Home

Help

Search

Feedback

Contents

Full text of this article Reprint (PDF) of this article Send a response to this article Online Subscription

Temporal course of the basic components of love throughout relationships. In a recent work (C. Yela, 1996) a structural theoretical model of love was proposed, introducing some variations on Sternberg's model (1986), and affirming four basic dimensions: Erotic Passion, Romantic Passion, Intimacy and Commitment. In this work, we aim to test the dynamic side of the model, that is, the temporal course of the components throughout a relationship. Thus, we applied the same scales (including items from Sternberg and other authors) to a sample of 412 subjects. Data analysis confirmed to a considerable extent the hypothesis derived from specialised literature. Nevertheless, we obtained some unexpected results, for which we offer different explanations. Global analysis of the evolution of the components seems to suggest the existence of three main stages in love relationships: "Being in love", "Passional love" and "Companionate love". Finally, we consider some limitations and consequences of our study, making some suggestions for further research.

En un trabajo reciente (C. Yela, 1996) se puso a prueba un modelo teórico estructural del amor, fruto de la introducción de ciertas matizaciones en el modelo de Sternberg (1986), verificando empíricamente cuatro dimensiones básicas: Pasión Erótica, Pasión Romántica, Intimidad y Compromiso. En éste, contrastamos la parte dinámica del modelo, es decir, el curso temporal de dichos componentes a lo largo de la relación de pareja. Para ello aplicamos las mismas escalas (con ítems de Sternberg y otros autores) a una muestra de 412 sujetos. El análisis de los datos confirmó en buena medida las hipótesis planteadas aunque también obtuvimos algunos resultados inesperados, de los que ofrecemos distintas alternativas de explicación. El análisis global de los componentes parece sugerir la existencia de tres fases fundamentales en la evolución del amor: "enamoramiento", "amor pasional" y "amor compañero". Finalmente, se recuerdan algunas limitaciones del estudio, así como posibles sugerencias de cara a investigaciones futuras.

The original Spanish version of this paper has been previously published in Psicothema, 1997, Vol. 9 No 1, 1-15
...........
* Correspondence concerning this article should be addressed to Carlos Yela García. Universidad Complutense de Madrid. Facultad de Psicología. Dpcho. 2.313-B (Box 225. Campus Somosaguas. 28223 Madrid. Spain.
   

n a recent work (C. Yela, 1996) we formulated a theoretical model of the basic dimensions of love, basing this on Sternberg's (1986; 1988) Triangular Theory, and incorporating in it certain modifications, including, notably, the double dimension of the passional component, previously exposed with greater or lesser emphasis by diverse authors, from the classic treatises on the subject (e.g., Rougemont, 1938) to current researchers (e.g., Carreño, 1991, Fraia, 1991). Thus, we postulate the existence of four fundamental components, which we call Commitment, Intimacy, Erotic Passion and Romantic Passion, obtaining empirical support for the structure of the "new" model by means of factorial analysis. In this way, we aimed to try and overcome the contradictions of Sternberg's original model in terms of the definition, assessment and temporal evolution of the components he postulated.
Having tested the structure of the model, the next step is to consider empirically its dynamic, that is, the fluctuation of the components over time. Attempts to construct a systematic theoretical model of love that incorporates the absolutely essential dimension of time are far from abundant. The closest approximations can be found in the works of Kerckhoff and Davis (1962), Levinger and Snoek (1972), Altman and Taylor (1973), or Murstein (1977), though none of these specifically assesses changes in the intensity of love components (postulated as a theoretical model and previously verified empirically) over the course of the relationship.

HYPOTHESIS
An extensive review of the specialised literature permits us to confer a more comprehensive (and, we hope, more solid) theoretical base upon the hypotheses of the evolution of the loving components than that offered by Sternberg (1988). In the above-mentioned work (C. Yela, 1996), we obtained a factor we labelled as "Erotic Passion" (EP), whose factorial structure and content appeared clearly and strongly differentiated from "Romantic Passion", and which referred to the physical    physiological dimension of love: general activation, sexual desire, tachycardia, physical attraction, etc. This EP would grow rapidly under the influence of the stimuli which, both in an innate way and through learning in socialisation, we associate with these responses of physical attraction, physiological activation and erotic excitation. It soon reaches its maximum level, to descend immediately, due to psycho-biological phenomena such as the opponent process (Solomon, 1980) or the socalled "Coolidge" effect (preference for new sexual stimuli; Wilson and Nias, 1976; Dewsbury, 1981; Wilson, 1981; Liebowitz, 1983; Cáceres, 1986; Fisher, 1992; Buss and Schmitt, 1993), and due to certain processes related to general laws of learning, such as habituation and satiation (Skinner, 1953), as explained convincingly by the law of gain-loss (Aronson and Linder, 1965) -which the former's students wittily re-christened the law of infidelity.
The factor we call "Romantic Passion" (RP) groups together a set of ideas and firmly-held attitudes about relationships (peculiar to our culture's stereotype of romanticism, to which we shall return in a later work): intrusive thoughts, idealisation (of the other and of the relationship), belief in something "magical" in the relationship, identification of the couple with the romantic ideal, belief in the omnipotence of love (as a vehicle that should inexorably produce happiness), etc.
The growth process of RP would be more drawn out than that of EP (though briefer by comparison with the non-passional components). Important roles in its emergence would be played by: the attribution of physiological activation and/or physical attraction felt for the other (generally unconscious, and occupying a central role in the initial stages of the relationship -as demonstrated by the Bifactorial Theory of Berscheid and Walster, 1978); personal attraction (influenced by similarity, the obtaining of reinforcement, the halo effect, etc.; Byrne, 1971; Dion, Berscheid and Walster, 1972; Wilson and Nias, 1976; Cook and McHenry, 1978; Griffitt, 1979...); and our own romantic expectations (generally acquired in an unconscious way during the socialisation process; Averill, 1985; Good, 1976; Averill and Boothroyd. 1977; Iglesias de Ussel, 1987; Simon, Eder and Evans, 1992...).
RP's decrease, gentler and more gradual than that of EP, would be due basically to cohabitation as a couple, which implies the progressive reduction of uncertainty and of selective attention (Berscheid, 1983), with an increase in the effects of habituation-satiation (Skinner, 1953), the law of gain-loss (Aronson and Linder, 1965), the law of change of emotions (Frijda, 1988), the attraction of the new and the desire to seduce and to be seduced.
The factor we call "Intimacy" (I), meanwhile, groups together reciprocal aspects related to a "special bond of affective union" (C. Yela, in 1996), such as affective support, understanding, communication, trust, self-revelations, security, comfort with one's partner, and so on. This factor I will tend to grow with cohabitation and the passage of time, chiefly due to the continuous increase in the reciprocity of self-revelations and the number of shared experiences, goals, friendships, emotions, places, etc. (Rubin, 1973; Levinger, 1988; Sternberg, 1988...), becoming stabilised at its maximum point in a hypothetical asymptote. (It is not necessary to stress that we are concerned here with normative romantic love in Western culture of the twentieth century, and not with other styles from other cultures or historical periods -such as recreational love, or pragmatic love-, where such reciprocity of self-revelations, and the course of events here described in general, may possibly not exist).
Finally, the factor we call "Commitment" (C), refers to the decision to maintain the relationship over and above any problems that may result from it, due to the special importance conferred on the other person or on the relationship itself. This factor, conceptually and empirically distinct from the other non-passional component (I), will tend to show a more delayed and gradual increase, in line with the progressive growth of interdependence -both personal and material- between the two partners (Levinger and Snoek, 1972; Altman and Taylor, 1973), which is subject to the not inconsiderable influence of social norms and pressures.
Its high asymptotic level, either as a positive result of the cost/benefit balance (Levinger, 1979) -in other words, of the satisfaction in the relationship-, or due to the influence of processes such as cognitive dissonance (Festinger, 1957), the level of comparison of alternatives (Thibaut and Kelley, 1959), or attributive self-perception (Bem, 1972), will be the main reason for the relationship being maintained (although the relationship might well be sustained in the absence of a "love C", for reasons of other kinds: children, economic dependence, fear of solitude, "what people might say", lack of energy to begin all over again, etc.).
We shall see this more clearly in the following figure, which summarises our hypotheses with respect to the     evolution of the basic components of love over the course of the relationship (graph of combined evolution, which is offered neither by Sternberg nor by any of those who have tried to test his model empirically: Carreño, 1991; Fraia, 1991; Acker and Davis, 1992). We shall try to verify to what extent the graph of the temporal evolution of love components obtained empirically fits the theoretical proposal, and to determine the precise numerical values corresponding to the points of inflection, both on the x-axis (duration of the relationship; measured in years) and on the y-axis (intensity of the love components). In other words, between what values of love intensity the different components range over the course of the relationship, and at what points in time a change takes place, be it an increase, a stabilisation or a decrease in the evolution of each component's intensity. Our review of the specialised literature allows us to affirm, with the obvious caution implied by any such affirmation, that our analysis is unique, not only in Spain, but also beyond our frontiers.
As far as a global analysis of all the components is concerned, we can postulate the existence of three main phases, becoming successively longer, in love relationships, bearing in mind that this development will also depend on a large set of biological, historical-cultural, sociological, demographic, interpersonal and psychological variables which will in fact make every couple different (though what Psychology aims to do is discern certain regularities in human behaviour, over and above variations great or small, which can help us understand it better -and in the best of cases to be able to apply our knowledge for the improvement of well-being and satisfaction in our relationships):
A first phase of being in love -from the start until t2 in Figure 1- (basically EP + RP), a second phase -between t2 and t4 -, of passional love (I + RP + EP, with an increasing C), and finally a third phase -from t4 onwards-, more durable, of companionate love (I + C; with a moderate level of RP and a low level of EP). In terms of what is undoubtedly the most quoted love typology (Lee, 1973, 1976), this evolution would correspond to moving from the Eros style to the Storge style, passing through what is referred to as "Storgic Eros" (though the "being in love" phase also bears certain similarities to the style Lee calls "Mania").
From here on, the relationship may stabilise in this companionate love, it may move on to what could be termed "friendly love" (I + C, with no degree of any type of P), or to become a simple relationship of convenience -for the reasons previously outlined- (C only), or it may break up (with C disappearing completely). These last stages would correspond more to the phenomenon of "falling out of love", "becoming indifferent", the study of which is complementary to that of being in love and love itself, and which we shall attempt to deal with in future research.
In sum, then, we shall try to test the dynamic of our model, checking whether the temporal course of the basic components of love corresponds to that proposed by our hypotheses, and is composed of the three main phases described. It is clear that the methodology employed (sample, design...) obliges us to be extremely cautious, whatever the results. Let us consider the present work, then, as a first step on the road we have taken.

METHOD
Subjects
The sample is made up of 412 persons who, at the time of collecting the data, were involved in a love relationship (of whatever type or duration). It is a quite balanced sample in terms of gender (54% women), and made up primarily of university students, under 25 (mean age 22.9 years), middle class (91%), single (92%), with an average duration of the relationship of 2.9 years, living apart (88%), not economically independent (almost 68%), and without children (97%).    

Instruments
The variables assessed were the following: Existence of a love relationship (necessary condition for inclusion in the sample) Duration of the relationship (in years and months) Basic components of love: Erotic Passion, Romantic Passion, Intimacy and Commitment Socio-demographic variables (those mentioned in describing the sample) For the assessment of the love components, Likert-type scales were used, with 15 items for each component, and a 5-point scale, according to the level of agreement-disagreement with the statement in the item. The scales used, the same as those used for testing the structure of the model (C. Yela, 1996), were constructed with items from Sternberg's (1988) and other authors' scales (Hatfield and Sprecher, 1985; Critelli, Myers and Loos, 1986; and Fraia, 1991). The reliability indices of the subscales (internal consistency through Cronbach's a index) range, in our sample, between 0.89 and 0.93.

Procedure
The questionnaires were filled out voluntarily by Psychology and Social Work students from Madrid's Universidad Complutense (U.C.M.). The researcher made sure all the subjects understood the instructions, did not rush the task, and did not talk about their answers to their colleagues. Anonymity was guaranteed.

Data analysis techniques
There is no doubt that to test a model that includes the temporal dimension the most suitable approach would be to use a longitudinal strategy, in this case over tens of years, but, as so often is the case, the means for such an undertaking were not available (we hope to carry out such a project in the future). This limitation should be borne in mind when assessing the achievements of this study. With a cross-sectional design, then, we established a series of divisions in the variable "duration of the relationship", and made the calculations for the construction of the graphs of temporal evolution using a combination of the sub-program "Graph of SPSS for Windows" (version 6.12) and the program "Harvard Graphics" (version 2.3). We also calculated the analysis of variance, the linear fit or deviation of the curves of each component, and the appropriate correlation statistic in each case (r for the linear correlations and h for the curvilinear ones), which would indicate whether or not the relationship between the variables (love component and duration of the relationship) reflected by each curve was statistically significant (always with an N.C. of 99%, shown by an asterisk).

RESULTS
The values of the variable "duration of love relationship" ranged from a few days to 34 years. Table 1 shows the descriptive statistics for this variable, according to the "temporal divisions" applied for carrying out the analyses.   Considering these temporal divisions in the duration of the relationship on the x-axis (which, it should be remembered, vary in length, so that the scale is not uniform), and the intensity of the love components (range 1-5) on the y-axis, we can observe the curves obtained empirically (Figure 2), as well as the results of the different types of analysis (Table II).

DISCUSSION AND CONCLUSIONS
Evolution of each component
At first sight it is noticeable that there is considerable similarity between the curves predicted and those obtained, the most notable difference being the less pronounced decrease in EP. We can also observe that the averages of the real values range from 3.24 (in C after one month of a relationship) and 4.41 (in I after a 10-year relationship). This may not be surprising, considering that the subjects were in fact people involved in love relationships, and it is to be supposed that if the scores were to fall much below 3 -the mid-point- the relationship would be likely to break up (this could indeed be the subject of future research). The youth of the sample and the fact that it included few couples at later stages of relationships may also have contributed to these generally high averages. However, let us consider each point in turn:
The curve of C is very close to what was expected. On the one hand, the differences between groups are significant (F = 12.69 *); on the other, the goodness of fit analysis reveals that its deviation from linearity is also significant (F = 8.17 *) -there is a quadratic component-, indicating that there are, in this case, two periods of evolution: one of continuous growth, up to (approximately) 4 years, and another of stabilisation, beginning as the first period ends. Meanwhile, the curvilinear correlation is also significant (h = .37 *), confirming the statistical significance of the relationship between the two variables (that is, of the evolution of C over the course of the love relationship). We can see that C is the least important component at first, and that it gradually increases its intensity, even more rapidly than was expected (a result that should be tested in future studies, preferably using a representative sample of the Spanish population), until it becomes stabilised -approximately around the fourth year- at a markedly high level.
The curve of I, in our sample, is also similar to what was predicted. As in the previous case, we found to be significant the inter-group differences (F 4.79 *), the linear deviation or quadratic component (F = 3.26 *) and the curvilinear correlation (h = .24 *). As it can be seen, there is an initial stage of rapid and continuous increase (again up to around 4 years), and a second stage of moderate growth and stabilisation from then on.
It is interesting to note that, as was the case to a lesser extent with C, the increase in I is sharper than expected in the first period of the relationship. This result may possibly be due to the rapid establishment of the perception of the reciprocity of self-revelations and of a large number of shared experiences (factors responsible for the growth of I, according to the model), or to methodological aspects, such as certain characteristics of the sample (e.g., its youth or the low mean duration of the relationship) and/or biases in the self-reports (it is easier and safer for one's self-esteem to think/say that everything is going well, and that one has quickly become close to one's partner). Another plausible explanation lies in that the condition necessary to form part of the sample was precisely that one must already be involved in a love relationship, so that subjects will already have established a certain intimacy with their partner, as opposed to starting from zero (as one would in the interval between falling in love with someone and establishing -if possible- a love relationship with him/her). Future research, using representative samples and including people that describe themselves as being "in love", but not yet involved in a relationship, may point us in the direction of the correct explanation.
The curve of RP also resembles the predicted curve. It is also quadratic (F = 4.27 *), maintaining a systematic relationship with the variable "duration of the relationship" (h = .23 *), and establishing significant differences between the different temporal periods (F = 4.26 *). As it can be observed, RP initially increases gradually, and continues increasing progressively until well after EP has reached its maximum -around 4 years for RP and six months for EP, which, incidentally, constitutes another reason for differentiating between the two types of passion, as indicated by our theoretical model, previously tested empirically (C. Yela, 1996). RP does not descend to levels as low as EP. All of the above coincides, approximately, with what is predicted by the model.
However, a result that departs form the expected pattern is the smaller increase in RP in the initial phases of the relationship. This may be due to the fact that, in comparative terms, the level of I manifested in these phases is greater than expected, due to the factors previously adduced, or to certain methodological biases -characteristics of the sample or of the data collection instrument-, or that, indeed, RP establishes itself progressively, in a more gradual way in the first months of the relationship (an explanation that seems less plausible, but which must also be considered). As in the previous cases, it would be of great interest to see what average levels RP would reach both before the establishment of the relationship (unconsummated or unrequited love), and after the 7-year mark. This is a matter for future research.
Finally, the curve of EP is that which fits most poorly the hypotheses formulated, the analysis of variance not offering statistically significant results (with the N.C. of 99%), although indeed offering a linear correlation (since neither is the quadratic component significant -although there is a meaningful trend) that is negative and significant (r = -.20 *), which indicates the gradual reduction of this passional component.
We can see how in the initial phases of the relationship EP climbs above RP and C, as expected, but not above I (perhaps for the reasons suggested previously for this high initial value of I). As predicted, EP reaches its maximum level relatively early -around six months-, but this is followed, not by the expected drastic decrease, but by a stabilisation phase, followed in turn by a gentle but continuous descent, after approximately 4 years (numerous authors coincide on this point as being when a notable falloff of erotic passion begins, e.g., Fisher, 1992). The question immediately arises: if the theoretical bases for predicting the notable decrease of EP -be it around the 4-year mark or earlier- are as apparently solid as those presented in the hypotheses section (Aronson and Linder, 1965; Solomon, 1980; Wilson, 1981; Liebowitz, 1983, etc.), then why was a much more gentle decrease found in our study? There are several possible answers: On the one hand, it may be due to the influence of certain characteristics of our sample, such as its excessive youth (over 90% of subjects younger than 30), and the facts that more than 90% were single and that nearly 90% did not live with their partner. These factors, which characterise approximately 90% of our sample, are good predictors of EP, in contrast to their opposites: older, living with one's partner, and being married -all inversely related to EP, as we were able to confirm in our own research (age with p < .01, cohabitation with p < .05, and marital status with p < .10). These sample characteristics would appear sufficient to explain this unexpected result. Moreover, though, we know little about what happens from the 6or 7-year mark onwards, given that our sample includes few subjects from the long duration category (and, equally importantly, all of them young people), and a majority of subjects and subdivisions corresponding to shorter relationship periods. It may be that the curve would continue to descend in these later stages. Again, future studies will look into this matter.
On the other hand, the discovery in our sample of EP values greater than those expected in advanced stages of the relationship may be due to the influence of a series of biases in the responses, such as "the defence of the self": many people refuse to recognise this decrease in EP felt towards their partner, or at least they find it difficult to accept (due to the existence of certain romantic myths, not the least important of which is the belief that passion should last "if the love is true"), and the questionnaire does not appear to be the best technique for detecting this decrease. Other biases that may be influencing these higher values of EP are that of "social desirability" (since the tacitly -and sometimes not so tacitly- observed norm is to feel "eternally" the same passion as the first day), that of "cognitive dissonance" (the positive evaluation in all senses of something that has been chosen "freely" and involves considerable effort to achieve and maintain), or the "acquiescence" bias (positive evaluation of items related to I and C, between which are situated those of EP, may contribute to the maintenance of this response pattern in the items corresponding to EP).
Any of the mentioned reasons, or possibly the combination of several or all of them, could provide the explanation for this discrepancy between the results predicted and those obtained with respect to the decrease in EP. Certainly, there is always the possibility that -in the final analysis- all our theoretical assumptions about the sharp     decline in EP are false, but this seems to us the most unlikely alternative in this case. As is usual in psychology, we must wait -as we have pointed out previously- for future research to throw more light on these questions ¹ (see note at the end of the article).

Combined evolution of the components: stages of the love relationship
A very interesting reading of the empirical results emerges from the global analysis of the graph. The results obtained in our sample appear to lend some support to the hypothesis of the 3 principal phases of love relationships: a brief initial phase of "being in love", followed by a phase of "passional love", and the longer one of "companionate love". In our study, the specific points that mark the beginnings and ends of these phases, as can be seen in Figure 2, are 6 months and 4 years (always approximate, of course). Let us look more closely at each one of these phases:

Phase 1 would include the first months of the relationship; in our sample, until around the six-month mark. We shall call this phase, as is common in the specialised literature, "Being in love" (a reciprocal "being in love", since, we insist, for our sample we are ignorant of the intensity of the love components from the commencement of unilateral "being in love" until the establishment of the relationship, when this phase of reciprocal "being in love" actually begins). It is a relatively brief period, in which there is a vertiginous increase of all the love components, especially of EP, which reaches its maximum point, and of I, indicating that the person is subject to a wave of new and intense emotions towards the other, both of general and sexual physiological activation and of the gradually satisfied desire to establish a special affective bond with that person. In this period the least important component is C, especially in the early months.

Phase 2, the intermediate stage between the "being in love" period and the subsequent long phase, comprises, in our sample, the period from the six-month mark to around the 4-year point. We shall call this the "Passional Love" phase, since it is during this period that EP oscillates around its maximum point, while RP continues to increase gradually. It differs from Phase 1 mainly in that, together with these passional components, C and I increase steadily and become much more important.
Finally, Phase 3 begins, for our sample, around the 4-year mark, when the two passional components begin to diminish. We shall refer to this phase as that of "Companionate Love", a term employed by, among others, Berscheid and Walster (1978), Duck (1983), and Sternberg himself (1986, 1988), to substitute the classic term "conjugal love", which implies a matrimonial union that does not necessarily exist (even though in our society it continues to be the most common situation). During this phase, I and C reach their maximum level, while RP, and even more so EP, steadily decrease.
As we mentioned previously, in our study only 8% of the subjects were in relationships that had lasted more than 7 years (a total of 34 subjects, with an average of approximately 10 years duration), so that for relationships of longer duration we cannot state anything with sufficient certainty. We have already, in our hypotheses, pointed out possible alternative theories for long-term relationships. It would be of great interest to try to verify, in future research, the differential incidence of these alternatives, and to try to ascertain which are the factors chiefly responsible for a love relationship taking one road or the other (that is, in general: stabilisation in a phase of "companionate love" -with a greater or lesser steady decrease in the passional components-, or deterioration towards a phase of "falling out of love", "indifference" -with or without a final break-up, depending on other types of variables).

Cosequences
Among the theoretical consequences of the present study is the offer of certain empirical support to the model proposed (based on the introduction of certain modifications to Sternberg's model, 1986, 1988), which attempts to account for the structure (basic dimensions -C. Yela, 1996) and dynamics (temporal evolution -in the present work) of the phenomenon of love. The lack of sample representativeness, the specific characteristics of the sample (youth, short time as a couple...), and the lack of a longitudinal design mean that empirical support for the model is scant and uncertain. However, it does constitute a first step. The confirmation of this model in future studies, with representative samples and longitudinal designs, would contribute to better understanding, explanation, and even -to some extent- prediction, with regard to the phenomenon of love. Furthermore, in addition to its consequences for the understanding and theoretical explanation of the phenomenon of love, the confirmation of the model in future studies, and its relation to a series of relevant variables (jealousy, fidelity-infidelity, physical attractiveness, differences between men and women, sexual and love satisfaction...) may at the very least lead to potential practical applications, both in the wide field of relationships counselling and in a general way through the spread of knowledge of it.
If many of the problems in sexual and love relationships result from ignorance and false beliefs (which cause disappointment and frustration), surely the knowledge of what tends to occur in such relationships, and why, would lead to a questioning of those erroneous beliefs, and contribute to an increase in satisfaction within relationships (or to a reduction of dissatisfaction, depending on how we look at it).
If people (in general; or the couple that goes to therapy) recognised that in a love relationship there exist different types of factors, some passional and others not, subject to many variables (both internal and external to the relationship), which cause these factors to vary in intensity, they would perhaps be better prepared to cope with the changes that occur.
If people (in general; or the couple that goes to therapy) understood that the intensity of the different love components tends, in general, to fluctuate in a certain way throughout the course of the relationship, giving rise to three main phases (a romantic love phase preceded by a brief phase of passional love and followed by a long phase of companionate love, non-passional but with maximal Intimacy and Commitment), then perhaps there would not be so many unfulfilled expectations, and nor would so much grief and disappointment be caused by processes that are absolutely normal (such as, for example, the progressive decrease in passion felt in the initial months -or years).
If people (in general; or the couple that goes to therapy) learnt to accept that the reduction in passion is something natural and inevitable, and that later the love relationship enters another phase in which this passion (though it may not disappear completely) is substituted by a series of positive factors that did not exist previously, or were present but with less intensity (feeling of a stable union and bond, trust in the other's commitment, mutual understanding, commitment to unconditional support, total involvement, maximum rapport and mutual knowledge, total openness, many shared experiences...), then they would not suffer -as is usually the case- when they find after "x" years (two, five or twenty-eight), they no longer feel the unbridled passion of the first day; nor should they deduce from this -as also often occurs- that they no longer "love" the partner, and that the relationship should be ended, since there is no "true love".
For the same reasons, if one is aware of the fleeting nature of "being in love" and passion (but not of love), it is to be supposed that a stable union (marriage or cohabitation) should not be established on the basis of this passional love (ephemeral by definition), but should be delayed until the relationship has entered a more solid and stable phase, in order to avoid the all-too-frequent disappointments that occur once the initial passion has subsided. We are aware that the above goes against the traditional beliefs of the great majority of the population about romantic-passional love as the basis of marriage, but we are also aware of the enormous suffering that many of these beliefs produce, such as the belief in founding marriage -or a stable union- on passional, inevitably fleeting, love, a paradox referred to with more or less emphasis by a wide variety of authors (Ortega, 1917; Rougemont, 1938; Wilson and Nias, 1976; Kinget, 1979; Masters, Johnson and Kolodny, 1982; Liebowitz, 1983; Simpson, Campbell and Berscheid, 1986; Iglesias de Ussel, 1987; CIRES, 1992; Hendrick and Hendrick, 1992...).

Limitations
As in any research, there are certain basic limitations in our study. Apart from other more specific ones, it is important to stress three main limitations: firstly, and as we have repeatedly pointed out throughout this work, the characteristics of the sample -incidental, and not representative of the Spanish population- imply a severe lack of generalisability of the results (as is the case in the great majority of studies in our discipline).
Secondly, as we have also pointed out, an important limitation within the temporal analysis of love components over the course of a relationship consists in the use of a cross-sectional design instead of a longitudinal one, which would be more appropriate for the objective set. Unfortunately, this is another common limitation in psychosocial -and psychological in general- research, given the scarcity of resources (both economic and temporal).
Finally, we must take into account the limitations of the self-report technique (in spite of its various advantages) with respect to other data collection methods, stemming mainly from the fact that what is assessed is not actually the behaviour of the subjects, but rather their opinion on their behaviour (external or internal), with all the possible biases that this involves. Among the techniques of     self-report, the use of the questionnaire involves certain specific limitations (again, despite its many advantages): it is less in-depth than other types of instrument (such as the interview); there is uncertainty as to the motivation and sincerity of the subjects; there are doubts as to whether the subject understands each item in the way intended by the researcher; and there is the possible influence of certain response biases (social desirability, defence of the self, cognitive dissonance, acquiescence...).

Suggestions
We would not wish to conclude without making, if only briefly, some suggestions arising directly from our study, with a view to future research.
The most important task is perhaps that of selecting a representative sample of the Spanish population, from which to obtain results that are reliable and generalisable to the whole population. This is a question to be dealt with by researchers in love behaviour in our country. Also, the inclusion in the sample of each member of a couple (as, for example, in the study by Serrano and Carreño, 1993) would allow us to test all of the hypotheses about the prevalence and importance of similarity throughout a relationship.
Likewise, it would be especially interesting to use a longitudinal design when assessing the evolution of love components throughout the relationship, to try and eliminate possible biases caused by the well-known "cohort effects."
Another interesting suggestion would be to employ the highly-regarded multimethodological strategy in the collection of data. With a little imagination (subject, of course, to the rigour of scientific method), and always respecting the necessary ethical and technical considerations, it is possible to produce experimental designs which, although of quite limited reach and doubtful ecological validity (as always, we are faced with the conflict between internal and external validity), represent a sound advance in the obtaining of causal relationships between variables.
Classic examples of this approach can be found in the work of Dutton and Aron (1974), Dermer and Pyszczynski (1978), Fisher and Byrne (1978), or Clark and Hatfield (1989). Other possibilities, not mutually exclusive, are to carry out systematic observation (e.g., in a pub, in a discotheque, at a party... -though it is obvious that certain intimate interaction behaviours of the couple would not be accessible to the researcher), the recording of psychophysiological responses, in-depth interviews, and the analysis of documents (love letters, private diaries -obviously with the consent of those involved-, adverts, popular song lyrics, etc.).
Another task that lies ahead is the improvement of the psychometric properties of the scales of the love components: for example, to reduce the number of items in each subscale, maintaining the reliability, to achieve greater construct validity and predictive validity, to purge the scales -obtaining items with greater saturation in their factor and less in the rest-, etc.
It would also be interesting to try and check whether or not our study's unexpected results are verified, chiefly the greater than expected intensity in the non-passional components (especially I) in the initial stages of the relationship, the lower intensity of RP in the same stages, and the smaller decreases than expected in the passional components (especially EP) during the advanced phases of the relationship. Thus, it would be appropriate, as we have pointed out, to include in the sample:

1) people "in love", but who have yet to begin a love relationship, in order to confirm whether, as deduced from our theoretical assumptions, they present higher levels of both types of Passion (especially EP) than of I.
2) a numerous group of people involved in relationships of 5, 10, 15, 20 and more years' duration, with the object of confirming whether, as suggested by our theory, the passional components continue to decrease steadily (and to what extent, and when they tend to stabilise).
3) people whose relationship has just broken up, with the aim of checking whether, as our theoretical assumptions suggest, this is related to a decrease in C, and/or a decrease in some component (and which one) to an intensity level below the mid-point (which in our Likert scale, with range 1-5, was obviously 3).
Finally, it would be of great interest to study the reciprocal impact of certain variables on the intensity of each love component (satisfaction -in love, sexual and general-, physical attractiveness -one's own and one's partner's-, fidelity -attitude and behaviour-, jealousy, sexual permissiveness, belief in romantic myths, desire for promiscuity, self-esteem, link between love and sex, etc.), as well as to look into whether differences exist between men and women both in the structure and the dynamics of the relationship (and if they do exist, to try to explain them). All of the above requires independent studies, for which we consider it necessary first to verify the theoreti    cal model proposed with regard to the structure and dynamics of love relationships, as we have done in this work and that mentioned previously (C. Yela, 1996).
As a final observation, we believe there is ample reason to affirm that the famous words of Harlow in his presidential address to the American Psychological Association, several decades ago, is no longer so true: "As far as love is concerned, psychologists have failed in their mission. The little we know about it amounts to no more than simple observation, and the little we write about it has been better written by novelists and poets" (Harlow, 1958; p. 673). For our part, we feel that over the last few decades of research we have learnt something -something beyond mere personal reflection, the obvious, the common sense, and simple asystematic observation. We hope to go on contributing in some way to that.

ACKNOWLEDGEMENTS
I would like to thank the late D. Mariano Yela, Dña. Rosario Martínez Arias, D. Amalio Blanco and D. Florencio Jiménez Burillo for their generous help throughout the years this research took to complete.

Final note (see p. 80):
Recently, we have obtained the first results of a study in which the author collaborated with other members of the Department of Social Psychology of the U.C.M., and which applied a national survey to a representative sample of the Spanish population under 65. The results show a clear and steady decline of Erotic Passion over time, descending to levels even below the mid-point (3.00), and reaching levels inferior to 2.5 in very late stages (from year 20 of the relationship, and even more so from year 35). At the present time we are completing the analyses and preparing the publication of the results (Jiménez Burillo, Sangrador, Barrón, De Paúl, and C.Yela; in preparation).

 

Acker, M. and Davis, M.H. (1992). Intimacy, Passion and Commitment in adult romantic relationships: a test of the Triangular Theory of Love. Journal of Social and Personal Relationships, 9, 21-50.

Altman, I. and Taylor, D.A. (1973). Social Penetration: the development of interpersonal relationships. Monterrey. California. Brooks-Cole.

Aronson, E. and Linder, D. (1965). Gain and loss of esteem as determinants of interpersonal attractiveness. Journal of Experimental Social Psychology, 1, 156171.

Averill, J.R. (1985). The social construction of emotion: with special reference to love. In K.J. Gergen and K.E. Davis (Dirs.): The Social Construction of the Person New York. Springer Verlag.

Averill, J.R. and Boothroyd, P. (1977). On falling in love in conformance with the romantic ideal. Motivation and Emotion, 1, 3, 235-247.

Bem, D.J. (1972). Self-perception Theory. In L. Berkowitz (Dir.). Advances in Experimental Social Psychology. Vol 4. New York. Academic Press.

Berscheid, E. (1983). Emotion. In H.H. Kelley (Comp.): Close Relationships (I 10-168). New York. Freeman.

Berscheid, E. and Walster, E. (1978). Interpersonal Attraction. Reading. Massachusetts. Addison-Wesley.

Buss, D.M. and Schmitt, D.P. (1993). Sexual strategies theory: an evolutionary perspective on human mating. Psychological Review, 100, 2, 204-232.

Byrne, D. (1971). The Attraction Paradigm. New York. Academic Press. Cáceres, J. (1986). Reaprender a vivir en pareja. Barcelona. Plaza & Janes.

Carreño, M. (1991). Aspectos psico-sociales de las relaciones amorosas. Doctoral thesis. Facultad de Psicología. Universidad de Santiago de Compostela.

C.I.R.E.S. (Centro de Investigaciones de la Realidad Social) (1992). La realidad social en España. Erandio (Vizcaya). Ellacuria

Clark, R.D. and Hatfield, E. (1989). Gender differences in receptivity to sexual offers. Journal of Psychology and Human Sexuality, 2, 39-55.

Cook, M. and McHenry, R. (1978). Sexual Attraction. Oxford. Pergamon Press.

Critelli. J.W; Myers, E.J. and Loos, V.E. (1986). The components of love: romantic attraction and sex role orientation. Journal of Personality, 54, 2, 354-370.

Dermer, M. and Pyszczynski, T A. (1978). Effects of erotica upon men's loving and liking responses for women they love. Journal of Personality and Social Psychology, 36. 11. 1302-1309.

Dewsbury, D.A. (1981). Effects of novelty on copulatory behavior: the Coolidge effect and related phenomena. Psychological Bulletin, 89, 3, 464-482.

Dion, K.L; Berscheid, E. and Walster, E. (1972). What is beautiful is good. Journal of Personality and Social Psychology, 24, 3, 285-290.    

Duck, S. (1983). Friends for life. New York. St. Martin's.

Dutton, D. and Aron, A. (1974). Some evidence for hightened sexual attraction under conditions of high anxiety. Journal of Personality and Social Psychology, 30. 4. 510-517. Festinger, L. (1957). A Theory of Cognitive Dissonance. California. Stanford University Press.

Fisher, H. (1992). The Anatomy of love. New York. W.W. Norton.

Fisher, W.A. and Byrne, D. (1978). Sex differences in response to erotica?: love vs. lust. Journal of Personality and Social Psychology, 36, 2, 117-125.

Fraia, G. Di (1991). La passione amorosa. Treviglio. Harlequin Mondadori.

Frijda, N.H. (1988). The laws of emotion. American Psychologist, 43, 5, 349-358.

Good, L.R. (1976). Belief in romantic love. Psychology, 13,1, 6-7.

Griffitt, W. (1979). Sexual stimulation and sociosexual behaviors. In M. Cook and G. Wilson (Dirs.): Love & Attraction. Oxford. Pergamon Press.

Harlow, H.F. (1958). The nature of love. American Psychologist, 13, 12, 673-685.

Hatfield, E. and Sprecher, S. (1985). Measuring passionate love in intimate relations. Journal of Adolescence, 9, 4, 383-411.

Hendrick, S. and Hendrick, C. (1992). Liking, loving and relating. Pacific Grove. California. Brooks Pub. Iglesias de Ussel, J. (1987). Sociología del noviazgo en España. Caja General de Ahorros y Monte de Piedad de Granada.

Kerckhoff, A.C. and Davis, K.E. (1962). Value consensus and need complementarity in mate selection. American Sociological Review . 27, 3, 295-303.

Kinget, G.M. (1979). The "many splendoured thing" in transition, or "the agony and the ecstasy" revisited. In M. Cook and G. Wilson (Dirs.): Love & Attraction Oxford. Pergamon Press.

Lee, J.A. (1973). The colours of love. Ontario. New Press.

Lee, J.A. (1976). Lovestyles. London. J.M. Dent & Sons Ltd.

Levinger, G. (1979). A Social Exchange view on the dissolution of pair relationships. In R.L. Burgess and T.L. Huston (Dirs.): Social Exchange in developing relationships. New York. Academic Press.

Levinge, G. (1988). Can we picture love?. In R.J. Sternberg and M.L. Barnes (Dirs.): The Psychology of Love (139-159). New Haven. Yale University Press.

Levinger, G. and Snoek, J.D. (1972). Un nuevo enfoque en la atracción interpersonal. In J.R. Torregrosa (Comp.): Teoría e investigación en la Psicología Social actual. (435-471). Madrid. Instituto de la Opinión Pública. 1974.

Liebowitz, M.R. (1983). The Chemistry of Love. Boston. Massachusetts. Little Brown.

Masters, W.H; Johnson, V.E. and Kolodny, R.G. (1982). La sexualidad humana. Barcelona. Grijalbo. 1987.

Murstein, B.I. (1977). The Stimulus-Value-Role Theory of diadic relationship. In S. Duck (Dir.): Theory and practice in interpersonal attraction. London. Academic Press.

Ortega y Gasset, J. (1917). Para la cultura del amor. In Obras Completas. Vol II (138-145). Madrid . Revista de Occidente. 1946.

Rougemont, D. De (1938). El Amor y Occidente. Barcelona. Kairós. 1978.

Rubin, Z. (1973). Liking and loving: an invitation to Social Psychology. New York. Hold, Rinehart & Winston.

Serrano, G. and Carreño, M. (1993). La Teoría de Sternberg sobre el amor: análisis empírico. Psicothema. 5, 151I 67.

Simon, R.W; Eder, D. and Evans, C. (1992). The development of feeling norms underlying romantic love among adolescent females. Social Psychology Quarterly, 55, 1, 29-46.

Simpson, J.A; Campbell, B; and Berscheid, E. (1986). The association between romantic love and marriage: Kephart twice revisited. Personality and Social Psychology Bulletin, 12, 3, 363-372.

Skinner, B.F. (1953) Ciencia y Conducta humana. Barcelona. Fontanella. 1970.

Solomon, R.L. (1980). The Opponent Process Theory of acquired motivation: the costs of pleasure and the benefits of pain. American Psychologist, 35, 8. 691-712.

Sternberg, R.J. (1986). A Triangular Theory of Love. Psychological Review, 93, 2, 119-135.

Sternberg, R.J. (1988). El Triángulo del Amor: Intimidad, Pasión y Compromiso. Barcelona. Paidós. 1989.

Thibaut, J.W. and Kelley, H.H. (1959). The Social Psychology of Groups. New York. Wiley.

Wilson, G. (1981). The Coolidge effect: an evolutionary account of human sexuality. New York. William Morrow.

Wilson, G. and Nias, D. (1976). Love's mysteries. Glasgow. W. Collins Sons & Co. Ltd.

Yela, C. (1996). Componentes básicos del amor: algunas matizaciones al modelo de Sternberg. Revista de Psicología Social, 11, 2, 185-201.

Home

Help

Search

Feedback

Contents