26 HarTuP enters, N. (1978). Reproductive spation in tanvie, 256 (Supp, Vougha, B., Egeland, B.. Wars, E., & Sroufe. L. A. (1979). Indiviat moter atschmert at 12 and 18 mors: Sibi and change in fonies unr stay, Deveopmen, 30, 971-975, Walersein 1S. & Kelly. J.B. (1981). Suv lvrce. New York: fase Books. Waser, E. (1978). The stability of individ! terse in infant metho ti opment, 49, 483-896 Water, E., Wippman J. & route, L.A, (1979. A 1 daugher Acts Psych ‘the breakup: Hos cre ‘edavioral problems, Jn tion Sympesiom 89). Land Youow, Ly & Goodwin, M. 3. (o 2 change in mcher gure, conpeten laf. Neo Yours, J. (1980). Parente a aro Relating Childhood Relationships to Individual Characteristics Robert A. Hinde Joan Stevenson-Hinde M.R.C. Unit on the Development and Integration of Behaviour, Madingley, Cambridge is chapter we present first an overview of some iss for the understanding of relationships in gener cular. These include the provi ing of the dialectics between ind lonships. on the ane hand, and between rel non the other. We then briefly sketch some f o study the effects of indi jionships and vice versa, PRINCIPLES OF DYNAMICS AND THE NEED FOR DESCRIPTION rependence; and of pos! xe principles is ubiquitously ap24 WARTUP 1. ON RELATIONSHIPS AND DEVELOPMENT — 25, t | Maceoby, &. Ea arin, JA. (983). Socialization in the content ofthe PH. Muss (Sere Ed) mow Dune J, Dale N. (1984). 13 dada: Twonyeu eld cosboratin in jin and wth moter. fa. Bretonen (6), Spal ply: Tropes ing. New York: Academie Pres 1. Social bekavior in youne siblings i rhs family Aitfret sex sys, Child Developmen, $2, 1265-1279, Dana, J. & Kendick, C. (ia pep). Stings w Eritson,E. (1950). Freed, $, (1933), New introductory leer ln paychianalyis. Mew Yotk: Noto, avthrn Gera anit, Page pesc ot nero ‘Tear Mincness, LS. & Nyeqger. CN uno. Piper presented bea Ri be Ime, $3. 372-395 and soe desert, Nex ‘pn rescaled a 'spespetive, American Peyohoant M, RSI-RSE | Hetberngse, EM, & Pac, RD. (19P%). Cid pug a eteopory cept, New i Va ene 4 ‘ote McGraw | JP in pr). Ey een: A reser ages | iat Wine. R A157) Toworde aerate a Hinde, R. A.. Tims, G., Easton, D., & Tampa, A } Posty) behavioe wih strong ssocaes wets mona & Seas, PS. (1953). Some Maccoby, EE. (1980). Suva! deefopmcn. New York: Hareout re Jovanovich Maceoby, £. €. ( fam Develpment during male edhood. Wasdington: Nalonsl Neakiny Pi Maccoby, EE, & Feldman, 8,8. (1972), Moiheratlschnien snd srangerteactions in the tind S.J. Horo H-F-( 1972). Sociale ‘al Paes, 6, 87-29 yea of le. Monographs of the Soci for Research in Child Developme, 17 (Whate No. Sut ah, & Wie Maccoby, E.E., & Gibbs, PK. (1856). Meth of ci nen Manin & C.'B. Stender (EX), eainge h levetayment, New Yorks Marcu & Bc22 HanTuP and the full range of the child's perspective ‘The same can be said of relationships and individual development in historic perspective. The data base on which this essay is based derives mainly f relationships as studied in the United St wee 1970, This has been a period of considerable change i ‘men and women in indus in the ecology of both fa is quite possibl contemporary social changes have not altered the most basic tionships and their impact on the development of individuals ployment outside che home hus become nott to have been major changes in the qu: (Hoffman, 1984). Nevertheless, the manner in which these relationships connect to individual differences and the manner in which ips are cartied for- ward in time may be modified by historical change. Thus, the most import Dutcomes of recent research may be the demo between relationships and development rather demonstrated jonships are placed in cross-ci CONCLUSION We have attempted to construct a chapters included or a compendium of cross-cutting themes and issues. incemmed with wi relationships. Key the other chapters have been individual development, The relevant studies vary greatly i actions were measured as opposed to the ‘elationships.”” We have sometimes made inferences about ce basis of measures that were actually taken o: have made is from trans each: k there are acceptable reasons for doing this, Nevertheless, we ‘acknowledge that the available data do not always * ave raised. ‘The evidence makes clear that cogr fected in various ney do different things with their with respect to control a and communicate differe fecting beth in 1. ON RELATIONSHIPS ANO DEVELOPMENT — 23 2iso indieates that experience in well-functioning relation ‘good functioning in individuals and that important continu between eel: lopment © conceptual schen We nevertheless assu ips are knowab ‘and none are complex. relatively involved in relat ACKNOWLEDGMENT was supported by Gi Manat Bowlby,20 HaRTUP «sis has almost always been stated with reference to the child’s relationship with the primary caregiver and its sequelae, although occasior the constraining effects of early sibling Father-chitd ir importance in middle childhood and adoles- cence has been suggested (Hetherington & Parke, 1979). The importance of the mother-child relationship, however, has been enaphasized again and again in ion to the child's subsequent friendship-cboices, and to second-tenerat relationships. Rescarch on these issues seems now to be Uncertainty and doubt, Earlier theories were probatsly processes thought to undertic these continuities. Neverthe! ty, empathy, und effectiveness in chil id fourth years, a Popularity (Waters, Wippman, & Sroufe, 1979). (b predict either negative interactions with other children, (Pastor, 1981). These studies, however, center on the carly relationship and kuter characteristics of the individual child; their in ions for the formation and functioning of i would expeet, on the basis of the evidence, ‘attachment, as compared with insecurely attached children, would: (a) be more frequently involved in mutual friendships; (5) dships that Tunction ‘more smoothly, and (c) have friendships that ure mor To date, none of but the interest of investigators is ‘elationships predict teacher-chil rl led (see Sears, Whiting, Nowlis, & Sears, 1953) Data are also beginning to emerge that link relationships across yener Frommer and O'Shea (1973) found that mothers reporting major di their year-old infants were mor ‘own mothers and fathers than were mothe: infants. Uddenberg (1974) interviewed 95 new mothers 4 months after the baby was bom, their mothers with respect to parent-child relations some 20 years ccrlior, and their children after they Rad reached the age of 4¥% years. Fi ture of the mother's relationship with her own mother predicted psy Uifficulties in the postpartum period, f a ambivalent or negative reactions tothe infant with her parents were correlated with her chiléren’s perce} after delivery. These associations were sex-linked? Good rel ‘mother and her father were associated with positive descriptions of the mot sons: poor relationships with the father were associated tions by sons. On the other hand, the early re 1. ON RELATIONSHIPS AND DEVELOPMENT t by their sons. foreshadow recent evidenc ly of on re mother's fa mothers press) ad whase own + dependent varia vary avcordi “These isvaes are heya Of this book. Quest 218 HARTUP {erms of manifold qualities was the key to significant advance those relationships and their impact on the ttue for peer relations and their lar state of affairs marks studies indicate that having a si development according to the sex of the child and ing. Rela- ships with s ings, for example, seem to extend the sex-t havior in normative directions whores re! older ones, attenuates these Current studies are focused carly sibling interaction rather th explored in terms of their seque! ‘The importance of doing th ings than between ui than between vale significance of these findings is considerable since itis know emerges in. mother-chi ction before bei be supported not so of the sibling RELATIONSHIPS AND RELATIONSHIPS Across Time Changes occurring within re {quent changes in these same re emergence of secute aituchments marked by effective proxi seeking in one another's presence and com te distal modes of contact for the proxi s used eal iman, 1972). ‘The progression is different for other mothers and their Gren: From “avoidant” interactions during reunion und caretaking, these infants ar 1. ON RELATIONSHIPS AND DEVELOPMENT — 19 ness in confronting problem that pass dyad to dyad, then, cert progressions, ‘As Maccoby and Martin (1983) point out Across Relationships ‘The notion th of ch16 HanTUP 1. ON RELATIONSHIPS AND DEVELOPMENT 17 ‘hrough maternal variables such as physical fatigue and lowered self-esteem, 10 later breakdown, The connection to psychiatric carly difficulty with the relationship ind maternal individuals other than the atric symptoms, ‘There are many other reasons for b the economic and pr , but aot among, jobs (Maccoby & Gibbs, 1954), But a clearer under ies underlying these correla ships appear to heve long-term consequent childhood are characteristic of children “at emotional ‘and behavioral disturbances in adolescence and adulthood. Early childhood ase sessments may not be strong prediciors, but individu differences during middle childhood are correlated with subsequent adjustment. Negative reput social rejection among third- and fourth-graders predict poor ment Peychoseaual ficulties; beginning in carly ada what children Jearn and how much they learn in the peer context is diffi specify. Peer socialization and its relevance to sexual social tion of aggression, und a prosoci however, about the cycles of soci We are only beginning to document selves make (0 child development role i the socislization of children who have friends14 HarTup substance of what we have learned through recent empirical studies forms a large par of the present book and so will not be repeated here. Sulfice to summarise this material: Firs, tbe (ailure to become attached has devastating consequences for individual development. The classic animal studies (Harlow and others) have. {0% been contravened in demonstcaling both short- and long-term conssquences ‘Tuming to instances of these have been difficult 10 determine in case studies catly attachments produce stcong and pers - ' (monkey studies) although reasonably adequute adaptations may eventuate if familiar animals are present to adopt them. Among children, inst ‘caring extending over the first several years creates evelopment of relationships with adoptive puren work themselves 1 ‘through adoption have similar consequences (Yarrow & Goodwin, 1973). There is Some indication in these studies, however, individoats (adults other than the ehild’s parents or other long-term effects. ‘Third, infants who are securely at priate adaptations, both infants who are not sect solving, a5 well as ime goes on, fer as 2-year ols (see Sroufe & Fleeson, :his ‘hese findings, the most experi that ure carried forward more or less directly ture adaptations? Thi investigators question this hypothesis. Do persist over time, serving to iavolve the child in a “ch ‘cumulates toward the individual differences we observe certain qual in the interpretation ‘ones concerning the nature of the mechanisms 1. ON RELATIONSHIPS AND DEVELOPMENT 15, relationships. In turn, global or summary asses ‘measure general dimensions in mother neluding use mother, rather12 HAATUP i 1. ON RELATIONSHIPS AND DEVELOPMENT 13. depending on a variety of factors (Hetherington, 1979). The weight of the evi- is & major challenge in middle childhood. Child.child relations, in some ways, dence suggests that divorce itself stresses the mother-child cal frontiers” of the classroom (Minuchin & Shapiro, 1983), with sess imposed by the confi i tothe divorce does not account ent the nexus defined by the peer culture and the schoo} nat the changes in mother-child interaction (Hetherington, 1979; Wallerstein & Kel- ly, 1981), Economic instability, crises imposed by life-threatening illness, and bereavement are other exam relationships this time. Schooling effects on cl weresting issues, however, can inan (1976), for example, found relatively ‘ional classrooms, as compared with open th more clear-cut consens One example consists of the child's entrance into school, an event that is i ‘elated and requires considerable restructuring of family routines. School en trance is a universal event in Western culture, Few good opportunities exist for ‘examining schooling effects per se (e.g., by comparing children who schoo! with similar children who do not) but schooling seems toe the parent-child re persis seem 10 ships fo greater extent us pressures and their RELATIONSHIPS AND THEIR IMPACT ON THE jon “together then apar INDIVIDUAL, nty as to each other’s whereabouts, normative hievement (See above), In addition, time spent Since relationships are enduring rather than is actually the relationship history dividual variation in personality and social developr hese “conclusions” are among developmental psychology. A consensus about ious di stable over begin w ry act parents inguage development in the early years (Bloom, 1970). The content and patterning of parent-child relationships thus change as a function of school entrance in diverse and su iodes of do develop- ciarlier adaptations? Does the individual carry Forward templates of schemata of relationships or does the individual carty ‘mental change occurring in the two partners. Cross-cultural research would forward “bits and pieces" of relationships into later social ex is volume). At this time, however, we know unswers 10 none of events such as school entrance in terms the size of ‘of interpersonal expectations, partner attitudes, and time-use tionships, And more is needed than an assessm ionships that and the manner in which these transformations ace assimilated into previous relationships that constitutes another. Exogenous events such as school entrance have implications for child-« «elationships as well. Peer interaction becomes sal the preschaot years, but more extensive "coming inctioning attachments. The experimental evidence conces tions and their effects derives en "as supporting the conclusion that relationships bear a causal relation to individual differences in personality development. Th terms” with the peer context10 waaTuP changes in the relation mental press sures at work on the chi fant "drives" the imterac- ‘on with its mother to a greater extent than does the younger infant as evidenced ty differences in configurations of action and reaction in exchanges between them (Stern, 1977). Ordinarily, development co ns to these changes but not the mother's since 6 or 9 months is too short a period to capture anything very signifi: is not co say that the mot tes nothing 10 Rather, we assume that observed changes 's development (or in an effort to stimulate it) rather than being manifestations of developmental change in the mother herself, nee maternal development is so slow in relation to the child's development seldom possible to exar consequences in the short term. For example, the implications of adolescent mothe:hood ca ied by means of Studies that also include new mothers who ar Ke 20s and 30s. Since the children of adolescent mothers and older women may differ, however, easy to unt This problem is even more complex and child-child relationships. One can contrast friendship format year-old and 8-year-old childeen, but itis always necess 4-year old's companions are most commonly other 4-year olds while the com- panions of 8-year olds are other 8-year olds. The confounding between the child's age and the age of the child's companion is complete, of course, wien ‘elationships are studied over time. One child (e.g., a younger one) may develop, ‘more rapidly than its companion (e-g., an older one) but, in each case, both are ‘changing. The difficu ings who were, on average, 25 months older than themselves. ame-sex pairs, there was a significant inergase in the frequency of the infa positive and negative behavior and an increase in the positive behavior of the fferent-sex pairs, there was an increase in negative behavior by both children but no change in the positive behavior of either. It would be 4 competencies for enguging in both positive and negative third your (the time extending between { I | ' i } { 1. ON RELATIONSHIPS AND DEVELOPMENT 11 age changes in the behavior of either the younger or the older sibling observed by Dunn and Kendrick can be attributed only 10 the development of the younger cones, Concord unlikely snce of gender in these ted ony to the emergence of gender norms among the older sage, the younger sib- this study make the younger chi younger el router spect ‘members ot het-understanding, in infants and very young cl 1983). Thus, we conclude ing such measures could advance our efforts to is tionships among children. Exogenous Pressures (Age-Related) snships are sensitive to ‘ionship does not change between 1 laboratory—e ships are not entirely buffered, however, from the offec tional exigency, Profound changes in the relationship (Vaychildren in their late 30s and early 40s reported themselves es: satisfied and ready to be parents but that, overall, there is no single right time” to have children, Other interesting and theoretically relevant issues in Iways considered cent (Hill, in press). And yet, children move through this period in close re tionships with mothers who are themselves entering middle age, ‘Transitional events, such as menopause, have never been examined in relation to these stresses, even though both cross-sectional and longitudinal methods could be ress factor in mother-adolescent re! jn general, generate reworking or restructuring of relationships between mothers and children th ‘mapped onto es ‘Some of the same considerations apply to the development of Fathers a relationships: Idren. The literature in this area is sparse, althou, two investigative teams have studied differences in father-child function of the age of the father. Nydegger (1975, 1981; Mitteness & parental experience iat are naturully confounded with paternal age. Older men erg than younger mien in some respects—they are warmer, ns with fl achievement and individuality, and are less rejecting. On the other hand, ‘older men, as compared with younger men, are less demanding for socialization enforcing di and Weingarten (1980) who found greater partic ‘care by ing fathers than early: a beginning has been made that provides information on the developmental status ofthe father at the outset of the relationship and the father’s behavior toward the child. We do not know, howev’ ‘nd young adul How does the developmental status of a sibling affect his or her relationships 1? Certain investigators have been interested in whether the Wg relations varies as a fu of the aye of the older le, t has been reported that 6-year-old first boms provide (0 their younger siblings than 5-year olds do (Pepler, ‘moviteh, & Corter, 1981) and that second bors acce ‘when the first-born child is 4 rather than 2 years older than the target 1. ON RELATIONSHIPS AND DEVELOPMENT 9 1973). These duta ore difficult (0 interpret, however, OW ing of chronological age with the spacin 183). New analyses are needed in order to argue ferences he developmental status of one sibling or the other as opposed to the jon about sibling age as a deter the target child at the outset of the younger 4 ider ones (Hartup, 1983; Hol socialized more than 10% of the preferred partners who were a year of more ol -age friendships were re frequenily involves chi chronological age than assortments between children who differ in 1978). Exen so, mixed-og not unknown a relationships wit Concurrent Change in Both Partners ‘Thus far, changes in relationships have been considered as deriving from the child’s development or the development of the partner. which these both the At6 HaRTUP 4. ON RELATIONSHIPS AND DEVELOPMENT 7 the strategies used , relevant issues range from the extent to which childrer mainte preschoo years 9 QD supervise their 1D ing ‘the mother's act in regulating the is achieved I the children acknowledge their parents’ it teachers, ‘makes right” (Damon, 1977). Se each pa kis \ Changes in the Partner These changes may be pr guidance and the concomitant increas fecnee. Some writers regard the ma} \ opmental chu the cognitive and emotional development ofthe chi Developmental changes ae also reflected in peer interaction. These interac- tions change, over the first 2 years, from simple "interacts" to more complex ted exchanges to coosdinated needs of others to more complex es, Some investigators (e.g., Brow middle childhood; concr from those involving older women. F ing women in various age groups have not y hanges for interaction among friends, however, are not cleur. In fac rences in friendship interaction have not been closely scrutinized. Si suggest that preschool-aged friends are not nokably more ™’mutu in their interaction than nontftiends (Hinde, in press) whereas school-aged friends are (Newcomb & Bri (00, found that fourth-grade friends a: to share rewards with them than were is necessary to con: which is mon in the4 HaRTuP: INDIVIDUAL DEVELOPMENT AND ITS IMPACT ON. RELATIONSHIPS section, some of the evidence 'o this assertion is examined along with evidence the child's companions bas similar effects ypmental changes in children and in theie companions be seen, esse the develop. at of individuals is associated with the development of relationships Changes in the Child ‘The young infant possesses only rudimentary capaci actions with the actions of other persons, cannot use words, and apparently possesses a limited sense of “self.”” With time, these conditions change, and the nature of the child's interactions with the caregiver change concordantly. The sources of these changes are probably 10 be found both within and outside these interactions. That is, cerain changes may depend on substrutes whose behavioral expression does not depend heavily on experience. On the other hand, the hun- dreds of hours spent by the child with its caregiver inthe first few months of life may also engender developmental change. idies on the development of focused relationships contain examples: mul that infants find atractive change over the first several months for reasons that are probably independent of experience. ‘The infant increasing preference for complexity and contour in visual stimulation has ‘important implications for social interaction, and most writers connect these I changes to changes in the infant's preferences for looking at creasing preference for Lan inmportant to note that these changes, itate changes in the behavior of book caregiver in thei interaction with one another. Mothers respond with pl signs that the infant recognizes or prefers her to other people as represent 've smiling. Thus, changes in the chi behavior vis a vis the significant other but cl ras well AAs the infant grows older, its response 10 separation from the caregiver undergoes considerable change: Crying and distress in response to separation from the mother are relatively uncommon prior to 5 or 6 m hereafter, and reach a peak in the first part of the seco: 1. ON RELATIONSHIPS AND DEVELOPMENT 5. these developmental changes in @ major way (Kaen, Kear 1978). ""We can only conclude that the be governed by a co fon and whether they are young enough to fee! he! ‘of these uncertainties" (Maccoby, 1980, p. 58). The pr these changes for the manner in which mother Uionship are obvious. ‘Still later, children inerease the distance between themselves und their nx crs that they can tolerate in exploring the environment, The rch ‘and their place of proximal signals for mai sured by the faet thal contact can be established rather than its uewwal occurrence. le, by 30 months-of-age, children play more comfortably in a strange ment if they can see the mother rather than if she is behind a screen, even, took toward her more frequently when she 975), At jon away from ‘exploratory act 1984), The devel thus affect she they bave leamed to be reas off changes in We know ret je about the developmental changes in the generate transformations in the mother-child relationship during the preschool Idhood years. Some writers (Maccoby, 1980; 1977) ds growing app ¢ ppersons—perm ‘more assurance than boring changes in the chid’s intz, 1983) and in the abi these chan; ta the mother and, si2 HaRTuP belongingness that are centr 980). tn tis sense, relationships are more than mere contexts for social learning: the sateways or channels to an ever-widening range of experiences Third, the relationships of ehildhood—both those in which the child panic Pates and those thatthe child observes—serve as important temples of models that can be used in the construction of future relationships. Of course tionships do not replicate themselves in endlessly repeated cycle course; the quality of any relationship depends, in part, on the spec with whom the relationship is formed. But consequences of earlier relationships can freque ‘once again, the importance of wl transeends their role in the incu skills and osm Nearly everyone who has studied children has recognized the importance of close relationships in child development, Firs, eame notions tha these relationships is closely are estical periods for them (Freud, 1933). N picted relationships as bilateral which two individuats provide one another with a basis for cont a through selective icipatory tendencies (Sears, 1951). Relationships have also been conceived as goal-conected partnerships that operate as feedback systems (Bowlby, 1969) and, elsewhere, as associations based on “interdcpen- dence" and “reward and cost” in their and maintenance (Kelley & aut, 1978). These notions have enriched the child development literature, ly with respect to our understanding of the attchnient between the infant and its caregiver. The vicissitudes of these attachments have beea traced 13 well as some oftheir antecedents and sequel (Sroufe & Fleeso Moce recently, a beginning bas been made in studying some volved in the emergence and maint But many gaps temaia: For ex jonships between caregiver and chi od and adolescence. The measu vanced: Intimacy, reciprocity, commitment, and disillusionment remain elusive aotions. We know |i the ways in which specific encounters, interac- tions, or exchanges are organized in relationships that extend over time, Finally, \we are just beginning o understand the resources that certain relationships (e.g an the ways in which one relations ges ‘means 10 adopt 2 ‘developmental perspective’ tudy of close relationships. Four assuimp- tions must be made in order to do this: Fist, relationships are describable wi i content, qualities, structure, and patterning (Hinde & Steven- 1, ON RELATIONSHIPS AND DEVELOPMENT 3 volume). It is sometimes useful to describe than one Jeve] of anal ‘when Patterson (this volume) sir wactions occurring between fa Second, we make the assumption tones. Whether examined in terms of “stages” that include “maintenance,” and “term lume) or whether described in some other manne fMuetuate with the passage of time. ‘Third, individuals change over time. The search for the best wa; terize developmental change has bee: forma: iation”” (see Levinger & Levinger, this vol: relationships is clear ionships are, at one and the same time, both independent and dependent variables in social experience. One ean, for e tionships as determinants of change in the individual also consider changes in individuals as det tionships. Because these events occur si three-fold scheme is used to examine these di sess what we know about development viduals as determinants of change in their relationships. together various studies dealing with the impuct of zelationships on inci developmen jon is given to the assessment of relationships future workings as well as the workings of other been exami the young child and the care, ss Been examined as one i sought th, ‘contemporary lection; specimens have besOn Relationships and Development Willard W. Hartup University of Minnesota ips as well asthe inculcation norms. The study of relationships, -ar understanding of their functions and dynamics (Hinde, nly, relationship outcomes remain difficult to specify among both in contact es of the self-system also have their origins ers (Berscheid, this volume; Sullivan, 1953). We do ot know jonships affect these acquisitions becouse most of the child's time is ignificunt others, or because the child’s experience within rel tionships contains unique demands and challenges. Its widely be! function independe wider world, Thus, 2 secure sttact the young child and the mother promotes exploration of the e from the mother (Ainsworth & Bi 975), and soci