sup1.('.2‘) ...........OOOOOOIOO0.0.00.0... Ites-Scale total Correlations and Scale Reliabilities of the RRSI for Rosen ........... Cluster Analysis of Sussed Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Hesories Scores - UPGHA Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, I - 28 ....................... 83 84 85 89 92 94 95 156 173 193 199 J2 J3 J4 J5 J6 J7 J8 J9 J10 J11 J12 Cluster Analysis of Sussed Attanucci Categories, Sussed RRSI Scale Scores, and Quality of Object Relations in Early Hesories Scores - UPGHA Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, K - 24 .......... .. ......... Cluster Analysis of Sussed Attanucci Categories - UPGHA Hethod, Jaccard Sisilarity Coefficient -Cosplete Sasple, R I 28 ........................................ Cluster Analysis of Sussed Attanucci Categories - UPGKA Hethod, Jaccard Sisilarity Coefficient -Restricted Sasple, .- 2‘ 0000000000000000000000000 O ....... O ...... Cluster Analysis of Sussed Attanucci Categories, UPGHA Hethod, Cosine Distance Coefficient -Cosplete Sasple, R - 28 .......... Cluster Analysis of Sussed Attanucci Categories, UPGHA Hethod, Cosine Distance Coefficient -Restricted Sasple, K I 24 ........ Cluster Analysis of Sussed RRSI Scale Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, K - 28 ......... Cluster Analysis of Sussed RRSI Scale Scores - UPGHA Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, K I 28 ......... Cluster Analysis of Sussed RRSI Scale Scores - UPGHA Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, R I 24 ....... Cluster Analysis of Sussed RRSI Scale Scores Vard's Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, K . 24 ....... Cluster Analysis of Sussed Attanucci Categories and Sussed RRSI Scale Scores - UPGKA Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, K I 28 ....................... Cluster Analysis of Sussed Attanucci Categories and Sussed RRSI Scale Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, l I 28-....................... xi 201 204 205 206 207 209 210 212 214 215 216 J13 J14 J15 J16 J17 J17 Cluster Analysis of Sussed Attanucci Categories and Susned RRSI Scale Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, I I 24 .... ..... . ........... Cluster Analysis of Sussed Attanucci Categories and Sussed RRSI Scale Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, R - 24 ........ . ...... ...... Cluster Analysis of Sussed Attanucci Categories and Quality of Object Relations in Early Kesories Scores - UPGHA Rethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, I . 28 ..... .................. Cluster Analysis of Sussed Attanucci Categories and Quality of Object Relations in Early Hesories Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Cosplete Sasple, K - 28 ................ ....... Cluster Analysis of Sussed Attanucci Categories and Quality of Object Relations in Early Hesories Scores - UPGHA Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, K - 24 ..................... Cluster Analysis of Sussed Attanucci Categories and Quality of Object Relations in Early Hesories Scores - Vard's Hethod, Squared Euclidean Distance Coefficient - Restricted Sasple, l - 24 ....... ...... ... ..... xii 218 219 220 221 222 223 LIST OF FIGURES Figure l Attanucci's Hodel of Vosen's Developsent ................ ................... xiii Page 21 INTRODUCTION AND REVIEW 0? LITERATURE Introduction This research seeks to address the general question: what is sature, consciously experienced self-love as it is subjectively felt by siddle-aged wosen? The question, of course, can be addressed--and sose would address it--as an investigation of norsal or healthy narcississ. Nevertheless, this research also strives to address the question in a contextual frasework. Partly this is due to another concern--for the work of Carol Gilligan and colleagues which proposes that wosen's growth and developsent takes place in a context of relationships. Gilligan's work leads one to consider that a wosan can achieve an integrated, free sense of care and responsibility for herself at the sase tise that she cares for others in a relatively unanxious, yet responsible sanner. However, the kernel of these ideas, nasely that a wosan's personal developsent takes place in relationships, and should be exasined in that sase context, also appears in a very different intellectual frasework, that of psychoanalysis. Psychoanalytic theorists of the interpersonal school, and, to sose extent self psychology, and object relations theory, also consider the isportance of personal developsent in context. The seaning of context can vary fros the ”real" interpersonal world throughout infancy, childhood, and adulthood, to an intrapsychic representation of self and other which takes on its own life independent of the ”real” world. To a certain extent, though, psychoanalytic 2 theorists stress the function that persons in one's environsent play in one's growth and developsent. Here recently sose psychoanalytic theorists have cose to stress a sutuality of realistic perceptions and espathic concern which characterize a developsentally advanced capacity for object relations. It is proposed that these two intellectual fraseworks, the work of Carol Gilligan in a developsental-contextual perspective and the work of sose psychoanalytic theorists in an object relations perspective, converge on a single idea, nasely, that personal growth takes place in relationships which are sutually enhancing. Given this consideration, it is possible to describe the proposed research in sore specific detail. Riddle-aged wosen who can be defined as involved in self care in what seess to be a responsible, healthy sanner, were studied and the seaning of their behaviors in the context of wosen's present conscious experience of self and of recollections of past developsent and relationships was investigated. This self involvesent is operationalized, for the purposes of this research, in two ways: first, in terss of concern for developing oneself intellectually which say be evidenced by engaging in acadesic studies; second, in terss of care for one's physical well-being which say be evidenced by engaging in regular physical exercise. Engaging in these activities say precipitate a sense of crisis about spending tise on self rather than caring for isportant others in sose obvious way. On the other hand, these activities say also be a response to an experienced crisis that the wosen are not spending enough tise on thesselves. Both activities, studying and exercising, are apt to be sotivated by a cosplex range of reasons, 3 engaging different levels of understanding. Vosen choosing such activities say well be quite conscious of at least sose of these reasons. Having set out the general purpose of this inquiry, it now resains to investigate in sose detail those considerations which bear on the choice of a suitable fors of research. To this end, interest is directed to the work of Carol Gilligan and her colleagues who contribute an isportant frasework to consider fesale developsent. On the other hand, Gilligan's work has not been integrated into the other, psychoanalytic object relations frasework, at this tise. There have been efforts by fesinist psychoanalysts in this direction, however. Gilligan's developsental theories are presented, her research sethods, and sethods of those who tried to study her developsental hypotheses. Object relations theories also present developsental nodels for consideration, with the points of potential overlap of these two types of theories focusing on advanced forss of personal developsent. The review also touches the siddle years for wosen fros the point of view of continued developsent of the self. These concerns are reviewed under the conceptualization of narcissistic issues in the siddle years of life. The review considers a broad scope of espirical seasures extant for seasuring developsent in an object relations frasework. However, only one concept speaks to both concerns of healthy self interest and sutuality with the other, Hutuality of Autonosy; thus seasures of that construct are is reviewed in sose detail. 4 Possible Overlap of Gilligan's Theories with Psychoanalytic Theories Freudian sodels of husan developsent have been subject to criticiss by psychoanalysts, especially for the unfavorable cosparison of fesale to sale developsent of conscience. It occasions, therefore, no surprise that psychoanalysts, particularly fesinist therapists, have seized upon recent work by Harvard developsental psychologist Carol Gilligan to support their thesis that psychoanalytic theory is sale- dosinated. Still, whatever the serit of their reasons, they have done little to develop Gilligan's works, to appraise then, or integrate then with other psychoanalytic constructs. Gilligan had herself puzzled over Kohlberg's theory of soral developsent which, in early studies, seesed to show that wosen consistently reached lower soral cognitive levels than sen; in her developsental-contextual approach, she constructed an alternative to Kohlberg's sodel. At first she - and later her students - exasined the conflicts experienced by adolescent girls and young wosen at isportant tises in their lives. In explaining wosen's soral developsent, Gilligan retains as a key theoretical concept justice, the basis of Kohlberg's work; but she also proposes developsent through care, as a second, equally valid, theoretical concept which say in fact be sore salient in wosen's soral deeds than in sen's. In works subsequent to 1982, she develops both of these soral perspectives and the interplay between thes. Each perspective, she observes, is subject to its peculiar distortion: if the agent, in reasoning about justice, is able to ignore the individuality of the other, the agent who reasons about care has the potential to deny the individuality of his or her self. Rogers (1987) 5 has reviewed the developsent of these perspectives and their definitions, and her work will be referred to later in this review. The ”different voice” of wosen which Gilligan stresses does seen to be one unappreciated by current psychoanalytic theories. Still, sose questions arise about her project. One say ask what overlap, if any, exists between her descriptions about wosen's conscious experience and various psychoanalytic theories which stress the unconscious. And, in particular, one say look at her idea of wosen's highest fors of developsent, nasely, the crowning achievesent which integrates justice and care towards both self and other, and question just how different that is fros the concepts of saturity which object relations theory has advanced. Gilligan has considered sose of these satters, and she asserts that object relations theory sakes developsent hinge on separation and individuation. As a consequence, object relations theory undervalues, in her estimation, both one's connection with others and one's ability to think and feel with others. Indeed, she affirss that the esphasis on separation and individuation sakes it theoretically ispossible to value a self which is experienced in the context of attachsent with others; and, she concludes, this in turn sakes it theoretically ispossible to value her concept of husan developsent. She cossents that the approach and tersinology, beginning with the ters ”object" relations, is wholly separate in perspective. Further, Gilligan claiss object relations theories, which esphasize the isportance of selflessness, distort the role of sother. Tying selflessness to sotherhood, in her opinion, sisply strips wosen of their agency (Gilligan, 1987). What happens to a wosan's own sense of 6 self and personal goals when she responds to an isperative to be “selfless” or ”good-enough" for her infant? Finally, while psychoanalytic theories value capacity to think objectively and fairly, Gilligan proposes that such an attitude can be a tors of detachsent; objectivity, in her view, can be a soral probles. There can be little doubt that Gilligan has seized upon significant issues--that she has pointed to sose lacunae in psychoanalytic theory, sose unfortunate forsulations in object relation theory. However, psychoanalytic theory in general, nurtured by Freud and others for close to a century and over several continents, is richer than she suggests. And object relations theory, in particular, it is subsitted, is not as inisical to her project as she states. For exasple, the sutuality in autonosous relationships on which Gilligan and fesinists rightly place high value also finds chaspions asong object relations theorists. Thus, if it is fair for Gilligan to catalog disagreesents with object relations theory, it is also just to appreciate their agreesents. In this way one can see that Gilligan, in theorizing about soral developsent and the achievesent of a balanced care for self and others treads fields already worked by object relations theorists. It will be necessary to review in sore detail the ispact of Gilligan's work on wosen's studies fros an object relations perspective in order to lay ground for sy argusent that, despite Gilligan's reservations, there are sose overlaps between her theory of wosen's developsent and object relations theories of developsent. Gilligan's work is frequently referenced in psychoanalytic studies. A recently edited voluse on wosen and psychoanalysis (Alpert, 7 1986) contains twenty-four references to Gilligan. In sany instances her ideas are assisilated presaturely into psychoanalytic theories. For exasple, in a chapter on sasochiss and love, Benjasin (1986) writes that ”...deflation of early osnipotence say be viewed positively as generating the ego's capacities for sociability and sublisation (Roiphe and Galenson, 1981; Gilligan, 1982)...” (p. 125). One say note that Gilligan does not use the terss ”ego,” "osnipotence,” ”sociability," or "sublisation” in her work. Nor is she particularly concerned with the phenosena these terss label. Another probles is that psychoanalytic writers quote her inaccurately. For exasple, in that sase voluse, Litwin, writing about fesale autonosy, cossents: ”Gilligan (1982) feels that relationships take precedence over soral reasoning and logic for wosen.” Litwin has sisunderstood Gilligan: Gilligan's position is not that relationships take precedence over soral reasoning, but that the conflict between self and other constitutes the central soral problem for wosen. Nevertheless, the large nusber of references to Gilligan in this text--her ideas were cited sore often than the developsental work of Nahler--indicate that a desire and interest exist in utilizing her theories to address issues with which fesinist psychoanalysts grapple. There is a need to study espirically the relationship of her ideas to those of object relations theory: such is, in fact, the overall intent of the present proposal. Before proceeding further, however, it is necessary to review in sose detail Gilligan's view, as of 1982, of the central issues and factors in wosen's soral cognitive developsent. 8 Gilligan's Theory of Noral Developsent In 1982 Gilligan published In a Different Voice. In this book, while arguing against views presented earlier by Kohlberg, she also presented her own views of soral developsent of care in wosen which she presented in terss of three soral perspectives. She distinguishes between justice and care voices within these perspectives. Justice and care voices say be distinguished as follows. Vhen an individual speaks fros a justice voice, he or she sees inequality in relationships as problesatic, highlights soral concern for equality, fairness or justice, and values rules and principles as protections of fundasental husan rights. Vhen speaking fros a care voice, an individual understands relationships in terss of attachsents and sees detachsent and abandonsent as problesatic. He or she highlights soral concerns of caring, that is, of not hurting, paying attention, helping, and sustaining connection, and values activities of care as responsive to husan needs (Rogers, 1988, p. 4). It should be noted that Gilligan portrays these soral orientations as asyssetrical, coherent perspectives, each representing a particular way to address-soral probless (Rogers, 1987, p. 4). These voices denote, however, the basics of husan soral judgsent which includes a consideration of self, of others, and of the relationship between thes. If a soral voice shifts fros justice to care, then the isages of relationship also shift fros hierarchy or balance to network or web. If and when they do change, there is a significant change in the whole systes of perception. There is an interplay between soral voices within the three soral perspectives which will be discussed, for they show progression fros 9 lower to higher thought. Based on the work of Johnston (1985), Gilligan concludes that soral judglent should not be lapped along a single, linear, stage sequence. Gilligan's developsental sodel, of course, does not speak to issues which are pre-verbal, although she does see antecedents in early childhood relationships in experiences of oppression and abandonnent. These are the basis for subsequent sensitivity to justice and care concerns (Gilligan, 1987). She presents these issues as becoaing salient in siddle childhood. These are presented in the following section. The Three floral Perspectives These soral perspectives nay be sunnarized as: (1) initial concern with self and personal survival: (1a) transition period (2) focus on care for others as one's responsibility; (2a) transition period,: and (3) care for self and others chosen freely due to reflective understanding of care as the most adequate guide to the resolution of conflicts in hunan relations (1982, p. 105). These perspectives and the transition periods between then are now presented in sore detail. The First Perspective. In the initial perspective, the individual focuses on care of self in order to ensure survival, feeling, at the sane tine, a sense of aloneness. It is not so such that there are no relationships in life, but ”relationships are for the lost part disappointing.” floral issues are not generally considered, unless the question of serving one's own needs are in conflict. Then the individual would have to decide which needs take precedence. There is a feeling of lack of personal power which is caused, according to Gilligan, by feeling disconnected fros others (p. 15). This felt 10 powerlessness extends to her soral code, which she feels isposed on herself rather than freely esbraced (p. 79). In addition, Gilligan says that wosen in this perspective ”in sose instances deliberately choose isolation in order to protect thesselves against hurt” (p. 75). This first perspective say be followed by a transition period in which the wosan is unable to integrate previous focus on self with a new sense of responsibility to care for others. Her forser focus on self is now judged selfish. In this transitional period, the new sense of attachsent to others is accospanied by a view of herself as a person who is capable of doing ”the right thing,” and who becones a worthy nesber of a con-unity. She redefines her own self-interest based on a change in priorities. In Gilligan's words: ”the dilensa assunes a soral dinension and the conflict (esphasis nine) between wish and necessity is cast as a disparity between 'would' and 'should.‘ In this construction, the 'selfishness' of willful decision is counterpoised to the 'responsibility' of soral choice (p. 77).” During the first transition period, the individual still desires to take care of self: however there is a change in the way the individual perceives her self--as a person who desires to do good for others-and this new view brings with it ”an enhancesent of self-worth" (p. 78). Without a nove to a sense of social inclusion, there can be no transition fros the first to the second perspective. 0n the other hand, the wosan who seriously doubts her own goodness is blocked in her transition to the second perspective (pp. 78-79). The Second Perspective. The second perspective in wosen is characterized by an overall sense of cos-unity with others which necessitates care for others; in fact, the soral good is equated with 11 such care. Gilligan says that the conventional feninine 'voice' is in the second perspective, defining self and proclaising her own worth on the basis of the ability to care for and protect others. In the second perspective the wosan ”validates her clain to social sesbership through the adoption of societal values” (p. 79). This brings a sense of freely choosing her soral values rather than having the. isposed. 0n the other hand, her personal survival now appears ”to depend on acceptance by others" (p. 79). As Gilligan describes this position: "The strength of this position lies in its capacity for caring; the linitation...lies in the restriction it isposes on direct expression” (p. 79). If there is a transition fros the second perspective to a third, it is caused by a wosan's awareness of a new conflict: she say hurt herself in the course of giving to others. There nay cone a tine when the wosan believes that there is no option that is in the best interests of everybody. By its nature, the conflict can be resolved by regression, progression, or the individual nay stay in the second perspective, feeling paralyzed by personal dependence on others while at the sane tine feeling the need to give to others. She is faced with a dilessa: she is responsible for others, but she also wishes that others take responsibility for her. She say becone aware that this causes both parties to feel ”nanipulated and betrayed” (p. 82). A wosan nay find it in herself to change, as she considers the lack of justice in serving others but not herself. She say reconsider and now begin to take on a new forn of responsibility to include her own needs within the "co-pass of care and concern” (p. 82). This disequilibriun nay cause a change to a new, integrated focus, in which 12 the inequality between her for-er cossitnent to other and relative neglect to self is addressed. Theighird Perspective. If there is a successful transition period after the second perspective, the wosan becoses able to separate her own needs fros those of others so that it becoses "possible to be responsible to self as well as others and thus reconcile disparity between hurt and care” (p. 82). This final perspective appears to be balanced: it is also characterized by a change in sense of self. As Gilligan puts it, there is ”an increasing differentiation of self and other...” (1982, p. 74). There is a shift in the wosan's values fros a conventional soral sense of goodness to one which acknowledges her own self and accepts responsibilities for this self. Thus, the criteria for judgnent of her own actions soves towards honesty with herself rather than judging herself by others' criteria. She becones capable of assessing her own intentions and now finds it isportant to try to assess the probable consequences of her acts. Thus the criterion of honesty replaces the criterion of goodness for judging her own actions (pp. 82-83). In the third perspective ”care then becoses a universal injunction, a self-chosen ethic which, freed fros its conventional interpretation, leads to a recasting of the dilensas in a way that allows the assusption of responsibility for choice” (1982, p. 90). TragsitiogAPeriods. Gilligan focuses on the isportance of transition periods in her perspectives. There are tises when the current way of viewing the world and one's place in it, one's rights and responsibilities, is questioned because of a conflict that cannot be resolved satisfactorily based on the current way of thinking. Such a 13 conflict occurs between perspectives two and three when a wosan realizes that there are things she needs, really needs, that no one other than herself can or should do for her . This confluence of circunstances and soral perspectives can occur at any tine after the second perspective-- of care for others--is reached. In Gilligan's research it could occur in wosen in their late teens and twenties when they are confronted between care for self (have an abortion) and care for others (the fetus; the father of the child: others). Speaking of the Iain study in regard to which she discusses those stages, in which wosen were interviewed while voicing their decisions of whether or not to have an abortion, Gilligan writes: The abortion study suggests that wosen ispose a distinctive construction on soral problems, seeing soral dilensas in terss of conflicting responsibilities. This construction was traced through a sequence of three perspectives, each...representing a sore cosplex understanding of the relationship between self and other and each transition involving a critical reinterpretation of the conflict between selfishness and responsibility (p. 105). These progressions are not considered the universal experience of wosen, and even novesent fros perspective one to perspective two is not thought to occur in every individual. Thus, this is perhaps the weakest of her claiss. One could guess that her theory of the progression fros perspective one to perspective two reflects her earlier dependence on Kohlberg's developsental stage sequences which she has now loved beyond. This progression, though, can be conpared to psychoanalytic theories of developsent which posit an initial narcississ beyond which an individual progresses in norsal developsent. 14 Gilligan's Research Hethod One say turn now to the consideration of the espirical sethods Gilligan used to develop her developsental perspectives of justice and care. ”These different views of care and the transitions between then energed fros an analysis of the ways in which the wosen used soral language - words such as ”should, ought, better, right, good, and bad," by the changes and shifts that appeared in their thinking, and by the way in which they reflected on and judged their thought” (Gilligan, 1982, pp. 73-74). Thus, her methodology sarks a shift fros the Kohlbergian structured sethod of scoring soral thought to a sore herseneutic sethodology. Her rationale for naking this shift was that the truth of wosen's voices was lost in a pre-forned set of criteria that sifted through the soral words of wosen and let slide through the sieve those ideas which did not comprise a part of the sequence in the sale (Kohlbergian) viewpoint. As part of her shift in sethodology she began to ask wosen to describe in their own words what they defined as noral dilennas, rather than decide this for then. At the tine she used this sethodology, her ideas were themselves in transition. She began to de-esphasize the organic sodel behind her thinking in a developsental frasework, and to esphasize sore the contextual sodel toward which she was loving. In a paper presented during the Hurray Lecture Series at nichigan State University, April 8, 1988, Gilligan offered her research sodel as the herneneutic one of studying wosen's self in relationships in a care voice. A herneneutic sethod assuses that a person's orientations to structure of relationships and to conflicts are to be discovered by clarifying the 15 explanatory framework, or context, which then reveals the person's meaning. In this new methodology, those narratives which describe relationships in terms of attachment or detachment are called care narratives. Those which describe relationships in terms of inequality or equality or reciprocity are called justice narratives. Both narratives deal with perceived vulnerability; care narratives reveal vulnerability to abandonsent, while justice narratives reveal vulnerability to oppression. If everyone is vulnerable to both oppression and abandonment, then this is why the two moral visions recur in human experience. The reader may infer intuitively that there is some overlap between Gilligan's progression and some theories of developsent extant in psychoanalysis. Since self is the sole object of concern in the first perspective, this perspective might be comparable to a stage of narcissistic preoccupation, which would also imply that the individual experiences only poor object relationships. The ”selflessness" of the second perspective could be viewed in a number of ways, including empathy, on the positive side, or as resulting, on the negative side, from a form of sasochiss. The final perspective could be compared to the achievement of ego identity--achieving a healthy sense of self, but also participating in relationships in which the person works for the enhancement of the well-being of the other who is perceived as a complex individual in his or her own right. The individual serves the needs of the other in a richer fashion, one that is not tied to the previous less cosplex view of the other and of his or her needs. Although the first perspective in Gilligan's theory and the third stage appear comparable to the beginning and and stages of developsent in object relations 16 theories, the explanation of what takes place in between obviously differs. Still there is one additional point of overlap, for psychoanalysis also appears to be moving towards an integration of an organic developmental model with a contextual (relational) model of personal development. In the next section, the review turns to a more complete discussion of psychoanalytic object relations theories of development as they concern the potential to develop the capacity for what Gilligan calls a new potential to integrate the interconnection between self and other, in which there is concomitantly a new differentiation of self and other and a growing comprehension of the dynamic of social interaction. Object Relations Theories of Human Development Psychoanalytic theorists call the capacity for experiencing human relatedness the capacity for object relations. Past relations between self and others, it is assumed, give rise to the development of internal psychic structure. There are, in fact, several distinct theories, each of which may correctly be termed object relations theory. Depending on which particular theory is used, the model of the internal psychic structure changes. Still, all models generally presume that relationships in the external world have internal representation in both conscious and unconscious forms. Further, the models presume that present relationships are interpreted by the self in light of its internal organization of all past experience, particularly early childhood experience. In sum, models of internal psychic structure propose to explain how individuals organize and understand the cognitive 17 and affective components of their relationships with others (Urist, 1980). While Gilligan claimed to discover her perspectives of the self in the larger meanings reflected in narrative, the object relations theorists fashion from persons' words and actions theories about their capacity for object relations and the internal structure in which these capacities are aligned. Indeed, to assess this capacity is to assess ”the ways in which feelings and conscious and unconscious ideas about the self, about other people, and about the relations between self and others are organized in an individual's mind” (Urist, 1980, pp. 821- 822). The capacity for object relations, it is presumed, is complex, composed of several related capacities. Its developmental progression corresponds to the increased complexity of internal mental structures (Kernberg, 1966). In general, the outcome of development is normally considered to be autonomy (Vinnicott, 1960: Hahler, Pine, & Bergman, 1975). The child's development is largely the outcome of the context in which he or she develops. The parental figure(s) must respond appropriately to the changing needs of the child in order for the child to gradually build up the necessary psychic structure to both take care of self and relate successfully to others; otherwise, less than optimal development ensues. Although relatively unknown in larger psychoanalytic circles, psychologist Jeffrey Urist's theorizing about object relations is reviewed next for the reason that he is the originator of the concept of mutuality of autonomy, which is central to this research, and he has also written reviews of object relations theory for psychologists 18 (Urist, 1980). Urist based his model of developmental progression in object relations which culminated in achievement of mutuality of autonomy mainly on Rernberg (1966) who postulated an initial stage of introjection, in which mental images are organized as self or not-self, depending on their potential for comfort or discomfort. The second stage concerned identification; it is here that early, split-off, and affectively opposite self and object fragments begin to achieve some degree of cohesiveness. Increased structuralization thus enables specific functions, which hitherto had been performed by the external object, now to become part of the capacities of the self. The third stage is that of ego identity; it culminates with the achievement of what Urist calls "mutuality of autonomy.” Kernberg's overall model is similar to others which Jacobson (1964) and Hahler (1971) have developed. Of course there are differences in the timing of developmental shifts and in relative emphasis each places on various cognitive, affective, or instinctual considerations. Still, it is evident that all these combined models use concepts drawn both from ego psychology and object relations theory. The work of Kohut, especially that produced between 1965 and 1971, belongs in this camp. Subsequent to 1971, Rohut moved some distance from an ego psychology-object relations view to formulate his own views. After 1971 he did not, for example, view narcissism as the earliest stage in object relationship, rather he considered narcissism and object love to establish distinct developmental lines. Where Preud had a line of development from autoerotism to narcissism to object love, Rohut, writing of ”The Formation of the Self,” postulated something which would look like a side by side development of archaic object love (parts) to 19 archaic whole object to mature object love as a model to consider. Concurrently there developed autoerotism (archaic parts) to nuclear self (archaic narcissism) to mature self (mature narcissism). That is, Rohut believed that the capacity for regulation of self-esteem (and self care) developed separately from the ability to love others (Kohut, 1974/1978, p. 765. Rohut's theorizing focused particularly on the developmental precursors of inadequate self esteem and various problems in the adult person's capacity for self care. Rohut postulated that these result from various developmental deficits due to inadequate responses from parental figures. Bis theoretical work no longer focused on potential to achieve mature object love. It focused, instead, on the development in the narcissistic (self care) line, and later, on the development of a healthy self. Kernberg, on the other hand, stayed within an object relations-ego psychology framework. Thus, in summary, object relations theories in psychoanalysis focus on the subject's inner world of self and object representations, and how these develop. Having completed reviews of both Gilligan's theories of female development, and object relations theories of human development, it is appropriate to turn now to empirical concerns. Here several questions suggest themselves. To what extent have Gilligan's theories on development of self in connection with other been tested - especially with samples of adult women? To what extent have object relations concepts of development been tested, especially as they relate to mutuality of autonomy? The review found that there were only two 20 empirical studies which addressed Gilligan's model in this developmental framework, and they are reviewed next. Empirical Study of Gilligan's Developmental Perspectives: Attanucci's Study One of Gilligan's students, Jane Attanucci, researched adult women's perspectives on self and mothering. Since her work comes closest to the conceptual framework which the present study adopts, it will be reviewed in some detail. Attanucci studied adult women in her doctoral dissertation in terms of their relationship to themselves and their relationships to others. She conceptualized her research as a bridge between the work of Lyons (1983) and another way of viewing justice and care perspectives. She began with a "logical” set of categories, two of which referred to self and two to the other. She analyzed interviews of mothers of young children to more clearly define their care perspective towards self and other, specifying to which ”other” the statements referred. Her work is important for the present research because her methodology is directly applicable to this research. Attanucci defines the feminine self not solely in terms of the internal organization of qualities and dispositions but also in terms of the interpersonal reality of ongoing relationships" (1988, pp. 2-3). Thus, Attanucci defines self not only intrapsychically, or in an object relations perspective, but also interpersonally, from the viewpoint of role theory. Attanucci considers that her research builds upon the work of Chodorow (1982), who called for research on the experience of conscious, 21 planned mothering, as well as speaking to Gilligan's call for studies of women in women's own terms. (Gilligan, 1982, p. 90). Attanucci based her dissertation on her own earlier study (1982) of young mothers' responses to the question: ”now would you describe yourself to yourself?” Attanucci expected that these women would define themselves predominantly in terms of others. However, she also found that they described themselves in ways previously attributed to men in the work of Lyons (1982), that is, in terms of the separate, objective self. Therefore, it was necessary to develop a model which showed not only women's development in the care perspective but also their development in the justice perspective (see definitions of these perspectives on pp. 7-8). This, among other things, led to her subsequent four category model, which is presented in Figure 1 (Attanucci, 1984, p. 104). It should be recalled that these were presented as "logical" categories at which she looked to see where women's responses would cluster. Categories In Attanucci's Hodel Catgggry I. When self description reveals an understanding of self and other in reciprocal roles, the self and other are described from an objective, third person perspective. Wo distinction is made between self and role, or other and role. This understanding is conveyed as mutually reciprocal and functioning in a rather closed and static system. Conflict between self and other is not explicitly acknowledged. (Attanucci notes that it cannot be concluded from the interview material in her study if this self description truly represents the women's understanding of self in relation to others, or indicates an unelaborated response not adequately probed and pursued.) 22 Figmre 1 Attanucci's Hbde] of Women's Devqupment (from Attanucci, 1984, p. 104) Self instrumental to Self in self's others terms Perspect- Others "in their own Others in their ive terms" own terms 11 IV toward Self instrumental ”Self in self's others to others terms” Other instrumental Other instrumental to self to self I III Perspective Toward Self In sum, persons show reciprocal roles, and the self is instrumental to others, and the other instrumental to self. There is an unreflective and unelaborated quality about the relationship. Also, the woman does not acknowledge any conflict between self and other caused by different roles. Subsequent to Category I come Categories II and III. Each of these has a positive and negative side, and neither is projected to be developmentally superior to the other. Category II. This category shows the feminine role of self subordinated to the needs, demands and expectations of others. Aware of the expense to themselves, women often describe the backfiring of their generous intentions to respond to the other in the other's terms. This resulted in harm to thesselves and ultimately to the other. "Other in their own terms" placed in quotations indicates the failure to express 23 authentic care and also failure to acknowledge their own responsibility and power in a relationship. Another aspect of this category is that a woman denigrates the self for failures in fulfilling the requirements of the selfless role; the other can be seen to eclipse the self. Category III. Self description in this category is both self- assured and self protective. Others seem to be instrumental to the self rather than persons in their own right. Women take a stand against others who might seek to dominate, use or abuse them. However, their uncompromising stance is described as threatening to the relationship, denigrating the importance of the other. The woman in this position is vulnerable to losing sight of the other person and the relationship they share. This position is described as ”self in self's terms" in quotations to emphasize that "when the self's terms emerge to the contrary of the other’s terms (or unaware of them), they are not an authentic representation of the self. Category IV. Self description in this category reveals an understanding of self and other in a relationship that requires dialogue between the two for consideration of each other's terms. Self description in this category recognizes self and other each in their own personal terms. Women describe themselves honestly as neither selfless nor selfish, Attanucci asserts. They see themselves both as being capable of hurting others and being hurt, and strive to minimize such hurt. The virtue of this position is authenticity, a self-in-relation- to-others perspective, and reflectivity. 24 Atgggucci's Research Design Citing criticisms by Dromley (1977) and others of existing methods of research, she proposed his option of studying persons in case studies which delineate person's ”ordinary language.” Specifically, she utilized an open-ended, unstructured approach to research interviewing involving home visits. Her subjects were 20 women who were presently part of a longitudinal study of parenthood and child development over an approximately six month period. Women were recruited through pediatricians when their infants (their second, third, or fourth born) were either 4 or 10 months of age. Infants were 10-16 months old at the time of the last interview which provided data for her research. Women were an average of 31 years old, ten years married, and had 2-4 children. Interviews were made by an independent interviewer (who appears to have been a graduate student) who developed the format for the final interview, and included Attanucci's questions as part of the final, terminating interview. Those questions which pertained to this research are listed in Appendix F, Interview 1, questions 2-6, 10-23. In general, they ask women directly about their experience of self in important relationships (self in relation to husband, children, own sother, etc.). Coding of data was done from transcriptions of interviews in which the specific types of self statements were identified and classified according to what person they refer to (husband, child, etc.) in a relationship with the self (in which category they fall). Since this method will be used in the present research, it will be reported in the subsequent ”Hethods' chapter. 25 Attanucci's Researcthesults A cluster analysis found three clusters in the data, which were Category I statements (cluster 1), Category II and III statements (cluster 2) and category IV statements (cluster 3). Thus, her data support the notion of a developmental progression in three broad stages. Cluster 1 finds maternal self description in the idealized terms of reciprocal roles. Cluster 2 is characterized by the conflict between self and other, the dilemma of in whose terms the self will be defined. Cluster 3 features a perspective on self and other based on dialogue and mutual recognition of terms. She proposes that there is a sovement in self description from reciprocal roles (Category I) through a transitional conflict concerning roles (Categories II and III). The author compares Category III responses to the liberated woman's insistence on her own needs and rights, which may lead her to lose of sight of the other. On the other hand, the traditional mother suffers from the loss of sight characteristic of Category II. She notes that persons who use solely category II or III statements to describe their relations with their husbands appear clinically depressed. Those who achieve the final level show a new perspective including self and others (Category IV). Espirical Study of Gilligan's Developmental Perspectives: Based on use of the Revised Relationship Self Inventory The Relational/Connected Self and its variants was studied by a research group from Hichigan State University which developed a pencil- and-paper self-report instrument, the Revised Relationship Self 26 Inventory. The first version of this instrument was presented to colleagues in a set of two papers presented several years ago (Reinhart et al., 1985; Pearson et al., 1985). At that time it contained scales measuring the Relational/Connected self in women, as well as the developmental perspectives of: (1) Self Care from Reed, (2) The Primacy of Other Care, and (3) Self and Other Care Chosen Freely. The instrument was developed from responses of a sample of 525 women ranging from ages 21 to 83 who were primarily married. At that time the instrument contained 27 items and items which were developed to measure Gilligan's model of the separate/objective self did not measure consistently any concept meaningful to their sample. Thus, such items were not included in the original version of the RSI. At that time, the ”Self and Other Care Chosen Freely” variable was found to be sore characteristic of separated or divorced women than for married women. Pearson and colleagues concluded that: ”...some women, regardless of marital status, perceive themselves to be in more than one focus area at the same time. Although these data are cross-sectional in nature, one implication is that Gilligan's developmental sequence is not a stage sequence. Second, neither Gilligan's data nor ours tell us if a life transition causes a change in perception of care, or a change in perception of care precipitates a life transition.” (Pearson et al., 1985, p. 3). After the initial work with the RSI, its item pool was expanded and new items were written which might be expected to tap the Separate/Objective orientation. It was then administered to a large sample of more than 1000, about two-thirds of whom were women. Based on inter-item and inter-cluster correlations, and on Cronbach's alpha, the 27 researchers then retained 60 items which clustered in four scales: (1) the Separate/Objective Self (18 items), (2) the Relational/Connected Self (12 items), Primacy of Other Care (14 items), and Self and Other Care Chosen Freely (16 items). The researchers note that: ”The scale Self Care from Weed was dropped from the revised RSI for several reasons. Relatively few items fell into this cluster, and its reliability was below acceptable levels for either men or women. Items expected to make up this scale clustered instead with Separate/Object Self and (to a lesser extent) with Self and Other Care Chosen Freely.” (Strommen et al., 1987, p.2). The internal consistency of this instrument and scale reliabilities for men and women were quite adequate and are reported in the subsequent ”Hethods” Chapter, as this is one of the instruments used in the present research. The RRSI is presented in the Appendix I along with its scales and scoring. The researchers found that ”at all ages, correlations between Separate/Objective Self and Connected/Relational Self are negative and weak: and correlations between Primacy of Other Care and Self and Other Care Chosen Freely are positive but moderate..." (Strommen et al., 1987, pp. 2-3). Two interesting aspects are noted about the revised instrument. First, it now contains a way of measuring the Separate/Objective Self, an mode of relating that is considered to be more characteristic of men than women. Second, it has dropped the scale measuring ”Self Cars from Weed." This was for statistical reasons as results suggested that this earlier scale was not statistically distinct from Separate/Objective Self. However, it is interesting that Attanucci also dropped a mode of measuring this perspective in her research plan. It might be that 28 researchers coming from different ways of measuring Gilligan's developmental ideas were in fact moving away from thinking of this as an initial stage, but were not yet ready to openly contradict Gilligan. The RRSI contains the following means and standard deviations for the entire sample and for women in the age group of 30 to 78 as shown in Table 1. It should be noted that the Relational/Connected Self is the orientation most clearly embraced by women. The researchers report that their results continue to support the concept of the Relational/Connected Self and its importance for women. At all ages they found a negative relationship between the two major orientations, that of Separate/Objective Self and Relational/Connected Self (Strommen et al, 1987, p. 3). In their 1987 presentation of results, they do not comment specifically on the relevancy of Gilligan's developmental perspectives to their work. A recent study connected with the use of the RRSI was that of Blank (1988) in her doctoral dissertation on the relational self in women. Her sample contained over 300 persons, composed mainly, for validation studies, of women attending an on-campus enrichment program. She studied persons who attained a minimum score of 3.5 on the scale measuring Relational/Connected Self, which was liberal, because the mean for the entire sample was about half a scale point higher. Her research raises questions about the meaning of being simultaneously ”low” or ”high” on POC and SOCCF. She raises several possibilities. One is that these foci are so important to women that they have trouble saying clearly how they make distinctions in care for self and other. Another is that the women in her sample may be caught between what might be seen 29 Table 1 Revised Relationship Se1f Inventory: ScaIe means and Standard Deviations ‘ Separate/ Relational/ Primacy of Self and Objective Connected Other Care Other Care Group Self Self Chosen Freely Whole Sample (I I 600) Hean 2.6 4.1 3.2 3.9 SD 0.53 0.51 0.50 0.49 Women Ages 30 - 78 (H . 320) Hean 2.5 4 1 3.2 3 8 SD 0.51 0 56 0.53 0 54 ‘ From unpublished data provided by Ellen Strommen, 1989. 30 as a socially traditional role for women, the POC focus, and what she called a more feminist position, the SOCCF focus. Women may find trouble synthesizing these positions, which might explain the high incidence of depression in women who are high on both scales. She also suggests that the perspective used, focus on care for self or care for both self and other, may be context-specific, or relationship-specific, and thus any attempts to measure a general orientation which guides actions will fail. In her review of the research using the RRSI, she concluded that ”the data suggest that women perceive themselves to be in more than one focus area [developmental perspective] at the same time. This implies that the developmental sequence is not a stage sequence. It is also not clear whether a life transition causes a change in perception of care and/or such a change precipitates a life transition" (Blank, 1988, p. 20). In conclusion, the empirical studies from Gilligan and her students at Harvard do not rule out the possibility of a developmental stage sequence in women comparable to that suggested by Gilligan in 1982, while the work emanating from Hichigan State suggests that these perspectives, while important, do not form a developmental sequence. It would be useful to combine both types of research on a single group of women and to see if the results can clarify the issues. Having reviewed Gilligan's theories of development of the care perspective, and the two sources of empirical studies of the connected self, we now turn to diagnostic issues in object relations theory, and a brief review of projective tests (content and structure of object 31 relations) and non-projective tests which are used to study object relations in general, and mutuality of autonomy in particular. Empirical Study of Development in Object Relations Theory Those diagnostic instruments which exist for describing a subject's inner world of self and object representations represent general ideas unless they are specifically cited as built upon a certain theory. In general, those projective tests that have described ego functions and defensive operations (the Rorschach, the Thematic Aptitude Test [TAT], Human Figure Drawing Test, and Early Hemories Test) have been adapted to the more phenomonological requirements of object relations theory. Host research comes from two main sources, namely, researchers connected with Hartin layman at the University of Michigan and researchers working with Sidney Blatt at Yale. The categories used for review are: (1) projective measures of content of object relations: (2) projective measures of structure of object relations; (3) measure of ego development and object relations; and (4) non- projective measures of object relations. Finally, one particular theoretical model which has been adapted to several specific instruments, is presented--the model and measure of mutuality of autonomy developed by Jeffrey Urist (1973, 1980). Projective Heasures of Content of Object Relations layman (1967) outlined the use of the Rorschach for explaining self and object representations. He also developed The Early Hemories Test (1968) to elicit a more subjective sense of the meaning of the individual's experience of self and other. He conceptualized object 32 representation as internalized images of the self and of others around which the phenomenological world is structured and into which ongoing experience is assimilated. He proposed that the manifest content of dreams, early memories, and Rorschach tests was more than a screen that both expressed and concealed deeper and more significant levels of unconscious meanings. He argued that the manifest content could reflect levels of ego functioning, the capacity for object relations, and the nature of interpersonal strivings. This is based on his argument that the ambiguity of the projective test situations calls forth the individual's expected way to experience'his or her phenomenal world, and that persons' first impressions tell much about the inner world of self and object representations and the quality of relationships between them. layman's methodology was applied by different students of his at the University of Hichigan to autobiographical material, dream interpretation, and to new ways of scoring the TAT and Rorschach (Krohn, 1972; Urist, 1973: Ryan, 1973). Projective Heasures of Structure of Object Relations The other main academic source of work on assessing object relations in a projective manner came from colleagues of Sidney Blatt at Yale, with the exception of a few unrelated studies. Coonerty (1986). for example, developed a scale for measuring Separation-Individuation Themes on the Rorschach based on Hahler's developmental model. Her work showed impressive ability to distinguish borderline from schizophrenic responses. The major source of work on the structural dimension of the object representation comes from Blatt and colleagues. Their attention 33 has focused, for example, on the establishment of ego boundaries between self and nonself, and between fantasy and reality (outside and inside). Thus, they have researched boundary disturbances in schizophrenics and in neurotics. They have also studied capacities for reality testing, quality of interpersonal relations, and nature of object representations. Blatt, Brenneis, Schimek, and Glick (1976) developed a method to rate Rorschach responses based on their interest in boundary disturbances. This is called The Concept of the Object Scale. The system calls for scoring human responses in terms of differentiation, articulation, and integration. These, in turn, are each rated as in a continuum based on developmental levels. Although the work of Blatt and colleagues could be described in more detail, because of the initial conceptual interest differences between Blatt and Gilligan (boundary disturbances versus continuity in relationships), this will not be done at present. Heasures ofAEgp Development and Object Relations Some would consider the work of Jane Loevinger with the Sentence Completion Test (SCT) in the realm of assessment of object relations. Her work spans that of ego psychology and object relations theory (which is also true of the work coming from the University of Hichigan and from Yale). Since she emphasizes the conscious presentation of self as seen in the individual's response to the semi-structured form of her sentence stems, her work would seem to be closer to the phenomenological interests of Gilligan. Nevertheless, although the recent doctoral dissertation by Rogers (1987) bridges Gilligan's theories with Loevinger's, the specific method used by Rogers would be inappropriate 34 to the present study. Specifically, Rogers developed a method to score a response on the SCT as in either the justice or care orientation of Gilligan. However, her analysis involved counting the number of responses in either justice or care orientations, rather than an analysis of the greater or lesser developmental maturity of the response. Thus, this work, while providing an important linkage of Gilligan's work to psychoanalytic theory, does not address the questions of the present research. Hon-Projective Heasures of Object Relations Horking from Bellak and colleagues' description of levels of object relations functioning (1973), Bell, Hetcalf, and Ryan (1979, 1980) developed a true-false self-report questionnaire composed of items adapted from patients experience of their experience of relationships and their characteristic patterns of relating." (p. 734). After extensive development, a factor analysis produced four factors in the final scale, which are: alienation, insecure attachment, egocentricity, and social incompetence They used their instrument to differentiate between borderlines (who show relatively high scores on alienation, insecure attachment, and social incompetence) and other types of personality disorders. They report that the instrument has been used to document improvement in interpersonal relations before and after therapy. Scores on this instrument show a ”strong linear relationship...between the severity of eating disorder and degree of object relations pathology in bulimic college women” (Bell, Hetcalf, and Ryan, 1986, p. 734). 35 The Comprehensive Object Relations Profile (CORP) developed by Burke and colleagues (1986) is a semi-structured projective test that asks a subject to answer specific questions about the relationships depicted in six specific stories. This instrument measures three dimensions: object constancy, object integration, and empathy. Burke and colleagues report that the empathy dimensions are the most sensitive to group differences. This instrument appears to show considerable promise in general, but as an instrument to be used in comparison with Gilligan's work, it would be less appropriate than a mode of inquiry which gave individuals the opportunity to speak from their own experiences. Heasures of Hutuality of Autonomy Although important aspects of object relations were measured by these already-mentioned projective and non-projective tests, for our purposes it is only that aspect of object relations,the Hutuality of Autonomy, developed by Urist (1973), which is relevant to the present research plan. Thus it will be summarized here. Urist developed a scale based mainly on the conceptual framework of Rernberg (1966). He clearly thinks of object relations as a multi-dimensional capacity, but he devised a separate scale to measure one aspect of it, mutuality of autonomy. This seven point ordinal scale is described as follows (Urist, 1980, pp. 830-831). 1. Relationships are characterized by a clear sense of the autonomy of each of the partners, where the overall tone is one of mutual respect. Interactions are portrayed as mutually enhancing: the autonomy of the partner is appreciated and valued. 36 2. Relationships reflect a sense of individuals mutually interacting, but the emphasis is on a goal rather than on the relationship itself. People are clearly perceived as separate. 3. There is no serious disparagement of mutuality or of the autonomy of individuals, but there is no consistent, enduring sense of engagement. Relatedness fluctuates back and forth with more functionally defined, need-satisfying level of relatedness. 4. Relationships reflect an underlying functional orientation, with the emphasis on the function rather than on the person. Helping others is seen as a justification for making others provide a function for the self. 5. People are portrayed as getting along with each other only insofar as they are alike. This implies more than shared appreciations but the tone is of people needing to act as reflections of the other. 6. Relationships are characterized by an overriding absence of any real sense of people as active, autonomous agents in relations with each other. A predominant theme is one of coercion, manipulation, or control. Helping others is experienced as a justification to control them. 7. Relations between people are portrayed in terms of malevolent, overpowering envelopment. Autonomy is deteriorated to the level of an experience or fear of incorporation. The Hutuality of Autonomy Scale as measured on the Rorschach has been correlated with independent measures of the same dimension by written autobiographies, and behavioral ratings of ward staff (Urist 1973). Correlations for highest mutuality scores ranged from .09 to 37 .40; correlations for lowest mutuality scores ranged from .09 to .63. (Ryan, 1973, p. 100) Looking at a combined sample of inpatients and outpatients, Urist found that scores of object relations on the Rorschach tended to be lower; he said that ”particularly in the case of the healthier outpatients, the Rorschach tended not to be the most representative measure of the integrity factor, since it was prone to give a somewhat sicker picture of object relations. The problem for measurement of relatively healthy object relations as one might expect for the outpatients was that his scale has little discrimination on the healthy end in its present form. Another researcher who did his doctoral work at the University of Hichigan, Edward Ryan, developed the Quality of Object Relations in Early Hemories Scale with a broad range for measuring object relations including Hutuality of Autonomy. The scale may be described as a measure of the quality of the object relationships through an analysis of the person's early memories. This is a 20-point scale, with five points within each of the four categories. The low point of the scale describes a quality of object relations in which the object world is unreal, nightmarish. The approximate mid-point involves a depiction of other persons in the life space as important, but their characters are defined almost solely as need-satisfying or need-frustrating objects or beings. The high-point involves multiple, bilateral role relationships with a variety of different people, with a clear articulation of the distinct individuality of the self and others in these relationships. There is a sense of belonging to a community of separate individuals, and a spirit of positive interaction: even negative aspects are 38 presented in a wider context of mutual trust, acceptance, regard or affection. Please refer to Appendix B for the instrument, Quality of Object Relations Scale for Early lemories. This scale has been used less than Urist's scale, but it has several advantages for the present research. First, it is an advantage to measure mutuality of autonomy by scoring early memories as if they were projective materials. This means that the method is more simple than administering the Rorschach. Second, the Rorschach measure of mutuality of autonomy is biased in favor of a lower clinical range. Third, the greater scale range of the Ryan scale makes it more applicable to less psychiatrically disturbed populations. In the pilot stages of scale development, layman and Ryan independently rank ordered the earliest memories of 28 randomly selected patients for "quality of object relations." The reliability coefficient was found to be .86 (p < .005). lore specific information about the Ryan scale will be presented in the "lethods" chapter. This completes the review of empirical measures for Gilligan's theories and for object relations theories. The instruments of choice for women's perspective on self and other, are the Attanucci research protocol and the Revised Relationship Self Inventory. There is one additional consideration about the Attanucci measure, and that is that her questions were embedded in a larger study of women in relationships and that she coded responses from these questions as well as from her own. This leaves a vacuum in the present study: what questions should be used to fill in the gaps using her method? One answer suggested itself: namely, to devise questions to reflect issues and concerns 39 about the self that have been identified by Colarusso and lemiroff (1981) and Rernberg (1975/1985) as narcissistic issues of the middle years of life. The concerns they raise can be summarized under the following categories: (1) Significant transitions and re-assessment of relationships: (2) Concern about authenticity of the self; (3) Experience of change in one's body; (4) View of oneself around the possibility of further development; (5) Presence or absence of [female] mentors: (6) Re-assessment of the meaning of time; and (7) Coping with losses and death. The object relations measure of lutuality of Autonomy which appears most appropriate for this purpose is the Quality of Object Relations in Early lemories Scale. We turn next to the specific research question, and subsequently, the research method. Research Question Tb what extent are achievements of "SeIf and Other Care Chosen Freely," the epitome of development according to Gi11igan, and flutuality of Autonomy," the epitome of development in one model of object relations, related, in middle-aged women? Basically this question assumes that healthy self care in middle- aged women is addressed from a viewpoint of the context of relationships. Two major and heretofore not related theoretical viewpoints converge in addressing the question: the developmental- contextual work of Carol Gilligan, and an object relations perspective which describes mutuality of autonomy. The specific plan is to study middle-aged women who are engaged in some activity which could be defined in a relatively clear fashion as 40 doing something for the self. For our purposes, this is defined by engaging in academic studies, and engaging in regular physical exercise. The question of the present research concerns the possibility for balance: to what extent can these women care for self and at the same time, care for the other in a mutually enhancing manner? A group of 28 women will be studied in depth, providing an opportunity to look at answers to this question in both a quantitative and qualitative manner. As an adjunct to the question, some women who score either relatively high or low on the measures of interest will be described in some further clinical detail. Finally, some general concerns will be addressed which are considered salient for middle-aged women from an object relations framework, based on narcissistic themes of the middle years. lore specifics about the research method are described subsequently. RESEARCH NETHOD Overview liddle-aged women who can be defined as involved in self care in what seems to be a responsible, healthy manner, were studied, and the meaning of their behaviors in the context of the women's present conscious experience of self and of recollections of past development and relationships were investigated. The research is composed of individual interviews with 28 middle-aged women. Research subjects were identified through advertisements and offered a modest stipend for participation. The selection criteria targeted women between the ages of 35 and 55 who have ever had children, and who are now either students or exercisers. Persons were selected through a telephone screen; volunteers subsequently were interviewed twice. The second interview was spaced two weeks to one month from the first. Questions about the conscious experience of the self and of relationships, in the past and in the present, were asked by the interviewer following a general protocol administered in an open-ended clinical fashion. Participants completed a general information form and the Brief Symptom Inventory (BSI) at the time of the first interview, and completed the Revised Relational Self Inventory (RRSI) at the second interview. Interviews were audiotaped and information pertaining to the research study was transcribed. All identifying information was removed or changed in transcripts and written materials. There were two coded 41 42 transcripts per case. One had the Early lemories responses: the second had the responses to the Attanucci protocol plus some additional general interview questions. The additional interview questions were developed to reflect those issues and concerns about the self that were identified by Colarosso and lemiroff (1981) and Rernberg (1975/1985) as narcissistic concerns of the middle years of life. The instruments for studying the women's ”care for self and other chosen freely” are: the Attanucci research question protocol and the Revised Relational Self Inventory. ”lutuality of autonomy in object relations” is investigated by the Quality of Object Relations in Early lemories Scale. The research question comparing developmental level in Gilligan's theory and in object relations theory is addressed by examining the relationship between membership in the categories in the Attanucci model, in perspectives on the RRSI, and membership in levels in the Ryan Quality of Object Relations Scale. Pearson Product loment Correlations were used, as well as cluster analyses. Case material provided some additional descriptive information of the women in this pilot research. Some themes common to the women's self descriptions were also investigated. lore specific information about this research plan is provided subsequently. Subjects Selection of subjects Twenty- eight middle -aged women were selected from two categories: women who are intellectually doing something for themselves, that is, women who are going back to school: and women who 43 are physically doing something for themselves, that is, women who are doing regular physical exercise. The important issue was that women were doing something for themselves, not which kind of activity they engaged in. If women were both students and exercisers, they were accepted as subjects. Women chosen for this study were in the age range of 35- 55 years of age. lature women were selected since it seems more likely that they would more consciously consider they are doing something for themselves when they go back to school or exercise, since both of these activities are less common in this age group than among younger women. Students were identified through their participation in community college or university classes. Exercisers were located through their participation in either organized physical exercise programs or personal fitness activities, either using local community services programs, including health clubs, or utilizing the physical fitness facilities at their college or university. Persons were asked to participate in a study of normal, middle- aged women: and they were told that the study involves two individual interviews, totaling between two and four hours. Individuals were offered a small financial incentive to participate. Fliers were distributed asking for volunteers and also the study was advertised in a campus newspaper (see Appendix C). The researcher used a telephone screen to take phone numbers and demographic information and to rule out persons who did not meet study criteria. Persons who were suitable were given information about amount of time entailed and amount of personal revelation required. The personal nature of interviews was explicitly stressed so that callers 44 could make informed choice about further participation. The telephone screening protocol is presented in Appendix H. Demographic Description Twenty-eight women were selected into the study from the 42 persons who responded either to posted notices or to a campus newspaper advertisement. The average age of these women was in the forties; ranging between 36 and 45 years of age. Twelve of the 28 women were both students and exercisers. Table 2 provides a summary of demographic information about the sample. It should be also noted that four of the 28 women were later excluded from some analyses due to their high scores on one or more subscales of the Brief Symptom Inventory. The women who were excluded included persons who were both students and exercisers, and did not appear demographically unusual in terms of education or occupation. The demographic description of the restricted sample of 24 women (excluding the four women) appears in Appendix D. Instruments Brief Sygptom Inventory (BSI). For the purpose of further identifying the subjects of this study as ”normal,” persons were asked to fill out a copy of the Brief Symptom Inventory. This is an abbreviated form of the Symptom Checklist 90 (Hale, Cochran, 8 Hedgepeth, 1984). A pencil-and-paper instrument, it takes about 10 minutes to complete. It provides information on nine scale scores indicating some aspect of psychological distress. It was standardized on a sample of middle-aged adults, and so is appropriate for use in the present study. Although the instrument is not provided 45 Table 2 Demographic Description of the Simple (HHZS) lean Range Age (years) 43 36-55 lumber of years married 19 9-35 lumber Percent Student 3 ll Exerciser 13 46 Both 12 43 Currently married 1st marriage 21 75 2nd marriage 3 11 Currently divorced or separated 4 14 Education High school or less 1 4 Some college 8 29 College graduate 4 14 Some graduate study 6 21 Graduate or professional degree 9 32 Occupation. Homemaker 5 18 Education, research 8 29 lanagement 8 29 Sales 1 4 Secretarial 2 7 Other health and human services 2 7 Student 14 50 Religion Protestant 12 43 Jewish 4 14 Roman Catholic 4 14 Other 2 7 lone 6 21 lbte. All women were married at least once. ‘Based on most recent employment, or on any employment in the past 10 years. Some persons indicate more than one occupation, thus percents total greater than 1008. 46 Table 2 (cont'd.) lumber of male childrena Age of male children (years) lumber of female children8 Age of female children (years) Husband's (former husband's) education High school or less Some college College graduate Some graduate study Graduate or professional degree Husband's (former husband's) occupation) Education, research lanagerial, administrative Health and human services Skilled trade, factory work Electronics 8 computer Law or criminal justice Student lean Range 2 0-4 23 3-34 1 0-5 21 3-35 lumber Percent 4 l4 4 14 4 14 3 11 13 46 6 21 8 29 2 7 4 14 2 7 2 7 4 14 iIncludes adopted and step-children. Although a person may have no children of one sex, they may have one or more of the other. 47 Table 2 (cont'd.) lumber Percent Perceived health status of subject Good 26 93 Average 2 7 Poor 0 0 Perceived health of mother Good 8 29 Average 6 21 Poor 5 18 (not applicable- deceased) 9 32 Perceived health of father Good 10 36 Average 4 14 Poor 1 4 (not applicable- deceased) 13 46 Perceived health of husband/former husband Good 20 71 Average 7 25 Poor 1 4 Perceived health of children (by family)a Good 24 86 Average 4 14 Poor 0 0 'Includes adopted and step-children. lost mothers globally rated their child(ren)'s health in one category. 48 Table 2 (cont'd.) lumber Percent Exercise frequency 3x/week or more 20 71 1 - 2x/week 5 18 lo regular exercise 1 4 (did not answer) 2 7 Type of exercise9 (of those who do exercise) Aerobic type 22 79 Strengthening 5 18 Toning ll 39 Other (such as golf, bowling) 3 11 Reasons for exerciseb Enjoyment of physical movement 19 68 Health (weight control, stress reduction, etc.) 23 82 Social reasons 14 50 Competition 4 14 Other responses 2 7 (Did not answer) 3 11 'Persons may indicate more than one kind of exercise done on a regular basis, so percents add up to more than 100 A. bPersons checked all that applied, so percents add up to more than 1008. 49 here, because of copyright restrictions, it is available, along with directions for scoring, through its author (Derogatis, 1977). Clearly any person could volunteer for this project for many reasons; thus, the mere fact that a woman exercises now or is a student now does not preclude her from feeling significant psychological distress. But such stressed persons were assumed, before the study, as unlikely to achieve higher levels on either the Gilligan measures (the Attanucci categories or the RRSI) or the object relations measure. Therefore, the BSI was used as a conservative measure of the likely ”normality,” at least compared to the norms for other middle-aged persons, of the women in the sample. Persons who scored three standard deviations or more above the standardized norms for persons aged 46 (Hale, Cochran, & Hedgepeth, 1984) on at least one subscale of the BSI were considered psychologically distressed as compared to the other women. The process identified four of the twenty-eight subjects in this category. Subsequent analyses were then conducted with and without the data from these four women, and were so labelled in the results. Information on the results of all subjects on the BSI compared to standardized norms is given in Table 3. Attanucci_interview geasures of "Self ig_Relationyto Others" Attanucci's open-ended questions were designed to elicit information about the person's conscious view of her self and of current and past relationships with important persons in her life. For the purposes of this study, the same questions were asked of middle-aged women participants. Attanucci's questions were included in a longer interview, and she coded relationship statements that were elicited in 50 Table 3 comparison of Scores of Full and Restricted samples to Nbrms for Persons with an Average Age of 46 on the BSI Full sample Restricted sample lormsa (l I 28) (l I 24) (l I 565) lean SD lean SD lean SD Subscaleb SOlA .21 .23 .20 .24 .43 .47 OB-COlP 1.08 .84 .85 .58 .44 .49 IlT SElS .68 .68 .50 .53 .35 .43 DEPRES .51 .58 .36 .33 .46 .52 Alx .60 .58 .45 .43 .37 .43 HOSTIL .64 .58 .53 .54 .33 .42 PHOB Alx .17 .36 .06 .22 .19 .37 PARAl ID .48 .42 .38 .34 .34 .46 PSYCHOT .44 .59 .23 .29 .15 .25 Grand Total (GSI) .53 .39 .40 .24 - - ‘Based on norms provided by Derogatis, 1977, as presented in Table 1 of Hale, Cochran, 8 Hedgepeth (1984, p. 321). ”Where: SOlA I Somatization OB-COlP I Obsessive-compulsive IlT SElS I Interpersonal sensitivity DEPRES I Depression Alx I anxiety HOSTIL I Hostility PHOB Alx I Phobic anxiety PARAl ID II Paranoid ideation PSYCHOT I Psychoticism 51 answer to other questions as well as her own. The same procedure was followed in this research, except that the "other questions” were the ones that were designed by this researcher about issues in the middle years of life. The Attanucci questions are presented in Appendix A. The coding procedure is presented next. Coding Procedure. All initial interviews were transcribed and coded using Attanucci's coding manual which provided the basis for coding open-ended questions into relationship categories, I to IV, which she argues reflect the developmental perspectives of Gilligan (1982). Transcriptions contained the subjects' responses to the Attanucci questions and the other questions as adapted for middle-aged women. Identifying information was changed or deleted. Two undergraduate research assistants were trained in coding according to the Attanucci manual; they completed this task, as described next, in a six month period, along with the researcher. One assistant received academic credit for two quarter terms for her time; the other assistant devoted similar amounts of time, but did not take it for credit as she was graduating that term and did not need it for graduation. Approximately twenty-five hours of discussion were devoted to the use of the manual prior to its application to the first pilot subject. The manual appears in Appendix A. It should be noted that the researcher attempted to contact Dr. Attanucci for further clarifications, but Dr. Attanucci was unable to provide more time than one phone conversation. Thus all procedures are based on what was spelled out in the Attanucci dissertation supplemented by what seemed reasonable. 52 The researcher and the research assistants read the protocol of a pilot subject (whose data was not included in subsequent analyses). First the transcription was read through as a whole. lext it was read looking for relationship statements which dealt with husband, mother, child, and ”other." These are the same categories which Attanucci used. However, since the interview was a clinical one, the relationship statements were not all together in one place, and it was necessary to decide what was the beginning and end of a codable unit. Each person tried to define codable units, and this process was discussed, with a final decision on codable units reached by consensus. Clarifications or elaborations were added to the original coding manual. These may be seen in Appendix A, along with the original manual. lext, each person coded the units according to the decision rules given by Attanucci in her manual. Again, there was a general discussion, and again the final decision on appropriate coding was reached by consensus. Finally, results were transferred to a data summary sheet. At this point the researcher divided subsequent interviews into codable units: the research assistants confined themselves to checking her accuracy, and did the actual coding in the agreed-upon method. All three persons coded the next interview. This was, in fact, the first one actually used in subsequent data analysis. Final decisions on coding were made after discussion and gaining consensus. Then each assistant was assigned one interview to code on her own: the researcher also coded it. The separate codings were discussed, and disagreements were resolved by consensus. Percent agreement was 53 calculated for two pilot interviews for the research assistants vs. the researcher. The basis for calculating agreement arose naturally from Attanucci's work. The cluster analysis she planned for her data turned simply on whether any statement of a Category I through IV occurred at all in describing a relationship (and not the frequency of its occurrence). Table 4 shows the person-by-category-of-relationships scheme according to which data were coded. Reliabilities were calculated on agreement between coders in locating subjects' responses in these categories. The data reduction, then, took place in two steps. First, each research assistant and the researcher independently coded the codable units (which varied, of course, in each subject's interview). Then they recorded their own responses. Where there were disagreements, they discussed the unit and came to a consensus. In some cases this involved a double-coding, such as coding a unit as both a Category II and a Category III example. This could not be avoided because often descriptions are quite complex. In addition, at times two persons are referred to in the same or almost the same codable unit. For example, a woman might refer to an incident which involved relating to both her husband and her child. In such cases, the codable unit is presented once under all the units referring to the husband, and repeated again under the listing of all units referring to the child. Attanucci used a similar method for coding. The first coder achieved a 75 percent agreement overall with the researcher; the second, an 81 percent agreement overall. This seemed 54 adequate and research assistants were randomly assigned the subsequent interviews to code. One exception to the randomness was that the student graduating at the end of Spring term, pressed for time, had to be relieved of one coding. The other research assistant coded that as well as her own set. The research assistants did not code all transcripts independently of the researcher. This procedure, which would result in a "blind" analysis, was not done because the process of discussion appeared invaluable. Reliability was calculated differently from Attanucci's method. Attanucci had used Cohen's Kappa, but this process is not appropriate for the type of data obtained. Cohen's Kappa should be used when units are assumed to be independent (Cohen, 1960, p. 38), which is an untenable assumption in the present research because the comments about any one person are not independent, nor is it reasonable to assume that relationships are independent of each other. In addition, the interrelationship of units to be coded made for different assumptions about the meaning of chance agreement, and also of perfect agreement, than the assumptions made by Cohen. Data obtained. Thus the final data set for each subject was a set of 16 scores, as previously shown in Table 4. For analyses comparing Gilligan's highest level of development as measured by Attanucci's model with its operational measure in the Revised Relational Self Inventory and with the Quality of Object Relations in Early lemories, a summary score for Category IV was used. This variable was created by counting the total number of relationship areas in which there was the presence of a Category IV response. Thus this variable could range from 0 to 4. 55 Table 4 Schema for Data Obtained from the Coding of Attanucci categories for Self in Relationship for Research Women Category of relationships described by research women I II III IV Person in relationship Husband a a a a lother a a a a Child a a a a Other a a a a Rbte. Where: a I present (1) or absent (0) And: Category I I Self instrumental to other, other instrumental to self Category II ISelf instrumental to other, others "in their own terms" Category III I”Self in self's terms," other instrumental to self Category IV ISelf in self's terms, others in their own terms 56 Prior to the final step, percent agreement for each item in each category was calculated. For the final 16 scores, percent agreement was calculated on presence or absence of the four categories within the four types of relationship. Further information on the coding process and the calculation of reliability overall appears in Appendix A. That appendix also contains the enlarged coding and training manual based on conventions used to decide ambiguous coding questions. The Revised Relational Self Inventory (RRSI) This instrument is used to assess Gilligan's model. The RRSI was designed to measure concepts such as the connected or separate selves described by Gilligan (1982) and her colleagues, as well as variants of the connected self. A pencil-and-paper instrument, it consists of 60 items to which the respondent can agree or disagree on a 5-point Likert scale (”like me” to ”not at all like me;" see Appendix I). The RRSI can be scored to obtain scale scores on Separate/Objective Self, Relational/Connected Self; Primacy of Other Care; and Self and Other Care Chosen Freely. The Separate/Objective Self describes a ”justice" orientation in which a self is defined through individual achievement and objective reciprocity. The Relational/Connected Self orientation is one in which the self is defined through connection with others, and concerns of ”care” are central. Reliabilities and scale intercorrelations for men and women are presented in Table 5. Item-scale total correlations are presented in Appendix I. The instrument appears to have an adequate face validity. 57 The most important RRSI variable, for this research, is the measurement of ”Self and Other Care Chosen Freely," a variant of the Connected Self, and a construct descriptively similar to what was measured by Attanucci more clinically in her coding for Category IV, "self in self's terms and other in other's terms.” Within the overall framework of a relational or connected self, there is a particular way of viewing the world which can be described by another variable, ”primacy of other care.” This appears to be somewhat comparable to Attanucci's Category II, while the view of the Separate/Objective Self appears to approximate Attanucci's Category III. This instrument is self-administered, and took about 20 minutes to complete. It was given at the end of the second interview. Data obtained. The RRSI produces 4 scale scores which are for: Separate/Objective Self (SO), Relational/Connected Self(RI) , Primacy of Other Care (POC), and Self and Other Care Chosen Freely (SOCCF). Theéguality of Object Relations in Early lemories Scale Quality of Object Relations (including lutuality of Autonomy) is scored from early memories using the Ryan Scale (Ryan, 1973). Each point in the 20-point scale represents a salient quality of object relations. The 20 points are divided into four main categories (each containing five graded levels). The complete scale and directions for scoring are given in Appendix B. The following abbreviated description of scoring categories is given in Ryan and Bell (1984, p. 211): A. (Levels 1-5): Prototypes of severe disturbances in object relatedness, such as occur in psychotic or borderline states. B. (Levels 6-10): Prototypes of disturbances characteristically more nearly representing depressed or pathologically narcissistic 58 Table 5 Revised Relationship Self Inventory (RRSI) Reliabilities and Scale Intercorrelations ‘ Reliabilities (alpha) Separate/ Connected/ Primacy Self and Objective Relational of Other Other Chosen Self Self Care Freely Women .77 .76 .68 .78 (l I 930) len (l I 228) .85 .76 .67 .77 Scale Intercorrelations b Separate/ Connected/ Primacy Self and Objective Relational of Other Other Chosen Self Self Care Freely Separate/ 1.00 c .23 .09 .40 Obj. Self Con./Rel. -.33 1.00 .56 .52 Self Primacy of -.01 .73 1.00 .10 0th. Care Self & .26 .58 .19 1.00 Oth. Chos. Freely ' From Strommen et al., 1987, Tables 2 and 3. b Intercorrelations for women above the diagonal: for men, below the diagonal. c Corrected for attenuation. 59 states, such that no vestige of a sense of good objects appears in the subject's inner world. C. (Levels 11 - 15): Prototypes of neurotic-level disturbances in relatedness, such that people do appear as important foci in these memories, but relationships are childishly conceived, self- centered, and limited. D. (Levels 16-20). Prototypes of more nearly normal, mutual interactions. According to Ryan and Bell, the continuum depicts the emotional quality of the representation of the self and others, the integrity of these representations, and the quality of the interaction between self and others. ”At each level, qualities of object relations in the OR scale are consistent with the dimensions described by Urist (1980) in his review of object-relations measurement instruments: richness and complexity of representations, quality of differentiation and individuation among them, and degree of mutual respect that exists in the depiction of the self-object world" (Bell and Ryan, 1984, p. 210). In the pilot stages of scale development, layman and Ryan independently rank-ordered the earliest memories of 28 randomly selected patients for ”Quality of Object Relations.” The reliability coefficient was found to be .86 (p < .005). Subsequent ratings of the scale have shown adequate interrater reliability. In a recent study, raters trained on practice sets of memories and then scored (blind) a sample of 80 transcripts. Rater reliability among three raters using interclass r was .86 (Ryan and Bell, 1984, p. 211). In the present study, the researcher and another advanced clinical psychology graduate student trained themselves in the rating of the memories by first obtaining the most recent version of the ElS manual from Dr. Ryan. His advice was to train on some early memories obtained through our own resources, and to arrive at conventions and 60 clarifications through this process, as he was unable to provide sets of memories for training purposes. In order to achieve reliability, 65 earliest memories in written form were obtained as pilot data from a larger study being conducted by Drs. Strommen, Donelson, et al., of which this researcher was a participant member. These memories were divided into three sets in which the memories appeared to have roughly the same range of potential scores. This first division was made by the researcher based on a quick assessment of scores which used roughly the criteria of the Ryan lanual. These three sets were scored independently, and, when scoring differed, results were discussed, with final scores for individuals achieved by consensus. The memories were then re-arranged in a single set of 65 written memories in order of the scores they had received (in a scale from 1 to 20, with highest possible score a 20), as suggested by Ryan in his scoring manual. These were then re-scored. There were still differences in opinion from the original scores, and between raters, and again final scores were arrived at by consensus. The results of this second scoring were graphed, and visual inspection of the graph of number of persons receiving any single score reveals what appears to be a normal curve, with the mode at 11. This would be scoring in the ”neurotic” range. The earliest memories from the pilot subject (one woman) in the present research were discussed and then coded. There was some change in length of the protocol, due to the fact that these memories were elicited in an interview and transcribed from an audiotape. Consequently, this and all other earliest memories were longer than 61 those coded for the pilot, and more complex. In addition, there were three "earliest memories” elicited in the actual research: earliest memory, then separately ”earliest memory of mother" and ”earliest memory of father.” When an individual had referred to mother and/or father in the ”earliest memory,” she was asked for another specific ”earliest memory” for that parent. Four additional subjects' sets of early memories were coded next. These were the four subjects whose Brief Symptom Inventory Scores led to their being labelled as psychologically distressed, and not likely to produce ”normal” protocols in the present research. Earliest lemory, Earliest lemory of lother, and Earliest lemory of Father were all coded and reliabilities of the average of two raters were calculated on this small sample. They ranged from .82 to .97, well within the range of acceptability. The earliest memories were then arranged for the 24 remaining research subjects in random order and scored in three sets of nearly equal size. In each case a record was kept of the initial score of each rater, as well as a consensus score, when that was necessary. The reliability for Earliest lemory was .75, for Earliest lemory of lother was .48, and Earliest lemory of Father, .67. Due to the rather low reliability of the Earliest lemory of lother, subsequent analyses used the consensus score rather than the average score of two raters. Data obtained from Early memorie . Coding resulted in three scores (1 through 20) which indicate Quality of Object Relations for: Earliest lemory, Earliest lemory of lother, and Earliest lemory of Father. Higher scores indicate higher levels of quality of object relations. 62 Case laterial Case material is presented for women who scored either relatively high or low compared to other women in the sample on both the Attanucci category IV measure and higher levels of object relations, compared to other women in this sample. Basically this is to provide some descriptive, clinical information about the kind of person who scores highly on both measures; it may assist in descriptively evaluating the hypothesis that Category IV and lutuality of Autonomy in Object Relations might refer to similar phenomena. Data obtained The data were clinical vignettes which were sought to clarify the possible nature of the relationship (or lack thereof) between constructs coming from Gilligan's perspective and constructs in object relations theory. Content Themes about Issues in the liddle Years of Life Semi-structured questions about life history and significant transitions were included in the interviews, as a qualitative adjunct to the main research question. The researcher developed these questions to reflect narcissistic issues of the middle years of life as defined by Rernberg (1975/1985) and Colarusso 8 lemiroff (1981). The questions were designed to elicit concerns about: authenticity in the nature of the adult self, reassessment of the meaning of important relationships: reassessment of the meaning of time; attempts to cope with losses and 63 death; concern with shift in time perspective; and facing the limits of one's own creativity. Data obtained Each woman's transcript was reviewed for the presence or absence of reference to the hypothesized themes for narcissistic issues in the middle years of life. Where an issue was clearly addressed, the woman's own words are presented to describe the issue as she sees it. Process Recruitment of Subjects Women responded to the posted notices and to an advertisement in a campus newspaper; and they were given a brief explanation of the purpose of the study and what would be expected of them, as indicated in the telephone screen protocol given in Appendix H. At the beginning of the study, some subjects were accepted immediately, based on the phone screen, and their first interview scheduled. Five women scheduled for an initial interview failed to show up or cancelled. It should be noted that every woman who came to the first interview showed up for the second. On the other hand, some women who appeared to be suitable were not scheduled for interviews. These were women who called the researcher rather late in the process and after most women had already been interviewed. At this time these women were told that their names would be kept on file, and if additional funds were procured, they would be called for an interview. (The researcher was unsuccessful in obtaining additional funds, and so some women were not accepted into the study who appeared quite appropriate.) 64 Two women were not accepted into the study because they seemed inappropriate. One woman described herself as literally as a ”bag lady,” going around a university campus and taking things out of trash cans. She described this activity as exercise, but this activity was not accepable for the purposes of this study. A second woman was not accepted because she was the sister of another subject. Sometimes one subject would suggest to a friend that she volunteer for this project. This was not a cause to eliminate a subject, but it did seem that sisters would share too much of the same type of life story, and so this woman was not accepted. Contact 1 The first interview began with explaining consent forms (Appendix G), obtaining signatures, and the completion of a general information form (Appendix E) and the Brief Symptom Inventory. All interviews were conducted by the researcher. The interview started with broad questions designed to help the subject introduce herself to the interviewer. It proceeded to other questions designed to help her present her conscious views of her self and of her relationships - the "Attanucci” questions - about relationship to husband/former husband, mother, child or children, and ”others.” There were certain questions which ideally would all be asked in the first interview, but the interviewer proceeded in a clinical manner, which meant that if the subject showed considerable interest in pursuing a certain question in detail, the interview then went in the direction of the subject's interest. The ideal format for the interview is given in Appendix A, Interview I. At the end of the first interview, women were thanked and the second interview was scheduled. 65 Contact 2 The second interview began by requesting the woman to bring up any issues that seemed important from last meeting. If the woman had no issue to raise, then the interview proceeded. The second interviews were more free-flowing, with the content to some extent dictated by any material that had been omitted in the first interview. Women were also asked to give their early memories in the manner described by Ryan. The ideal form for a second interview is given in the Appendix A as Interview II. The RRSI (Appendix I) was administered at this time. The subject was thanked at the end of the second interview, and informed how she may receive general results of this study if she desires them. The subject was then paid a ten dollar stipend. The researcher administered and audiotaped both interviews. Data Analysis The research question is: To what extent are achievements of the epitome of development according to Gilligan, and the epitome of development in one model of object relations theory, related, in middle- aged- women? The research hypothesis is that they are positively related and that this relationship will hold true both for the measure of Gilligan's highest developmental level in the Attanucci measures of ”Self in Relationships" and in the measure in the Revised Relational Self Inventory. lore specifically, the hypotheses are: 1. ”Self in self's terms and other in other's terms" will be positively associated with "lutuality of Autonomy," an aspect of quality of object relations. __———* -- mm”- 66 2. ”Self and Other Care Chosen Freely” will be positively associated with ”lutuality of Autonomy," an aspect of quality of object relations. To examine the research question comparing the highest developmental level in Gilligan's theory and in object relations theory, two methods were used. The first was examination of Pearson Product- loment correlations which were computed between assignment to Category IV (and the other categories) of the Attanucci measure, of choice of ”Self and Other Care Chosen Freely” in the Revised Relationship Self Inventory as self-descriptive, and of membership in higher levels of the Quality of Object Relations in Early lemories Scale. The second method was examination of cluster analyses for women's results on these variables. lo further explanation is needed for the correlation procedure, but some comments on the process of cluster analysis appear germane to the present research. Attanucci's analysis involved an investigation of category scores through a cluster analysis, and she came up with clusters which supported Gilligan's developmental hypothesis. For each case (person) there was a vector of single scores to represent presence (1) or absence (0) of statements in each of the four categories (I to IV) for each relationship (husband, children, own mother, and others). She reported using Ward's method of cluster analysis. This method begins with each case as a cluster and produces a hierarchical tree structure building from each subject as a separate unit to the top where the sample is one cluster. The purpose of cluster analysis is to identify a set of characteristics that can significantly differentiate between the groups. The procedure begins with computing a similarity matrix. Then the 67 matrix is searched for the most similar pair. These are merged to form a cluster. The matrix is searched again for the next closest pair. This process is followed until all entities are in one cluster. There are several ways to perform a hierarchical cluster analysis. The two preferred methods suggested by lilligan and Cooper (1987, p. 351), are Ward's method and the Group Average method. Both were used in the present study. Attanucci used Ward's method only and she reported her results by means of a dendrogram. Visual inspection of the dendrogram provided the initial support for three groups of women: women mainly characterized by category I responses (Self instrumental to others; others instrumental to self); women categorized by both use of category II (Self instrumental to others, others ”in their own terms") and III responses (”Self in self's terms,” other instrumental to self), and women mainly characterized by category IV responses (Self in self's terms, others in their own terms). She further investigated differences among these groups by means of analysis of variance. Hierarchical cluster analytic techniques are frequently used to test developmental hypotheses. However, the method may produce clusters based on criteria which ultimately produce practically meaningless clusters (Borgen and Barnett, 1987, p. 461; Blashfield, 1980, p. 457). To avoid this problem, clusters are frequently made using several different methods to see if the same clusters are produced. In addition, there is a problem of deciding which clusters or groups of clusters in the analysis are really meaningful ones. Any person who uses this method must decide how many clusters should be made from the 68 input data, and come up with a rational decision rule about how many clusters to examine. Since Attanucci used Ward's method of cluster analysis, this was chosen as principal method for the present research, using the statistical package in SPSS-x. Attanucci did not report the resemblance coefficient used, although presumably it was the squared Euclidean distance, which is the method recommended when using Ward's method of cluster analysis (SPSS-x Users Guide, 3rd Edition, 1988, p. 406). There is one problem, though, about Ward's method, which Attanucci does not address. The problem arises from the kind of input data, which in her case (and in this case, at least for one analysis) is binary. There is a set of sixteen scores per person which indicates presence or absence (0 or 1) of categories I to IV relationships within each person category (husband, mother, child/children, other). According to Romesburg (1984), when working with binary data such as in the present case, one should investigate the value of using a special type of similarity coefficient for qualitative data. It is also justifiable, according to him, to use one of the more general methods without making any special accommodations for binary data. The data in the present research were grouped and analyzed by case (Q-analysis), which is the same method as used by Attanucci. However, in the present research it was decided to use three methods for cluster analysis for one set of clusters investigated, those for the sixteen zero or one scores for each person. Additional analyses were done using two methods. First, the analysis was done with Ward's method, using squared Euclidean distance as a resemblance coefficient. lext, the analysis was done by first using the Jaccard coefficient of resemblance, 69 which is appropriate for binary data (Romesburg, 1984, pp. 143-144). The Jaccard coefficient indicates maximum similarity when two objects have identical values, and maximum dissimilarity when there are no 1-1 matches. The cosine of vectors of variables is another pattern similarity measure appropriate for use with binary data, and this was used with the Group Average method of cluster analysis (it would not be appropriate to use with Ward's method). Because the data set was small, no attempt was made to eliminate outliers. This meant that there should be some tolerance in the number of clusters investigated, and no more than six clusters per analysis were selected by visual inspection of data output. In the present case, decisions were made by visual inspection of the vertical icicle plot. It should be noted that in some cases, clusters were formed based on a single case, and there was no practical way to decide if these single- case clusters were outliers or were theoretically meaningful clusters, but under-represented in the present sample. Romesburg (1984) suggests presenting results of a cluster analysis by giving the mean and standard deviation of the data for each cluster, which is done for the present analyses, although Attanucci did not provide this. Attanucci, rather, provided histograms of the proportion of cases in each cluster that use statements from the four categories. The cluster analysis was not followed by an analysis of variance, as Attanucci did for ”confirmation” of clusters, because when these analyses [discriminant analysis, lAlOVA, or multiple univariate F tests] are performed on the variables originally used to form the clusters, the results have no meaning. The reasoning behind this is that if, for example, the cluster solution implied a certain number of clusters which 70 were known to have no meaning, "a one-way analysis of variance on these three groups...[will find] the F test will be highly significant, even though no true clusters exist in the data.” (Blashfield, 1980, pp. 457-458). Cluster analyses were also used to investigate the clusters which might be formed from the RRSI variables and the Early lemories variables. The was by Ward's method and the Group Average method. Additional Results The following were also be investigated in an exploratory manner: a. Case material from interviews for persons who score highly on the Attanucci measure of Gilligan's highest perspective as well as on the object relations measure. b. Case material from interviews for person who score relatively low on both the Attanucci measure of Gilligan's highest perspective and on the object relations measure. c. Themes in the middle years of life related to issues around "narcissism.“ Basically this was an investigation of interesting or particularly well described responses to the questions about normal narcissistic concerns in the middle years of life, and the ebb and flow of these concerns in relationships. RESULTS Relations between Highest Levels in Gilligan's Theory and Object Relations Theory The research question is: To what extent are achievements of the epitome of development according to Gilligan, and the epitome of development in one model of object relations theory, related, in middle- aged women? The research hypothesis is that they are positively related, and that this relationship will hold true both for the measure of Gilligan's highest developmental level in the Attanucci measures of "Self in Relationships” and in the measure in the Revised Relational Self Inventory. lore specifically: 1. "Self in self's terms and other in other's terms" will be positively related to ”lutuality of Autonomy,” an aspect of quality of object relations. 2. ”Self and Other Cars Chosen Freely” will be positively related to ”lutuality of Autonomy,” an aspect of quality of object relations. To address the hypotheses, first we look at the results of a Pearson Product loment correlation analysis examining possible significant relationships, and then we look at cluster analyses including all three measures. The Attanucci method of arriving at category scores provides four scores for each of the four categories. These 16 scores were collapsed over persons in relationship, giving a summed score for each 71 72 developmental perspective category. The rationale for collapsing the scores is that Gilligan's theory does not assume that any specific relationship should be at a higher or lower level than other relationships. It seemed likely that a weak effect for a Category IV relationship, if present, would be assisted statistically if these were summed into a single variable. Summations resulted in four variables named Sum I to Sum IV. Table 6 presents the intercorrelations for these summed scores with the RRSI and Early lemories scores. Table 7 presents the results for the restricted sample excluding women who scored three standard deviations or above on any scale of the Brief Symptom Inventory. Probability levels of .05 or less were required for considering a relationship significant. Hypothesis 1 The first hypothesis was not rejected in that Sum IV was positively correlated with one measure of Quality of Object Relations in Early lemories, that of Earliest lemory of Father. Hypothesis 2 The second hypothesis was rejected, in that ”Self and Other Care Chosen Freely” was not correlated with any of the measures of Quality of Object Relations in Early memories. These findings hold both in the complete and restricted samples. In a parallel examination of relationships, cluster analyses were run for scores from the three instruments. The Attanucci variables were Sum I to Sum IV, as they were in the correlations. The analyses used squared Euclidean distance coefficients of similarity matrices as input. Results for complete and restricted samples are shown in Tables 8 and 9. The first cluster in Tables 8 and 9 shows high early memory scores, Sum 73 IV of over 2, and ”Self and Other Care Chosen Freely” (SOCCF) of a value equal to or lower than SOCCF in any other cluster. On the other hand, the final cluster in both tables consists of very low early memories scores, Sum IV of less than one, and a high SOCCF (compared to all other clusters). Similar patterns are found when the cluster analyses were done by the Group Average method (UPGlA) as shown in Appendix J. However, in one case, that of the restricted sample (J-2), a six cluster solution was chosen. This had the same pattern of results in that the first two clusters had the same pattern as compared to the last cluster. The intermediate clusters contained what might be called a ”mixed bag” of results: with both high and low results for the Attanucci Sum IV and Els scores. However, the cluster analyses confirm the hypothesis of relationship between Attanucci Sum IV and Quality of Object Relations in Early lemories Scores, but do not support the hypothesis of relationship between "Self and Other Care Chosen Freely” and Quality of Object Relations in Early lemories. Other Findings of Interest The first set of additional findings refer to the correlations (Tables 6 and 7). Within the ElS scales the different scores (earliest memory, earliest memory of father, earliest memory of mother) all correlate with each other, which is expected. Within the RRSI scales, however, some interesting findings occur in both complete and restricted samples. First of all, score on RI correlates with score on POC. This is not surprising, either, because persons whose way of thinking comes from a relational framework would be expected to include some persons 74 Table 6 Pearson Product lbment correlations for Summed Attanucci Categories, Summed RRSI Scale Scores, and Quality of Object Relations in Ebrly lemories Scores (Complete sample, R’I 28) SumI SumII SumIII SumIV so RI sun: SUlII -.26 SUlIII -.03 .15 sunrv .00 .14 -.12 so -.31 .05 .07 -.18 xx .17 -.1s -.16 -.04 .13 90c .33 -.01 .03 -.03 .17 .55“ soccr -.29 .00 .00 .00 .49“ .37 mm .08 .05 -.10 .11 .29 .03 rn-n .02 .35 -.19 .25 .18 .05 rn-r .11 .27 .29 .56" -.23 .07 Rbte. Two tailed. And the following abbreviations were used: Sum I'ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sum Il’ISummed score for Category II statements (Self instrumental to other, others “in their own terms” Sum Ill ISummed score for Category III statements (”Self in self's terms,” other instrumental to self) Sum IV'ISummed score for Category IV statements (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI’I Relational/Connected Self score POC IPrimacy of Other Care score SOCCF’ISelf and Other Care Chosen Freely score D! IEarliest lemory score fl-l! Earliest lemory of lother score E'l-F IEarliest lemory of Father score 75 Table 6 (Cont'd.) POC soccr m El-lOl soccr .12 mm -.12 -.17 rm-n —.11 -.23 .63*“ sm—r -.11 -.30 .53“‘ .70*** lbte. Two tailed. p < .05 p < .01 p < .001 76 Table 7 Pearson Product lament Correlations for Summed Attanucci Categories, Summed RRSI Scale Scores, and Quality of Object Relations in Early lemories Scores (Restricted Sample, I I 24) SumI 30-11 SumIII SumIV so RI sour SUlII -.20 SUlIII -.03 .27 suwrv .07 -.02 -.03 so -.23 -.03 .03 -.15 31 .14 -.1s -.20 .01 .21 30c .33 .03 .04 .03 .07 .50** soccr -.27 .15 -.03 .07 .43* .41* ms .27 -.25 -.07 -.03 .25 .19 mn-n .13 .12 -.14 .13 .22 .17 zn-r .33 -.10 -.27 .47‘ -.30 .31 Rota. Two tailed. And the following abbreviations were used: Sum I'ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sum ll'ISummed score for Category II statements (Self instrumental to other, others ”in their own terms" Sum Ill'ISummed score for Category III statements ("Self in self's terms,” other instrumental to self) Sum IT'ISummed score for Category IV statements (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI'I Relational/Connected Self score POC IPrimacy of Other Care score SOCCF'ISelf and Other Care Chosen Freely score El IEarliest lemory score n-a Earliest lemory of lother score 01-! IEarliest lemory of Father score e p ( .05 ::e p ( '01 p ( .001 77 Table 7 (Cont'd.) 90c soccr m El-lOl soccr .06 an -.07 -.20 mn-n .02 -.13 .54“ mu-r .16 -.23 .50“ .30" Rote. Two tailed. t it .tfi .05 .01 .001 ‘3‘ AAA 78 Table 8 Cluster Analysis of Summed Attanucci categories, Summed RRSI scale scores, and Quality of Object Relations in Ehrly'lemories Scores - Word's Hothod, Squared Eoclidean Distance coefficient - Complete Simple, R’I 28 Cluster SUlI SUlII SUlIII SUlIV SOCCF SO RI POC Cluster 1 (lI13) lean 2.8 3.3 3.6 2.4 3.97 2.72 4.26 3.11 SD 1.2 1.0 0.6 1.2 0.45 0.35 0.42 0.31 EH-E Efl-l El (Cont.'d) lean 11.0 11.8 11.8 SD 1.1 2.0 0.9 80711 801111 8011111 SUHIV SOCCF SO RI POC Cluster 2 (lI5) lean 2.8 SD 1.2 (Cont.'d) lean 11 5 SD 1. . 0. 79 Table 8 (cont'd.) SUHI SUlII SUlIII SUlIV SOCCF 80 RI POC Cluster 3 (lIS) lean 3.2 3.6 4.0 3.0 4.00 2.43 4.13 3.34 SD 1.0 0.5 0.0 1.1 0.11 0.25 0.36 0.39 El-F El-l El (Cont.'d) lean 9.4 6.2 8.2 SD 0.8 0.4 2.4 SUlI SUlII SUlIII SUlIV SOCCF SO RI POC Cluster 4 (lIS) lean 2.4 2.6 4.0 0.8 4.25 2.67 4.15 3.18 SD 1.4 1.0 0.0 0.7 0 30 0.38 0.39 0 42 El-F El-l El (Cont.'d) lean 3.4 4.0 4.2 SD 1.0 1.1 1.2 lots. The following abbreviations were used. Sum I'ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sum Il'ISummed score for Category II statements (Self instrumental to other, others "in their own terms” Sum III'ISummed score for Category III statements (”Self in self's terms,” other instrumental to self) Sum IV'ISummed score for Category IV statements (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI’I Relational/Connected Self score POC IPrimacy of Other Care score SOCCT'ISelf and Other Care Chosen Freely score I! IEarliest lemory score ”-8 Earliest lemory of lother score Ell-F IEarliest lemory of Father score 80 Table 9 Cluster Analysis of Summed Attanucci categories, Summed RRSI scale scores, and Quality of Object Relations in Early lemories Scores - Word's lethod, Squared Euclidean Distance coefficient - Restricted Sample, I I 24 Cluster SUlI SUlII SUlIII SUlIV SOCCF SO RI POC Cluster 1 (lI12) lean 2.9 3.3 3.6 2.3 3 92 2.67 4.26 3.08 SD 1 1 1.0 0.6 1.2 0 42 0 30 0.44 0 31 El-F El-l El (Cont.'d) lean 11.0 12.0 11.8 SD 1.2 2.0 0.9 SUlI SUlII SUlIII SUlIV SOCCF SO RI POC Cluster 2 (lI5) lean 2.8 3.8 3.6 3.2 3.97 2.36 4.02 3.16 SD 1 2 0.4 0.8 0.7 0.51 0.56 0.49 0.36 El-F El-l El (Cont.'d) lean 11.0 10.4 5.8 SD 1.1 0.8 0.7 81 Table 9 (cont'd.) SUlI SUlII SUlIII SUlIV SOCCF 80 RI POC Cluster 3 (l-S) lean 3.2 3.6 4.0 3.0 3.97 2.46 4.11 3.38 SD 1.0 0.5 0.0 1.1 0.10 0.28 0.40 0.43 El-F El-l El (Cont.'d) lean 9.4 6.2 8.2 SD 0.8 0.4 2.4 SUlI SUlII SUlIII SUlIV SOCCF 50 RI POC Cluster 4 (lI2) lean 1.0 3.5 4.0 0 5 4 22 2.67 3 79 2 75 SD 1.0 0.5 0.0 0 5 0 41 0.34 0 21 0 04 El-F El-l El (Cont.'d) lean 4.0 3.5 3.5 SD 1.0 0.5 0.5 lots. The following abbreviations were used. Sum I’ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sum ll'ISummed score for Category II statements (Self instrumental to other, others ”in their own terms” Sum IlI'ISummed score for Category III statements ("Self in self's terms,” other instrumental to self) 50- IV'ISummed score for Category IV statements (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI'I Relational/Connected Self score POC IPrimacy of Other Care score SOCCF'ISelf and Other Cars Chosen Freely score El! IEarliest lemory score 311-)! Earliest lemory of lother score fl-I' IEarliest lemory of Father score 82 whose form of relating is at what Gilligan calls the conventional feminine mode - that of Primacy of Other Care. However, some other interesting patterns occur. Persons whose way of thinking comes from a relational framework also include persons whose form of relating is at the level of ”Self and Other Care Chosen Freely,” but this holds only for the restricted sample which excludes persons with indicators of psychological distress. In both restricted and complete samples, there is a relationship between holding to a Separate/Objective mode of thinking and to endorsing ”Self and Other Care Chosen Freely, ' which was an unexpected finding. This raises the question that, at least in this group of women, they may hold both kinds of beliefs simultaneously. Of course, if this were so, then RI should correlate with SO. Although scores on SO and RI do not correlate significantly with each other, they are in the same direction ( r I .18, complete sample: r I .21, restricted sample). Descriptive Statistics After examining the main results of the cluster analyses, it seemed useful to look at the women's scores on these measures in more minute detail. What, for example, is the average score of women on each variable used in the analyses? The tables 10-12 provide descriptive statistics. The average score on Sum IV was 2.4 (2.5 in the restricted sample). Average ElS scores in the three categories were between 8.8 and 9.4 in the complete sample, and between 9.1 and 10.1 in the restricted sample. It is somewhat surprising that these women score 83 Table 10 comparison of Summed Category Scores for Attanucci variables Between complete Sample (R’I 28) and Sample with High symptomatic Scores Excluded (Restricted Sample, R'I 24)) lean SD Range Sum I Complete sample 2.8 1. 0 - 4 Restricted sample 2.8 1. 0 - 4 Sum II Complete sample 3.3 0. 1 - 4 Restricted sample 3.5 0. 1 - 4 Sum III Complete sample 3.8 0. 2 - 4 Restricted sample 3.6 0. 2 - 4 Sum IV Complete sample 2.4 1. 0 - 4 Restricted sample 2.5 1. 0 - 4 tote: Sum I I Summary score for Category I relationships (range 0 - 4); Sum II I Summary score for Category II relationships (range 0 - 4): Sum III I Summary score for Category III relationships (range 0 - 4): Sum IV I Summary score for Category IV relationships (range 0 - 4). 84 Table 11 comparison of RRSI Scale Scores between complete sample (R'I 28)and Sample with High symptomatic Scores Excluded (Restricted sample, l’- 24) lean SD Range SOCCF Complete sample 4.03 0.42 3.25-4.69 Restricted sample 3.97 0.42 3.25-4.69 80 Complete sample 2.60 0.42 1.72-3.39 Restricted sample 2.55 0.40 1.72-3.39 RI Complete sample 4.17 0.44 3.25-4.92 Restricted sample 4.14 0.46 3.25-4.92 POC Complete sample 3.17 0.37 2.43-3.93 Restricted sample 3.12 0.36 2.43-3.93 lote. SOCCF I Self and Other Care Chosen Freely SO I Separate/Objective Self RI I Relational/Connected Self POC I Primacy of Other Care 85 Table 12 comparison of Quality of Object Relations in Early lemories Scores between Complete Sample (N’I 28) and Sample with High Symptomatic Scores Eacluded (Restricted Sample, l’I 24)) lean SD Range Earliest lemory Complete sample 8.8 3.4 3 - 13 Restricted sample 9.1 3.2 3 - 13 Earliest lemory of lother Complete sample 9.2 3.5 3 - 15 Restricted sample 9.8 3.3 3 - 15 Earliest lemory of Father Complete sample 9.4 3.0 2 - 13 Restricted sample 10.1 2.2 3 - 13 86 below the expected averages of about 11 to 13 which Ryan found in his work using this scale. It can also be seen that the relational perspective (and "Self and Other Care Chosen Freely”) are much more clearly embraced by these women (both samples) than is the Separate/Objective perspective. The fact that women, on average, clearly embrace ”Self and Other Care Chosen Freely" but score below what would be expected in early memories, is puzzling. Cluster Analysis of Attanucci Categories Cluster analyses were run based on the 16 scores as used by Attanucci (4 categories by 4 types of relationship), but the results are reported in terms of means and standard deviations for summed categories I - IV. Because the former involved binary data (1 for present, 0 for absent), the analysis was tried a few different ways to see if the kinds of clusters formed would vary by method. The two methods used were Ward's method (used by Attanucci) and the Group Average method. Several ways of forming similarity matrices were tried, as recommended by Romesburg (1984). Tables 13 and 14 show the results using Ward's method, choosing four cluster solutions. (Appendix J, tables J-3 to J- 6, contains results from the Group Average method, and using similarity coefficients more specific to binary data. These provide five and six cluster solutions. The first cluster (Tables 13 and 14; Ward's method ) has a Sum IV of IV. In the UPGlA methods in Appendix J, the first cluster contains a Sum IV that ranges from 3 to 4. However, in the supplementary methods, one other cluster would have a Sum IV of 3. these clusters (fifth on J-3; fourth on J-4: sixth on J-5, and fifth on J-6) had a Sum I of 4 (high), but relatively lower sums II and III 87 (compared to the first cluster). Are these different types of women? Since Sum III tended to be high in most clusters, this might indicate two types of persons who score highly on Sun IV, those who are also high on Sums II and III, but not on Sum I, vs. those who are lower on Sums II and III, but high on Sum I. In a practical sense, this might mean that people who are in a transition period to more completely acting at a Sun IV level (still high on Sums II and III) would show up in the first cluster, while those who have more completely integrated Sum IV behavior show up in the later cluster in the additional analyses. Cluster Analyses of RRSI variables Another type of comparison looked at the clusters formed by the RRSI scale scores (Tables J-7 to J-10 ) for the complete and restricted samples, using the Ward's and UPGlA methods, and four and six cluster solutions. In the Ward's method, the first cluster had the highest value for SOCCF, while the fourth cluster had the highest SOCCF in the UPGlA method. The UPGlA method showed lowest values of Primacy of Other Care (P00) in the same cluster that had the highest value for SOCCF, but this was not so for results using the Ward's method. But in the Ward's method, the lowest values for SOCCF were in the last cluster, which also had lowest values for the Separate/Objective Self. There were differences in values of the Relational Self and Primacy of Other Care between the complete and restricted samples, with restricted samples having higher averages on these variables. Cluster Analyses of the Two leasures of Gilligan's Highest Perspective Tables J-11 through J-14 (Appendix J) show cluster results of analyses of summed Attanucci Categories and RRSI scale scores. In each case the first cluster has the highest value of Sum IV, and also a high 88 Table 13 Cluster Analysis of Summed Attanucci categories, Ward's lethod, Squared Eoclidean Distance Coefficient, complete Sample (l’- 28) Cluster Sum I Sum II Sum III Sum IV Cluster lumber 1 (lI6) lean SD mm Cu 0 ~24 O. 00 ON Ow 0'0! Cluster lumber 2 (lI6) lean SD mo 04» e hm H” H“ OH 04» (J's) Cluster lumber 3 (lIS) lean SD Cu .00 OH Ow ON GO Cluster lumber 4 (lIll) lean SD Ow 0 e we o-H-I 0 0 N04 04» ~20! Ow fine 89 Table 14 Cluster Analysis of Summed Attanucci categories, Ward's lethod, Squared Euclidean Distance Coefficient, Restricted Sample (N'I 24) Cluster Sum I Sum II Sum III Sum IV Cluster lumber 1 (lI6) lean SD ON 04» mm 04» O. OO Cluster lusber 2 (HIS) lean SD 00 0'09 Ch! 0 O .Q HN Nfi OH 04» Cluster lumber 3 (lI10) lean SD 04» I 0 4m Ow 0 Oh) O O wso r-u-o O 0 H4 Cluster lumber 4 (lI3) lean 4.0 SD 0 0 OH e e Md 0:» e ON a“ 90 value of SOCCF. Then there is another cluster, the last or next to last, which has the lowest value of Sum IV, but it will also have a high value of SOCCF. The implication here is that persons who score high on SOCCF seem to be of two types - those who would also score high on Sum IV, as expected from the theoretical comparability of those variables - and also of another type, which is composed of persons who score quite low on Sum IV. These persons may be persons who want to score high on SOCCF, who value such an attitude highly, precisely because it is so hard to achieve at this time. This is speculative, of course. Cluster Analyses of Sum IV and Quality of Object Relations in Earliest lemory of Father Finally, Tables J-15 to J-18 show cluster results of analyses cosparing Sum IV to scores on Quality of Object Relations in Early lemories. It is interesting to note that one cluster which has the highest value for Sum IV (Table J-15, cluster 4: Table J-16, cluster 3; Table J-17, cluster 4: and Table J-18, cluster 3) has also an Early lemories of Father score of about 10. On the other there is another cluster in each table which has a low Sum IV of one or less (Table J-15, cluster 6; Table J-16, cluster 4: Table J-17, cluster 6; and Table J-18, cluster 4), but which also has a very low Early lemories of Father score (of 4 or less). This final cluster is the only one which has a very low score for Early lesories for Father. Case Studies Some case material is presented for women who placed in the highest category of "Self in Relation to Other” in the Attanucci scoring system, and who also were in higher levels of object relations, compared 91 to other women in the study. In addition, women who score lowest in both categories, compared to other women in this sample, are investigated further. The highest level of ”Self in Relation to Other” is defined as achieving Category IV, ”Self in Self's Terms and Other in Other's Terms," in all four types of relationships. And since the cluster analyses showed in both complete and restricted samples that there was a significant relationship between the score on Category IV and the score for Earliest lemory of Father, that score was taken as the criterion for the object relations measure. Persons who scored above 10 on that measure (which puts them in the ”neurotic to normal” range) were considered "high,” for the purposes of this study. Four women set these criteria. Table 15 presents a summary of their scores. Another four women are presented for contrast; two scored in the lowest categories of the Object Relations in Early lemories scale and also had no score in the Category IV range of the Attanucci measure (Table 16). The results of two additional women are presented in Table 17 because they scored high on the BSI (an index of psychological distress), low on object relations, and relatively low on Attanucci Category IV. Some brief case material follows from the interviews of the wosen who scored the highest on Attanucci Category IV and on the score for Earliest lemory for Father. The excerpts are intended to provide information on the complex ways that women describe important relationships. 853 is a highly verbal, married, professional woman in her early 40s who describes her relationship with her husband as follows: 92 Table 15 Summary of Scores for High-Scoring Women for the Attanucci category IV (Sum 17 I 4) and for Quality of Object Relations in Early'lomories El-E El-l El sun I SUl II SUE III Subject lo. 53 11 11 11 3 3 4 SOCCF SO RI POC 4.19 2.78 4.67 2.93 El-F El-l El SUl I SUl II SUl III Subject lo. 55 11 13 13 3 4 4 SOCCF SO RI POC 3.75 2.72 3.92 3.07 Note. The following abbreviations were used. Sum I ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sua II'ISummed score for Category II statements (Self instrunental to other, others "in their own terms” Sum III'ISummed score for Category III statements (”Self in self's terms,” other instrumental to self) Sum IV'ISummed score for Category IV statements (Self in self's terms, others in their own terms) ' SO ISeparate/Objective Self score RI'I Relational/Connected Self score POC IPrimacy of Other Care score SOCCT'ISelf and Other Care Chosen Freely score El IEarliest lemory score Ell-l! Earliest lemory of lother score Hf-F IEarliest lemory of Father score Table 15 (cont'd.) 93 El-F El-l El SUl I SUl II SUl III Subject lo. 64 13 11 12 2 4 4 SOCCF SO RI POC 4.50 2.51 4.33 3.43 El-F El-l El SUl I SUl II SUl III Subject lo. 67 11 10 5 2 4 2 SOCCF 80 RI POC 4.69 2.50 4.58 3.21 94 Table 16 Summary of Scores for Low-Scoring Romen for the Attanucci category IV (Sum IW'I O) and for Quality of Object Relations in Early lemories EH-E Ell-l El SUE I SUN II SUl III Subject lo. 52 5 4 4 2 4 4 SOCCF SO RI POC 3.81 2.33 3.58 2.71 El-F El-l El SUl I SUl II SUl III Subject lo. 73 2 6 7 3 3 4 SOCCF SO RI POC 4.13 3.22 4.50 3.71 lots. The following abbreviations were used. Sum I ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sum Il'ISummed score for Category II statements (Self instrumental to other, others ”in their own terms” Sum Ill'ISummed score for Category III statements ("Self in self's terms,” other instrumental to self) Sum IV’ISummed score for Category IV statements (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI’I Relational/Connected Self score POC IPrimacy of Other Care score SOCCF’ISelf and Other Care Chosen Freely score A?! IEarliest lemory score Eli-l! Earliest lemory of lother score Elf-F IEarliest lemory of Father score 95 Table 17 Summary of Scores for Low-Scoring Women for the Attanucci Category IV (Sum IV I 1 or 2), for Quality of Object Relations in Early'laaories, and high score on Brief Symptom Inventory (BSI) El-E El-l EH SUE I SUE II SUN III Subject lo. 60 4 4 5 3 1 4 SOCCF SO RI POC 4.56 2.50 4.00 3.07 El-F El-l El SUl I SUl II SUl III Subject lo. 66 3 3 3 4 2 4 SOCCF 50 RI POC 4.13 2.28 4.67 3.64 lots. The following abbreviations were used. Sun I ISummed score for Category I statements (Self instrumental to other, other instrumental to self) Sua Il’ISummed score for Category II statements (Self instrumental to other, others ”in their own terms” Sua Ill'ISummed score for Category III statements ("Self in self's terms," other instrumental to self) Sum IV’ISummed score for Category IV statesents (Self in self's terms, others in their own terms) SO ISeparate/Objective Self score RI'I Relational/Connected Self score POC IPrimacy of Other Care score SOCCT'ISelf and Other Cars Chosen Freely score D! IEarliest lemory score Ell-l! Earliest lemory of lother score Hl-l' IEarliest lemory of Father score 96 ly husband is, um, is brilliant. He's, um, (pause), has much greater capacity for patience than I. He's much more patient. luch lower key. I would describe him as laid back. Um, what makes me feel guilty is that I have progressed more quickly than my husband has, uh, academically. He's older than I am, but, but, he's a lower faculty rank because he has never, he hasn't published anything since he published his book. But his book was published by [prestigious publishing house]. I could never publish a book with them. Um, and, you know, it was a brilliant book. He just really has made a choice not to get into the academic rat race. So he is different in that respect, uh. You know, he teaches his classes, he's very sensitive, um, and he's great. I know he is. He's very good as well, as being a good teacher, and a, and having a marvelous mind. He's also very good, good with his hands. He, um, completely renovated a house, an old house that we bought. And, um, he's good at carpentry work. And I've talked to other women about this, by the way, uh, who have husbands like mine, uh, to some extent, like mine. And we find that very attractive, very sexy. It's one thing for someone to be a carpenter by trade. It's another thing for someone to be an intellectual and also to be able do certain kinds of, of carpentry work. And that I find very attractive. Umm, even the plumbing. (laughs). So.... Um, on the other hand, I get very impatient with my husband. He, uh, he, at the same time as I admire his decision not to, to pursue academic (unintelligible), um, he, uh (pause), he does, um, he's very introverted, or, uh, he does a lot of analysis. Self analysis....[on the other hand, I find it very frustrating....that he's not a kind of go-getter. I like, I like certain aspects of him, and they certainly smooth me down, but they also ruffle my feathers, too. Um, he's a late night person, and I tend to go to sleep a little bit earlier than him. I'm a morning person. So we're not exactly on the same schedule. Um, I'm much more of a socializer, than he is. Although he has his friends, too, and enjoys them, and is a very good entertainer when we have people over. Very charming and funny, um, so, but that's pretty much how I would describe my husband. [Interviewer] And your relationship with him? ly relationship with him, you know, is kind of hot and cold. There are times...hot and cold are extreme words...there are times when we go for long periods of time where we're, we're very intensely related to each other, and, um, that may or may not be sexual, uh, intensity. There are times when we are um, often somewhat indifferent to each other. And there are times when there's some hostility, either that I feel or that he feels, and if one of us feels it, it ends up being both of us. But we do try to talk it out... 97 455 is a highly verbal married woman who is employed part-time and values her family above her need for a career. She describes her present interaction with her mother as follows: Yeah. Yeah, well, um, my mother is interesting. ly mother used to always say when we were kids, if you can't say something nice, don't say anything. And yet, the older she gets, the more she kinda looks for the negative. And we had the first grand children, we had the only grandchildren at the time. And we got along beautifully with my family. We would go over there at least once a weekend, take all whoever was a baby and whoever. We had cribs over there and diapers and everything. And we'd spend at least Saturday, maybe Saturday and Sunday, um, got along fine. Once we moved here, as soon as we moved here, my dad had died, and my mother has since come here every year except for last year when we went to visit her. ....ly mother has moved into a real nice, um, retirement place, and a lot of the people there are very, you know, dress nicely and all. ly mother never cared that much about it.....[And] this Christmas when she came, I was going to take her shopping, and she ended up getting sick right on Christmas. She got a real bad cold. But I had bought her a sweatshirt, and she insisted she wore large. But when she tried it on, of course, it didn't fit. And I finally said. I'm going back to the store. I'm going to keep the large for myself. I will go to the store. I said, lom, you don't wear a large sweatshirt, you need an extra-large. And I said, why don't we just face that and I will go and get one. And I thought, oh, that was terrible to do. And she wrote me a letter later. She said, I know I'm hard to shop for. And I thought, (makes noise of aggravation) ....But, uh, I try to be tolerant. (laughs). It's hard, though. Especially when you see the mother-child relationship turning around. 864 is a highly verbal wosan also, married, in her 50s, who does volunteer work, although she has some professional training in her field. She describes her relationships with her children as follows: I think [I was] overprotective and judgmental. Uh, yeah, I think somewhat of a perfectionist, maybe. I had really high goals for 98 these kids. I mean, you know how you always hear that you don't know how it is until you finally experience it (laughs).... Well I think I learned some hard lessons being a mother, in that respect. That your children are not perfect, even though you think they are. Uh, and that the more praise you give them, the more response you get. But if you're judgmental, you're not going to get any response. But I'm not sure when that all evolved. I really enjoyed my children. I'm not sure that I enjoyed babies. Babies are kind of blah. But I really, when they were about 4 or 5, I think I, and through their teenage years, we had a great time, a really good time. I think I wanted them to talk back to me. I think that's when I first started really liking kids, when they could respond. And talk to me and do things. Cause I had a good time with my kids, I really did. Really enjoyed them. Um, I liked to read, and so we went to the library from day 1. Everybody. And my kids were all avid readers. I mean, they were all reading by the time they were 5, and they knew... And I think its because we all went to the library and I read. And if I couldn't read my Robert Ludlum (refers to book on table), which wasn't out on that time, but I mean, I read my books to them in order to read. And so it wasn't always their books that got read. (Both laugh). I was a little self-centered at times. Uh, and we would take walks. Well, you gotta win somehow. But we did a lot of camping with our children when they were young. Did a lot of things out doors. Things that didn't cost anything, because we didn't really have a lot of money when they were young. 867 is a divorced woman, highly verbal, in her early 40s, who has this to say about relations with her forser husband as they tried to raise their sons: But getting to the way my ex-husband was a father, he felt like, well, it's so hard during different stages of my boys' lifes, you know. I think of them in terms of their teenage years cause that was the most recent. But, um, growing up he wasn't with them a lot or when they were very small, he was around them a lot and catered to them a lot. I think during a lot of their more impressionable years, he wasn't around a lot to, to do a whole lot with them. He was busy with ball and all this stuff. Um, but he was, he was a good father. I mean, he wasn't a drinker and he wasn't a womanizer, and when he was there, he, um, he often 99 had, you know, he had a temper, and he would often, I think, over reacted I think to some of their situations, but then, so did I. Um, but I think he was a much better father to them during their high school years. He's always there for advice and support, and he's not, he's a very good father now. luch more so than when the kids were younger. He sees...he sees the importance of being there and being, serving as an advisor. And that's a difficulty that we both had to assume as serving the role of advisor instead of punisher or making sure that they did what we told then to do. On, I would tell them sonething to do in their teenage years, and, you know, I just had to follow through with it, that they did it. And now I had to realize that my role as they got older was to give them advice. And if they didn't take it, they had to assume the consequences of their actions, whereas, before it was always me assuming the burden of them not doing their homework, not doing this. If you don't do this. But that was hard to let go of them. And Joe is learning to do that, to let go, serve as advisor to them, give then counsel. If you don't do this, then you're assuming that now. You're an adult. You're 18, 20. It's important that you assume that responsibility. And I think that as a parent, that's important, that we let go of them and not always assume the responsibility for their actions. The two wonen who had no responses which could be coded IV in the Attanucci coding system are presented next in terms of some of their discussions of relationships. 452 can be described as ”forty-something:' she is married and has a family. She describes her relationship with her mother as follows: Oh. On, my mother, I just have a, um, have always been at odds with. We have never had a good relationship. I don't like her as a person. Um, I envy her energy and her talent, um, I don't like her disposition. I don't like her attitude. I don't like her, um, the way she thinks about other people. I don't like the way she looks bodywise. I think she's, um, I don't know. I don't like her in a lot of ways. It's been difficult. I would like to know her and get a relationship with her. She's not at all interested in that. She's never been. Um, she does not want to have a friend or a relationship. She's, she, um, not sure she even wants to be a mother. She likes to dominate people, but she, um, she won't even let people a crack into her own life. 100 I mean, uh, when I was 20, I'd say I knew nothing about her former life at all. Um, I knew she was a seamstress , and, because when I was in my 20s she went back to work (laughs). I didn't know that before. Um, she never told me anything about her romance life. The only thing I knew of her life was that her mother had died of cancer when she was 10. her father had re-married a woman who had an illegitimate child. The woman was [religion], and, um, my mother totally hated this woman. She claimed the woman was really mean. Um, and that's all I know about her whole life. Um, I sort of put things together through the years, but it's all speculation. If I try to ask my aunts or anyone, they'll change the subject. They won't tell me anything. So I don't know her. What I do know about her is that she, she's got an incredible amount of hate in her that's always been expressed. Its seething under the surface all the time. Um, I just don't like her at all. 873 is a married woman, in the process of getting a divorce, with teenage children. As a child she was a victim of incest by her stepfather. She discusses her current attitude as follows: Um, I wanted to survive. I just felt like it was something in my life that stopped...was stopping me. Like I couldn't really go out and be what I really wanted to be. You know, until I could clear everything. It is kind of hard to describe. Un, oh, I don't know, it is just very hard to describe. Uh, I think the big...the biggest thing that started it was that it came out in the family. Uh, my brothers found out and one of my brothers confronted my father and, uh, its...there now is a lot of splits in the family. It's caused a lot of problems. We were never really close--any of us. I have seven brothers and two sisters. Uh, there is only eight of us left living now. (Clears her throat.) But, all through my life it was like something that stopped me. You know, I could never get close to my sisters. Um, my brothers were more like just somebody else that lived there. We were never really close or anything. But when this came out, uh, my sister felt she had to tell one of my brothers--is how the whole thing came out. Who was contemplating...uh, he's retired from the army. He has two sons and a daughter. They were going to move next door to my mom and dad. And my sister out of concern for my brother's daughter, felt she had to tell him. And I backed her. I fully supported her, up to the talking to my brother about it. I told her, I said, “I will do anything but I cannot talk to him about it. But I will support you, you know.” 101 And, uh, we didn't realize at the time that he would confront my father with it. And this was my dad's favorite son. And just totally destroyed the family. And was just...we are not...we were not close and we are even farther apart now. I60 is a divorced student with teenage children who also was a victim of incest as a child. She talks about her relationship with her teenagers as follows: Um, we had a real difficult during the divorce and since then. And, you know, just, things were in turmoil. I was crazy (laughs) for a long time, I really was. I was not helpful to them, I don't think. I tried to be, but I was so emotionally drained and distressed and a combination there that, uh, I feel like I didn't help them a lot adjusting to the divorce. I did what I could. But, on the other hand, we've talked about it since then and looked back over the last two years, and we said, well, we got through it. And we all helped each other out at different times. And that's made us all a lot stronger, I think, as a family. I think we're all, it changed our relationships, I think, a lot. but I think we've all kinda grown from it. Grown and grown up. Yeah, I just...woke up one morning very literally and said, this is it, I don't want to be married any more. And I called my friend who works for an attorney and I said, give me the names of sone lawyers, and I called, and got an appointment, and that was it. I just, that's the way it happened. (laughs). And, uh, I think that was kinda the catalyst for all of this. And it was just like...that. And I didn't really think about the incest or anything, for quite some time, until I started therapy, actually, in which..you know, and the sequence of events. There was this turmoil, and I felt very unhappy, and I knew I was unhappy, but I didn't know why. And I didn't know...anything about it. So I just tried to continue on with my life...and, but, I just kept realizing that I was more and more and more unhappy. 466 is a woman in her late 30s who is a married student, going for advanced degree. She describes some of the relationships in her life follows: Um, I don't get close to people very easily. I've had some bad experiences that have caused me to be very, um, non trusting--I guess that is probably the best way to put it. Um, (sighs), I was very naive when was 18 and went away to school. Remember I had had one date. I mean, I had dated for two years but it was 102 really, um,...I went away from home at 18 to school I had only dated one person. And they had been very limited in the double dates or him coming to my house was all that I had done. So you now have the freedom of supposedly a curfew but I learned how to break the curfew really quick. ....So it ended up giving me more freedom than I knew how to handle. So I had a lot of passing, not friendships, acquaintances, a lot of people I got to know because of this new freedom, you know, that I didn't know how to handle. I never had a close girlfriend. (And about her experience of her woman's body when she was an adolescent:) When my father noticed that I was blossoming, I felt dirty and ashaned. And I can remember being, 13 or 14, I filled out quite young. And he would make a conment, and I felt very bad. In summary, women who score relatively highly on the Attanucci measures of Sun IV,”Self in self's terms, others in their own terms," and also on Quality of Object Relations in Early lemories, describe relationships with others in complex terms, but inevitably as contributing both to their own sense of self and to the well-being of the other. Homen who score relatively low on these same measures describe relations with others that reveal frustrations and emotional turmoil. Two women described no relationships with anyone that met criteria for Category IV. Two other women had some Category IV responses, but had low Quality of Object Relations in Early lemories Scores and were high scoring on the Brief Symptom Inventory. These women had described very difficult, even abusive, relationships with their parents when they were growing up. One of the two women was currently in theraPY. and the other was focusing intensively on her graduate school career. 103 Qualitative Analysis of Themes of the liddle Years of Life Although interview questions were developed to reflect the narcissistic issues which may be salient for the middle years of life, based on the analytic writings of Rernberg (1975/1985) and Colarusso & lemiroff (1981), in fact the interview process became much less structured than the protocol might indicate. Consequently some persons spoke at length on certain issues, and, for the sake of time, other questions which could have been asked were omitted. The review of interviews for material on these issues then was considerably complicated. In addition, it became evident that to do justice to the way one woman might describe significant transitions in her life, and to compare it with the way another would do so became a Herculean task. The overall plan to analyze the content of the interviews then became streamlined to reflect instead a broad general overview of those issues which were described well in at least some interviews. It became obvious that one issue which perhaps is more a male issue was not dealt with at all: that is, women did not address in any way a concern with facing limits of their own creativity. In general, when asked if they were still developing as women, they tended to say either ”yes” or else that they wanted to do this. Women tended to discuss significant transitions in their lives in most interviews, but these transitions were frequently tied up with changing relationships. For example, several women noted that they became more assertive in their relationships taking more time to meet their own needs, and that their husbands (or other important persons in their lives) adjusted in certain ways, usually positively. One woman 104 talked eloquently about how she was sure that her husband was proud of her when she left her full-time homemaking activities and went out into the work force. One of the issues which Rernberg and others mention as appropriate to middle aged persons is that of coping with losses and deaths. This issue did come up, of course, with a number of women, but the losses they talked about were not necessarily ones which happened recently. They might easily be very much moved by losses which occurred some years ago. One rather discouraging finding is the relative scarcity of women talking about female mentors (or even male mentors) who helped them at different points in their lives. Some women talked about the changing experiences of their bodies, but they tended to talk about changes experienced in adolescence rather than nore recently. Some women, however, talked about how their bodies were in better shape in recent years (because of exercising) than they were when they were in their 20s. One woman talked about changes she experienced when she lived through cancer surgery, and another talked about losing her teeth and getting dentures, which she found a horrible experience. Almost all of the women talked about changing relationships with family members - husbands, mothers, fathers, and children. A sample of some of the particular comnents follows. Significnnt Trnnsitions and Re-Assessnent of Relntionships 469 comments on how she went from an unhappy childhood into an ~ unhappy marriage. Eventually she started working outside the home and asked herself: ”Why am I always putting up with this for?” 105 Subsequently she and her husband went to a religious-based group experience called larriage Encounter, and she said about that experience: ”We just came back different people." During the larriage Encounter itself and later they worked through issues around faithfulness to the marriage and to a more satisfying (for her) sexual relationship. 868 said she has been in her second midlife crisis. Her divorce when she was in her early 30s was the first crisis. The second is because ”The relationship with my husband is changing somewhat, and that's really...I think it has to do with the fact that he's getting older....I feel like my life is sort of on hold." She reported that her life was very tunultuous this past year. Her mother had a stroke, and the woman herself was diagnosed with high blood pressure and cholesterol. Her husband has an ulcer and "he has this and he has that....I don't like not being 100 percent well. I think what's getting to me is his telling me over and over again he's getting old, he's getting old, and he's always sick. And if you tell yourself that long enough, you're going to be that way.” At a later point in the interview she predicted that, despite her desire to avoid it, this second marriage too might end in divorce. But she has considerable reservations: “Only the older you get, the harder it is to do that....because you have to really take stock of yourself and say, OK, are you going to spend the rest of your life alone? 'Cause that's what it now comes down to. And an I ready for that? And I am not sure I am ready for that. I have to feel 100 percent ready to do that because if I left this husband....I would definitely face the rest of my life by myself. First of all, I...don't have the energy to get, try and 106 find another relationship. And at (age), it's not very easy to do....I'm not there yet, not, not, not by a long shot." ”I do look for satisfactions. And so far its working.” 067 discussed positive transitions. In the last few years she says she has grown a lot and gained in confidence. She used to doubt she could handle things that take a lot of juggling - through the man who is in her life now she has gained confidence in herself. She also feels proud being a student, more than she did in her former job which was part-time and low-paying. Another transition was recovering from a devastating blow to her relationship with the current man in her life; she found out that he had considered asking another woman to marry him very shortly before he asked her to marry him. She felt that this meant a lack of real commitment on his part, and she then cancelled their present marriage plans. She describes a happy relationship with him in the present, but does not know if they will ever eventually marry. 467 had divorced her first husband after a series of trial separations. Her decision came when she felt "like I was in limbo constantly.” She described this feeling as ”married, but I wasn't married.” She described how her first husband became afraid of being emotionally hurt and pulled back away from a deeper involvement with her. Then she would pursue him. At times they would reverse roles and she pulled back and he pursued her. After some time this seemed to become an insurmountable problem, and they divorced, but in a friendly fashion. 065 spoke dramatically of how she came to love her first horse. She said that in her early 20s that her horse encompassed her entire life at that age. She discussed also the dramatic loss of her 107 horse by accident due to the fault of a drunk driver. She said about her horse: "You can never replace them." 464, a woman in her 50s, talked about her changing relationships with her mother, who is in her 80s. She described her attempts to get her mother to be "more verbal" in telling her that she loves her. She also talked about her lifelong sense that her mother thought that a sister was more intelligent and more accomplished than she was. Finally, she said, her mother realized that she had done well with her life, had married a fine man and raised a family to be proud of, and her mother became able to praise her, also. 463 discussed how she adjusted to the lower sex drive of her husband than herself. 460 discussed her transition from marriage to the status of single parent. She described the relations with her mother and former husband deteriorating at around the same period of time. In terms of transitions, she says: "What has shaped my life is the fact that I am a victim of incest.” She discussed how a stepfather molested her, and, when she had finally told her mother (as a child), the molestation stopped, but her mother ”never came back and talked to me or anything at all...then it was totally ignored...” In recent years she has entered therapy and worked through a number of issues around the incest and its lasting effects on her. 459 describes a change in the relationship to a teenage daughter. She spoke of a particular occasion on which she lost her temper, to her considerable shame and embarrassment, and how she and her daughter talked about their mutual responsibility to avoid such heated exchanges in the future. 108 These are not, of course, all the issues described by all the women, nor even necessarily the most dramatic ones. They do show, however, that change in relationships and personal transitions are very salient issues. The changes, however, do not seem to be a simple result of chronological age, but rather the outcome of working through issues with important persons in their lives. gnngern about Authenticity of tne Self 457, a divorced woman, discusses how she wants to be her "own person.” This came after a lengthy discussion of the problems in her former marriage and how she had consistently taken care of others: her husband and her children, without really taking care of herself and her own needs. She describes herself at this time, however, as ”selfish” because she is taking care of her own needs. 452 describes how she wants to find out what it means to be a woman. She describes a difficult relationship with her mother, and how she is trying to find role models of feminine behavior that differ from that of her mother, whom she describes as selfish. To some extent she has been successful, finding role models in women's literature and biographies. 469 described how her experience at larriage Encounter made her ask herself: ”What is your meaning of life? Why are you living?” Women were less likely to ask themselves in a theoretical way if they were authentic than to ask themselves if their relationships were authentic, and to seek to make relationships better. Enpnrience of Change in One's Own Body Some women, such as 464, talked about pleasure in how their bodies are now functioning. 464 said she is more physically active and 109 energetic now than she was in her 20s. She attributes this to becoming a runner in her 30s, and continuing this habit for about 20 years. She also changed the exercise habits of her husband and children to one of regular physical exercise. low joint exercise is a way of life for herself and her husband. 468, however, said about her body: ”I sure remember not feeling terrific about the body that I was blessed with....at 13 I weighted 160 pounds....so that was not a beautiful body as a child.” This woman, now in her 40s, currently appears to have a lean, muscular body. She talked about loosing her teeth and having to wear dentures. This she found one of the most devastating events of her life. 467 said that how her body looks and feels has a great influence on how she perceives her self and on her self image and confidence. She said that only after the age of 30 did she really enjoy sex. low she says that she ”totally enjoys sex," unless she is really tired or there is something on her mind. This woman has successfully lived through cervical cancer surgery and recovered her sense of pleasure in her body. Again, not all comments are reported here. It is clear, however, that the body is clearly a source of both pleasure, when it is functioning (and looking) well, and of pain (because of illness or some losses in function). Women who exercise regularly appear to have more pleasure in their bodies, on the whole, than those who don't, in the middle years. View of Oneself nnnund tne Possibilitynof Furtnnn Development as a Woman Women seem considerably optimistic about the possibility of further development. They conceptualize development both in terms of 110 personal experiences: "I can always learn something more...." or in terms of development in relationships. Only one woman said that she thought that her time of further development was over, and that she had ”peaked” in her mid 30s. 4 51 said: ”I never stop learning and stop changing. lever. I see myself as being a better lon.” (Because she has learned to balance her busy schedule of work with the demands of her teenagers.) 455 says: ”Oh, yes, definitely. [Every year she finds things she wants to do.] She has learned that: ”I need to do something for myself. ly weight is going crazy, and I'm riding my broomstick a lot...I found that once I start doing the exercise, it gets rid of a lot of that....If I can get that two or three times a week, it really helps me keeping myself on a more even keel....And I feel like I'm doing something for myself. I said I don't look any better, but I feel a lot better, you know. I feel like I'm taking care of myself and more and doing this for myself.” 467 says: ”Yes. I've just learned not to care what other people think of me and it's such a liberating feeling....But things seem to work out much better for me when I adopt that attitude....I guess I just have more confidence as a person and I feel that I am more worthy than I used to be.” Presence or Absence of Fennle lentors Very few wonen mentioned female mentors, although this may be a related to the small number of women who were in either high corporate levels or else in advanced ranks in an academic profession. One woman called a woman friend who helped her at the time of her divorce from her first husband a ”mentor,” but this would not be the ordinarily accepted 111 meaning of mentor by most women in the study. One woman described her mother's influence on her in such a way that this might be seen as the equivalent of a ”mentor.” Re-assessment of the leaningnof Time Only one woman, 465, talked in any detail about re-assessing her use of time. She talked about trying to make priorities among the current professional activities in her life, and especially trying to make time to do the fun things that she and her husband wanted to do some years ago. Other than that, only one woman mentioned time, and this was in regards to trying to figure our a new career for herself when she retires from her current one which she enjoys but which she plans to retire from in about ten years. Coping with Losses and Death This was an issue that most women discussed, but the losses were not necessarily connected with their age. The deaths of parents, siblings, or friends at any age brought forth significant feelings of mourning. Losses in relationships due to separations and divorce were also discussed, as were losses of bodily functions. levertheless, this was not as central an issue for this sample of women as one might expect. In summary, many of the issues raised as salient for middle-aged persons by Kernberg and others were in fact discussed by the women in this sample, but the issues did not seem to be strongly correlated with just chronological age, but, rather, certain events needed to happen to precipitate these issues. A death of a loved one, for example, can happen to any one at any age, although wonen are more likely to lose persons of the older generation when they are themselves in their middle 112 years of life. It is an empirical question, not answered by the present inquiry, if the same content would be elicited from younger (or older) women, or if the pervasive importance of the theme in a woman's life made her look back on examples of it in her former life because it is salient now in her present state of mind. DISCUSSION This research studied middle-aged women who were involved in self care in what seems to be a responsible, healthy manner. It investigated the meaning of their behaviors in the context of the women's present conscious experience of self and of recollections of past development and relationships. The research was composed of individual interviews with 28 middle-aged women between the ages of 35 and 55 who have ever had children, and who are now either students or exercisers (or both). The subjects were selected through a telephone screen and subsequently interviewed twice. The second interview was spaced two weeks to one month later than the first. The first interview began with administration of a general informational form and the Brief Symptom Inventory. Questions about the conscious experience of the self and of relationships, in the past and in the present, were asked following a general protocol administered in an open-ended clinical fashion. Participants completed the Revised Relational Self Inventory (RRSI) at the second interview, which also involved open-ended questions about issues considered salient for the middle years of life. Based on examination of BSI results, four persons were identified as psychologically distressed, and of questionable inclusion in a group which is hypothesized to possibly exemplify ”normal” development. Results were subsequently reported both with and without the scores of these four women. 114 There were two coded transcripts per case. One had the Early lemories responses: the second had the responses to the Attanucci protocol plus some additional general interview questions which were developed to reflect those issues and concerns about the self that were identified by Colarusso and lemiroff (1981) and Rernberg (1975/1985) as narcissistic concerns of the middle years of life. The research question addressed the potential relationship between the highest form of development in Gilligan's model (1982) of the development of care in women and the achievement of advanced forms of relationship in psychoanalytic object relations theory. The two specific hypotheses varied the mode in which Gilligan's highest perspective was measured. The Attanucci research question protocol and the Revised Relational Self Inventory (RRSI) provided the means of studying the women's highest developmental perspective from Gilligan's model (1982). The Quality of Object Relations in Early lemories Scale (Ryan, 1973) provided the measurement of "lutuality of Autonomy in Object Relations," which was hypothesized to be similar to Gilligan's highest perspective. The analysis of some case material provided some additional descriptive information of the women in this pilot research. Thematic material was reported in clinical fashion. A correlational analysis revealed that Sum IV (which is the total of types of relationships which can be described as containing the perspective of "Self in self's terms and other in their own terms”) does correlate significantly and positively with one measure of object relations: ”Quality of Object Relations of Earliest lemory of Father.” This finding was true for both the complete sample and the restricted 115 sample, which excluded the women with high psychological distress. This provided some confirmation for the first hypothesis. This measure of Gilligan's perspective does not, however, correlate with the two other object relations measures: ”Quality of Object Relations of Earliest lemory” and ”Quality of Object Relations of Earliest lemory of lother." lo person in the sample achieved an object relations score in the range of 15 to 20, which is the range in which ”lutuality of Autonomy" is proposed to occur. Those relatively healthy scores that were obtained were in the 10 - 15 range, which would be considered "neurotic" rather than "normal.” Thus, there is support for stating that Gilligan's highest developmental perspective does relate to higher forms of object relations. ”Self and Other Care Chosen Freely," Gilligan's highest perspective as measured by the RRSI, does not correlate with any object relations measure. Thus the second hypothesis was rejected. Cluster analyses were done with the two Gilligan measures and the object relations measure. Analyses of the Attanucci category scores suggest that there are two types of persons who score highly on Sun IV: women who are also high on Sums II and III, but not on Sun I: and women who are lower on Sums II and III, but high on Sun I. This suggests that women who are in a transition period could show behaviors at several levels, while there are also some women who show more consistent relationships at the highest level, where they have presumably consolidated their capacities. Cluster analyses of the RRSI variables gave mixed results, not clearly supporting a developmental hypotheses. When the two measures of Gilligan's highest perspective were examined in cluster analyses, an 116 interesting pattern emerged. There appear to be two types of women who score highly on SOCCF: women who are also high on Sun IV, and women who score quite low on Sun IV. The latter may be persons who want to score highly on SOCCF because they value such attitudes highly, but this may be precisely because they are having difficulty achieving Self and Other Care Chosen Freely at this time. This is speculative, of course. Finally, cluster analyses of Attanucci categories and scores on Quality of Object Relations in Earliest lemories of Father show clear differences between persons who score highly on both types of variables and who score very low in early memories. The data are preliminary, but support theoretical overlap of Gilligan's highest developmental perspective and advanced forms of object relations. The findings can be discussed from both theoretical and operational perspectives. It is fascinating to consider why women's conscious ability to relate in a caring way both with self and other might be related to the quality of her object relations with her father, as she remembers them in her earliest memory of him. First of all, I am sure it does not mean that the woman's relationship with her mother is unimportant. I presume that a good relationship with her mother is necessary for the woman to develop her capacities for object relations. levertheless, it might be easier for her to have good object relations with a wide range of people if she had not only a good relationship with her mother, but also a good relationship with her father. There might even be a qualitative difference between the ability to care for and relate to persons like oneself (one's mother) and the ability to care for and relate to persons who are different from oneself (one's father). If there are problems in 117 relating to the person who was (hopefully) closest to you, but different (one's father), then it might be extremely hard (but not impossible) to develop the capacity to relate to a wide range of different people as an adult. I say ”wide range” of people because the measure, Sum IV, refers to all the kinds of relationships discussed by women. There are other theoretical concerns which relate to the object relations' measure and its problems in administration and scoring. First of all, it was not possible to obtain standardized early memories to train with, and so the raters used a pilot set of early memories from a written portion of another research project. Although the reliabilities on this pilot study were quite acceptable, when the raters turned to the transcriptions of early memories in the research sample, reliabilities went down. Specifically, there was some concern about the low level of average of two raters for scoring of Quality of Object Relations of Earliest lemory of lother, and the consensus score was used in lieu of an average score. This appeared to be a result of the fact that the early memories in the research study were obtained from transcription of interviews and were rich and complex (and ultimately harder to score than the more simply phrased set of written memories on which training took place). Another theoretical issue was that the object relations measure showed a variability across its three components: frequently there was as much as a five or six point spread in rating across the different memories. What does this mean in a theoretical sense? Is this then measuring something which is not stable? Can one's level of object relations vary depending on who one is in a relationship with? It was assumed in the present research that it should be relatively stable. 118 However, the data are more consistent with considering object relations a function not only of the individual's inner object world but also of the person with whom one is in interaction. Theorists who score the object relations content of Rorschach and TAT protocols come up with a range of from lower to higher functioning, and those who rate the manifest content of dreams (which has been coded by other students of layman for level of object relations) also find variations and fluctuations. It is not unusual for persons to show a range of possible scores on object relations measures in different instruments, especially projectives, and researchers have argued that it is useful to consider median scores rather than to accept any one score, or to decry the range of scores. Thus, in the present case, it might have been preferable to use a median score as a single object relations score. This was not done in the present study, as it seemed to imply a waste of what could be other rich data for interpretation. After all, the early memories question and its answer takes very little time to administer, in a practical sense, conpared to the lengthy Attanucci protocol. There are other theoretical considerations which should be addressed. Why do the operational measurements of the Gilligan perspectives not correlate with each other? Both of these place importance put on the woman's conscious (as opposed to unconscious) experience of her phenomenal self. Cluster analyses indicate that there are some women who score highly on Sun IV as well as on SOCCF, but results are difficult to interpret because there are also some women who score highly on SOCCF but low on Sun IV. It is possible that the latter 119 group of women are persons who value SOCCF as an ideal, but have not yet achieved it. As a beginning mode of addressing this issue, some problems with coding the Attanucci categories are addressed. Clearly the Attanucci measures take the woman at her word: if she says there is a mutual respect and care-taking in a relationship as well as mutual healthy care for self, then it is so coded. (This is true for other measures of less interest in the Attanucci coding, except it is not true of the coding of Category I, which is a rater judgment of a stereotypical role response). However, it is harder to give a rounded out presentation of mutuality in a relationship than it is to check off a behavior on a questionnaire. There are specific problems with coding the Attanucci categories, and they are described as follows. (1) The coding referred to relationships both in the present and in the past. The analyses did not reveal the picture some women painted of a happier present and more troubled relationships in the past. All relationships were coded as if they were all happening simultaneously, whereas it might have been better to have separate analyses by chronological time periods in a woman's life. (2) The coding of Category III included a wide range of foci on the self, which seemed to include what Gilligan called her first perspective in 1982, as well as parts of the self responsibility which could be considered present in the third perspective. Phrased in other words, this category seemed to include what psychoanalysts would call aspects of both healthy and less healthy narcissism. (3) Category I seemed to be out of synchronization with the other categories because its coding involved a rater judgment of a 120 stereotypical role relationship, while all other categories were coded on the basis of the woman's conscious elaboration of them. (4) There seemed to be a fairly large number of persons for whom Category I responses were coded about statements made in the beginning of the first interview. This led to a larger than expected number of Category I responses compared to what was found by Attanucci. However, considering the fact that the interviewer was presumably known to the research subjects in her work, whereas I was not known to the women in this case, this seems an ample explanation of why stereotypical responses might occur at the beginning of a new relationship. Only two persons had no Category I response at any point in the interview. (5) Every woman had at least one Category II and Category III response. The raters were particularly apt to disagree about the occurrence of these responses. The younger (student) raters tended to see Category II responses in a statement that the older (researcher) rater might see as reflecting a Category III response. They would compromise and discuss, but the pattern continued. It is not known in what way this may have biased results. (6) It was difficult to decide what was the beginning and end of any particular relationship incident. Thus the coding for presence or absence overall of the category in a relationship seemed to be most fair. (7) Sometimes it seemed that when there is a three-person interaction it should receive a different coding than a two-person one. For example, the relationship of a woman to her child is almost certainly influenced by her relationship to her husband, but there was no way to account for this in the present coding. 121 (8) Because the interviewer allowed the subject to have considerable leeway on the most important relationships or developmental issues to discuss, sometimes a very thorough discussion of one relationship precluded (because of time) a more thorough discussion of other relationships. In general, the less time spent discussing a relationship, the less likely it was that a Category IV coding mas made. Thus some persons may seem to have less Category IV responses than they were in fact capable of having. Taken all together, then, the Attanucci coding was quite difficult and these problems need to be considered when weighing the overall study results. The other measure of Gilligan's perspectives, the RRSI, is easier to administer and code, but it also has its unique problems. A woman's report is also taken at face value in the pencil-and- paper administered RRSI. This instrument, however, has the usual kind of problem associated with a very straightforward measure, namely, it can be answered in a socially (or personally) desirable fashion. There was a group of women who were high on SOCCF but low on Sun IV, as indicated in cluster analyses, and these seemed to be the women who answered the instrument in the manner that indicated they valued the attitudes described, but, it is argued, that since they scored low on Sun IV, they do not yet possess these behaviors. Also, there is a problem in any instrument that tries to measure a complex variable. Perhaps the need to chop up the whole idea of ”Self and Other Care Chosen Freely” leads to expressing the complex idea in a series of statements which, if endorsed as a whole, indicate that one has achieved the highest perspective. But it may also be possible to endorse a number of items which make up the scale without endorsing the 122 whole idea of "Self and Other Care Chosen Freely." The item which has the largest correlation with the total scale for women is ”I want to be responsible for myself." This can be endorsed, I argue, without having achieved Gilligan's highest perspective. But this does not mean that it is not a necessary component of Gilligan's highest perspective: it is necessary, but not sufficient, for evidence of having achieved Gilligan's highest developmental level. This might explain why certain brave women, who are struggling against considerable odds of very stressful life histories, and who also scored relatively high on "Self and Other Care Chosen Freely," may have done so because of a firm desire to overcome their past and present problems, and a desire to be seen as having conquered over considerable odds. Such women did not achieve Gilligan's highest perspective as it was measured in the Attanucci schema. Another theoretical concern is whether the kinds of responses given to the two measures of Gilligan's highest perspective might in fact have been measured satisfactorily, but that these perspectives do not in fact form a developmental continuum. The research was not set up to examine this, but the results still can be examined to see if they support a real difference between women who seem to be at Gilligan's second perspective, that of focus on care for others, and women who have achieved the third perspective, that of caring for both self and others in a mutually beneficial fashion. Operationally, this would imply that there would be a difference between women who endorse Category II statements (Self instrumental to other, others ”in their own terms”), and women who endorse Category IV statements (Self in self's terms and other in their own terms). In the restricted sample there was, in fact, 123 a small negative correlation (but not significant) between Sum II and Sun IV. There was a small positive correlation between the two in the complete sample. Similarly, POC (Primacy of Other Care) and SOCCF (Self and Other Care Chosen Freely) showed small positive (but not significant) correlations in both sanples. The results, then, are equivocal about the possibility of a developmental continuum. There are some other interesting findings for the Gilligan theories as measured by the RRSI. Both the measure of the Relational/Connected Self (RI) and the measure of the Separate/Objective Self (SO) show significant correlations with "Self and Other Care Chosen Freely” (SOCCF) in the restricted sample. SO is correlated with SOCCF in the complete sample, and the relationship of RI to SOCCF approaches significance (at p < .06). RI is positively related to S0, but not significantly. These results differ from the the much larger study of Strommen and colleagues (1987) in that no subgroup of theirs showed a significant relationship between RI and SO. However, this finding could be an artifact of the conditions under which this sample took the instrument. The RRSI was administered at the very end of the second interview, and they had been asked questions about their interest in personal achievements, and they may well have responded to what they thought was the researcher's expectation that they be both autonomous and caring for others. The lack of relationship between achievement of the highest Gilligan perspective on the Attanucci measure and the RRSI still remains to be addressed from another viewpoint. Perhaps some of the more recent theoretical work from Gilligan and her colleagues may lend another way of interpreting the findings. First of all, since 1982 Gilligan has 124 moved away from a relatively bipolar view of moral cognitive perspective, either from a care or a justice perspective, to one in which she explains a relatively predominant care perspective which includes its own version of justice in its purview, as well as a justice perspective which has its own version of care. The RRSI was developed from her ideas as expressed in 1982. Perhaps the statements which comprise SOCCF in its present form represent mainly ”Self and Other Cars Chosen Freely" from a justice ,permpective.” This would explain its importance to women, especially for those women who had been victims of injustice (e.g., victims of incest). The Attanucci model for Category IV may in fact reflect mainly ”care for self and other from a care perspective.” These both would validly reflect a mutuality, but yet explain their variation. The developers of the RRSI provide infornation on item-total correlation, corrected for item overlap, for women in their sample (from Appendix I). In those items that make up the SOCCF scale, the two items with the highest item-total correlation are: ”I want to be responsible for myself.” (.63 item-total correlation) "I want to learn to stand on my own two feet.” (.53 item-total correlation) These statements, if they were made in the interview situation and coded with the Attanucci manual, would be coded as Category III statements ("Self in self's terms,” other instrumental to self). On the other hand, the next highest item of the SOCCF scale reads as follows: ”In order to continue a relationship, it has to let both of us grow." (.52 item-total correlation). This statement would be coded as the highest level, Category IV. Other items which make up the SOCCF also 125 would be coded differently by the Attanucci criteria if they were said spontaneously. For example, another item in the SOCCF scale is: 'If I am really sure that what I want to do is right, I do it even if it upsets others.“ (.42 item-total correlation). I suggest that this item reflects mutuality fron a justice perspective. This would be coded a Category III (by Attanucci) unless it was clear that the person had thought through the effect of their actions on the other. In other words, this is the type of statement which I propose reflects ”care from a justice perspective” which I attribute to the RRSI measure of SOCCF. This is not to say that SOCCF does not reflect ”care from a care perspective”. That, I propose, it also reflects in such statements as "True responsibility involves making sure my needs are cared for as well as the needs of others " (.38 item-total correlation). I argue, based on the importance of the above items to the total scale, that women who are in a more feminist position would embrace Gilligan's perspective as measured by SOCCF, yet any women for whom oppression was (or is) a significant issue would embrace such a perspective. I propose that this explains why women with psychological distress due to oppression by others would score just as highly on SOCCF as those persons who have much less psychological distress. Case studies provided additional descriptive information about the relationship between the constructs from the differing psychological views. In general, women who scored more highly on the Attanucci Category IV tended to give very complex and rich responses to requests to describe their relationships, in which caring and frustration abounded, but caring was predominant. Problems and misunderstandings 126 were acknowledged and faced openly, on the average, and relationships stayed together without breaking under the strain of examining their premises. Women who scored in the lowest levels, with no Category IV responses for any type of relationship, tended to describe more uniformly negative relationships with others, especially parents (who may have been neglectful or abusive). The highest and lowest scorers on Level IV tended to be in clusters (in the cluster analysis procedures) in which persons also had correspondingly high or lower levels of object relations. However, the middle clusters were relatively uninterpretable, with persons with a mixture of high and low scores on the various measures. Of course, all the other scores on the other categories of the Attanucci model, also contributed to production of clusters. lo mention was made of the variation in other scores in the results chapter. That was due to the fact that they were not an object of specific research hypotheses. The transcripts provided considerable information about the themes of the middle years of life, although the content of these was considerably abridged for reporting here, due to the enormity of doing justice to the women's words. 0n the whole women considered that they either were still developing as women, or wanted to, in the way that they personally define development. Women tended to see capacity to learn as equivalent with capacity to develop; in addition, they saw constant ways of improving their relationships, which also contributed to a sense of personal development. Finally, issues such as facing inevitable losses and deaths were described in considerable detail, but a woman might be just as verbal in describing a loss she experienced 20 or so years ago as a loss another women experienced in the present year. 127 In other words, losses are a salient issue for middle-aged women, at least in this sample, but they do not have to have happened literally in the recent past. Women also come up with a number of different ways of coping with losses, or anticipated losses. Losses seem to be defined mostly in terms of relationships, although losses in one's person, such as losing one's teeth and needing to wear dentures, are also quite important concerns. lo woman talked about coming to terms with the limits of her creativity, which is interesting, as this is hypothesized by Rernberg to be an important issue for middle-aged persons. Women in the sample did show creativity in a number of areas in their discussions, so this does not seem to be omitted because the women weren't creative in the first place. lost likely the women define their creativity in terms of the ways they can still develop - which means in personal learning and experience and in relationships - and perhaps in the use of their body in exercise - and so they would not think they had lost this capacity. Time does not permit doing full justice to the range of issues that women discussed. levertheless, the inclusion of these concerns in the women's own words reflects a tendency towards which Gilligan and her colleagues appear to be moving, namely, to move away from criticisms and judgments of women's words and instead to work to explicate and clarify the words, and how the women say them. In other words, Gilligan's movement toward a more hermeneutic perspective clears the way for what psychoanalysts would call ”clinical examples" which in fact may teach far more than the theoretical words in which they are encapsulated. In other words, each woman says more than our research categories can capture, and it is in the best research interest to try and capture the 128 issues of the middle years of life as they are experienced by this group of wonen, and not as one would expect them to be from pre-formed categories. A number of issues were raised by the existence of women whose scores on the BSI indicated significant psychological distress. One woman had no evidence of significant distress outside her score on the BSI. She was one of the younger women, and she still had young children at home. Although she evidenced an obsessive style which she humorously described as getting in her way in her relations with her children, and as she also mentioned that she had a chronic illness, there were points of stress in her life, but nothing otherwise that would seem to rate excluding her from an analysis of ”normal” women. Also, one woman told of a history of hospitalization for serious depression and suicide attempts, yet the BSI, at least in the conservative manner in which it was used in this study, did not flag her for exclusion from a group of "normal” women. In conclusion, like any useful study, this research raised at least as many questions as it answered. The women seemed to enjoy it, in general, and it is a point of pride that every woman who came to the first interview showed up for the second interview, a zero dropout rate. This seemed to be partially due to the open-ended nature of the first interview, and that the women who volunteered for the study seemed to want to talk about themselves, and they had plenty of opportunity to do just that. In addition these women were lively, humorous, hard-working, serious, and a pleasure to interview, even when the material in the interview was painful. 129 If the research could be done again or followed up, it would be useful to play with different ways of coding the Gilligan perspectives which would be variations on the Attanucci model. For example, Category I seems to be an oddball in her coding system because it involves a judgment on the part of the raters, whereas all other codings come more directly from the woman herself. It would be useful to investigate the flow back and forth, if it exists, of Category II and III responses, and also to examine those women who are high on Category IV and also Categories II and III, or else high on Category IV and high on Category I. Can these women, in fact, be placed into transitional stages, with Category I explained differently? It is not a one-to-one match for what Gilligan called her first perspective. That was perspective was aptly labeled ”self care from need,” by the lichigan State group which created the original RSI, however, this instrument also lost its capacity to measure this variable when it was revised and expanded. This seems to be a loss, because the raters using the Attanucci Category III frequently remarked on how the focus on taking care of the self seemed to vary from relatively weak and defiant self isolation to a more mature and healthy emphasis on the self's needs. However the possibility of investigating healthy narcissism, as it would be called in self psychology, was impossible with the present instruments. This is true, also, with the measure of object relations, which also did not have what would be called a way of coding for healthy narcissism. There is a need for an instrument which measures both healthy self love and self care and also healthy relationships, and how the care of the self and the other interrelate in the present and across time. 130 The object relations scale did not seem to have a good midpoint. In general the scale tended to have a lot of pathological ways of viewing the self and relationships at its low end, and very advanced forms at the other end, with not much in the middle. This was true even though it was a 20-point scale. It probably would have been better to use its 6 or 7 iten analog developed by layman and Rrohn in the early 1970s. Also, there were markers for pathological narcissism, but not for healthy narcissism. It claimed theoretical roots from Rohut, but really Rohut had not developed self psychology much by the early 1970s, and so the scale does not reflect his further ideas on transformations of narcissism. If time and financial support permitted further analysis in the present study, each woman's personal story would be studied in detail to investigate the intricate ways in which the woman expresses her sense of self in relationships and how the self experience varies across time and across relationships. It would also be useful to look for gaps and omissions, for what women avoided talking about, either because of personal reluctance or because of lack of probing on the part of the interviewer. The research question itself, about the comparability of Gilligan's highest perspective and higher levels of object relations, still invites further investigation. Do the Attanucci measure of Category IV and the RRSI measure of SOCCF both measure Gilligan's highest perspective, as she described it in her work in 1982? If so, why don't these measures correlate with each other? Can the lack of correlation really be explained by discussing ”Care from a care perspective" vs. "Care from a justice perspective?" 131 Why is it that the quality of the relationship with the father seems to be more important to a woman than that of her earliest memory of mother? Is this just an artifact of a pilot study, or will it hold in other situations? Could it possibly mean that if women solve their Oedipal issues with their fathers they will have achieved a higher developmental level in their current relationships? Since the data on relationship with fathers was buried, in the Attanucci coding, in with "other," the present research did not correlate "Quality of Object Relations of Earliest lemory of Father” with Category IV statements with regard to father. This is certainly a useful next step in teasing out some meaning to these findings. And now, having reached the limits of my creativity (even though no one in the sample seemed to worry about that), I'll end. LI S'l' OE REFERENCES LIST OE REFERENCES Alpert, J. L. (Ed.) (1986). Psychoanalysis and women: contemporary remppraisals. Hillsdale, lew Jersey: The Analytic Press. Attanucci, J. S. (1982). How would you describe yourself to yourself: lothers of infants reply. Unpublished qualifying paper, Harvard Graduate School of Education, Boston. Attanucci, J. S. (1985). lothers in their own terms: A developmental perspective on self and role. (Doctoral dissertation, Harvard University, 1984). Dissertation Abstracts International, 46, 6708. Bell, l. 0., Billington, R., and Becker, B. (1986). A scale for the assessment of object relations: reliability, validity, and factorial invariance. Journal of Clinical Psychology 42, 733-741. Bell, l. 0., letcalf, J., 4 Ryan, E. R. (1979, September). Reality testing and object relations: A self report instrument. Paper presented at the 87th Annual Convention of the American Psychological Association, lew York. Bell, l.D., letcalf, J., 4 Ryan, E. R. (1980, September). Reality testing-object relations assessment scale: Reliability and validity studies. Paper presented at the 88th Annual Convention of the American Psychological Association, lew York. Bellak, L., Hurvich, l., R Gediman, H. R. (1973). Eye functions in schizophrenics, neurotics and normals. lew York: Wiley. Benjamin, J. (1986). The alienation of desire: Women's masochism and ideal love. In J. L. Alpert (Ed.), Psychoanalysis and women: contemporary reappraisals (pp. 113-128). Hillsdale, lew Jersey: The Analytic Press. Blank, L. S. (1988). The relational self in women: Perspectives on self and others. (Doctoral dissertation, lichigan State University, 1988). Blashfield, R. R. (1980). Propositions regarding the use of cluster analysis in clinical research. Journal of consulting and Clinical Psycholoar. 48 (4), 456-459. Blatt, S. J., Brenneis, C. B., Schimek, J., 8 Glick, l. (1976). lormal development and psychopathological impairment of the concept of the object on the Rorschach. Journal of Abnormal Psychology 86, 364-373. 133 Borgen, F. H. 8 Barnett, D. C. (1987). Applying cluster analysis in counseling psychology research. Journal of counseling Psychology, 34(4), 456-468). Bromley, D. B. (1977). Personality description in ordinary language. lew York: Wiley. Burke, W. F., Summers, F., Selinger, 0., R Polonus, T. W. (1986). The comprehensive object relations profile: A preliminary report. Psychoanalytic PschOIOEY. 3(2), 173-185. Chodorow, l. (1982). The reproduction of mothering: Psychoanalysis and the sociology of gender. Berkeley: University of California Press. Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological leasurement, 20 (1), 37-46. Colarosso, C. A. 4 lemiroff, R. A. (1981). Adult development, a new dimension in psychodynamic theory and practice. lew York: Plenum. Coonerty, S. (1986). An exploration of separation-individuation themes in the borderline personality disorder. Journal of Personality Assessment 50, 501-511. Derogatis, L. R. (1977). The SCI-9O lanual I: Scoring, adninistration, and procedures for the SCI-90. Baltimore, lD: Johns Hopkins University School of ledicine, Clinical Psychometrics Unit. Engelman, E. (1985). Level of object representation in depression: A longitudinal study. (Doctoral dissertation, Adelphi University, 1985). Dissertation Abstracts International, 46, 1334B. Gilligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, lass.: Harvard University Press. Gilligan, C. (1987, June). Two moral orientations: Justice and care. (pp. 8-34). In: Brown, L. l. (Ed.). A Guide to reading narratives of moral conflict and choice for self and moral voice. Unpublished manual, Harvard University. Gilligan, C. (1988, April). Psyche embedded; the place of body, relationship, and culture in personality theory. Henry A. lurray Lectures in Personality. lichigan State University, East Lansing, lichigan. Hale, W. B., Cochran, C. D., 4 Hedgepeth, B. E. (1984). lorms for the elderly on the Brief Symptom Inventory. Journal of Consulting and Clinical Psycholocy, 52 (3), 321-322. Jacobson, E. (1964). The self and the object world. lew York: International Universities Press. 134 Johnston, R. (1985). Two moral orientations--Two problem-solving strategies: Adolescents' solutions to dilemmas in fables. (Doctoral dissertation, Harvard University, 1985). Rernberg, O. F. (1975/1985). Borderline conditions and pathological narcissism. lew York: Aronson. Rernberg, O. F. (1980). lormal narcissism in middle age. In, Internal world and external reality: Object relations theory mpplied (pp. 121-134). lew York: Aronson. Rernberg, O. F. (1980). Pathological narcissism in middle age. In, Internal world and external reality: Object relations theory mpplied (pp. 135-155). lew York: Aronson. Rernberg, O. F. (1966). Structural derivatives of object relationships. International Journal of Psychoanalysis 47, 236-253. Kernberg, O. F. (1976). Object relations theory and clinical psychoanalysis. lew York: Aronson. Rohut, H. (1971). The analysis of the self. International Universities Press. Rohut, H. (1974/1978). Remarks about the formation of the self. The search for the self ( pp. 737-770). Vol. 2. lew York: International Universities Press. Rrohn, A. (1973). Level of object representation in the manifest dream and projective tests--A construct validation study. (Doctoral dissertation, University of lichigan, 1972). Dissertation Abstracts International, 33, 55208. Litwin, D. (1986). Autonomy: A conflict for women (pp. 183-312). In: Alpert, J. L. (Ed.): Psychoanalysis and women: contemporary reappraisals. Hillsdale, lew Jersey: Analytic Press. Loevinger, J. (1979). Theory and data in the measurement of ego development. In J. Loevinger, Scientific weys in the study of ego development (pp. 1-14). Worcester, HA: Clark University Press. Lyons, l. P. (1982). Two perspectives: On self, relationships, and morality. Harvard Educational Review 53, 125-145. lahler, l. S. (1971). A study of the separation-individuation process and its possible application to borderline phenomena in the psychoanalytic situation. Psychoanalytic Study of the Child 26, 403-424. lahler, l. S., Pine, F., 4 Bergman, A. (1975). The psychological birth of the human infant. lew York: Basic Books. 135 layman, l. (1967). Object-Representations and object relationships in Rorschach responses. Journal of Projective Techniques and Personality'Assessment 31, 17-24. layman, l. (1968). Early memories and character structure. Journal of Projective Techniques and Personality Assessment 32, 303-316. layman, l. and Ryan, E. R. (1973/1974). Quality of Object Relations in Early lemories. In E. R. Ryan. The capacity of the patient to enter an elenentary therapeutic relationship in the initial psychotherapy interview (pp. 241-255). (Doctoral dissertation, University of lichigan, 1973). Dissertation Abstracts International, 35, 522B. lilligan, G. W. 8 Cooper, l. C. (1987). lethodology review: Clustering methods. Applied Psychological leasurement, 11(4). 329-354. lorusis, l. J. (1985). Cluster analysis. In, SPSS-X: Advanced statistics guide (pp. 165-191). lew York: chraw-Hill. Pearson, J. L., Reinhard, l.A., Donelson, E. F., Strommen, E. A., R Barnes, C. L. (1985, April). larital status and Gilligan's developsental theory of care of self and other. Paper presented at the Biennial meeting of the Society for Research in Child Development, Toronto, Ontario, Canada. Reinhart, l. A., Pearson, J. L., Ramptner, l. L., Cornwell, R., Barnes, C. L., Strommen, E. A., a Donelson, E. F. (1985, April). Assessment of Gilligan's model: Developsent of the relationship self inventory. Paper presented at the Biennial meeting of the Society for Research in Child Development, Toronto, Ontario, Canada. Rogers, A. G. (1987). The question of gender differences: Ego development and moral voice in adolescence. (in press). Roiphe, H. 4 Galenson, E. (1981). Infantile origins of sexual identity; lew York: International Universities Press. Romesburg, B. C. (1984). Cluster analysis for researchers. Belmont CA: Lifetime Learning Publications. Ryan, E. R. (1974). The capacity of the patient to enter an elementary therapeutic relationship in the initial psychotherapy interview. (Doctoral dissertation, University of lichigan, 1973). Dissertation Abstracts International, 35, 522B. Ryan, E. R. (1984). Changes in object relations from psychosis to recovery. Journal of Abnormal Psychology 93, 209-215. SPSS-x'user's guide. (3rd ed., 1988). Chicago, SPSS, Inc. 136 Strommen, E. A., Reinhart, l. A., Pearson, J. L., Barnes, C. L., Blank, L. S., Cebollero, A. l., Cornwell, R., & Kamptner, l. L. (1987, April). Assessment of Gilligan's model: The Revised Relationship Self Inventory. Paper presented at the Biennial leeting of the Society for Research in Child Development, Baltimore, lD. Urist, J. (1973). The Rorschach Test as a multidimensional measure of object relations. (Doctoral dissertation, University of lichigan, 1973). Dissertation Abstracts International, 34, 1764B. Urist, J. (1980). Object relations (pp. 821-833). In: Woody, R. H. (Ed). Encyclopedia of clinical assessment. Vol. 2. San Francisco: Jossey-Bass. Winnicott, D. W. (1960). Transitional objects and transitional phenomena. In, collected Papers. lew York: Basic Books. APPENDIX A CODING NANUAL FOR SELF IN RELATION TO OTHERS 137 Coding lnnnnlnfprn§e1f_in Relation to Others Based on Jane Attanucci, 1984 1 Figure 1 indicates the general schema for considering the four categories for coding the self in relation to others. The categories are described next. First there is the description provided by Attanucci. lext there are comments about coding which are keyed to subject numbers. These are clarifications and elaborations developed in the present study. lext, there are examples of coding for the types of relationship: husband, children, mother, and ”other.” Figure 1 Attanucci's Hodel of Women's Development (from Attanucci, 1984, p. 104) Self instrumental to Self in self's others terms Perspect- Others "in their own Others in their ive terms” own terms II IV toward Self instrumental "Self in self's others to others terms" Other instrumental Other instrumental to self to self I III Perspective Toward Self General notes: 1. Reference to influence of,parents on the subject as she was growing up will be coded for a relationship, so as not to lose information. This may be coded to mother, father, or both, depending on the level of continuity of the statement about psrents with other statements about mother or father (or lack of them.) If in doubt, it will be coded for mother. 2. Double coding is possible when discussing events happening concurrently (not developing from lower to higher relationships over time) when this appears to be a real change in the form of the 1 Paragraphs in italics are elaborations and clarifications added by this researcher. Considerable help was obtained from discussion with the research assistants Sharon Spryszak and Annie lahle. 138 relationship. Be most careful about double coding with IV.because if more information were known, it might be entirely coded IV. 99t¢V°{Y_IE Self instrumental to other (role), other instrumental to self (role) When self description reveals an understanding of self and other in reciprocal roles, the self and other are described from an objective, third person perspective. As such, there appears to be no differentiation between the self and the role (nor between the other and the role). This understanding of self and other in reciprocal roles conveys a sense of self and other performing mutually beneficial functions in a rather closed and static system. Conflict between self and other is not explicitly acknowledged. Category I: o reciprocal roles: self instrumental to other and other instrumental to self 0 objective, third person perspective 0 unreflective and unelaborated quality 0 conflict between self and other not explicitly acknowledged Husband: I. This category may be the repository for statements which are not adequately elaborated for the reader to understand how the woman sees herself in relation to others (”in whose terms"). Responses in this category have the objective, third person quality that describes the traditional or equalitarian roles, wherein, the people playing the roles are rather faceless or interchangeable. The role descriptions are often idealized. Statements in this category suggest that the couple apparently performs these roles with little personal or interpersonal conflict. 1. ”Our marriage is traditional. I am flexible and agree with his decisions about family finances and I make decisions concerning the house and the children. It just naturally works out that way.” I [454, 9 (1), Page 5] ”I felt finally'that I had that person rho was there 1008 for me...that I always had needed." comments: Because of simplistic view'of marriage. 139 I [454, 9 (4), Page 9] Son rides on Daddy's back and on her stomach when they have sex. comments: They deny any conflict in their behavior. I [454, 9 (9), Page 17-18] Husband has never liked any of the jobs he's had. She doesn't know shy. They moved back to lichigan because she is from here. comments: 2 brief statesents, back to back, which would each have been given the same code. Reither elaborated. I [454, 9 (13), Page 24] [lbout if he objects to her going out with girls.) "I don't know. I don't think [so]. He's never really said anything, so I really don't know. I don't think so. comments: They don't talk about it. I [455, 9 (8), PAGELBJ "I am just as willing to jump in and help my husband work on the car..." Comments: light be coded IV, if context of all she said before were taken into consideration--about how each accommodates to the other. But she did not elaborate, so coded 1. Children I. As this category contains third person, impersonal, reciprocal role statements with regard to self in relation to children, there are relatively few examples. Women who describe themselves as "by nature, motherly” or "instinctively” motherly would fall in this category. lost often, the women describe themselves as not good enough mothers, struggling to be more patient and more giving. They struggle to be better mothers, better people for their children, unable to accept themselves and include themselves in the caring environment they seek to create. I. "When I feel good, the kids feel good, and when I'm upset, they're upset. That's natural.” Children I [154, 18 (4), PhGlLlSl [Son is bright 3 year old who can already blow bubbles and balloons.) comnents: A non-reflective statement about the main characteristics of a bright 3 year old boy. 140 lother: I. The descriptive statement of self in relation to own mother, like ”I am just like my mother", which do not elaborate the meaning of the identification to the woman, will be assigned to this category. Statements which reveal an unreflective identification, such as ”I haven't even thought out how I am like my mother" also belong in this category. I. ”We are just alike. Carbon copies of each other." "I don't really think about myself in relation to my mother. I don't really know.” I [150, 11 (l), Page_8-9] TPart of it...not usual thing to do" [somewhat like journalist mother in behavior] comnents: Esplains why she thinks she feels isolated from other people - is like mother in behavior. Tries to explain that identification with a negative aspect of the lother on self causes her probless, but, extremely unreflective in her actual explanation, which led to coding it a I, rather than a III. I [454, 10 (2), Page 7-8] Sexual activity in family described as "open and very warm and very loving" but also, as a source of hilarity. However, says "I do wish my mother had told me a little bit more about my body." comments: Possibly inappropriate sexual activities described as Thilarious." Denial. Also critical of mother's role. Denial outweighs criticiss, which would be more III. Father I [455, 12b1, PAGELIBJ "I see some traits that I picked up from my father...[he] was very sarcastic." comments: Although critical of both self and father, this is too brief to be coded a III. lo further elaboration. Descriptive statements of self in relation to others (general) usually occur in responses to the describe yourself questions and the describe yourself in the past question. These statements can be categorized in a parallel fashion to the other relationships. The examples of statements drawn from the interviews are the best description of this rather infrequent style of self description. I. ”How can I turn around and shut and turn my back on somebody that needs me because I feel that maybe down the road a piece I may need help. So you always need somebody, so you can't turn your back on people.” 141 Other I [454, 21(1), PAGIL20] Ruth is lay] best friend now. we exchange Christmas presents and enjoy each other's company. "She's just another person a lot like syself." comments: Ispersonal view of best friend as one with whom one exchanges Christmas presents. I [454, O (2), PZGELJ, 11, and 15] "In my family,.sy sisters and I quite often talk...[on p. 11] "I tell my sisters, but I don't tell my husband" (about horny dreams). [About family]. I think of all of us as individuals...we're all loving and caring about each other...have good times when get together...1augh and tease and goof and have a lot of fun." comnents: Denial that father was abusive [revealed elsewhere in transcript], while emphasizing only'positive facts of family interaction. I [455, O (2),PAG3;8] About her relation with the Black maid in childhood, and telling her not to kiss her brother or she would make the brother turn Black. ‘ Comnents: Stereotyped and biased. Does not seem to reflect on how this comment would cause emotional pain to the Black said. Comnents about coding in Category I 2 1. Relationship statements that seem obviously thoughtless or insulting, not clearly recognized by the speaker, are coded I. 2. When a woman describes a series of what seems to be superficial or inconsequential aspects of a relationship, code I. 3. Humorous and/or flippant references to a relationship are coded I, despite the fact that one expects that the woman can in fact relate to the other at a higher level. 4. Although ordinarily each conceptually different relationship idea is coded separately, an exception is nade for 2 or more brief statements back to back which would each have been given the same code. The justification for this is that only presence or absence of responses within a given category is coded for statistical analysis, not absolute number of responses. 5. Double coding occurs when 2 clearly separate ideas are presented back to back within the same relationship story. Double coding is omitted when there is one clear idea and a non-elaborated second idea which might have been coded differently. 2 Added by the researcher. 142 6. Although criticism of the other person in the relationship is normally coded as 111 (for counter-identification), when she identifies herself as possessing also the negative characteristic, but does not elaborate, this is coded a I. 7. When a close relationship is claimed with an other, based on what seems to be an inadequate reason, this is coded I. 8. Several brief statements making the same point, plus having denial of conflict, would be coded I. II. Self instrumental to other (role), ”Other in their own terms" Self description in this category characterizes the feminine role of self subordinated to the needs, demands and expectations of others. Aware of the expense to themselves, wonen often describe the boomerang of their generous intentions to respond to the other in the other's terms resulting in harm to themselves and ultimately to other. (Therefore, the expression of ”other in their own terms” in quotations to indicate the failure the express authentic care without awareness of one's own terms in the relationship.) When women express intentions to respond to the other in "their own terms” they simultaneously abdicate or deny their own responsibility and power in the relationship. A related group of self description denigrates the self for failures in fulfilling the requirements of the selfless role. The woman who expresses herself as instrumental to others is vulnerable to losing sight of herself, and in some cases this loss or absence is apparent (i.e., ”I don't know who I am" or ”I can't describe myself”). Category II: o self subordinated to the needs, demands and expectations of others o other eclipses self o for others in ”their own terms" 0 ”selfless” Husband; -H_-m II. In this category, the self is defined in service to the husband's particular needs, demands and expectations. It is often easier for the woman to describe him than herself, as she is so tuned into him (although not always in active dialogue with him). The women describe themselves as compromising and compromised, ~losing a sense of themselves. 143 II. ”I do everything I can to help my husband with what he wants to do.” "I expect a lot of strength from my husband, which he doesn't give me, which is difficult. I think it was been difficult learning to cope with a man who has this terrible temper. I try to put myself in someone else's place and he seems to totally lack that.” II [454, 9 (3), Page 10] THe got a really big layoff...his self esteem must of really went down because we used to have sex quite frequently... he just fizzled down...and it never has quite come back to what it was. . . " comments: Because frequency depends on him, she appears to accede to him on this. 11 [454, 9 (16), Page 21] flayhe my husband would be happy if we had a farm....He's an old farmer from way back anyhow. In fact, we'd like to have a small farm. That's one of the things we'd very much like to have." comments: Possibly this could also be coded a III or a IV as a separate response because she says elsewhere that she loves a farm. They could have talked about this and both agreed they would like a farm. But lacking further infornation, it is only coded for the emphasis on looking for what husband wants. II / III [454, 9 (11, 12), Page 19] Tries to talk to husband about his lack of interest in things, but "he doesn't say; Just doesn't say. costs too much or something like that....I tried to get point across that I enjoyed it and everything,.but I don't know." comnents: doublnngoded for II and III because of both trying to do for him, and also trying to get her own point across. However, there does not appear to be a mutuality in the interaction, so it would not be coded a IV, nor is it a Level I. 991mg}! II. In this category, the self is defined in service to the needs of the children, often at the expense of the woman herself. Women describe their responsiveness and attentiveness to the children in an effort to protect the best interest of the children. But the woman subordinated to others risks resentment and anger toward the very children she seeks to protect. II. If there is anything, this intensity that you feel towards your children, if there was anything I could do, if there was any pain I could take from them, or anything like that, I most certainly would.” 144 ”I have a woman friend, she is terribly patient. I always admire her patience and her ongoing, well, 'why don't we do this now and why don't we do that now' with the children." (IDEAL). II [454, 14 (1),.PAGE_15] THe [3 year old son] told me the other night that I was a good mommy. [he thinks] I give him a lotta love and attention and that, and do a lot of things with him. So I think he's pretty satisfied with the lomma he's got. comments: coded for self-sacrifice and giving as reflected in her view of son's words. lother II. ”She's a giver; she's always there for us. I guess, I get that from her; my urge to give.” "I admire my mother's patience. I wish I could be as patient as she is sometimes.” (IDEAL) Fnther [Attanucci gives no examples] Other 11. I am always wanting to be all things to all people." "I'm just very close to my family. I will do anything for my family, anyone in my family." ”Two years ago I used to make commitments that I would regret, that I would have to follow through with because I said I would. In other words, I allowed myself to be used." (PAST) II [450, O (2), Page__11] [Refers to love affair in her past] "I‘m very,proud...[he was] dominating and all-enveloping...[but] a wonderful experience." comments: II because experiences self subordinate to the other II [155, 20 (1), RAGE_32] [In high school had several good male friends. They and her brothers would want to talk to her. Described herself as open and not uncomfortable listening to them.] comments: Eaphasis on her listening to them rather than mutual give and take. Tlot uncomfortable" is too weak to be considered mutual give and take. 145 Table A-1 Reliability of Rater Codings for Attanuci Categories: Percent Agreement before Discussion Percent Agreement Total l Total Percent by Categories of Items Agreement I II III IV Rater 1 and Subject lumbers 53 100 25 100 100 49 81 67 75 50 25 50 24 50 66 50 75 75 100 36 75 80 75 50 75 75 44 69 52 75 75 100 100 36 88 56 75 50 50 75 42 69 78 50 100 100 50 42 75 65 50 100 100 100 51 88 70 50 100 75 75 38 75 75 50 100 100 75 39 81 79 25 100 100 50 51 69 71 100 100 100 75 32 94 146 Table A-1 (Cont.d) Percent Agreement Total N Total Percent by Categories of Items Agreement I II III IV Rater 2 and Subject lumbers 60 75 50 100 75 33 75 59 50 50 100 100 36 75 51 100 100 50 25 30 69 64 100 75 75 75 53 81 63 100 75 75 100 38 88 76 75 100 75 50 56 81 77 75 100 100 100 41 94 68 75 100 100 0 45 69 73 75 100 75 100 41 88 57 100 100 100 100 39 100 58 100 100 100 50 59 88 69 0 100 75 100 43 69 74 75 100 100 75 49 88 72 100 100 100 100 53 100 lots. Subjects numbered 60, 67, 73, and 77 were eliminated in analyses of the restricted sample. APPENDIX B CODING NANUAL FOR QUALITY OF OBJECT RELATIONS IN EARLY NENORIES 147 QUALITY 0r osarcr RELATIOlS scan: 1 (Ryan, 1970, 1973. 1974)2 This scale is an attempt to measure the quality of a person's object relations through an analysis of his/her early memories.3 This analysis is based on the assumption that we may learn much about an individual's character structure and inner object world if we treat his/her early memories not as historical truths (or half-truths) but as thematic representations of prototypical dilemmas, life strategies, and role paradigms around which he or she defines the relationship to self and to one's personal world. This scale is divided into four major categories which seem to reflect the natural breaks in a continuum of quality of object relationships. (While no attempt was made to construct a simple health- sickness scale, it is assumed that this continuum will be significantly correlated with such a scale.)‘ In terms of quality of object relations, the ”psychotic” memories are different from the “borderline” memories in the way that totally alien experiences are different from a sense of alienation from 1 Copyright by Edward R. Ryan, Ph.D. Ryan, E. R. (1970). Ohject relations and ego coping style in early memories. Unpublished laster's thesis, University of lichigan. Ryan, E. R. (1973). The capacity of the patient to enter an elementary therapeutic relationship in the initial psychotherapy interview. Doctoral Dissertation, University of lichigan. Ryan, E. R. (1974, June). The capacity of the patient to enter an elementary therapeutic relationship in the initial psychotherapy interview. Paper presented at the Fifth AnnualoConvention for 'Psychotherapy Research, Denver, Colorado. 3 While the scale was developed for use with early memories, it is assumed that it nay be used with any projective production (dreams, test esponses, autobiography, etc.) Ryan, E. R. and Bell, l. D. (1984). Changes in object relations from psychosis to recovery. Journal of Abnormal Psychology 93 (2): 209-2150 148 otherwise ”normal” experiences. The ”psychotic" memories are generally characterized by qualities of chaos or other-worldliness or objectlessness, setting them apart from the ”borderline" memories, which have a much more coldly narcissistic character. There are people in the "borderline” memories whom we can recognize, but one has the sense that the subject lacks the ability to make a warm, interactional, human contact with them. One senses a self-contained, essentially affectless detachment from people in the ”borderline" memories, a detachment from people with whom the subject is unable to or unwilling to become engaged. The ”neurotic” and ”normal" memories differ from the "borderline" and ”psychotic” memories in one essential respect: in the former categories, one feels the presence of human objects with whom the person is involved, in an affectively charged human interaction. However, the ”neurotic” memories represent this quality of relationship at an essentially regressed stage. The person is engaged in painful, conflictual, crisis-laden interactions with the objects of his childhood. The assumption here is that the objects have some real character for the subject, but that this character remains fixed at the subject's infantile experience of these objects, seen through the affectively biased and developmentally imnature eyes of his or her childhood. The listener, in turn, experiences these objects emphatically as figures who are in some ways larger than life, protagonists of infantile conflicts, but figures with whom it is possible to make an affective contact that isn't possible in borderline‘ or psychotic memories. 149 The ”normal” memories allow one to have a sense of human engagement with essentially real people at a level of interaction much closer to the present-day adult world of the subject, of the objects, and of the listener. In the normal memories the person perceives, experiences and responds to the other person in a way that another adult, observing the interaction might have responded to him or her as well. This consensual adult perception is different from the emotional distortion of the characters of the ”neurotic" memory, in that the "neurotic" object is pulled into being a figure in a transference conflict rather than an object in his or her own right. From the point of view of affective valence, the "psychotic" memories are pre-ambivalent: the die has been case in the direction of a belief in the existence of only malevolent objects. Good objects are simply wiped out of the inner world of the psychotic. ” Borderline” nemories also tend to be pre-ambivalent in which objects tend to be "all good" or ”all bad.” The relationship here is not to objects in a real world, but rather to a projection of the narcissistic sense of the all- good or all-bad self. From the ”neurotic” memories one has a sense of ambivalent conflict which does not blot out the reality of the other person. The subject can recognize these conflicting aspects of the other person but is unable to resolve them by himself/herself. The ”normal” memories range from not quite post-ambivalent, in which the subject does not pull the object into gross distortions, struggling with an unmanageable ambivalence, to post-ambivalent, in which more attention is available for investment in the object as a more fully integrated 150 human being whose many-faceted nature one can tolerate, become engaged with, and enjoy. An attempt has been made to differentiate each scale point from the next by offering a conplex, multi-dimensional definition for it. However, an attempt has also been made to insure that the richness of each scale point is directly referent to an object relations configuration, a level that defines a person engaged in or interaction with a psycho-social world. The attempted result is a scale of the multifaceted definitions of narcissism and object relatedness as manifested in ”remembered" self-other interactions.5' 5 A. Prototypes of severe disturbances in object relatedness such as occur in psychotic or borderline states. These memories express and absence of any sense of real human objects in a real interpersonal world, and depict instead a malevolent object world, which at worst is nightmarish, and at best offers only an ephemeral glimmer of hope of rescue in an otherwise paranoidly evil world. 1. The object world is unreal, nightmarish, other-worldly. In El Paso in the Hilton Hotel...and I was there with another by, I guess it was a boy, and he wanted to leave and get out. And I couldn't get him to stay. So I followed him. We went out on the corner. And I held him back and waited for the lights. And we went down the block. And we went into another hotel...He was all lost. I was just following him. He started crying, so then I took him back to the hotel. (?) I remember my mother said it's a wonder you didn't get run over. I told her I waited for the lights. But she wouldn't believe I knew the lights. The more I told her the more she wouldn't believe. so I just shut up and let her have her way. . 5 Because this scale measures relatively undisrupted, self-reported early memories, the lowest scale points represent the object world of those people who can remember and who can communicate their memories zhen asked. For scoring purposes, the scale may be regarded as a 20 point continuum. 151 2. The self is the object of malevolent attack in a hostile world or "bad” objects. lo vestige of a good human object anywhere to intercede, stave off, or mitigate the threat. lore archaic than the feeling of loneliness or deprivation. The experience is not so much of being deprived of good objects as of being beset by destructive forces in a psychic world devoid of good objects filled exclusively with bad objects. About five years old. Got bit by a dog...the dog was eating and I took a bone away from the puppy and I had to get rabies shots...(Where bit?) On the finger. (Painful?) What the fuck! It didn't tickle...and the rabies shots didn't either. low I'm getting sarcastic. (Why?) Why? What the fuck...how can I remember...I'll get ”crazy" and tel you it was fun, I loved it! This reminds me of the Senate investigation. 3. lot control over potentially devastating events, and no sense of having any control over them. The self is represented as a victim of the unexpected and uncontrollable occurrence, not at all able to influence or forestall destructive events. People are not experienced as agents but rather as elements in a field of forces--self as well as others--and these forces originate outside of the self. The story is told with a matter-of-fact acceptance of the evil or dangers depicted. ly early memory is riding in a baby buggy down a hill and cutting my eye open. Somebody pushed me down a hill in the buggy. They say I was very young...about 2 or so...very young. (?) I can still eel myself going down that hill and I can see the hill. (Feeling tone?) I made quite a fuss about it, I guess. I was too young to have feelings of anything but fright. (?) It was either sone relative or some neighbor child who pushed me down...older children did it. As I grew up they talked abut it and that helped me to remember it. 4. As best these memories involve a cry of protest out of the field of forces depicted in A3. There is a hint of an emerging assertion of self against a cold, or hostile, or uncaring, or uncontrollable object world. 152 Another one I remember. I was very upset and I was telling my mother that I got blamed for everything that happened around there. I yelled and then I couldn't get my breath. She spanked me with a lathe. (?) A thick, rough yardstick with slivers on it. And I remember getting so mad that I couldn't breathe. I wondered if I was going to die. 5. The cry for help, or the plight itself, finally brings about some glimmer of hope in the form of a seemingly fortuitous intervention: the 'good object' who might come to one's rescue is more magical than real. I think the earliest I remember is being in the convent when I was about five. lother put me in the convent because of my brother's illness. I remember being surrounded by a lot of nuns and being terrified. they looked like big black crows--very menacing. One in particular had big black eyebrows and I think she gave me a lollipop which helped to lull me a little bit. B. These nemories depict a level of disturbance in object relatedness closely akin to the malevolence of the A-category memories. The B- memories, like the a-memories, show no vestige of a sense of real, good objects in the subject's inner world. lemories Bl to B4 convey instead a sense of empty aloneness with, at best, some wholly self-invested satisfactions filling the people-less world. B5 depicts a stage of chronic object-hunger, i.e., a bleak, hungry, deprived sense of separateness or aloneness. 6. The world is not so much ”bad" as it is empty, essentially devoid of ”good" human objects, past or present, and equally devoid of good self- feeling. Or the memory may also be a purely narcissistic expression of well-being, unrelated to the presence in the person's object world of other people as instrumental to his/her well-being. I remember when I was 4, or maybe 5, at my great-grandmother's house, in Denver. Behind the house there was a garden, or a terrace with a garden. I used to go out there and sit and watch the birds and the neighbor's cat. It was a very peaceful setting...perfect. (?) There was a white picket fence all around. (?) I feel like I'm at peace with the world...I can see it clearly. 153 7. The self is narcissistically self-absorbed due to an illness, wholly invested in the fact of being ill or in the experience of the pains, symptoms, or deprivations associated with being ill. There may be some allusions to an attendant care-taker, but these people exist only as props in the subject's wholly self-involved state. lo idea how old, still in a crib. Always sick at Christmas. Sick all my early life. Sick in a crib and could see a Christmas tree. Could have been two and that's all I have, the impression of the Christmas tree and that I was sick. (Feelings?) I don't know. Sometimes I thought about it. lever could remember much of my childhood. lo feeling. I could imaging having a feeling. (?) Feeling I wish I hadn't been ill. (age?) two. (?) lo, must have been in a living room. Brought into the room with the tree, but I was ill, don't know whether a cold or illness or what. 8. Alone except for things rather than people. The self is related to possession rather than people. The inanimate objects acquire a very special value which gives them the status of transitional objects. (There is in these relationships to things a central component of "primary narcissism.”) I can remember it was summer and I was lying in the bed. I'm not sure whether I was about four months old or what. I can just remember lying in a crib and looking out a window. I'm not sure whether that's how old I was. I think that's about it. Sometime when the snow was off the found anyway. (Saw?) Oh, I just saw the stuff in the backyard--clothesline, and stuff line that-- nothing much, but I must happen to remember the incident. (Feeling tone?) Well, no, I was all by myself. (Feeling?) OH, I think it was a kind of good feeling. Of course, most kids when they're only four months old, they shouldn't feel too bad. (Age?) I'm not sure whether it was four months or whether I was a year old, but I know it was summer and we lived in the same house for only two summers. I know it was in that house. I figure it was about four, four to five months. (Thought of before?) Oh, year, I remember that all the time. I doubt that I could forget it. I mean once it sticks with you, it sticks with you. (lOTE: If the object is enjoyed or treasured because of its nngnndnry narcissistic values, i.e., for its socially defined, mutually enjoyed 154 and other-rewarded qualities, it belongs in Category C., e.g., the “yellow sunsuit" example in C2.) 9. Other people if present at all, are incidental, shadowy, two- dimensional in character. They may not be bad objects, but are not yet good or real objects. They remain essentially impersonal props rather than participant others. Self and other come across as vague, detached, impersonal, almost shadowy in character (despite inquiry aimed at eliciting more convincing expression of interpersonal involvement). Both self and others could just as well not have been there at all as far as any real effect it would have had on others. Or, the self may have been only the inertly participating object of others' initiatives. There may be what seems like an interaction between self and other, but the separate participants evoke no empathy from the examiner. I remember...I don't know where I was or how old I was, but they took me in this place to have my picture made and the people were...plate glass window with people passing by. There was a stand with artificial grass they put me on and took my picture. I remember standing up there on that green and people passing by looking in. I think I remember that. (Feeling?) lo...(Old?) I think I asked my mother once how old I was and I think it was under two years old. They have the picture hanging on the wall. (?) When I was a little kid and see the picture I guess. 10. A painful yearning for unavailable or lost objects in an otherwise bleak world, or a euphoric basking in a diffuse goodness. The other persons, if present, are anonymous figures, interchangeable one with another. They are conveyors of the global feeling-quality, or pegs onto which to hang the feeling state. It's hard to place them in time. I remember sitting in a kitchen with a large black stove...I was sick with something because I was wrapped up in a blanket. This may have been the time I fell in a fish pond. I've never been sure. And the kitchen was full of 155 friendly people. I don't remember who they are. Chiefly, I remember the blanket and the stove. (Feeling tone?) Of comfort, and of people paying attention to me. Being taken care of. C. Other people do appear as important foci of one's relatedness to the world, but relationships with them are childishly conceived, neurotically defined, or self-centeredly limited. The sense of others as objects in their own right is stunted or warped. People in the stories take on their significance only in terms of the subject's pressing needs or intrusive transference paradigms. 11. the other person is present in the life space, important to the subject, but his character is defined almost solely as a need-satisfying or need-frustrating object or being. (If that object takes on traces of a more individually distinctive person, the score would move up to 12 or even 13). Then I was sent to the orphan's home. While there I can remember my brother and I were both there. I remember just looking toward the building where I knew he was. Realizing that he was not far away. 12. The self is caught up in some special, interpersonally relevant but nonetheless self-centered interest of its own in relation to others. The self may be doing something with others, but it would be essentially a parallel activity rather than a full-bodied interaction with them. I think the first thing that I remember is playing with dolls in the back yard, under an apricot tree with a little girl. We were aware of each other but we each played with her own doll and didn't have much to say to each other. A very nice time being together, enjoyed it very much. (?) I can remember in a vague way what she looked like and I remember that she moved away. She wasn't living there too long. 13. Both self and other are more distinct people, but they are defined exclusively by the subject's immediate intrapsychic conflict or affect state, not as unique persons in their own right. generally they simply 156 Table A-1 Reliability of Rater Codings for Attanuci Categories: Percent Agreement before Discussion Percent Agreement Total l Total Percent by Categories of Items Agreement I II III IV Rater 1 and Subject lumbers 53 100 25 100 100 49 81 67 75 50 25 50 24 50 66 50 75 75 100 36 75 80 75 50 75 75 44 69 52 75 75 100 100 36 88 56 75 50 50 75 42 69 78 50 100 100 50 42 75 65 50 100 100 100 51 88 70 50 100 75 75 38 75 75 50 100 100 75 39 81 79 25 100 100 50 51 69 71 100 100 100 75 32 94 157 Table A-1 (Cont.d) Percent Agreement Total l Total Percent by Categories of Items Agreement I II III IV Rater 2 and Subject lumbers 60 75 50 100 75 33 75 59 50 50 100 100 36 75 51 100 100 50 25 30 69 64 100 75 75 75 53 81 63 100 75 75 100 38 88 76 75 100 75 50 56 81 77 75 100 100 100 41 94 68 75 100 100 0 45 69 73 75 100 75 100 41 88 57 100 100 100 100 39 100 58 100 100 100 50 59 88 69 0 100 75 100 43 69 74 75 100 100 75 49 88 72 100 100 100 100 53 100 lots. Subjects numbered 60, 67, 73, and 77 were eliminated in analyses of the restricted sample. APPENDIX D CODING NANUAL FOR QUALITY OF OBJECT RELATIONS IN EARLY NENORIES 158 QUALITY OF OBJECT RELATIONS SCALE 1 (Ryan, 1970, 1973, 197412 This scale is an attempt to measure the quality of a person's object relations through an analysis of his/her early memories.3 This analysis is based on the assumption that we may learn much about an individual's character structure and inner object world if we treat his/her early memories not as historical truths (or half-truths) but as thematic representations of prototypical dilemmas, life strategies, and role paradigms around which he or she defines the relationship to self and to one's personal world. This scale is divided into four major categories which seem to reflect the natural breaks in a continuum of quality of object relationships. (While no attempt was made to construct a simple health- sickness scale, it is assumed that this continuum will be significantly correlated with such a scale.)‘ In terms of quality of object relations, the ”psychotic” memories are different from the "borderline" memories in the way that totally alien experiences are different from a sense of alienation from ..__.__-_ ..-. .... ........ 1 Copyright by Edward R. Ryan, Ph.D. 2 Ryan, E. R. (1970). Object relations and ego coping style in early memories. Unpublished laster's thesis, University of lichigan. Ryan, E. R. (1973). The capacity of the patient to enter an elementary therapeutic relationship in the initial psychotherapy interview. Doctoral Dissertation, University of lichigan. Ryan, E. R. (1974, June). The capacity of the patient to enter an elementary therapeutic relationship in the initial psychotherapy interview. Paper presented at the Fifth Annual Convention for Psychotherapy Research, Denver, Colorado. 3 While the scale was developed for use with early memories, it is assumed that it may be used with any projective production (dreams, test esponses, autobiography, etc.) Ryan, E. R. and Bell, l. D. (1984). Changes in object relations from psychosis to recovery. Journal of Abnormal Psychology 93 (2), 209-215. 159 otherwise “normal" experiences. The ”psychotic" memories are generally characterized by qualities of chaos or other-worldliness or objectlessness, setting them apart from the ”borderline” memories, which have a much more coldly narcissistic character. There are people in the "borderline” memories whom we can recognize, but one has the sense that the subject lacks the ability to make a warm, interactional, human contact with them. One senses a self-contained, essentially affectless detachment from people in the ”borderline" memories, a detachment from people with whom the subject is unable to or unwilling to become engaged. The ”neurotic” and "normal” memories differ from the "borderline" and ”psychotic” memories in one essential respect: in the former categories, one feels the presence of human objects with whom the person is involved, in an affectively charged human interaction. However, the ”neurotic" memories represent this quality of relationship at an essentially regressed stage. The person is engaged in painful, conflictual, crisis-laden interactions with the objects of his childhood. The assumption here is that the objects have some real character for the subject, but that this character remains fixed at the subject's infantile experience of these objects, seen through the affectively biased and developmentally immature eyes of his or her childhood. The listener, in turn, experiences these objects emphatically as figures who are in some ways larger than life, protagonists of infantile conflicts, but figures with whom it is possible to make an affective contact that isn't possible in ” borderline" or'psychotic' memories. 160 The ”normal” memories allow one to have a sense of human engagement with essentially real people at a level of interaction much closer to the present-day adult world of the subject, of the objects, and of the listener. In the normal memories the person perceives, experiences and responds to the other person in a way that another adult, observing the interaction might have responded to him or her as well. This consensual adult perception is different from the emotional distortion of the characters of the ”neurotic” memory, in that the ”neurotic” object is pulled into being a figure in a transference conflict rather than an object in his or her own right. From the point of view of affective valence, the ”psychotic" memories are pre-ambivalent: the die has been case in the direction of a belief in the existence of only malevolent objects. Good objects are simply wiped out of the inner world of the psychotic. ” Borderline” memories also tend to be pre-ambivalent in which objects tend to be ”all good” or ”all bad.” The relationship here is not to objects in a real world, but rather to a projection of the narcissistic sense of the all- good or all-bad self. From the "neurotic” memories one has a sense of ambivalent conflict which does not blot out the reality of the other person. The subject can recognize these conflicting aspects of the other person but is unable to resolve them by himself/herself. The “normal” memories range from not quite post-ambivalent, in which the subject does not pull the object into gross distortions, struggling with an unmanageable ambivalence, to post-ambivalent, in which more attention is available for investment in the object as a more fully integrated 161 human being whose many-faceted nature one can tolerate, become engaged with, and enjoy. An attempt has been made to differentiate each scale point from the next by offering a complex, multi-dimensional definition for it. However, an attempt has also been made to insure that the richness of each scale point is directly referent to an object relations configuration, a level that defines a person engaged in or interaction with a psycho-social world. The attempted result is a scale of the multifaceted definitions of narcissism and object relatedness as manifested in ”remembered” self-other interactions.5' 5 A. Prototypes of severe disturbances in object relatedness such as occur in psychotic or borderline states. These memories express and absence of any sense of real human objects in a real interpersonal world, and depict instead a malevolent object world, which at worst is nightmarish, and at best offers only an ephemeral glimmer of hope of rescue in an otherwise paranoidly evil world. 1. The object world is unreal, nightmarish, other-worldly. In El Paso in the Hilton Hotel...and I was there with another by, I guess it was a boy, and he wanted to leave and get out. And I couldn't get him to stay. So I followed him. We went out on the corner. And I held him back and waited for the lights. And we went down the block. And we went into another hotel...He was all lost. I was just following him. He started crying, so then I took him back to the hotel. (?) I remember my mother said it's a wonder you didn't get run over. I told her I waited for the lights. But she wouldn't believe I knew the lights. The more I told her the more she wouldn't believe. so I just shut up and let her have her way. 5 Because this scale measures relatively undisrupted, self-reported early memories, the lowest scale points represent the object world of those people who can remember and who can communicate their memories when asked . For scoring purposes, the scale may be regarded as a 20 point continuun . 162 2. The self is the object of malevolent attack in a hostile world or "bad" objects. lo vestige of a good human object anywhere to intercede, stave off, or mitigate the threat. lore archaic than the feeling of loneliness or deprivation. The experience is not so much of being deprived of good objects as of being beset by destructive forces in a psychic world devoid of good objects filled exclusively with bad objects. About five years old. Got bit by a dog...the dog was eating and I took a bone away from the puppy and I had to get rabies shots...(Where bit?) On the finger. (Painful?) What the fuck! It didn't tickle...and the rabies shots didn't either. low I'm getting sarcastic. (Why?) Why? What the fuck...how can I remember...I'll get "crazy” and tel you it was fun, I loved it! This reminds me of the Senate investigation. 3. lot control over potentially devastating events, and no sense of having any control over them. The self is represented as a victim of the unexpected and uncontrollable occurrence, not at all able to influence or forestall destructive events. People are not experienced as agents but rather as elements in a field of forces--self as well as others--and these forces originate outside of the self. The story is told with a matter-of-fact acceptance of the evil or dangers depicted. Hy early memory is riding in a baby buggy down a hill and cutting my eye open. Somebody pushed me down a hill in the buggy. They say I was very young...about 2 or so...very young. (?) I can still eel myself going down that hill and I can see the hill. (Feeling tone?) I made quite a fuss about it, I guess. I was too young to have feelings of anything but fright. (?) It was either some relative or some neighbor child who pushed me down...older children did it. As I grew up they talked abut it and that helped me to remember it. 4. As best these memories involve a cry of protest out of the field of forces depicted in A3. There is a hint of an emerging assertion of self ‘mgainst a cold, or hostile, or uncaring, or uncontrollable object world. 163 Another one I remember. I was very upset and I was telling my mother that I got blamed for everything that happened around there. I yelled and then I couldn't get my breath. She spanked me with a lathe. (?) A thick, rough yardstick with slivers on it. And I remember getting so mad that I couldn't breathe. I wondered if I was going to die. 5. The cry for help, or the plight itself, finally brings about some glimmer of hope in the form of a seemingly fortuitous intervention: the 'good object' who might come to one's rescue is more magical than real. I think the earliest I remember is being in the convent when I was about five. lother put me in the convent because of my brother's illness. I remember being surrounded by a lot of nuns and being terrified. they looked like big black crows--very menacing. One in particular had big black eyebrows and I think she gave me a lollipop which helped to lull me a little bit. 8. These memories depict a level of disturbance in object relatedness closely akin to the malevolence of the A-category memories. The B- memories, like the a-memories, show no vestige of a sense of real, good objects in the subject's inner world. lemories 81 to B4 convey instead a sense of empty aloneness with, at best, some wholly self-invested satisfactions filling the people-less world. 85 depicts a stage of chronic object-hunger, i.e., a bleak, hungry, deprived sense of separateness or aloneness. 6. The world is not so much ”bad" as it is empty, essentially devoid of 'good” human objects, past or present, and equally devoid of good self- feeling. Or the memory may also be a purely narcissistic expression of well-being, unrelated to the presence in the person's object world of other people as instrumental to his/her well-being. I remember when I was 4, or maybe 5, at my great-grandmother's house, in Denver. Behind the house there was a garden, or a terrace with a garden. I used to go out there and sit and watch the birds and the neighbor's cat. It was a very peaceful setting...perfect. (?) There was a white picket fence all around. (?) I feel like I'm at peace with the world...I can see it clearly. 164 7. The self is narcissistically self-absorbed due to an illness, wholly invested in the fact of being ill or in the experience of the pains, symptoms, or deprivations associated with being ill. There may be some allusions to an attendant care-taker, but these people exist only as props in the subject's wholly self-involved state. lo idea how old, still in a crib. Always sick at Christmas. Sick all my early life. Sick in a crib and could see a Christmas tree. Could have been two and that's all I have, the impression of the Christnas tree and that I was sick. (Feelings?) I don't know. Sometimes I thought about it. lever could remember much of my childhood. lo feeling. I could imaging having a feeling. (?) Feeling I wish I hadn't been ill. (age?) two. (?) lo, must have been in a living room. Brought into the room with the tree, but I was ill, don't know whether a cold or illness or what. 8. Alone except for things rather than people. The self is related to possession rather than people. The inanimate objects acquire a very special value which gives them the status of transitional objects. (There is in these relationships to things a central component of ”primary narcissism.”) I can remember it was summer and I was lying in the bed. I'n not sure whether I was about four months old or what. I can just remember lying in a crib and looking out a window. I'm not sure whether that's how old I was. I think that's about it. Sometime when the snow was off the found anyway. (Saw?) Oh, I just saw the stuff in the backyard--clothesline, and stuff line that-- nothing much, but I must happen to remember the incident. (Feeling tone?) Well, no, I was all by myself. (Feeling?) OH, I think it was a kind of good feeling. Of course, most kids when they're only four months old, they shouldn't feel too bad. (Age?) I'm not sure whether it was four months or whether I was a year old, but I know it was summer and we lived in the same house for only two summers. I know it was in that house. I figure it was about four, four to five months. (Thought of before?) Oh, year, I remember that all the time. I doubt that I could forget it. I mean once it sticks with you, it sticks with you. (lOTE: If the object is enjoyed or treasured because of its nnnnndany narcissistic values, i.e., for its socially defined, mutually enjoyed 165 and other-rewarded qualities, it belongs in Category C., e.g., the ”yellow sunsuit" example in C2.) 9. Other people if present at all, are incidental, shadowy, two- dimensional in character. They may not be bad objects, but are not yet good or real objects. They remain essentially impersonal props rather than participant others. Self and other come across as vague, detached, impersonal, almost shadowy in character (despite inquiry aimed at eliciting more convincing expression of interpersonal involvement). Both self and others could just as well not have been there at all as far as any real effect it would have had on others. Or, the self may have been only the inertly participating object of others' initiatives. There may be what seems like an interaction between self and other, but the separate participants evoke no empathy from the examiner. I remember...I don't know where I was or how old I was, but they took me in this place to have my picture made and the people were...plate glass window with people passing by. There was a stand with artificial grass they put me on and took my picture. I remember standing up there on that green and people passing by looking in. I think I remember that. (Feeling?) lo...(Old?) I think I asked my mother once how old I was and I think it was under two years old. They have the picture hanging on the wall. (?) When I was a little kid and see the picture I guess. 10. A painful yearning for unavailable or lost objects in an otherwise bleak world, or a euphoric basking in a diffuse goodness. The other persons, if present, are anonymous figures, interchangeable one with another. They are conveyors of the global feeling-quality, or pegs onto which to hang the feeling state. It's hard to place them in tine. I remember sitting in a kitchen with a large black stove...I was sick with something because I was wrapped up in a blanket. This may have been the time I fell in a fish pond. I've never been sure. And the kitchen was full of 166 friendly people. I don't remember who they are. Chiefly, I remember the blanket and the stove. (Feeling tone?) Of comfort, and of people paying attention to me. Being taken care of. C. Other people d9 appear as important foci of one's relatedness to the world, but relationships with them are childishly conceived, ‘ neurotically defined, or self-centeredly limited. The sense of others as objects in their own right is stunted or warped. People in the stories take on their significance only in terms of the subject's pressing needs or intrusive transference paradigms. 11. the other person is present in the life space, important to the subject, but his character is defined almost solely as a need-satisfying or need-frustrating object or being. (If that object takes on traces of a more individually distinctive person, the score would move up to 12 or even 13). Then I was sent to the orphan's home. While there I can remember my brother and I were both there. I remember just looking toward the building where I knew he was. Realizing that he was not far away. 12. The self is caught up in some special, interpersonally relevant but nonetheless self-centered interest of its own in relation to others. The self may be doing something with others, but it would be essentially a parallel activity rather than a full-bodied interaction with them. I think the first thing that I remember is playing with dolls in the back yard, under an apricot tree with a little girl. We were aware of each other but we each played with her own doll and didn't have much to say to each other. A very nice time being together, enjoyed it very much. (?) I can remember in a vague way what she looked like and I remember that she moved away. She wasn't living there too long. 13. Both self and other are more distinct people, but they are defined exclusively by the subject's immediate intrapsychic conflict or affect state, not as unique persons in their own right. generally they simply 167 represent starkly set, inconpletely integrated needs or restrictions or conflicts. Although the people often seem more alive, the essentially infantile transference of elements in their characterization is readily apparent. I recall during the first world war when the troops were coming back. I can remember a night when it was raining and troops were marching down the street in the rain. Then I recall my uncle...I mean my father's younger brother...coming in the front door dripping wet. He seemed imnensely tall, he took off his sidearm and laid it on the table in the front hall. The thing fascinated me. When I thought no on was looking, I reached up for the pistol. ly uncle's voice from behind me said, ”Look out! That will bite.” It seems like I could almost see the thing move and it scared the hell out of me. I backed away from it. (Age?) Two and a half. 14. Some give and take in relation to others, but the overriding focus is on the self. Others are aware of and responsive to the self, but are decidedly secondary figures in the cast of characters. The self seems somewhat aware of and responsive to events, people, and things in a real world, and this world shows the beginning signs of a reality and validity of its own, one which exists independently of the event being reported. Sometimes there is an inversion of this relationship between self and others, such that the self is seen as the more shadowy, secondary object, and the others seem more alive and real enough to empathize with. Well, I know quite well because I've thought about it. they say I was two and a half at the time and we were beside a lake some place in the States and I as playing with this ball and while I was playing with it, it rolled into the lake. Did I say I was with my mother and sisters? And the current carried it away. ly mother told me it was going to flow into the ocean and never come back. And I was fairly astounded by this. I just stood there and watched sort of dramatically. (?) I thought it was sort of dramatic. (?) It's quite often true of things that happen in life, you can't bring them back. (?) Well, you could say it represents a sort of fatalistic philosophy of my own, and you 168 can't change it or do anything about it, just resign yourself to it. I didn't cry, I just stood there thinking and was just very surprised. 15. The self-other interaction depicted in the story involves traces of a real relationship: it is more than an exclusively transference-based or conflict-drenched representation of self and other. lonetheless, the characterizations, especially of other persons in the story, remain thin. the other person does not quite emerge with a full-bodied identity of his own. The interactions at this level begin to seem more palpable and real, as increasingly believable motives are ascribed not only to the self but to the others as well. When I was sick, about three years old. It was the next day after my mother and father had a big party during prohibition. I was sitting under the piano eating cheese and crackers and drinking stale booze--trying to hide from my mother. I took a drink from a glass--spit it out. ly mother came and caught me and scolded me. In fact, I got a few pokes from that. I don't know if it's the earliest one but it's the one that came to mind right away. 0. As depicted at this level, interpersonal relationships reflect a more sharply defined differentiation of a real self from real others in real mutual interactions. Other persons come across as people with their own personalities, motives and emotional postures; they are clearly more than mere extensions, props, or projected facets of the self. The store presents distinct characterizations of self and others. Even in those stories which are built around prototypical unconscious themes (usually ”phallic” or ”Oedipal"), the figures emerge as separate and distinct people with identities which clearly transcend their roles as transference objects. The listener is provided with a sound basis for empathizing with the motives and viewpoints of each figure in his or her own right, not just as a transference object who really provides the listener empathic access to the subject himself or herself. 169 16. Although the memory is a self-centered one, other people and circumstances are brought onto the scene in a way which adds realism and puts the self-other interaction into a broader, more realistic, less self-centered perspective. The events seems real and permits empathy ' not only with the self, but to a somewhat lesser extent, with the somewhat contrasting separate posture(s) of the other person(s) in the story. Thoughts and feelings of self and other)s) are recounted by the subject, or are clearly implied. When I was in crib--I must have been two years old--my mother was saying goodnight to me and I had my finger on my penis and my mother said don't ever touch that. That's the first memory I can tell with any assurance. (How did you feel?) Well, my mother hadn't told me the name of that part of my body. I just had an impression that there must have been something wrong with it. But I think she nust have said something about it before to me or I wouldn't have felt the way I did. (?) Seemed to sink in so much. I don't know that one saying would have had that effect. Though maybe it could have. Our own son has developed a special interest in his sexual organs at the same age. ly mother was looking at me when I put my hand on it, and I think I had some misgivings about it when I did it--fear that what she would say she actually ended up nnying--that she would actually say what she did. 17. Events are reported with a less self-centered bias. The memory involves sharing with or joining with others in a common interest in something outside the self. Together, self and other(s) experience some interest, activity, person, thing or event which does not necessarily have more immediate bearing for the subject than for any of the others who share that experience with him. The subject's life space seems less narrowly, less egocentrically delimited than in the preceding examples. 18. The self-other interaction reflects clearly the separate, inner- directed motives of self and others. Thoughts and feelings of all participants in the event are well enough to allow the listener to empathize vicariously with both the self and the others. One can 170 readily participate vicariously in the event as it was experienced by the significant others in the story, as well as in how it seemed to the subject. The others' point of view is adequately grasped and effectively mediated in the subject's report of the incident. 19. There is active, two-way comnunication of separate, or even contrasting, two-way communication of emotional and interpersonal messages. The other(s) seem real, very much there, with well- differentiated thoughts, feelings, and motives of their own. Each of the characters emerges in a way which makes him or her seem real and recognizable as a person in his or her own right, someone whom we can imagine in situations other than the one recounted. 20. lultiple, bilateral role relationships with a variety of different people, with a clear articulation of the distinct individuality of the self and others in these relationships. A sense of belonging to a community of separate individuals. A spirit of positive interactions; even negative aspects are presented in a wider context of mutual trust, acceptance, regard or affection. I remember going on vacation with my family (smile). I always looked forward to going camping..we usually went camping in a tent when I was a kid...I still do it today with my family. Well, I remember one time in particular. We were all filling the car, getting ready to go. And each of us had our own jobs. ly brother and I were horsing around with the sleeping bags and as usual my mother and father got into an argument about how much to take. They always started something when we were just about ready to leave. ly mother was always fussing at the last minute and this always got my father grumpy. When they saw my brother, J and me laughing--he was about five then--my father chased us down the driveway. (?) About 7 I think. I remember him scooping us up and carrying us back to the car and we were yelling and laughing at the same tine. We still kid about it today. (?) I don't know, but I guess my brother and I must have seemed like rascals at the time because they really got mad at us when we laughed. (?) lo, this happened lots of times. (?) At the time I just 171 remember feeling good and as I look back at it now, I feel I as pretty lucky. APPENDIX C POSTED NOTICE AND ADYERTISENENT 172 ‘ii~o . sac: 206 Table J-5 Cluster Analysis of Summed Attanucci categories - UPGlA lethod, cosine Distance coefficient - complete sample, R’I 28 glnstgr Sun I Sum II Sum III Sum IV :l a 1m ml H” 0 04» .“ Ow e “a 094 e OH Cluster lumber 2 (lI4) lean SD ON .@ 04» e e Cluster lumber 3 (lI3) lean SD C. e 00 OH Old 0U e e 5". OH e m“ Cluster lumber 4 (lIl) lean SD O‘ GO ON e 0° 04» e 00 C” e co Cluster lumber 5 (lI2) lean 4.0 SD 00 00 Oh» O 040' O O 04-5 O Cluster lumber 6 (lIl) lean SD 0. e ON e 0° G” e O“ e co 207 Table J-6 Cluster Analysis of Summed Attanucci categories - UPGlA lethod, cosine Distance coefficient - Restricted Sample, R’I 24 Cl ster Sun I Sum II Sum III Sum IV Cluster lumber 1 (lI15) lean SD 04» 0 0'” HQ) 0U e 94“ 04.4 e ‘d Cluster lumber 2 (lI2) lean 0. SD 0 0'0! 0‘ OO O. O CO CO Cluster lumber 3 (lI3) lean SD 00 ON e e 0“} 0:» e e 0“) 0:» e e Uld Cluster lumber 4 (lI2) lean SD Ob e co OH e UIUI Ow e e UIU'I ON e 0° Cluster lumber 5 (l-l) lean 4.0 SD 0.0 C” e 00 C” e 00 GU e 0° Table J-6 (Cont'd.) Cluster lumber 6 (III) lean 0. SD 0. 209 Table J-7 Cluster Analysis of Summed RRSI Scale Scores - Ward's lethod, Squared Euclidean Distance coefficient - complete Sample, l’- 28 Cluster SOCCF SO RI POC Cluster lumber 1 (lI4) lean 4.50 2.82 4.31 2.81 SD 0.18 . . Cluster lumber 2 (l-7) lean 4.00 3.10 4.10 3.33 SD 0.33 0.25 0.34 0.24 Cluster lumber 3 (lI12) lean 4.03 2.38 4.43 3.41 SD 0.40 0.22 0.25 0.21 Cluster lumber 4 (lI5) Nean 3.69 2.21 3.55 2.68 SD 0.30 0.26 0.27 0.10 210 Table J-8 Cluster Analysis of Summed RRSI Scale Scores - OPGHI lethod, Squared Euclidean Distance coefficient - complete Sample, l’- 28 Cluster SOCCF SO RI POC Cluster lumber 1 (lI10) lean 4.28 2.51 4.45 SD 0.22 0.15 0.29 0.28 Cluster lumber 2 (lI3) lean 4.27 3.33 4.42 SD 0.25 0.08 0.18 0.12 Cluster lumber 3 (lI10) Nean 3.78 2.58 3.96 3.08 SD 0.23 0.31 0.30 0.26 Cluster lumber 4 (lI2) Nean 4.63 3.00 4.29 2.61 SD 0.00 0.00 0.29 0.18 Cluster lumber 5 (lIl) Nean 3.31 1.78 4.42 3.36 SD 0.00 0.00 0.00 0.00 Table J-8 (Cont'd.) 211 Cluster lumber 6 (lI2) lean 3.38 1.97 SD 0.13 0.25 212 Table J-9 Cluster Analysis of Sumned RRSI Scale Scores - UPGlA lethod, Squared Euclidean Distance coefficient - Restricted Sasple, R I 24 Cluster SOCCF SO RI POC Cluster lumber 1 (lI8) lean 4.27 2.54 . . SD 0.22 0.14 0.27 0.28 Cluster lumber 2 (lIl) Nean 4.06 3.39 4.58 3.43 SD 0.00 0.00 0.00 0.00 Cluster lumber 3 (lI10) lean 3.78 2.58 3.96 3.08 SD 0.23 0.31 . . Cluster lumber 4 (lI2) Nean 4.63 3.00 4.29 2.61 SD 0.00 0.00 0.29 0.18 Cluster lumber 5 (ll-1) lean 3.31 1.78 4.42 3.36 SD 0.0 . O O O O O O O O O O Table J-9 (Cont'd.) 213 Cluster lumber 6 (lI2) lean 3.38 1.97 SD 0.13 0.25 214 Table J-10 Cluster Analysis of Summed RRSI Scale Scores - Ward's lethod, Squared Euclidean Distance coefficient - Restricted Sample, I I 24 Cluster SOCCF SO RI POC Cluster lumber 1 (lI12) lean 4.28 2.67 4.42 3.25 SD 0.26 0.31 0.28 0.37 Cluster lumber 2 (lI5) lean 3.69 2.21 3.55 2.68 SD 0.30 0.26 0.27 0.10 Cluster lumber 3 (lI4) lean 3.80 2.93 3.86 3.16 SD 0.20 0.19 0.21 0.16 Cluster lumber 4 (lI3) Nean 3.46 2.17 4.42 3.29 SD 0.21 0.28 0.13 0.06 215 Table J-11 Cluster Analysis of Summed Attanucci categories and Summed RRSI Scale Scores - UPGll lethod, Squared Euclidean Distance coefficient - complete Sample, R'I 28 Cluster SUlI SUlII SUlIII SUlIV SOCCF SO RI POC Cluster Number 1 (lI10) lean SD 0'04 04» 0‘4 DU 0 0‘s) 0 fl» he 0 fi 0' O O fi 0 O N \0 ON 0 O ‘44-. Cu O 0 Cluster lumber 2 (lI7) lean SD 900 On. 00 ON ~24» O u ‘0 O O 0 u .5 O h 0| 0 w .5 04» O. OO Cluster lumber 3 (lI4) 4.02 2.46 4.46 3.34 0.45 0.12 0.36 0.22 lean 3 SD 0. bu OH 010! Ow O 0 .00 OH 0 0 15¢ Cluster lumber 4 (lIl) lean 4.0 3.75 2.50 3.75 2.79 SD 0.0 . ON GO ON 0 Os» 0 00 O O O O O O O O O O O C O O Cluster lumber 5 (lI4) 3.8 0.3 4.08 2.71 4.31 3.04 0.35 0.41 0.42 0.50 lean 2.8 SD 0.4 cu e UIUI Q e . o e . Table J-11(Cont'd.) 216 Cluster lumber 6 (lI2) lean 0. SD 0 217 Table J-12 Cluster Analysis of Summed Attanucci categories and Summed RRSI Scale Scores - Ward's lethod, Squared Euclidean Distance coefficient - complete Sample, l’I 28 Cluster SUNI SUNII SUNIII SUNIV SOCCF SO RI POC Cluster lumber 1 (lI10) 4.11 2.64 4.09 3.07 0.41 0.45 0.40 0.29 lean SD 0'09 OW 04.1 Ow I O 0“! ON e e “4". Oh) e e Cluster lumber 2 (lI5) 4.32 3.23 0.42 0.11 0.43 0.29 lean 3 SD 0. m0: Oh! e e a. ON e e GO C” e 0 01 N e . ‘H .O OH Cluster lumber 3 (II?) lean 3.9 SD 0.3 O. e OO O. OO ON e d“ O u 90 O e w O O e O 0! O w ‘ Cluster lumber 4 (lI6) 4.17 2.87 4.30 3.06 lean 0 5 0.4 0.5 0.35 0.42 0.38 0.45 SD o-Ic-I C 04» O 218 Table J-l3 Cluster Analysis of Summed Attanucci categories and Summed RRSI Scale Scores - UPGlA lethod, Squared Euclidean Distance coefficient - Restricted sample, R'I 24 Cluster SUNI SUNII SUNIII SUNIV SOCCF SO RI POC Cluster lumber 1 (lI16) lean SD HM OW mm 04» OH O b O O u N O 0 fi . O u 0" Oh! bu OU e e Cluster lunber 2 (lI3) lean 4 SD 0. 04» O 0 GO ON on» O U H O O s! O O O O N ~l e OO OH e e 4". Cluster lumber 3 (lI3) lean 2 SD 0. (”‘4 Oh! e e as) OU e e 5"" OO OH» O e e O O O u Cluster lumber 4 (lI2) 4.35 3.20 4.29 3.11 0.29 0.19 0.29 0.3 lean 0 0 SD 000 O9! 0‘04 O. OO OH O O OO 80 219 Table J-14 Cluster Analysis of Summed Attanucci categories and Summed RRSI Scale Scores - Ward's lethod, squared Euclidean Distance coefficient - Restricted Sample, R I 24 QJBBtQE SUNI SUNII SUNIII SUNIV SOCCF SO RI POC Cluster lumber 1 (lI9) 4.05 2.55 4.08 3.02 0.39 0.40 0.42 0.27 lean SD ”'0‘ Os» 4‘. OED e O“ ON a e ON Ob) e e Cluster lumber 2 (HI?) lean SD can o. co ON ~24.» O 4» VO 0 O u .5 O .5 0' O u .5 Oh! O. OO Cluster lumber 3 (lI3) lean SD OH O O OO Q's-I ”O 3.71 2.52 4.31 3.14 . . . 7 OH 04» ON mu O (as ..n O O O ‘4 O . O O N Cluster lumber 4 (lI5) lean 1.6 .6 4.18 2.80 4.26 2.93 SD 1.4 5 0.38 0.43 0.41 0.37 O“ e Ola 04» e O“ OO e 220 Table J-15 Cluster Analysis of Summed Attanucci categories and Quality of Object Relations in Early’lemories scores - UPGlA lethod, squared.Euclidean Distance coefficient - complete sample, l’I 28 Cluster SUNI SUNII SUNIII SUNIV E-DAD E-NON EN Cluster lumber 1 (lI10) lean SD ON 0 0 ~00 HO 0 0 OH O4» 0 ~20! HH 0 O U'IH Cluster lumber 2 (lI3) lean 2. SD 1 ~44» :3.- c343 c>»- 0'4 caso ~o~4 cacm Uld Cluster Number 3 (lI5) lean SD HO H HN O ease O O .5 OH O coca O4» 0 0 a» HH 0 O0! 0 I!” Cluster lumber 4 (lI4) 9.8 lean 4.0 3 SD 1. O4» e W“ O e O Ow e UIUI O e O O09 bu Cluster lumber 5 (lIl) lean 4.0 4.0 4. SD 0.0 0 OH e OO O“ e OO OO e Cluster lumber 6 (lI5) lean 2.4 2. 4.0 SD 1.4 1 221 Table J-16 Cluster Analysis of Sussed Attanucci categories and Quality of Object Relations in Early Rosaries scores - Fard's lethod, Squared.luclidean Distance coefficient - cosplete Sasple, N'- 28 Cluster SUlI SUlII SUlIII SUlIV E-DAD I-HOH ll Cluster lunber l (l-IO) lean SD HN O 0 ”0" 04 H” O 0 HQ pa NO as I-‘N O O 0.0" ..a Cu Out 0 h“ HH Cluster lunber 2 (la!) lean 3. SD 0. HM O I H“ b” Ow cow ran- Cluster lunber 3 (l-9) lean SD h‘h) peso th\ 0 0w 0 0w 0 s at“ OH 0 s ‘1'“ N” a ”at Cluster lunber 4 (l-S) lean SD H» . ca 0. 00 00 Hu 0 ”N 222 table J-l? Cluster Analysis of Sussed Attanucci categories and Quality of Object Relations in Early laaories scores - arena lethod, Squared Euclidean Distance Coefficient - Restricted Sasple, R'- 24 gluster SUlI SUNII SUlIII SUNIV E-DAD B-lOl Bl Cluster lunber 1 (l-9) lean SD Cu 0 O ~IO-O mu 0 O HO OH! O O ‘1. HM we. a... HH 0 C an Cluster lunber 2 (l-3) lean 2. SD 1 O. CO CH a.“ 0‘0 0 0 ON. ~19) O‘ OO Cluster lunber 3 (l-S) lean SD t-‘N Ow .Q Ow am 09-! a» hot-- H0 00‘ 400 Cluster lunber 4 (lI4) lean 3.0 9.8 SD 1 0 Cu O. o 0 0° 0U mm 0 a . cc. .9! Cluster lunber 5 (l-l) lean 4.0 4.0 4.0 SD 0.0 0.0 0 0 Oil-0 0” O OO 00! O OO Cluster lunber 6 (l-G) lean 1 SD 1. One O 09» O 223 Table J-ls Cluster Analysis of Sussed Attanucci categories and Quality of Object Relations in Barly'loaories scores - Hard's Hotbod, Sguared.Euclidean Distance coefficient - Restricted Sasple, R’- 24 Q19 t r SUlI SUlII SUlIII SUlIV B-DAD B-lOl EH Cluster lunber l (l-9) lean SD MN 0 0 DJ" O“ ‘1‘] cu O O” Cluster lunber 2 (lI4) lean SD H00 0 0 HQ” Cu O O has Ow as» HH aw Cluster lunber 3 (l-S) lean 2.9 SD 1 1 Gt.» “‘1 O“ a at“ OW a fill-d e Cluster lunber 4 (l-Z) lean SD U'IUI 4.0 0.5 4.0 0.0 0.5 1.0 OO 09» Ow O O U'IUI O9» 0 0"” HH