Indirect Parental Influence on Mate Choice: A Test of the Psychoanalytic Theory DAVOR JEDLIČKA University of Texas Indirect parental influence on mate choice was ascertained using a recent population Of brides and grooms of mixed parentage in Hawaii. With respect to nativity of parents, the influence of mothers on mate selection is greater than the influence of fathers. Simultaneously, mothers' influence on sons is greater than on daughters, and fathers' influence on daughters is greater than on sons. The data seemed to support the psychoanalytic theory of mate selection f or both sexes. Related literature is interpreted in the light of these findings, and suggestions for further research are offered. Studies of the long-term effects that parents have on the behavior of their children have been among the most difficult and important endeavors in social science. The research in this area emerged as a continuation of concern expressed in the Western literature. Emotional bonds between parents and their children have been stressed since the Greeks. Perhaps, the stories of Oedipus and Electra inspired Freud to theorize about filial relationships with the same- and with the opposite-sex parent (Freud, 1927, 1931, 1933, 1938, 1949).' Reviews of Freudian theory under the version of Oedipus and Electra complexes show that clinical studies verify the theory (Winch, 1949a, 1950; Neubaur, 1960), while empirical tests have been interpreted as inconclusive (Eckland, 1974; Murstein, 1976; Simpson, 1960; Udry, 1974). These authors based their conclusion on studies by Commins (1932), Kent (1951), Kirkpatrick (1937), Mangus (1936), Prince and Baggaley (1963), Strauss (1946) and Winch (1943, 1949b). A study by Aron et al. (1974), which was published about the same time as some of the reviews, was equally inconclusive. The psychoanalytic theory of Oedipus and The author is grateful to Dr. Thomas A. Burch, chief and to the staff of the Research and Statistics Office of the Department of Health, State of Hawaii, for provid-mg the data that made this study possible. Department of Social Sciences, University of Texas, Tyler, TX 75701. Electra presents the view that a man is likely to marry someone resembling his mother and that a woman would marry someone resembling her father. Because of the large number of possible resemblance indicators, let us focus only on those that males and females can share and that are important to the continuity of family traditions. Some variables that qualify under this restriction are religion, race, national origin, nativity, occupation, political orientation, and world view. These variables are associated with the lifestyles that parents impart to their children. In Farber's words, "socialization of children is aimed at making children duplicates of their parents as these children achieve adulthood" (Färber, 1964:105). Such surface variables are clearly different from the interpsychic variables of the original psychodynamic theory. Intrapsychic patterns, personality, character, needs and organization of defenses are also meaningful criteria of homogamy. Because these variables interact with the surface variables, it is doubtful that even the most prudent use of data will produce highly positive findings. The difficulty lies with perceptual changes that tend to invalidate objective measures of psychic structure. The surface variables, on the other hand, retain validity as they are less likely to succumb to perceptual changes and disagreement between objective measures and participants' perceptions. Therefore, only a small number of traits that children can duplicate through marriage can be ascertained reliably through the currently available methodology. February 1984 JOURNAL OF MARRIAGE AND THE FAMILY 65 . T"e Prem'se of duplication is intuitively appealing and easily verifiable. Decades of studies on endogamy and homogamy rule of marriage" attest to that. However, this theory fails when it comes to mate selection by children whose parents married exogamously. For example, whose side of the family are the children likely to duplicate when ethnic backgrounds of parents differ? The unchallenged answer to this question has long persisted: mothers more than fathers transmit family traditions from generation to generation (Flügel, 1929; Kirkpatrick and Stone, 1935; Mowrer, 1950; Nimkoff, 1942; Winch, 1943, 1946). Consequently, mothers more than fathers are expected to influence mate choice. Winch repeatedly argued that the greater influence of mothers on children is culturally determined. Mothers are expected to be affectionate toward their children, and children are expected to be more emotionally attached to them than to fathers. These cultural expectations, according to Winch, influence mate selection and interfere with empirical discoveries of oedipal tendencies in nonclinical populations. In general, the mother's influence on mate choice is greater than the father's influence. Simultaneously, her influence on sons is greater than on daughters. Though father's influence is weaker, it is more likely to be reflected m mate choice of daughters than of sons. The object of this study is to ascertain indirect parental influence on mate choice of sons and daughters according to Winch's hypothesis and the psychoanalytic theory. Because research assumptions are crucial to this theory, some of the more important assumptions are discussed first. THEORETICAL ASSUMPTIONS Simpson (I960) and Udry (1974) rejected the evidence offered by Strauss (1946) in support of the psychoanalytic theory. They argued that already-married subjects in Strauss's study changed their perceptions of parents to match the characteristics of their spouses. This criticism was based on the assumption that parental images are unstable. However, it is not known in which direction perceptions change. Do they change randomly, or do they systematically become compatible with the characteristics of the spouse? Without such knowledge it is illogical to assume that changes m perception influence only the positive tindmgs. It changes occur randomly or if images are forgotten, one could assume that negative findings also may reflect changing perceptions. In short, correspondence between theory and data should be judged according to how closely a given study approaches the following assumption: Assumption 1: Measures of parental images are robust, external characteristics that retain validity over long periods of time. On the surface, two studies appear to satisfy this assumption (Kirkpatrick, 1937; Commins, 1932). Both researchers used sibling order and age of wife as seemingly robust indicators of resemblance between husband's mother and wife. The rationale was that first siblings more than subsequent siblings will recall their mothers as being young; hence, their wives were expected to be younger. Because results were statistically insignificant, this evidence has been cited as invalidating the psychoanalytic theory of parental influence on mate choice. If the results were what they seemed, this evidence would indeed be convincing. However, until 1980 it passed unnoticed that Kirkpatrick and Commins did not use first and subsequent siblings born to the same mother (Jedlička, 1980). Consequently, some first siblings had older mothers than some higher order siblings. This error alone was sufficient to cast doubt on the importance of the statistical significance in these studies. Flügel (1929) identified another assumption. He cautioned that the expected idealization of the opposite-sex parent is likely to dissipate if parent-child relationships were hostile. Flügel argued that hostility toward the opposite-sex parent inhibits development of heterosexual relationships. Winch (1943) also suggested that sentiments toward parents are important and noted that his subjects were reluctant to admit hostilities toward their parents. On the other hand, Freeman (1955) reported that unpleasant experiences with parents motivated his subjects to date outside their ethnic groups. Freeman's finding can mean that hostility toward parents obscures oedipal tendencies. Therefore, another assumption is important in tests of the psychoanalytic theory: Assumption 2: Overall sentiment of the child toward the opposite-sex parent is positive, Finally, the most salient assumptions concern the selection of resemblance indicators. Researchers usually have selected indicators of r resemblance a priori. That is, they have not considered whether their indicators were indeed rele-. vant to the subjects. Without such a considera--" tion, two assumptions underlie interpretation of: earlier findings: Assumption 3: Indicators of resemblance be-_r tween parents and spouses are relevant to all. subjects. 66 JOURNAL OF MARRIAGE AND THE FAMILY February 1984 Assumption 4: Indicators of resemblance between parents and spouses are equally applicable for fathers and mothers. Both of these assumptions may be invalid without some notion of what is important to whom. It is unlikely that any researcher knows a priori what constitutes a universally salient indicator of resemblance. To avoid ambiguities one methodology proposed below has been designed to avoid unfounded assumptions in psychoanalytic theory testing in general. METHODS AND PROCEDURES Nativity of fathers and of mothers was compared with the nativity of their children's spouses on each of over 32,000 marriage licenses issued in 1978 through 1980 by the Department of Health, State of Hawaii. These records are maintained on magnetic tapes and contain information on state or country of birth for parents of each bride and groom. Other information pertains to brides and grooms only and includes marriage order, occupation, race, and place of birth. Nativity is defined as a place of birth anywhere in the continental United States, Hawaii, Asia (China, Korea, Japan, and Philippines), Europe, British Commonwealth (Australia, Canada, New Zealand), Latin America and Pacific islands other than Hawaiian Islands. This variable is associated with ethnicity; it is external and unlikely to change due to cognitive manipulation. As such, nativity meets Simpson's (1960) and Udry's (1974) strictest requirements for robustness. Marriages were selected only if parents were of mixed nativity and if their children married into one of the parents' nativities. The excluded subjects were more likely to view nativity as irrelevant or to have avoided marrying into parents' nativity as a result of hostilities toward parents. This procedure approximates the assumption that the indicator of resemblance is relevant to the subjects. As an additional precaution, all couples with at least one partner previously married were excluded. Remarried partners tend to -be older and perhaps less affected by childhood experiences. Another step minimized spurious relations. A spurious relation is possible when one's own nativity is identical to the nativity of the opposite-sex parent. In such cases it would be difficult to distinguish between the cross-sex parent's influence and the tendency to marry into one's own nativity. This problem was prevented by selecting subjects who were born in a different region from that of either parent. This rigorous filtering reduced the population to 3814 brides and 3357 grooms, all of whom married into the native groups of their parents. These subjects are a population of all who married in Hawaii during 1978 and 1980 and who possessed the following characteristics: (a) their parents were of mixed nativity, (b) they married into one of parents' nativities, (c) each partner was marrying for the first time, (d) marrying children were born in regions different from that of either parent. Using nativity of each parent and nativity of the spouse, four matricies were prepared: nativity of the bride's father by nativity of the groom, nativity of the bride's mother by nativity of the groom; nativity of the groom's father by nativity of the bride, and nativity of the groom's mother by nativity of the bride. Diagonals of each matrix contained observed numbers of marriages into native groups of parents. In Tables 1 and 2 these numbers are expressed as percentages out of total number of parents in each nativity group. The same tables show percentages of expected frequencies of marriage into nativity groups of parents. These frequencies are a product of row and column totals divided by the total number of subjects in each matrix. Differences between the expected and the observed frequencies indicate the degree of departure from random patterns of table 1. percentage observed and expected frequency of marriage into mother's native group, by children's gender Mother's Daughtera Sona Native Group N Observed Expected Difference N Observed Expected Difference Hawaii 1630 60.3% 49.4% 10.9% 1485 69.3% 38.9% 30.4% Asia 806 21.8 10.0 11.8 727 26.3 11.8 14.5 U.S. mainland 732 54.6 34.8 19.8 705 47.5 27.6 19.9 Europe 150 4.0 0.7 3.3 109 3.7 0.9 2.8 Pacific British Commonwealth 104 43.3 2.9 40.4 71 21.1 1.4 19.7 83 26.5 1.2 25.3 102 25.5 2,0 23.5 Latin America 58 0.0 5.2 5.2 105 3.8 _b 3.8 Total 3563 45.9 32.2 13.7 3304 48.5 26.1 22.4 .For daughters, chi-square = 45.0 with 6 df and p < .001; for sons, chi-square 'he expected frequency was less than one person. 18.3 with 6 df and p < .01. Feb ruary 1984 JOURNAL OF MARRIAGE AND THE FAMILY 67 marriage into parents' native group. Chi-squares in Tables 1 and 2 show that these differences are statistically significant for daughters and for sons. In other words, opposite-sex parent and the same-sex parent homogamy occurs with greater frequency than expected by chance alone. The total number of subjects should be equal in each table. Apparent discrepancies in numbers reflect unusable nativity codes. Missing value and nativity classified as "other" were unusable. RESULTS Table 1 shows percentage differences between the expected and the observed frequencies of marriage into the mother's native group. The tendency for sons to marry into mother's native group is greater than for daughters. Among sons the difference between the expected and the observed frequency was 22.4%; among daughters it was 13.7%. Sons and daughters of mothers born on the United States mainland showed only a slight preference toward nativity of the opposite-sex parent. Most daughters married grooms born on the mainland; but, because of higher expectations of such marriages for daughters, the difference between males and females was slight. Generally, the tendency to marry according to psychoanalytic expectations was evident in the three largest nativity groups: Asian, Hawaiian, and U.S. mainland. Children whose mothers represent small populations in Hawaii tended to depart from the expected patterns. Daughters whose mothers were born in Europe, Pacific islands, the British Commonwealth, or Latin America married into mother's nativity group more often than did the males. However, the differences were too small to offset more definite tendency toward marriage into the group of the opposite-sex parent among larger nativity groups. The most notable exception to the psychoanalytic theory is evident among Pacific islanders, in this nativity group mothers seem to have greater influence on mate choice than in any other group studied. Table 2 shows percentage differences between the expected and the observed frequencies of marriage into father's native group. The tendency for daughters to marry into father's native group is greater than for sons. The difference between the expected and the observed frequency is 13.8% for daughters and 10.9% for sons. This tendency was clear in six out of seven ethnic groups. Children whose fathers were born in Latin America departed from the pattern. Males in this group married more frequently into father's nativity than did females. This exception may be related to the very small size of the Latin American population in Hawaii, Under restricted availability of partners, nativity may lose theoretical significance. Perhaps, the less available a characteristic is in the total population, the less relevant such a characteristic becomes as an image of an ideal mate. DISCUSSION AND CONCLUSIONS Despite a few exceptions, two general tendencies are indicated by the data: First, at least with respect to nativity, mate choice is considerably more influenced by mothers than by fathers. Second, mate choice of sons is more influenced by mothers than is mate choice of daughters, and mate choice of daughters is more influenced by fathers than is mate choice of sons. In general, these data support the psychoanalytic theory of indirect parental influence on mate choice. The lack of empirical support in previous literature probably reflects the lack of attention to fundamental assumptions inherent in tests of psychoanalytic theory using aggregate data. These assumptions were: (a) the resemblance between parents and mate holds over time, (b) subjects come from reasonably happy homes, and (c) indices of parental images are relevant for the sub- TABLE 2. PERCENTAGE OBSERVED AND EXPECTED FREQUENCY OF MARRIAGE INTO FATHER'S NATIVE GROUP, BY CHILDREN'S GENDER Daughtera Son3 Native Group N Observed Expected Difference N Observed Expected Difference U.S. Mainland 1158 51.3% 34.4% 16.9% 935 44.3% 27.5% 16.8tt'o Hawaii 985 64.7 49.9 14.8 789 70.8 56.8 14.0 Asia 957 18.0 10.2 7.8 1060 16.3 11.7 4.6 Pacific 81 21.5 1.3 20.2 63 15.9 1.6 14.3 British Commonwealth 79 43.2 1.2 42.0 67 16.4 1.5 14.9 Europe 77 6.5 1.3 5.2 69 0.0 1.4 -1.4 Latin America 63 3.2 _b 3.2 128 4.7 _b 4.7 Total 3400 43.0 29.2 13.8 3111 37.7 26.8 10.9 aFor daughters, chi-square = 36.9 with 6 df and p < .001; for sons, chi-square ^The expected frequency was less than one person. 20.6 with 6 df and p < .01. 68 JOURNAL OF MARRIAGE AND THE FAMILY February 1984 iects Because these assumptions remained hidden in the related literature, expected theoretical relationships were obscured rather than absent. When the findings supported the theory despite unfavorable biases, the theory was probably valid though statistical significance was not achieved. Strauss' (1946) study deserves re-evaluation because his findings were consistent with those reported here. His subjects could have restructured their perceptions after marriage, as Simpson (1960) and Udry (1974) claimed; however, the results presented here were not affected by the change in perception. This study, then, can be viewed as a successful replication. Other studies also may be reinterpreted. For example, Commins (1932) showed a slight tendency among first siblings to marry younger women than subsequent siblings would. The image of mother might have influenced mate selection. Winch (1946) and Aron et al. (1974) found that males tend to be especially influenced by mothers but that females were no more influenced by fathers than by mothers. An explanation might be that daughters shift the idealized love object from mother to father. This shifting occurs during the resolution of the Electra complex when symbols of sexuality associated with parents are particularly important. If the idealized characteristics of the mother lack sexual connotations, they are less likely to be shifted and focused on the father. Certainly, to some subjects nativity symbolizes cultural continuity without sexual implications. In general, shifting of the ideal romantic image from mother to father is selective for females; for males, idealized images are expected to remain focused on the mother until the Oedipus complex is resolved. In any particular case this explanation can be contested. However, it is becoming more difficult to contest the cumulative evidence in support of the psychoanalytic theory. Findings that show resemblance between a man's wife and his mother and between a woman's husband and her father occur more frequently than expected by chance, implying that the presence of both parents may be fundamental to developing and maintaining heterosexual relations. For example, children raised by the same-sex parent are expected to depart from the cultural norms of mate selection more often than are children raised in two-parent households. In this situation the ideal image associated with the opposite-sex parent is lacking unless a parent substitute is available. Because parental image influences mate choice, those without well-developed images may exhibit a tendency toward random mate selection. Ethnic and religious out- marriages could occur more often than among those raised in two-parent families. Similarly, wider fluctations in age differences at marriage also could occur. Because sociologists have hastily rejected the psychoanalytic theory of indirect parental influence on mate selection, there has been little effort to link the psychodynamics of socialization with the consequences for adults. At the same time, because better theories have not emerged, an important area of research remains unexplored. Given the encouraging results of this study, it can be concluded that the opposite-sex parent serves as a guiding image in mate selection. Therefore, the resemblance between the opposite-sex parent and the spouse occurs more frequently than expected by chance. FOOTNOTE 1. The term Electra complex postdates Freud's formulation. Resolution of the oedipal conflict was applied by Freud to both sexes. Nevertheless, the parallel between Euripides' Electra and Freud's ideas is inescapable. REFERENCES Aron, A., Ain, R., Anderson, J. A., Burd, H., Filman, G., McCallum, R., O'Reilly, E., Rose, A., Stichman, L. and Tamari, Z. 1974 "Relationships with opposite-sexed parents and mate choice." Human Relations 27:17-24. Commins, W. D. 1932 "The marriage of oldest sons." Journal of Social Psychology 3:487-490. Eckland, B. K. 1974 "Theories of mate selection." In M. B. Sussman (Ed.), Source Book in Marriage and the Family (4th ed.). BostomHoughton Mifflin. Farber, B. 1964 Family: Organization and Interaction. San Francisco: Chandler. Flügel. J. C. 1929 The Psychoanalytic Study of the Family. Lon-don:Hogarth Press. Freeman, L. 1955 "Homogamy in interracial mate selection." Sociology and Social Research 6:369-377. Freud, S. 1927 "Some psychological consequences of the anatomical distinctions between the sexes." The International Journal of Psychoanalysis 8:133-142. 1931 "Female sexuality." International Journal of Psychoanalysis 13:281-297. 1933 New Introductory Lectures on Psychoanalysis. New York:Norton. February 1984 JOURNAL OF MARRIAGE AND THE FAMILY 69 1938 "Three contributions to the theory of sex." Pp. 551-629 in A. A. Brill (Trans, and Ed.), The Basic Writings of Sigmund Freud. New York:Modern Library. 1949 An Outline of Psychoanalysis. New York: Norton. Jedlička, D. 1980 "A test of the psychoanalytic theory of mate selection." Journal of Social Psychology 112: 295-299. Kent, D. P. 1951 "Subjective factors in mate selection: an exploratory study." Sociology and Social Research 35:391-398. Kirkpatrick, C. 1937 "A statistical investigation of the psychoanalytic theory of mate selection." Journal of Abnormal and Social Psychology 32:427-430. Kirkpatrick, C. and Stone, S. 1935 "Attitude measurement and the comparison of generations." Journal of Applied Psychology 19:564-582. Mangus, A. R. 1936 "Relationships between the young woman's conceptions of her intimate male associates and of her ideal husband." The Journal of Social Psychology 7:403-420. Mowrer, H. O. 1950 Learning Theory and Personality Dynamics. New York:The Ronald Press. Murstein, B. I. 1976 Who Will Marry Whom? Theories and Research in Marital Choice. New York: Springer. Neubaur, P. B. 1960 "The one-parent child and his oedipal development." Psychoanalytic Study of the Child 22:456-465. Nimkoff, M. F. 1942 "The child preference for father or mother." American Sociological Review 7:517-523. Prince, A. J. and Baggaley, A. C. 1963 "Personality variables and the ideal mate." Family Life Coordinator 3:93-96. Simpson, G. 1960 People in Families. New York:Thomas Y. Crowell (espec. pp. 159-163). Strauss, A. L. 1946 "The influence of parent-images upon marital choice." American Sociological Review 11: 554-559. Udry, R. J. 1974 The Social Context of Marriage. Philadelphia: J. B. Lippincott. Winch, R. F. 1943 '' Relations between courtship behavior and attitudes toward parents among college men." American Sociological Review 8:164-174. 1946 "Interrelations between certain social background and parent-son factors in a study of courtship among college men." American Sociological Review 11:333-341. 1949a "Courtship in college women." American Journal of Sociology 55:269-278. 1949b "The relation between the loss of parent and progress in courtship." Journal of Social Psychology 29:51-56. 1950 "Some data bearing on the oedipus hypothesis." Journal of Abnormal and Social Psychology 45:481-489. 70 JOURNAL OF MARRIAGE AND THE FAMILY February 1984 SELECTED STUDIES IN Marriage and the Family Edited by Robert F. Winch Robert McGinnis NORTHWESTERN UNIVERSITY HENRY HOLT AND COMPANY, New York 434 DETERMINANTS IN MATE - SELECTION men had a median frequency of about 16 to 18 standard city blocks' between themselves and their future wives, whereas the Unskilled men married girls living within three to five standard city blocks. > The findings of this study sustain the original hypothesis, namely, that residential propinquity is a factor in mate selection for white mates who were both city residents of Columbus, Ohio, 1938 and 1946 and that residential propinquity is explainable, in part, as a function of (a) age and (b) occupation of males in Columbus, Ohio, 1938 and 1946. Generalizations applied to age and occupational groups are not wholly correct unless one specifies which age group or which occupational group in a given year he means. What interpretation should be given these findings ? What social inference can we find here? If our findings are correct, then some of Bossard's original ideas that there are "social types in urban communities" who tend to marry may be correct. It is further suggested that because residential propinquity is operative in the city, the parents of boys and girls of marriageable age have unconsciously helped select their son's or daughter's mates by choosing to live in a given urban area. There appears to be a stronger than fifty-fifty chance that a young boy or girl in the city will marry someone living very close to his residence. Here, we might have a predictive device of great value. We must be very cautious, however, before we generalize too freely about residential propinquity. Thus far the studies have dealt with marriage license applications. The findings must be supplemented by additional research, such as interviews, to determine if the residential propinquity reported in the documents is more apparent than real. More research using similar methods to this one should be undertaken to check these findings in Columbus. More researches using different methods are also welcomed as they might reveal discrepancies ihat cannot appear in a statistical study. With Bossard then, we repeat, "Yes, Cupid has wings but he doesn't fly very far. . . ." COMPLEMENTARY NEEDS IN MATE-SELECTION 435 THE THEORY OF COMPLEMENTARY NEEDS IN MATE-SELECTION* Thomas and Virginia Ktsanes Who Marries Whom? The question of "who marries whom" is one which has aroused "common sense" as well as scientific interest. The common sense answer is paradoxical, for while everyone knows that "like marries like" and that "birds of a feather flock together," it is also equally clear that "opposites attract." As is frequently the case in folk wisdom, both assertions are probably true depending upon the characteristics considered. If by "like" one means similarity in regard to a variety of social characteristics such as ethnic origin, religion, occupation, residential location, and social status, then indeed the view that mates tend to be similar seems correct. If, on the other hand, "like" is used to denote similarity in a variety of psychological attitudes, traits, tendencies, or needs, then i the situation is by no means clear. This being the case, it is in order to take a brief look at some studies which have attempted to answer the question of the degree to which homogamy or heterogamy prevails in marital choice. The tendency of persons to select mates who have certain characteristics similar to their own is called homogamy or assor-tative mating. Conversely, heterogamy refers to the selection of mates who are opposites or are merely different. We shall begin with a brief review of the research literature on homogamy. Later we shall present the theory of complementary needs as a special type of heterogamy. Homogamy in Social Characteristics. Interest in the problem of assortative mating is probably an analogical extension out of the field of biology where for lower animals there seems to be a trend toward similarity in size and vitality. On the human level also there is some slight evidence for homogamy in physical characteristics.3 With human * Original Manuscript. The theory of complementary needs was first set forth by Robert F. Winch. For a more detailed exposition of the theory see his book The Modern Tttmily, New York, Holt, 1952, pp. 209-13, chap. 12, and esp. chap. 15. ■"■In Mary Schooley, "Personality Resemblance Among Married Couples," Journal of Abnormal and Social Psychology, 31 (1936), 340-47, some low positive correlations -. JFere found to exist between mates on height, weight, visual acuity, and appearance. 436 DETERMINANTS IN M ATE - S ELE CTIO N beings, however, physical similarity has not been the principal concern. Most work on assortative mating has concerned a variety of. social characteristics. We shall now briefly examine some of this evidence. In an early study by Marvin2 it was noted that there was a greater than chance tendency for marriages to occur between persons with similar occupations. More recently Centers3 has pointed out that there tend to be no wide differences in the occupational statuses of spouses. Burgess and Wallin4 have shown that there is homogamy in educa* tional level. Further, basing their conclusions on the ratings by the couple of the social status of their parents and on their report of the present income of their fathers, Burgess and Wallin state ". . . it is clear that there is a considerable excess over chance for young people to fall in love and become engaged to those in the same social and eco* nomic class."5 Kennedy6 has indicated that there is a strong trend toward homogamy in regard to religious affiliation and a tendency, though less marked, toward homogamy in ethnic origin. Bossard,7 in a study repeated by subsequent researchers, showed that people usually select their mates from those who live nearby. In Bos-sard's classic study more than half of the marriages in his sample were between persons living within twenty blocks of each other. However, the effect of this factor of mere spatial propinquity must not be overemphasized for it overlaps with the factors discussed before. The various ecological areas of the city are characterized by heavy concentratiofls of certain socio-economic classes, ethnic and religious groups; and these groups as noted above tend to be endogamous.8 In summary, the studies reviewed indicate that persons who many tend to be similar in regard to a variety of characteristics such as social class, ethnic background, educational level, religion, occupation, and area of residence. However, these findings actually bear little direct 2 .Donald Marvin, "Occupational Propinquity as a Factor in Marriage Sekctum,1' , Journal of the American Statistical Association, 16 (1918-19), 131-50. 3 Richard Centers, "Marital Selection and Occupational Strata," American journal oj ' Sociology, 54 (1949), 530-35. i E. W. Burgess and Paul Waliin, "Homogamy in Social Characteristics," America« Journal of Sociology, 49 (1943), 109-24. 5 Ibid., p. 114. ' 6 R. J. R. Kennedy, "Single or Triple Melting-Pot? Intermarriage Trends in Nw Haven, 1870-1950," American journal of Sociology, 63 (-1952), 56-59. > 7 J. H. S. Bossard, "Residential Propinquity as a Factor in Marriage Selection," American Journal of Sociology, 38 (1932), 219-24. & Endogamy refers to marriage within the group. COMPLEMENTARY NEEDS IN M ATE - SELEC TI O N 437 relationship to our problem. They are of some interest in that they give "Us a notion of the limits within which another principle of selection may operate. As we interpret ■them, these factors tend to define a field of eligibles from which a mate may be selected on psychological grounds. Homogamy in Psychological Characteristics. Psychological characteristics which have been studied with respect to homogamy include along and varied list. Characteristics investigated by means of "paper-and-pencil" personality inventories include neuroticism, dominance, self-sufficiency, etc. One early study9 found moderately high correlations between mates un neurotic tendency and dominance. Burgess and Wallin10 in their more recent study of 1000 engaged couples found homogamy in regard to a few traits. Their correlations, however, were of a rather low order and are therefore not too convincing. In regard to various "content" attitudes, e.g., religious and political attitudes, there Is some evidence for similarity.11 These similarities, however, may have developed after marriage. The results in this area arc thus considerably short of being definitive. Stagner in reviewing the studies on homogamy in psychological characteristics has pointed out that correlations indicating similarity are higher with respect to intellectual, interest, and attitude scores, but that measures of temperament do not show this tendency as clearly.12 The measures of temperament referred to by Stagner are those estimates of various traits such as dominance, self-sufficiency, etc., which are arrived at by means of paper-and-pencil tests. Confidence in paper-and-pencil tests is vitiated by the fact that subjects Can "fake" their responses and thereby create what they regard as favorable impressions.13 When we try to get behind the picture of personality which the subject wants us to accept, and more particularly, when we want to understand a subject's motivational patterns of which he may be only partially aware, we find no systematic research on the question of homogamous vs. heterogamous mate-selection.I4 In the SE. L. HolTeditz, "Personality Resemblances Among Married Couples," journal of Aknoimal and Social Psychology, 5 (1934), 214-27. 10 E. W. Burgess and Paul Wallin, "Homogamy in Personality Characteristics," Journal 0} Abnormal and Social Psychology, 39 (1944), 475-81. 11T. M. Ncwcomb, and G. Svchla, "In tra-family Relationships in Attitude," Soci-ometrv, 1 (1937), 180-205. iailoss Stagner, Psychology of Personality, New York, McGraw-Hill, 1948, p. 387. ' u Cf. Albert Ellis, "The Validity of Marriage Prediction Tests," pp. 494-95 below. 14 A few individual cases have been reported at this "deep" level of analysis, but they hale been neurotic patients and the authors' reports have lacked experimental control. Cf., eg., C. P. Oberndorf, "Psychoanalysis of Married Couples," Psychoanalytic Review, 35 (1938), 453-57. 438 DETERMINANTS IN M A TE-S ELE CTION absence of experimental evidence various writers have been theorizing on this problem. Toward a More Adequate Theory Ideas about types of harmonic intermeshing of needs have been suggested by various theorists and researchers. Many of these owe a debt to Freud, who made a distinction between "anaclitic" and "narcissistic" love.15 By the anaclitic type Freud meant a love which was expressed in attitudes of self-derogation and reverential admiration toward the love-object. In this type of love one is dependent on the loved one toward whom he can express his need to revere and admire. Narcissistic love is essentially self-love but the narcissist has a great need to be admired by others as well as himself. Thus in his formulation of the narcissistic-anaclitic typology, Freud posited a type of complementary relationship, i.e., the dependent person who has the need to revere and admire is attracted to the narcissistic person who has a great need to be admired and receive adulation. Following the suggestion that persons with complementary psychic make-ups are attracted to each other, several psychoanalysts have proposed that matching occurs between those who are complementarily neurotic.16 According to this hypothesis, for example, a dependent male with unresolved emotional ties to his mother would be attracted to an aggressive and dominant woman burdened with conflicts over her sex role. As a general theory of mate-selection, however, this literature is inadequate because the writers have explained attraction only in terms of the highly individualized neurotic patterns of their patients. What we are seeking is a theory which will be generally applicable, not merely to Freud's anaclitic and narcissistic types of persons, not merely to dependent people who marry nurturant people, not merely to neurotics, but to all kinds of personalities. Gray37 has used a broader approach to this problem. He hypothesized that mate-selection would be complementary with respect to the types 15 Sigmund Freud, "On Narcissism: An Introduction," in Collected Papers, vol. 4, London, Hogarth, 1925, pp. 30-59. 1C Of., e.g., C. P. Oberndorf, op. cit.; Edmund Bcrgler, Unhappy Marriage and Divorce, New York, International Universities Press, 1946; and Bela Mittlcman, "Complementary Neurotic Reactions in Intimate Relationships," Psychoanalytic Quarterly, 13 (1944), 479-91. 17 Cf., e.g., I-I. Gray, "Psychological Types in Married People," Journal of Social Psychology, 29 (1949), 189-200; and "Jung's Psychological Types in Men and Women," Stanford Medical Bulletin, 6 (1948), 29-36. COMPLEMENTARY NEEDS IN M ATE - S ELEC TION 439 of personality formulated by Jung (extrovert-introvert, etc.). His empirical findings, however, were not convincing.18 Other theorists have tried to identify various motivation-linked aspects of interaction. Bernard, for example, suggests various dimensions of love.10 She notes the usual dimension of dominance and also dwells upon the desire for response or acceptance and on the differential ability of persons to "give" as she calls it. As we shall see later, these are similar to some of the "needs" in our conceptual scheme. Bernard did not systematically state that attraction occurred between persons who were complementary in regard to these dimensions. Others, however, have come very close to this notion. Ohmann30 stated this idea by saying that we are attracted to those who complete us psychologically. We seek in a mate those qualities which we do not possess. Taking leads from all of the foregoing, Winch attempted to pull them together. He began by defining love in terms of needs: Love is the positive emotion experienced by one person (the person loving, or the lover) in an interpersonal relationship in which the second person (the person loved, or love-object) either (a) meets certain important needs of the first, or (b) manifests or appears (to the first) to manifest personal attributes (e.g., beauty, skills, or status) highly prized by the first, or both.31 Then he hypothesized that mate-selection would take place according to what he called the theory of complementary needs: In mate-selection each individual seeks within his or her field of eligibles for that person who gives the greatest promise of providing him or her with maximum need gratification.22 Perhaps this can be phrased more simply by hypothesizing that the personality needs of marriage partners tend to be complementary rather than similar. Two points require further clarification: (a) What 18 Winch applied rests of significance to some of Gray's data. These tests showed that the selection of mates in terms of Jung's types was not significantly greater than might have been expected by chance. *9 Jessie Bernard, American Family Behavior, New York, Harper and Brothers, 1942, pp. 435-56. 20 Oliver Ohmann, "The Psychology of. Attraction," in Helen Jordan (ed.), Yon and' Marriage, New York, Wiley, 1942, chap. 2. 21 Robert F. Winch, The Modern Family, New York, Holt, 1951, p. 333. 22 Ibid., p. 406. In the phrase "field of eligibles" Winch takes account of the previously noted homogamy with respect to such social characteristics as race, religion, and social class. 440 DETERMINANTS IN M ATE - S ELE CTIO N COMPLEMENTARY NEEDS IN MAT E - SELECTION 441 í* .1 are personality needs and which needs are germane to our problem? and (b) What exactly is meant by the term "complementary"? Needs. One can think of the term "need" as meaning a goal-oriented drive. Goal in this sense refers not only to such things as material objects and status in the social structure but more particularly to such things as the quality and kind of response desired in interpersonal situations. Examples of the latter are the desire to give help or adulation to others, the desire to take care of others, the desire to control, etc. When these goals are attained, the need is gratified. However, gratification is a dynamic process, and a need once gratified does not cease to function. Patterns of behavior which are tension-reducing tend rather to be reinforced. In a marriage, for example, a woman who finds in her interaction with her spouse gratification for a need to control will continue to want to control him. One further characteristic of needs should be noted. Needs function at both the conscious and unconscious levels. A person may be conscious, partly conscious, or not at all conscious of the goals he desires. Henry A. Murray has defined "need" in a more formal way: A need is a construct . . . which stands for a force . . . which organizes perception, apperception, intellection, conation, and action in such a way as to transform in a certain direction an existing, unsatisfying situation.23 Further, he has elaborated an extensive list of emotional needs. However, because Murray's list is so detailed, we found it necessary to depart from it in a number of ways. The following list of needs24 nevertheless based upon Murray's scheme. is , i -. Needs n Abasement25 To accept or invite blame, criticism or punish- ment. To blame or harm the self. n Achievement To work diligently to create something and/or to emulate others. n Approach To draw near and enjoy interaction with an- other person or persons. 23 H. A. Murray, et al., Explorations in Personality, New York, Oxford University <\ Press, pp. 123-24. 24 R. F. Winch, op. cit., pp. 408-409. 35 The notation "n" before the name of a variable is used as a shorthand form for the term "need," and where it is found on following pages, that is what it represents. n Autonomy n Deference n Dominance n Hostility n Nurturance n Recognition n Sex n Status Aspiration n Status Striving n Succorance Anxiety Emotionality Vicariousness To get rid of the constraint of other persons. To avoid or escape from domination. To be unattached and independent. To admire and praise a person. To influence and control the behavior of others. To fight, injure, or kill others. To give sympathy and aid to a weak, helpless, ill, or dejected person or animal. To excite the admiration and approval of others. To develop an erotic relationship and engage in sexual relations. To desire a socio-economic status considerably higher than one has. (A special case of achievement.) To work diligently to alter one's socio-economic status. (A special case of achievement.) To be helped by a sympathetic person. To be nursed, loved, protected, indulged. General Traits Fear, conscious or unconscious, of harm or misfortune arising from the hostility of others and/ or social reaction to one's own behavior. The show of affect in behavior. The gratification of a need derived from the perception that another person is deriving gratification. A study to test this theory has been undertaken with a group of middle-class subjects. Because striving for upward mobility (or higher socio-economic status) is so central to the middle-class value system, it was decided to include two variables pertaining to status. Complementariness. To explain this theory let us imagine two person, A and B, interacting with each other. Let us assume that both are deriving gratification from this interaction. Then the interactional sequence will be in accordance with the theory of complementary needs if: 1. the need or needs in A which are being gratified are different from the need or needs being gratified in B; or -342 DETERMINANTS IN MATE-SELECTION 2. the need or needs in A which are being gratified are very different in intensity from the same needs in B which are also being gratified. An example of (1) is found in the case of a person desirous of attention ;and recognition (n Recognition) who rinds gratification in relationship with a person who tends to bestow admiration on the former (n Deference). Alternative (2) is illustrated in the interaction between a person who wants others to do his bidding (high n Dominance) and one lacking the ability to handle his environment who is looking for someone to tell him what to do (low n Dominance). It will be recognized that this definition of complementarincss embraces two forms of heterogamy. Points Requiring Further Elaboration. At present the theory of complementary needs is a hypothesis enunciating a general principle of mate selection when both spouses are given some freedom of choice. (It is clear that the theory would not be applicable under such a system ■of arranged marriages as has been traditional in Japan.20) This principle is now under empirical investigation, but the results of this study will not be available for some time. There are a few points to be noted about the theory before the results of the research are known. First, although marriage is viewed as .a major source of gratification, it is a matter of common observation that most married people derive gratification from social interaction with other persons as well as with their respective spouses. To the degree that this is true it is not necessary to hypothesize that marriage partners will be totally complementary in their need-patterns. The theory also hypothesizes, however, that if there is not a minimum degree of complementaryness in the need patterns of the two persons, they will tend to regard the relationship as unsatisfactory. Their dissatisfaction would probably be registered as follows. Either the relationship would be broken during the dating or engagement periods, or if the couple should be married, their marriage would have more ■than the average probability of ending in divorce. At this time the minimum degree of complementariness, referred to in the above paragraph, is unknown, and some criteria are required •concerning the number of needs sufficient to hold a relationship to- 2(5 See the discussion of the system of arranged marriages in the traditional Japanese family, chap. 2 above. COMPLEMENTARY NEEDS IN M ATE - S EL E CTION 443 gether. Other questions which may be raised but which cannot yet be answered are as follows: First, can matching which occurs only on one need in each spouse hold the marriage together? It seems logically possible that only one need of each member of the couple might be met in a relationship. This need, however, might be so important that it would set the tone of the whole relationship. Second, when a person exhibits two needs which are in conflict, for which of these needs is gratification sought in marriage? For example, in the case of a woman who is upwardly mobile and is also very dominant, does she marry an aggressive type male who will get for her the status she desires but who will not submit to her domination, or does she marry a dependent male who will give in to her but who lacks the iaitiative to achieve status? It would be interesting to determine how frequently this type of problem is resolved by the individual's directing one need towards the marital partner and the other towards interaction with other persons. On the other hand, it may be that many persons with this type of conflict never achieve a satisfactory solution and that hence the intrapsychic conflict becomes a source of conflict in marital interaction. Third, in persons who show a marked disparity between needs which are expressed overtly (or directly) and those which are expressed covertly (or indirectly), on which level does matching occur? Persons may behave overtly in a fashion quite different from, or even opposite to, their more basic wishes. We all have known insecure persons whose bold and aggressive exterior is an attempt to convince themselves and others that they are really unafraid. In this situation it may be that matching at the covert level would be more important than at the overt level, but this we do not know as yet. Illustration of the Tlieery To illustrate the theory of complementary needs we have chosen a case from a sample of middle-class married couples and have attempted to show how these two partners complement each other need-wise. It will be noted that in this case the male shows some dependent trends. We do not feel that this case is atypical of our middle-class sample. Dependent needs in the personality of the middle-class male are prob- 444 DETERMINANTS IN M A T E - S ELECTION ably more frequent than is popularly supposed.27 It is to be emphasized that the man and wife discussed here are a normally functioning couple. The Case of Anne and Frank Hamilton.28 Before we can understand how individual needs function for mutual gratification in a marital relationship, it is first necessary to present the personalities involved. Wc shall consider first the wife and then the husband before we attempt to understand their relationship to each other. Anne Hamilton is best described in build as "hefty." Her outstanding features facially are her large mouth and rather prominent teeth. That her mouth is so noticeable the interviewer attributes to the fact that "it never seems to be still." She talks loud and fast. She punctuates her words by dramatic use of her hands and facial expressions. Even when she is listening, her face does not relax. She smiles broadly or raises her eyebrows or in some other way responds aggressively to what is said. Anne's energy is also evident in her capacity to work. To finish college in three years, she carried extra courses each term and still sailed through her undergraduate work. She earned most of the money to pay her college expenses even though her family was able and willing to pay them. But she just liked to keep busy, so not only did she work and keep up her grade average, but she also held responsible positions in numerous extra-curricular affairs. She was so efficient in getting ads for the school yearbook that for the first time that publication had a financial surplus. Going along with this terrific need to achieve, there is a high need to 21 For further elaboration on this point, ej., for example, Arnold Green, "The Middle Class Male Child and Neurosis," American Sociological Review, 11 (1946), 31-41; and Talcott Parsons, The Social System, Glencoe, 111., The Free Press, 1951, esp. pp. 262-69. 28 This case represents one of those being studied in a project under the direction of Dr. Robert F. Winch at Northwestern University. This investigation is supported by a research grant, MH-439, from the National Institute of Mental Health, U. S. Public Health Service. The material upon which the case analysis was done consists of a case-history type interview, Thematic Apperception Test protocols, and a second type of interview designed to get at the more behavioral aspects of personality. The full case analysis was made by the research staff of this project which consists of Dr. Winch, Mrs. Sandra K. Oreck, Dr. Oliver J. B. Kerner, and the authors of this article. The present report is a synopsis of their findings, which cannot be presented in their entirety because the analysis runs to about two hundred pages of manuscript. Much of the documentation for generalizations must be omitted. All names and identifying characteristics have been changed in order to preserve the anonymity of the couple without impairing the crucial facts of the case. It is our desire to present the case as simply as possible for the purpose of illustrating the theory. COMPLEMENTARY NEEDS IN M A TE - SELEC TIO N 445 dominate others, which Anne describes as "a certain element of bossi-ness in me." She feels that her way of doing things is best and she wants people to do things "in the manner I so designate."29 She does not like to be "stepped on".nor does she admire people who can be pushed around. Such people she cannot respect. "People that I cannot look up to, I have a tendency to shove out of my way or to trample on, just shove, push." Thus we see in Anne little need to feel sympathy for other persons (n Nurturance) but rather a hostile attitude towards them. She tends to be critical of other people and apparently because of this she has encountered some difficulty in forming close friendships. She says that people usually like her if they can overcome their first impression which frequently is one of antagonism. She says on this point, "I'm very quick spoken and rarely stop to think that I may he hurting somebody's feelings or that they are not going to take it just the way I meant it." But she needs people and she wants them to like her. The competitiveness and the need to manipulate people undoubtedly indicate compensatory behavior for feelings of insecurity at some level. There is some evidence to indicate that these feelings stem from her doubts about her being a feminine person. She tends to be jealous of pretty women. She is contemptuous towards them when their attractiveness and "poise" win them positions of prestige which they are not equipped to handle because of a lack of the "executive ability" that she possesses. All her life she states that she wanted to be like her mother who is pretty and sweet and "gives a lot, perhaps too much." She feels, however, that she has not succeeded in becoming this sort of woman. She regards herself as a person who is "quick, uneven-tempered and impatient, ambitious . . . ready to tell others how to do things." Evidence that she rejects this "masculine" component in her personality is her view that she would not want a daughter to be like herself, but "more like Mother." The postulation of such a conflict helps to explain why Anne did not continue with her career plans. She took a master's degree in advertising the year following her undergraduate work- She then set out to make a career in this field, but there were no jobs immediately available. Employers did not want college graduates who had their own 29 Shortly we shall note that this domination of others occurred very early in her life in her relationship to her parents and other members of the household. 446 DETERMINANTS IN MATE-SELECTION bright ideas about the business, and, according to her account, they were unwilling to employ her for menial jobs which she was willing to take because they felt she was too intelligent and soon would become disinterested. At this point Anne's career drive began to fluctuate. She took a job in an office. While there and while formally engaged to another man, she met Frank. She and Frank were married six months after their meeting, and they moved to a city where she had obtained a good job and where he enrolled in college. At the end of a year she became pregnant and stopped working for awhile. By the third month of her pregnancy, however, she became bored with "sitting around home" and took a job as a waitress, much against the doctor's orders. She lost the child three months later. She staled that she wanted the child very badly and that she was broken up over her loss. This wish would be consistent with the feminine desire to be a "mother." In addition to the conscious desire to be feminine, it seems probable that she had an unconscious wish to abort and to deny willingness to play a feminine (maternal) role. Perhaps if we look into Anne's background for a moment we can see more clearly the circumstances which led to the development of her pattern of aggressive behavior and the confusion over appropriate sex-role behavior. Anne was the only child in a family of four adults. Her father was a self-made man, one who built up a trucking business to the point where it netted him an income of around $700 monthly even during the depression years. She describes him as being a short man, one who was hot-tempered and stubborn. He was 30 when Anne was born and her mother was only 18. The mother is described as being even-tempered, calm and dependent. The third adult was Anne's maternal grandmother who came to live with the family shortly after Anne was born. She managed the house and Anne's mother and apparently Anne's father as well. Anne says her grandmother often warned the father against his outbreaks of wrath in front of the child. The grandmother brought with her one of her sons who was about the age of Anne's father and who was similar to Anne's mother in temperament. He was very good to Anne and gave her everything she wanted. He married for the first lime and left the household when he was 50 years old. Anne was the center of attention for these four persons. What she COMPLEMENTARY NEEDS IN M A TE- SEL E C TION 447 could not get from one, she could get from another. This pattern of relationships was conducive to her manipulation of persons and the need for recognition from them which we have noted earlier. Grounds for the competitiveness may also be found in this network of relationships. Anne's mother was very young and still dependent upon her mother who looked upon Anne as "her youngest child." Thus the relationship between mother and daughter resembled sibling rivalry, not only for the "mutual mother's" love but for the husband-father's love as well. Here were two bases for Anne to dislike her mother, but her mother was such a sweet young thing that she never gave Anne any rationalization for hating her. This left Anne with an unexpressed hostility which apparently has been partially sublimated into an achievement drive and partially displaced onto "feminine" women like her mother. Her mother was better looking than she, so Anne could not compete with her on these grounds but had to seek other means of achieving superiority. To strive in an aggressive manner was satisfactory in another way too because the father, who wanted a son, approved of such behavior in his little tomboy. Further, grandmother was a model of aggressive behavior. Anne's gratifying relationship with her fostered an identification. The aggressive pattern was fairly well set by the time Anne leached adolescence as is evident in her report that, in junior high school, teachers commented on it. One teacher advised her to change her ways or she would never get a husband. Father also changed his mind about what he wanted and began to look upon her as "feminine" and wanted her to become dependent on him while she was in college. These undoubtedly are the sources of some of the ambivalence we note in her picture, especially concerning career and motherhood. Although she had doubts about her "feminine appeal," Anne apparently had little trouble in rinding dating relationships. Though she confesses she was not the most popular girl on campus and that her weekend calendar was not always filled, she dated from the time she first entered high school. She had only one serious relationship before meeting Frank. This was an engagement to a man described as "suave and smooth . . . and with nice manners." It apparently was a stormy affair, off and on several times. The engagement was broken finally over the issue of whether or not there should be a formal wedding. Anne wanted one, but her fiance's family did not. Frank is unlike Anne in many ways. Whereas she gets much grati- 448 DETERMINANTS IN M A TE - SELECTION rkation from work and positions of responsibility, he much prefers just loafing and being with people. He is now in college, at Anne's request, and very much looks forward to the time when he will be through. College is just a means to an end for him; the less work he has to do to get through, the happier he will be. He wants the degree, however, because it will facilitate his getting a good job. He looks to the job to bring him status and prestige and to provide a large income so that he can buy sporty cars and a big house. Never-theless, he does not like to work for such a position and is just as content if someone gets it for him. Frank likes people and he gets along with them very well. It is important to him that they like him and give him attention. He loves to talk and to joke, and generally he is successful in winning friends. "I'm an easy person to get along with ... I do a fair job of amusing people although I feel that people don't regard me as entirely full of nonsense." His physical appearance contributes to his acceptability for he is a good-looking man, tall and slightly heavy. His build is somewhat athletic but his muscles seem to lack the firmness and tonus of a well-developed athlete. He is light-hearted, pleasure-oriented, and loves to eat.;,° To achieve acceptance Frank relates to people in a deferent manner. He consciously admires and accepts his allies almost uncritically. He shows no tendency to control them nor to compel them to do what he wants; in other words, he reveals no need to dominate. Though he likes very much to have the spotlight himself, he is willing to share it with others and even to concede it without resentment to people who are better attention-getters than he. He tends to establish friendships with such persons and to identify with them. Thus he receives vican-ous gratification for his own need for recognition. This is illustrated in the fact that he joined the fraternity to which most of the "big wheels" on campus belonged though he himself was not a big wheel. Merely through association he felt he was able to share in their glory. It is interesting to note that Frank does not limit his struggle for 30 In terms of the Freudian stages oř development, this aspect of his personality would place him at the "oral" stage, the stage at which the infant, for example, docs little more than receive love, care, and attention from the mother. The passive-dependent trends which we note in Frank's personality ai'e considered the psychological counterparts of this stage of development. We shall note, however, that this characteristic is by no means the whole picture and that he is considerably more active than is implied for this stage. COMPLEMENTARY NEEDS IN MA TE - S ELECTION 449 recognition to a few fields or a select group of persons as mature adults generally do. He is almost child-like in his willingness to perform. Once when drunk, he paid the singer in a night club twenty-five dollars to let him sing with her in front of the microphone. He still wears the badge that he received when he was deputized a sheriff for a week in his hometown. The importance of this incident was shown when Frank flipped his lapel so the interviewer could see the badge. In addition to recognition, Frank seems to want love and affection. He tells that he was the "mascot" of a sorority at the first college he attended, and he was chosen "king of the prom" one season. If he feels blue, which he says is rare, he can be cheered by having women, peers or the mothers of peers, tell him how handsome he is. Apparently since high school Frank always got along well with women because he always had a girl. He tended to date one girl at a time and to go with her pretty "seriously." He expected the same of her, and as a result most of these relationships broke up by his becoming jealous when the girl would date another fellow. He became jealous he says because he wanted "all her attention." The girls he dated were all short and very attractive. They conformed to his "ideal" of "one other fellows thought highly of, a popular girl in other words." Apparently a girl of this type brought vicarious recognition to Frank in the same manner as did the "big wheels" in the fraternity. Now let us consider Frank's background, Frank was the third son in a family of four boys, all of whom were born during a period of eight years. His father, who was 57 when Frank was born, was a successful salesman until the depression. After losing everything in the depression, the father stopped working. The major burden of supporting the family then fell upon his mother who was about 28 years younger than the father. In time this responsibility was shared by the oldest son. The mother was a petite and good-looking woman.31 She was a very hard-working, efficient sort of person who, besides working at a full-time job, kept her house, herself, and her sons immaculately neat and also found time to participate in a few club activities. She had considerably more education than her husband in that she had a B.A. degree whereas he completed only the eighth grade. Frank remembers her as being undemonstrative in her affections and as a reasonably impartial judge in the children's quarrels but with a tendency to side with tire underdog. Frank had little to say about his 31 It will be recalled that the girls he dated were o£ similar stature. 450 DETERMINANTS IN M ATE - SELE CTION father's personality. Though the man had died only two years before the interview, Frank gave the impression that his father had participated little in family affairs. Frank's few descriptive comments portrayed an opinionated man, harsh in his judgments. Among the seemingly more important aspects of this family is the absence of daughters. Having two sons already, both parents had desired that the next children be girls. Indeed Frank can remember the time when his mother gave him a girl's haircut. It would appear therefore that this attitude on the part of his parents, and especially his mother, laid the groundwork for the passive-dependent trends we have noted in his personality. It seems logical that Frank wanted the love and attention that is given to the baby. At the age of two years, however, he could no longer be gratified in these desires because of the arrival of the fourth and final brother. It appears that Frank resented this brother greatly. In one two-hour interview he mentioned both of the older brothers but not this one. Undoubtedly as a consequence of this situation Frank has developed a fear of rejection to which he has responded by always doing what is expected of him and by endeavoring to please people in order not to be rejected by them. Frank did not react to his feeling of rejection by rebellion. Perhaps this was because the mother never actually rejected him; she just did not give him all the affection he desired. To avoid losing what he did receive and to try to get more he reacted by being a "good boy." But Frank was not a sissy in the common use of the term. He was interested in athletics and became captain of his high school football team. He liked mechanics and cars. Currently he is studying mechanical engineering and hopes someday to become a salesman for some large engineering firm.'12 These masculine interests are very important for understanding Frank's personality. We have shown the tendency towards dependency in his personality which culturally is considered "feminine." Generally, males in our culture who tend to be passive experience some conflict if they are not able to live up to the cultural imperatives that they be assertive and "masculine." Frank shows little anxiety on this score, however, and appears to be very well adjusted. His not having developed a conflict on this score may be due to his having achieved such 33 It is not surprising that Frank wants to become a salesman because he enjoys so mach talking with people and feels certain that he is able to get along with them well. COMPLEMENTARY NEEDS IN MATE-SELECTION 451 successful identifications with male authority figures that he consciously never questions his "maleness." Undoubtedly, the oldest brother is a significant figure in understanding these identifications with males. Very early this brother became a counsellor to the mother. Frank felt ambivalent towards him. He was jealous because this brother played such an important role with the mother. On the other hand, if he hated his brother, then the mother would reject him completely; but if he were like his brother, he would get his mother's attention and at the same time establish a good relationship with the brother, who was moderately successful in his own business and popular with people. Thus, the brother became an ego-model for him and at the same time was a person who could meet some of Frank's dependent needs. Thus, we now see Frank as an amiable, non-anxious person who does not have a great deal of ambition but who has the knack of relating himself to people who can do things for him. Up to this point we have attempted to describe both Anne and Frank with very little reference to each other. Now we shall discuss their case with relation to complementary need theory. Frank says that he was attracted to Anne because "she's probably the smartest woman I've run into, and I admired her a great deal I think before I truly loved her." On the other hand, Anne admired his easygoing manner and his ability to get along with people. Knowing what we do about each of them individually, we can see in these two remarks alone some ground for their complementary matching. First of all, we have pointed out that Anne has had some difficulty in getting along with people and that she would like to be able to do so.more easily. Frank's ability to attract friends and to keep them facilitates Anne's social relationships in that he attracts their mutual friends. For Frank, Anne's initiative and her ability to attain the financial and other goals she sets for herself complements his lack of drive. The question is open, however, whether or not this particular pattern of i interaction which is now mutually gratifying will continue to be so if Frank becomes a successful salesman. In their interaction with each other we note that Anne has the authority. She handles their finances, and she decided that he should go back to school. As we have seen, this is the way she likes to do things and we have also noted that Frank shows little need to dominate and he accedes quite willingly to her plans. 452 DETERMINANTS IN MATE-SELECTION Anne tends to be a very emotional person who is easily aroused and upset. At such times Frank's calm and easy-going manner is consoling to her. Fie has a good shoulder to cry on and he is willing to listen to her problems. She feels that he is helping to calm her down. About the only thing that disturbs Anne about Frank's personality is that he does not have as much ambition as she would like to see. Indeed she has been somewhat bothered by his rather lethargic attitude towards school work. She would prefer to see him as excited about it as she has always been, but she feels that she is learning to accept his attitude that graduation is the important thing and that the level of one's performance in school is soon forgotten. Occasionally Frank is a little perturbed by Anne for sometimes he is embarrassed when she pushes ahead in a crowd and drags him along with her, but he goes along and says nothing about it. Undoubtedly he is ambivalent about her aggressiveness. On the one hand, her behavior and her drive facilitate the realization of such desires as the new car which they recently bought. On the other band, Frank fears that the same aspects of Anne's personality may put him in a position of stepping on other people which might result in their rejecting him. However, this aggressiveness does not constitute one of the things he would change about her if he could push a button to change anything. Fie would want to modify only her quick temper and her heaviness. Anne is very different from the girls that Frank dated. The other girls were like his mother in physical characteristics in that they were all short and attractive. Anne has none of these physical characteristics, but does resemble Frank's mother in her efficiency. Although very different from Anne's father, Frank tends to be more like Anne's uncle and Anne's mother who are calm, easy-going, and dependent. Both Anne and Frank desire considerable recognition from other people. F'rank is attentive to Anne and considerate of her. She undoubtedly regards his submissiveness to her as admiration. Anne does not pay as much attention to Frank as he would like. It would seem that although Frank would like more in the way of demonstrated "hero-worship," he does not feel too deprived because she facilitates his getting the symbols (e.g., the new sports car) which enable him to attract attention from other persons. There is one other thing about Frank which Anne finds gratifying and which is worthy of mention here. Frank's attractive appearance and engaging manner enable Anne to compete successfully on a femi- COMPLEMENTARY NEEDS IN M ATE - SELE C TIO N 453 nine basis with other women. Although this appeal on his part is gratifying to her in one sense, in another sense it threatens her. She mentioned that she is jealous if he pays too much attention to other women at parties. He also becomes jealous when she has occasion to lunch with another man. This mutual jealousy is understandable in terms of the marked need for recognition which each of them exhibits. On Frank's part, it undoubtedly is a manifestation of his fear of rejection; and from Anne's point of view, the insecurity stems from doubts about her feminine ability "to hold a man." The complementariness that is described in this couple can be summarized generally as a case of a passive-dependent male finding gratification in relationship with a striving aggressive woman (and vice versa). Indeed, they are not complementary on all counts, e.g., neither is willing to surrender his own desire for recognition in favor of the other. However, it would seem that the mutual choice that has been made satisfies the major, predominating trends within the personalities of each.