Journal of Social and Personal Relationships 2010 Vol. 27(5): 620–636.
Qué es el amor? Una explicación esencialista con base empírica
Kevin E. Hegi and Raymond M. Bergner
Illinois State University, USA
The authors wish to thank Sue Sprecher and Keith Davis for their valuable contributions in designing and critiquing this research. All correspondence concerning this article should be addressed to Raymond M. Bergner, Department of Psychology, Illinois State University, Normal, IL 61790–4620, USA [e-mail: rmbergn@ilstu.edu]. Sandra Metts was the Action Editor on this article.
RESUMEN Este estudio de dos partes investiga nuevas posibilidades en nuestra comprensión de la naturaleza del amor. La evidencia reunida presta apoyo preliminar a las siguientes asesveraciones: Prrmero, los conceptos de amor humano de varias clases de las personas pueden no ser, como se sostiene habitualmente, clases roschianas prototipicas, sino que en lugar de ello pueden ser clases esencialistas y definibles. Segundo, primordial entre las características esenciales del amor en términos de trascender cuatro diferentes clases de amor (romántico, progenitorial [parental], de compañerismo, y altruista) puede ser “inversión en el bienestar del otro por sí mismo o sí misma.” Tercero, los modelos o prototipos de relaciones románticas, progenitoriales, de compañerismo y altruistas de las personas pueden ser diferentes de sus conceptos Esencialistas del amor mismo de cada una de estas clases, y más amplios que ellas.
Palbras clave: amor de compañeros • amor compasivo • essencialismo • amor (naturaleza de) • amor progenitorial [parental love] • prototipo • amor romántico
Este estudio sugiere nuevas respuestas a la vieja pregunta “¿Qué es el amor?” Los datos presentados aquí provenientes de dos studios dan apoyo prliminar a las siguientes aseveraciones. Primero, los conceptos que tienen las personas del amor humano de varias clases puede no ser, como nosotros y otros hemos mantenido previamente, clases prototípicas roschianas (Aron, Fisher, & Strong, 2006; Fehr, 1993, 2006), sino que en lugar de ello puede ser clases definibles y Esencialistas. Esto es, estos conceptos, a diferencia de los conceptos roschianos prototípicos, pueden implicar para las personas que ciertos rasgos esenciales deben estar presentes si deben emitir el juicio de que una persona A ama a la persona B. Segundo, prevalente entre estas características en términos de trascender cuatro clases diferentes de amor humano (romantico, progenitorial, de compañeros y altruista), puede ser “Inversión en el bienestar del otro por él mismo o ella misma” (“IBO=IWB”). Tercero, los modelos o prototipos que tienen las personas de de relaciones románticas, progenitoriales, de compañeros y altruistas pueden ser diferentes de sus conceptos esencialistas de amor en sí mismo de aquellas clases y más amplios que ellas. Estos modelos pueden incluir características juzgadas esenciales para esa particular clase de amor, pero pueden también incluir otras características que se consideran importantes para tales relaciones pero no esenciales para el amor en sí mismo. El trabajo presentado aquí es parte de un cuerpo de investigación en curso emprendido dentro de la Sicología Descriptiva (Bergner, 2000; Bretscher & Bergner, 1991; Davis & Todd, 1982).
Amor: ¿Prototipo roschiano o concepto esencialista?
La costumbre intelectual, que se remonta al famoso análisis de Wittgenstein (1953) del concepto de “juego,” y que fue introducida en la sicología por la obra de Eleanor Rosch (Rosch, 1973; Mervis & Rosch, 1981), ha sido postular dos modos distintos de articular la significación de un concepto. El primero y más tradicionales de estos modos es el esencialista de dar una definición formal; i.e., de establecer las condiciones necesarias y suficientes para el empleo correcto de un término (Ossorio, 2006). El segundo modo es el basado en la demostración de Wittgenstein (1953) de que la mayoría de los conceptos del mundo real no pueden ser definidos formalmente porque no hay un único rasgo que todas las instancias del concepto tengan en común. Por tanto, carecen de la(s) condición(es) universal(es) necesaria(s) y suficiente(s) que se requieren para una definición formal. Lo que tienen en lugar de ello son “parecidos de familia” que los hacen pasibles de un procedimiento de articulación de un prototype (e.g., “en depresión, generalmente pero no siempre encontramos un síndrome que comprende tristeza, anhedonia, desmotivación, fatiga, etcétera”), parecido que es lo que justifica nuestro uso de ese término en cualquier ocasión dada.
En el caso del “amor,” la tendencia reciente en sicología ha sido considerarlo como un concepto prototípico (Aron et al., 2006; Bergner, 2000; Fehr, 2006). Aron et al. (2006) articularon esta perspective bien cuando afirman que “las cobntroversias de filosóficas de largo plazo sobre el significado del amor y la correspondiente diversidad de definiciones conceptuales y operacionales en la literature científica se deben a la posibilidad de que la gente común reconozca instancias de amor no por conformarse a alguna definición formal sino en lugar de ello por su parecido de familia con un ejemplar prototípico” (p.597).
En la investigación presente, reabrimos el tema de si el concepto del amore s esencialista o prototípico. Lo hacemos formulando la hipótesis de que pueda haber alguna característica o características del amor que son consideradas por personas como esenciales para que ellos juzgen si una persona A “ama” a la persona B. Tal como juzgarían que “si no tiene tres lados rectos no puede ser un triángulo” del mismo modo juzgarían que “if person A’s relationship to person B lacks characteristic(s) X, person A does not love person B.”
Además, formulas la hipótesis de que los prototipos pueden seguir siendo importantes, pero de un modo diferente al propuesto previamente. Las personas pueden tener un segundo modelo, más amplio, o prototipo, de los que deben corporizar las buenas relaciones de diferentes clases. Estos modelos incluirían características juzgadas esenciales para el amor de cualquier tipo (e.g., romantic o progenitorial), pero también contienen otras características consideradas importantes para tales relaciones pero no esenciales para el amor en sí mismo. En ausencia de estas últimas características, las personas harán juicios tales como, por ejemplo,“Sí. Su impposibilidad de comunicarse íntimamed es ciertamente un a importante deficiencia en su relación romantic, pero no significa necesariamente que no la ama.” Además, y de un modo aparentemente paradójico, si estos modelos de relaciones son prototípicos por naturaleza, nada ssería esencial en ellos, incluyendo la presencia del amor en sí mismo. Así las personas podrían hacer juicios como “Mi modelo de una Buena relación de Amistad contiene amor como uno de sus elementos, pero ciertamente hay casos que yo consideraría como amistades donde la persona A no ama a la persona B.” El contenido de estos modelos de relación sera consistente con el trabajo previo sobre concepciones prototípicas del amor (e.g., Davis & Todd, 1982; Fehr, 2006; Hassebrauck, 1997).
Methodological requirements
Historically, the following methods have been the most frequently employed in studies of love as a prototypical concept. First, researchers take a large number of participants and ask them to generate descriptors for the target concept (e.g., “romantic love”). Second, other participants are then asked to rate these descriptors with respect to how central vs. peripheral they are to the concept in question (see, e.g., Fehr, 1993; Fehr & Sprecher, 2004; Hassebrauck, 1997).
These methodologies, however, could be employed with any concept, regardless of whether that concept was essentialist or prototypical. One could,for example,employ them for “triangle,”“vertebrate,”and “bachelor,” all of which are classically definable concepts with necessary and sufficient conditions for their correct application. The fact that participants could proceed to follow these directions and generate data, and that researchers could statistically analyze this data, would not be sufficient to establish whether the target concept was a prototypical or an essentialist one.
An alternative means of capturing the conceptual meanings in actual use by persons is anticipated in Fehr’s (2006) observation that “people may not necessarily produce the full range of important features when asked (a recall task), but ‘know them when they see them’ (. . . a recognition task)”
(p. 242). It is an observational commonplace that people use concepts that embody criteria they are unable to articulate. Asked what time it is, they correctly respond by telling us the time. Asked to describe their mother’s personality, they tell us that she is shy, generous, and considerate of others. Asked to give an example of humor, they tell us a joke. However, asked to define or otherwise articulate the concepts of “time,” “personality,” or “humor,” they are for the most part at a loss. They possess the correct distinctions, they make correct judgments on the basis of them, but they cannot articulate well the conceptual criteria they are using or the manner in which they are using them.
Thus, methodologically, if we wish to identify the conceptual criteria in actual use by people, a logically compelling procedure would be to get research participants to make judgments in which they utilize the concept at issue (e.g., judgments about whether person A loves person B), and then to make inferences from their judgments regarding the criteria they are using and how they are using them. In the present instance, some way of determining whether the concept at issue is being employed in an essentialist or a prototypical way must be built into the research tasks. Essentialist concepts, as they have necessary criteria for their employment, are subject to judgments of the form, “If it lacks characteristic X, then it cannot be a case of concept Y”(e.g.,“If John is married,then he cannot be a bachelor.”). Prototypical concepts, per Wittgenstein and Rosch, have no such necessary criteria, and thus are not subject to contradiction based on the presence or absence of any feature. Nothing is a sine qua non for their employment. Accordingly, a research task that involves participants making judgments about the necessity or non-necessity of criteria will enable us to distinguish essentialist concepts from non-essentialist ones. The procedures employed in the present two studies are designed to enable us to determine (i) what, if anything, is seen as essential to judging that person A loves person B; and (ii) what, if any, other characteristics are seen as important, but not essential to love itself, in participants’ broader models of good romantic, parental, friendship, and altruistic relationships.
Característica crítica: Inversión en el bienestar del otro
Foremost among the proposed essentialist features explored in this research is “Investment in the well-being of the other for his or her own sake” (“IWB”). Singer (1984), in his classic three-volume study of the history of the concept of love in Western culture, concluded that such investment was love’s essential feature, and articulated it as follows:
“The lover takes an interest in the other as a person, and not merely as a commodity . . . He bestows importance on her needs and her desires, even when they do not further the satisfaction of his own . . . In relation to the lover, the other has become valuable for her own sake” (1984, p. 6).
In love, then, Mary is invested in the well-being of John for his own sake, and not merely for how his well-being might benefit her. He is for her an end and not merely a means to her ends. Such an investment in the wellbeing of the other is most compellingly expressed as a willingness to act on behalf of the other. This might include such things as supporting or assisting him in times of need, acting to further his interests and goals, and avoiding or preventing anything from happening that would harm or hurt him. In love, John is not for Mary a “commodity;” that is, is not an entity that has a place in her world which consists essentially of satisfying her needs and desires. (This is not to say, of course, that in any real world love relationship there is not some admixture of love and self-interest.)
Further bases for giving preeminence to IWB come from the work of Margaret Clark and her associates (Clark & Mills, 1979; Clark & Monin, 2006), and from that of Rempel & Burris (2005). The former, while they have not promoted it as an essential feature in a formal definition of love, have developed the concept of “communal responsiveness” that is very similar to our concept of IWB. Indeed, given the degree of similarity, the studies presented below can be seen as explorations of the essentialness of communal responsiveness in human love. Rempel and Burris, in their conceptually oriented 2005 account, argue extensively and persuasively for an actual definition of love as “a motivational state in which the goal is to preserve and promote the well-being of the valued object” (p. 299).
We hypothesize, consistent with the authors just cited, that Investment in the well-being of the other for his or her own sake (IWB) will be strongly endorsed by participants as an essential characteristic of the four kinds of love under consideration in this research. In a previous 4-part study exploring this relationship characteristic with respect to romantic love only (Bergner & Davis, 2007), it emerged as the single most strongly endorsed feature in all four studies, with no fewer than 75% and as many as 84.9% of participants endorsing it as essential to romantic love. In the present research, we explore the possibility that IWB might prove the single characteristic that transcends and is essential to four different varieties of love: romantic, parental, companionate, and altruistic.
Aside from the sources just cited, IWB enjoys a varied presence in current research and theorizing on love. In one prominent approach, love is conceived as a term that covers a diverse set of relationships, each of which embodies some combination of three factors, Intimacy, Passion, and Commitment (Aron et al., 2006; Sternberg, 1988, 2006). In this conception, something akin to IWB is present as one subfactor, termed “caring,” contained in the broader factor Intimacy, which also contains items relating to trust, intimate disclosure, understanding, and feelings of closeness, warmth, and comfort. Thus, on grounds of its tendency to modestly but significantly covary with these other subfactors, IWB is incorporated into the Intimacy factor in a manner such that it enjoys little visibility and little emphasis in this approach to love.
In the current literature, finally, some theorists discuss IWB as a type of love. For example, Fehr, Sprecher, & Underwood (2009) discuss it as “compassionate love,” while Lee (1973) and Hendrick and Hendrick (2006) discuss it as “agape” or “altruistic love.” In these accounts, IWB is posed as a distinct kind of love, but not as an essential defining feature of love itself or as one that possibly transcends many different kinds of human love. From the present perspective, it is true that there are relationships in which IWB exists in relative isolation from other relational qualities. Devoted caretaking relationships such as those established by many nurses, hospice workers, and in-home caretakers would be cases in point. IWB being their cardinal characteristic, there is good reason to characterize these relationships on the basis of this characteristic as “compassionate,” “altruistic,” or “agapic.” However, going beyond such cases, we wish here to explore the further possibility that IWB may well be, per Singer (1984) and Rempel and Burris (2005), an essential dimension of love that, perhaps uniquely, transcends a variety of important kinds of human love relationship.
Una nota sobre los conceptos de amor legos y científicos
La preocupación central de esta investigación es capturer las concepciones de amor y de relaciones amorosas de las personas. Tales concepciones a menudo son correctamente presentadas como concepciones “legas”, en cuanto opuestas a las concepciones “de expertos”. Sin embargo, decir que son concepciones legas es decir que son concepciones de uso real en la conducta de la vida de las personas (Kelley, 1983). Corporizan las distinciones que las personas en la realidad hacen cuando están tratando de decider asuntos tan vitalmente importantes como si parejas las aman o si ellas aman a sus parejas. Así que tienen un impacto critic en juicios, decisions, acciones y emociones vitals del mundo real. Por esta razón puede argumentarse que no hay major concepción científica del amor que la corporizada en las concepciones legas. No hay mejor concepción (ningún esquema o estructura cognitive major) para ayudarnos a predecir y entender cómo hacen las personas en la realidad juicios sobre la presencia o ausencia del amor, por qué sienten como sienten, y por qué toman las decisiones que toman (Kelley, 1983).
Hipótesis
Los estudios 1 y 2 ponen a prueba las siguientes hipótesis: Primero, en relación con el amor mismo de las cuatro clases bajo investigación (romantic, progenitorial, de compañeros y altruista), los participantes lo emplearán como un concepto esencialista i.e., juzgarán que una o más características relacionales son necesarias para considerar algo como un caso de cada uno de estos tipos de amor. Segundo, los participantes respaldarán la característica Inversión en el bienestar del otro ( “IBO”=“IWB”Investment in the well-being of the other) como algo esencial par alas cuatro clases de amor, y la respaldarán con más fuerte que a cualquier otra característica relacional. El etudy 2 examina una hipótesis adicional: tercera, en relación con las concepcciones de buenas relaciones románticas, progenitoriales, de compañeros y altruistas, los participantes tendrán modelos o prototipos diferentes y más amplios que lo que cada uno de ellos debe corporizar. Estos tipos comprenderán no solo aquellas características consideradas esenciales These will comprise not only those characteristics deemed essential to the kind of love at issue, but also further characteristics perceived as important for such relationships, but ones whose absence would not necessarily indicate the absence of love itself in those relationships.
Study 1
Methods
Participants. Students (N = 160) from a large midwestern state university participated in Study 1. They were solicited by means of a campus-wide email that went to all students who indicated at class registration time that they were available to participate in research projects. Initially, 250 students elected to participate; however, 90 of these failed to complete the study materials, resulting in a final sample of 160 individuals (131 females and 29 males) who ranged in age from 18 to 54 years, (M = 22.0 years). The majority identified themselves as Caucasian (88.1%), followed by Hispanic (3.8%), African-American (2.5%), Asian-American (1.9%), and “Other” (1.3%). Four participants did not provide their racial/ethnic identity.
Measures. Two revised versions of Bretscher and Bergner’s (1991) “Factors in Intimate Relationships” (FIR) scale were used for this research. These scales were presented to participants with the title “Personal Meanings of Love.” While the FIR focused exclusively on romantic relationships, the Factors in Loving Relationships (FLR) Scale was modified to include four different types of relationships: romantic, parental, companionate/friendship, and altruistic. These relationship types were included as representative of the four basic types of love identified by Berscheid (1985, 2006).
The FLR includes relationship characteristics derived from the earlier research of Davis and Todd (1982), Bretscher and Bergner (1991), and Fehr (1988): Trust, Respect, Acceptance, Investment in the Well-being of the Other, Enjoyment, Emotional Intimacy, Knowledge/Understanding, Exclusiveness, Freedom to be Ourselves,Affectionate Feeling,Sexual Desire,Similarity,Preoccupation, and Commitment. Each of these characteristics is first defined for participants (a list of all definitions is available from the second author). For example, the variable IWB is defined as follows: “In some relationships, we have a sense that each of us truly cares about the well-being of the other. We have a sense that each of us genuinely cares about, and is willing to make personal efforts when needed, to further the other’s welfare and happiness. Such caring may be expressed in various ways. For example, it might be expressed in a desire to give to the other in ways that will make him or her happy . . . or in wanting to help and to stand by each other when the other is hurt or ill or unhappy . . . or in being willing to do things to assist each other in important matters. In all of this, finally, our sense is that the other is not just giving to get. They are not just doing all of this because there is something in it for them. Rather, they are doing it because our welfare and happiness genuinely matter to them.”
Following their reading of the definition of each characteristic, the FLR contra scale calls for participants to consider hypothetical examples of each of four different kinds of relationship. The romantic relationship example portrays a young, engaged couple; the parental relationship example specifies a mother and her sixteen-year-old son; the companionate relationship example describes two adult female friends; and the altruistic relationship example portrays a caregiver and a chronically ill man in his care. In all these cases, nothing more is described than the nature of the relationship between the two persons.
For each type of relationship, participants were asked to consider a scenario in which a specified relationship characteristic (e.g., Trust, Respect, or IWB) is missing on the part of one individual in his or her relationship to the other. For example, for the factor Trust within the romantic relationship, the example reads as follows: “Consider a romantic relationship in which trust was missing on the part of one or both partners. For example, suppose that in the case of our engaged couple, Jill did not trust Jack. To what degree would you find it contradictory to say the following: “Jill loves Jack, but she does not trust him.” Each item is then accompanied by a fiveinterval scale ranging from “very contradictory” to “not contradictory at all,” and participants are asked to judge the case at issue on this dimension.
Items for symmetrical relationships (romantic and companionate) were systemically varied so that the featured person in the hypothetical situation (e.g., the male or female romantic partners) occurred 50% of the time. Items for nonsymmetrical relationships (parental and altruistic) were not varied; the parent and the caregiver, respectively, were the target relaters in all of the items.
The rationale for this form of question is that it distinguishes between whether or not the concept of love of each of these kinds is essentialist or prototypical. Essentialist concepts, since they have necessary criteria for their employment,are subject to contradiction (e.g.,“John is a bachelor,but he is married.”). Prototypical concepts, since they have no such necessary criteria, are not subject to contradiction based on the presence or absence of any feature. Accordingly, the selection of option “very contradictory” signifies that the participant views the concept of love as an essentialist one, in which the relationship characteristic in question is seen as necessary for love of that kind. Participants are in effect endorsing the proposition that, “You can’t say that person A loves person B in this way (e.g., romantically) if element X is missing from his or her relationship to B.” The endorsement of any other response option indicates that the factor in question is not viewed, in varying degrees, as essential to person A loving person B.
Procedures. Upon entering the online research site, participants encountered an informed consent page and selected an option button indicating that they had read this page and voluntarily consented to participate in the study. Participants then provided demographic information (age, gender, ethnicity, year in school, and current relationship status) and proceeded to the FLRcontra scale. Here, after reading the definitions of each relationship characteristic, they answered the four items pertaining, respectively, to the hypothetical romantic, parental, friendship, and altruistic relationships. If they failed to do so, the program did not permit them to progress to the next item. Participants could discontinue their participation at any point. They could log on to the FLRcontra scale only once, and were not allowed to change responses once submitted or the web page was closed. Upon completion, participants were provided with a debriefing statement concerning the purpose of the research as well as information about how to obtain class research credit.
Results
Given that the range of possible group mean ratings for any relationship characteristic could extend from 1.0 (essential) to 5.0 (not at all essential), any group mean of less than 1.5 would indicate that the entire sample of participants, on average, viewed that characteristic as essential. Operationally then, with respect to IWB being regarded as the single characteristic most strongly endorsed as essential, our hypotheses were: (i) that the group mean ratings for IWB will be less than 1.5 for all 4 kinds of love; and (ii) that these means will be significantly lower than those for any of the other 13 characteristics.
Romantic love. The group mean ratings (as well as the percentage of participants who rated each of the characteristics as essential to romantic love) are displayed in Table 1. As predicted, IWB (M = 1.07, 93.1%) met the criterion for a characteristic to be regarded as essential and was rated as the highest of all the features of love. In addition to IWB, Enjoyment (M = 1.18, 87.5%), Exclusiveness (M = 1.20, 89.4%), Acceptance (M = 1.23, 78.8%), Affectionate Feeling (M = 1.24, 83.8%), Respect (M = 1.26, 81.9%), and Commitment (M = 1.43, 70.0%) met the criterion for essentialness. A series of dependent samples t-tests indicated that IWB received significantly lower average ratings (indicating greater essentialness) than any of the other six characteristics that met the criterion.
Parental love. As shown in Table 1, IWB exceeded the criterion for being regarded as essential and received the highest ratings of all 14 features (M = 1.07, 93.1%). Only one other relationship characteristic, Commitment (M= 1.50,68.8%) equaled the criterion.A dependent samples t-test revealed that IWB received significantly higher average ratings for essentialness than Commitment, t(159) = 6.25, p = .001.
Friendship/companionate love. As shown in Table 1, IWB met the criterion for essentialness (M = 1.29, 73.1%). One additional feature, Enjoyment, also met this criterion (M = 1.38, 68.1%). A dependent samples t-test demonstrated that IWB did not receive significantly higher average ratings for essentialness than Enjoyment, t(159) = 1.65, p = 0.102.
TABLE 1 Group means and percentage of participants rating factor essential (Study 1)
Romantic Parental Friendship Altruistic
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Altruistic love. As shown in Table 1, IWB was the only characteristic to exceed the requirement for essentialness (M = 1.19, 85.6%). A dependent samples t-test revealed that IWB received significantly higher average ratings for essentialness than the next highest rated characteristic, Respect, t(159) = 9.32, p = .000.
Relative ability of characteristics to transcend relationship types. In order to assess the degree to which any given characteristic transcended the four different types of love relationship, group mean ratings for each were obtained across the four types of love relationship. On this analysis, only IWB (M = 1.15) exceeded the criterion for essentialness. A dependent samples t-test revealed that IWB received significantly higher levels of endorsement than the second most endorsed characteristic, Respect (M = 1.74, t(159) = 14.53, p = .000).
Study 2
Methods
Participants. For study 2, performed at the same university as Study 1, participants were again solicited by means of a campus-wide email solicitation. Initially, 250 students elected to participate; however, 83 of these failed to complete the study materials, resulting in a final sample of 167 individuals (137 females and 30 males) ranging in age from 18 to 57 years, with a mean age of 22.2. The majority of participants identified themselves as Caucasian (92.2%), followed by African-American (2.4%), Asian-American, (2.4%), “Other” (2.4%), and Hispanic (0.6%).
Measures. The FLR version used in Study 2,“FLRdef,”was identical to that used in Study 1, except that participants were posed a different final task. Following their reading of the definition of each relationship characteristic, they were asked to make judgments about the degree to which the absence of that relationship characteristic represented a “deficiency from your model of what a good relationship of this kind should be” For example, the query for Trust in romantic relationships now read as follows: “Consider a romantic relationship in which trust was missing on the part of one or both partners. For example, suppose that, in the case of our engaged couple, Jack did not trust Jill. Which of the following statements would be closest to what you would think?” Then, a five-item interval scale was presented in which options ranged from, (a) “If this were missing, I would have a hard time believing that Jack actually loved Jill,” to (e) “If this were missing, I could still believe that Jack loved Jill, and I would consider the lack of trust as being no deficiency at all in their relationship.”
This form of question was intended to provide an alternative way of establishing whether or not the concept of love was perceived as essentialist or prototypical.As noted previously,essentialist concepts have necessary criteria for their correct employment – criteria such that, if they are absent, the phenomenon at issue cannot be a case of X at all – while prototypical concepts have no such necessary conditions. Thus, the endorsement of response option “a,” in which a participant would indicate that the absence of the characteristic would lead him or her to judge that person A did not love person B, would demonstrate that the participant viewed the relational characteristic as essential to the type of love in question. The endorsement of any other response option would indicate the degree to which the absence of the factor in question would be viewed as a deficiency from their models or prototypes of what such a relationship should be, but not necessarily as indicative of the absence of love.
Procedures. Aside from the administration of a different version of the FLR, the procedures employed in Study 2 were identical in all respects to those employed in Study 1.
Results
Given that the group mean rating for any relationship characteristic could again range from 1.0 (essential) to 5.0 (not at all essential), any mean of less than 1.5 would indicate that the entire sample of participants, on average, viewed that characteristic as essential. Thus, our hypotheses were (i) that group mean ratings for IWB will be less than 1.5 for all four kinds of love, and (ii) that these means will be significantly lower than those for any other characteristic.
The wording of the items in Study 2 permits the drawing of some further conclusions regarding the distinction between participants’ conceptions of love itself (of any of the four kinds) and their broader models or prototypes of different kinds of relationships. Characteristics which receive overall group mean ratings greater than 1.5, but equal to or less than 3.5, may be categorized as “very important” or “moderately important,” though not essential, features of participants’ prototypes of these relationships. We make no specific predictions regarding the content of these prototypes, merely permitting the data to speak for itself in this regard.
Romantic love. Mean ratings and percentage of participants who rated each of the relationship characteristics as essential are displayed in Table
2. As predicted, IWB (M = 1.19, 83.8%) met the criterion for a characteristic to be regarded as essential. Also meeting this criterion were Enjoyment (M= 1.23, 77.8%), Commitment (M = 1.25, 79.0%), Affectionate Feeling (M = 1.32, 70.7%), and Exclusiveness (M = 1.43, 71.3%), Acceptance (M = 1.41, 62.9%), and Respect (M = 1.46, 57.5%). A series of dependent samples t-tests indicated that IWB was not significantly more strongly endorsed as essential than either Enjoyment, t(166) = 1.04, p = .298 or Commitment, t(166) = 1.14, p = .258. It was, however, more strongly endorsed than the other characteristics that met both essentialness criteria; e.g., Affectionate Feeling, t(166) = 2.70, p = .008, and Exclusiveness, t(166) = 3.17, p = .002.
In addition to the characteristics judged essential to romantic love, participants rated all of the remaining 7 characteristics as “very important” or “moderately important” features in their prototypes of a good romantic relationship. In order of perceived importance, these were Sexual Desire, Emotional Intimacy, Freedom to be Ourselves, Knowledge/Understanding, Trust, Preoccupation, and Similarity.
Parental love. As shown in Table 2, IWB (M = 1.29, 71.9%) met the criterion for essentialness, along with the characteristic Commitment (M = 1.35, 73.7%). A dependent samples t-test revealed that IWB was not significantly more strongly endorsed as essential than Commitment, t(166) = .99, p = 0.325.
In addition, participants regarded nine other characteristics as “very important” or “moderately important” features in their prototypes of a good parental relationship. In order of perceived importance, these were Affectionate Feeling, Respect, Enjoyment, Acceptance, Knowledge/Understanding, Freedom to be Ourselves, Preoccupation, Trust, and Emotional Intimacy. Not surprisingly, Similarity, Exclusiveness, and Sexual Desire were regarded by participants as neither essential to parental love nor prototypical of a parental relationship.
Friendship/companionate love. No characteristic met our essentialist criterion for companionate love. Enjoyment (M = 1.63, 47.9%) and IWB (M= 1.72, 45.5%) were the two highest rated characteristics. A dependent samples t-test revealed no significant differences between the response ratings for Enjoyment and IWB, t(166) = 1.30, p = 0.194.
TABLE 2 Group means and percentage of participants rating factor essential (Study 2)
Romantic Parental Friendship Altruistic
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Participants rated ten characteristics as “very important” or “moderately important” characteristics in their prototypes of companionate relationships. In order of perceived importance, these were Enjoyment, IWB, Respect, Commitment, Acceptance, Freedom to be Ourselves, Trust, Knowledge/ Understanding, Affectionate Feeling, and Emotional Intimacy. Only Similarity, Preoccupation, Exclusiveness, and Sexual Desire were not regarded as prototypical for this kind of relationship.
Altruistic love. No characteristic met the essentialist criterion for altruistic love (see Table 2). Although IWB narrowly missed criterion (M = 1.54, 57.5%), it did receive the lowest mean rating of all characteristics. Dependent samples t-tests indicated that IWB received significantly stronger ratings for essentialness than any other characteristic; e.g., figures obtained were the next most strongly endorsed characteristics were for Respect (M = 2.29), t(166) = 8.92, p = .000 and Commitment (M = 2.49), t(166) = 9.36, p = .000.
In order of perceived importance, IWB, Respect, Commitment, Trust, Enjoyment, Acceptance, Freedom to be Ourselves, Affectionate Feeling, and Knowledge/Understanding were all regarded by participants as “very important” or “moderately important” features in their prototypes of altruistic relationships. Only Emotional Intimacy, Similarity, Exclusiveness, and Sexual Desire were not regarded as prototypical.
Relative ability of characteristics to transcend relationship types. In order to assess the degree to which any given characteristic transcended the four different types of love relationship, group mean ratings for each were again obtained across the four types of love relationship. On this analysis, only IWB (M = 1.44) exceeded the criterion for essentialness. A dependent samples t-test revealed that IWB received significantly higher levels of endorsement than the second most endorsed characteristic, Commitment (M = 1.74, t(166) = 7.53, p = .000).
Discussion
The purpose of this research was to shed light on the perennial question, “What is love?” We were interested in determining whether or not certain relationship characteristics were viewed by people as essential to human love of four different kinds, and therefore if these kinds of love could be defined in a classical, essentialist way. Of particular interest was the relational characteristic stressed as most central to love by Singer (1984), Clark and Monin (2006), and Rempel and Burris (2005), which we termed Investment in the well-being of the other for his or her own sake (IWB). In addition, we wished to determine if people had prototypes of romantic, parental, companionate, and altruistic relationships that included, but were broader than, their conceptions of what is essential to love itself of each of these kinds.
Love as an essentialist concept
Overall, the present studies lend strong support to our hypothesis that people hold an essentialist view of four important kinds of love. In both studies, when certain characteristics were described as missing from person A’s relationship to person B, both group mean and percentage data revealed that participants consistently: (i) found it contradictory to assert that person A loved person B (Study 1); or (ii) judged that person A did not love person B (Study 2). Data from both studies were most supportive for romantic and parental love, but less so for companionate and altruistic love (strongly supported in study 1; less so in study 2).
Investment in the well-being of the other
As predicted, IWB was the characteristic most strongly and consistently endorsed as essential to the four kinds of love under investigation. In Study 1, participants strongly endorsed IWB as essential to all four kinds of love. In Study 2, the same pictures were obtained for romantic and parental love, and were robust but sub-criterial for altruistic and companionate love. When averaged across the four relationship types, IWB was the only characteristic that achieved essential status, with group means of 1.15 (Study 1) and 1.44 (Study 2); in both cases, these proved significantly lower than those for the next most strongly endorsed variable. Thus, IWB emerges here as the single factor most universally perceived as essential to these four kinds of love.
Two different concepts: Love itself and relationship prototype
Data from Study 2 supports the hypothesis that people have two distinct concepts pertaining to love. The first of these is definitional and has to do with what is essential to love itself of the four kinds under investigation. The second is prototypical, and has to do with what would appropriately be embodied in a good relationship of each of these kinds. The latter includes the essential aspects of love, but includes other features as well. Thus, these two conceptions are distinct, the former set of characteristics constituting a subset of the latter.
Differences across studies
Despite the overall strength and consistency of the above findings, results obtained for altruistic and companionate love in Study 2 were less supportive than those from Study 1. The same features were most universally endorsed as essential in both studies (IWB and Enjoyment for companionate love; IWB for altruistic love); however, they failed to meet the essentialness criterion in Study 2 (although IWB was quite close for altruistic love).
One reason for this discrepancy may lie in the fact that, in Study 1, love is stipulated at the outset of each probe (“Jack loves Jill . . .”), and participants are given an immediate possible contrast to consider (“but he does not trust her”). In Study 2, the word “love” does not appear in the item stems. Instead, participants are instructed to examine a potential “deficiency” from a relationship, then asked to consider a specific situation (e.g., “Jill does not trust Jack”), and only after two additional sentences called upon to judge if such a deficiency would represent an absence of love. The juxtaposition of the key concepts is far less sharp and immediate, thus perhaps impairing participants’ ability to consider the contrast at issue.
A second reason may lie in the fact that, in western cultures, romantic and parental relationships are almost tautologically considered love relationships. Friendships and caregiver relationships do not automatically carry this connotation. Told of a romantic or a parental relationship, we tend automatically to think of it as a love relationship; told of a friendship or caregiver one, we do not. Thus, when considering the romantic and parental relationships in Study 2, even though the word “love” does not occur in the initial probes, participants may far more readily read it in, thus creating more of the kind of immediate contrast present in Study 1. If so, this might help account for: (i) the consistency of the findings for romantic and parental love between studies 1 and 2; and (ii) the inconsistencies for companionate and altruistic relationships.
No firm conclusions can be drawn on these issues. However, the possibilities just cited suggest that designing clear experimental tasks is essential to future exploration of these issues.
¿Necesario y suficiente?
Las definiciones formales son especificaciones de las condiciones necesarias y suficientes para el empleo correcto de un concepto (Ossorio, 2006). En los dos estudios aquí presentados, la preponderancia de evidencia sugiere que las personas consideran ciertos rasgos como necesarios para que ellos juzguen que una relación dada es una instancia del concepto “amor” (de las cuatro clases examinadas). Lo que es menos claro es si esta investigación ha establecido o no condiciones suficientes. ¿Podría haber otras características relaciones que hubieran probado ser esenciales para una clase de amor u otro? Esta cuestión debería explorarse en la investigación future de este tópico, y lo será. Sin embargo, esta falta de finalidad en cuanto a la suficiencia no debería oscurecer la conclusión de que, una vez que uno ha establecido la presencia de condiciones necesarias, uno ipso facto ha demostrado que un concepto no es prototípico en su carácter. Los conceptos prototípicos, según Wittgenstein y Rosch, son por definición conceptos sin condiciones necesarias.
Limitaciones y direcciones futuras
Importantes limitaciones de la investigación presente, que deben ser enfocadas en nuestro trabajo future, son las siguientes. Primero, como se adviertó, si bien los resultados dan apoyo a que un cierto conjunto de características sean consideradas necesarias para cuatro clases de amor humano, no establecen suficiencia. Por tanto, se necesita considerer otras características. Segundo, la investigación explora cuatro clases importantes de amor humanio, pero no indaga en otras formas (e.g., el amor de un niño por sus progenitores). Tercero, nuestras muestras estuvieron restringidas a estudiantes universitarios de una gran Universidad pública de los Estados Unidos. Como tales, las muestras consistían fuertemente en personas jóvenes, relativamente limitadas en experiencia de vida, primordialmente caucásicas, socializadas en una única cultura occidental, y relativamente exitosas y con ventajas en la vida. En la investigación future, planeamos explorer la generalidad de nuestros resultados en una muestra de individuos de más edad, con más experiencia de relación, y más diversos demográfica y culturalmente. Una necesidad especialmente importante es replicar la investigación presente en diferentes contextos culturales (e.g. los asiáticos o del Medio Oriente) donde las concepciones del amor como algo que tiene características esenciales, como bien pueden ser éstas, pueden ser muy diferentes.
Conclusión
Los resultados aquí presentados sugieren un rettrato del amor humano que esperamos estimule a mayor investigación. Elo concepto de “amor” (de cuatro clases diferentes) puede ser, contrariamente a lo que se ha supuesto ampliamente, un concepto definable y esencialista, y la Inversión en el bienestar del otro por sí mismo es un importante candidato para ser considerado su característica más trascendente. Es la característica que aquí se encuentra como la más universalmente considerada como esencial en el amor romantic, progenitorial, de compañeros y altruista, y la única característica que trascendió estos cuatro tipos de relaciones. Como tal, la IBO puede ser la única característica que provee un elemento común entre el amor de un miembro de pareja romántica por la otra persona, de un padre o madre por su hijo o hija, de un amigo íntimo por otro, y de esas personas, tales como los cuidadores devotos, que en muchos contextos exhiben un cuidado sin egoísmo por otros en sus mundos. Finalmente, más allá del possible empleo del amor como concepto esencialista, las personas pueden poseer modelos o prototipos relacionados, pero diferentes y más amplios, con las diferentes clases de bien que las relaciones humanas corporizarían idealmente.
REFERENCES
Aron, A., Fisher, H. E., & Strong, G. (2006). Romantic love. In A. Vangelisti & D.Perlman
(Eds.) The Cambridge Handbook of Personal Relationships (pp. 595–614). New York:
Cambridge University Press.
Bergner, R. (2000). Love and barriers to love: An analysis for psychotherapists and others. American Journal of Psychotherapy, 54, 1–17.
Bergner, R., & Davis, K. (September, 2007). Lay conceptions of romantic love: Essential
components or only prototypical characteristics? Paper presented at the 29th Annual
Conference of the Society for Descriptive Psychology, Golden, CO
Berscheid, E. (1985). Interpersonal attraction. In G. Lindzey & E. Aronson (Eds.), The
handbook of social psychology (3rd. ed., Vol. 2). New York: Random House. Berscheid, E. (2006). Searching for the meaning of “love.” In R. J. Sternberg & K. Weis (Eds.), The new psychology of love (pp. 171–183). New Haven, CT: Yale University Press. Bretscher, F., & Bergner, R. (1991). Relational qualities as factors in mate selection decisions. In M. Roberts & R. Bergner (Eds.), Advances in Descriptive Psychology (Vol. 6, pp. 107–123). Ann Arbor, MI: Descriptive Psychology Press.
Clark, M., & Mills, J. (1979). Interpersonal attraction in exchange and communal relationships. Journal of Personality and Social Psychology, 37(1), 12–24.
Clark, M. S. & Monin, J. K. (2006). Giving and receiving communal responsiveness as love. In R. J. Sternberg & K. Weis (Eds.), The new psychology of love (pp. 200–224). New Haven, CT: Yale University Press.
Davis, K. E., & Todd, M. (1982). Friendship and love relationships. In K. E. Davis & T. O. Mitchell (Eds.) Advances in Descriptive Psychology (Vol. 2, pp. 79–122).Greenwich, CT: JAI Press.
Fehr, B. (1988). Prototype analysis of the concepts of love and commitment. Journal of Personality and Social Psychology, 55, 557–579.
Fehr, B. (1993). How do I love thee? Let me consult my prototype. In S. Duck (Ed.), Understanding personal relationships: Vol. 1: Individuals in relationships (pp. 87–120). Newbury Park, CA: Sage.
Fehr, B. (2006). The prototype approach to studying love. In R. J. Sternberg &K. Weis (Eds.), The new psychology of love (pp. 225–248). New Haven: Yale University Press.
Fehr, R., & Sprecher, S. (July 23, 2004). Compassionate love: Conceptual, relational, and behavioral issues. Paper presented at the Conference for the International Association of Relationship Research, Madison, WI.
Fehr, B., Sprecher, S., & Underwood, L. (Eds.). (2009). The science of compassionate love: Theory, research, and applications. Malden, MA: Blackwell Press.
Hassebrauck, M. (1997). Cognitions of relationship quality: A prototype analysis of their structure and consequences. Personal Relationships, 4, 163–185.
Hendrick, C., & Hendrick, S. (2006). Styles of romantic love. In R. Sternberg & K. Weis (Eds.), The new psychology of love (pp. 149–170). New Haven: Yale University Press.
Kelley, H. H. (1983). Love and commitment. In H. Kelley et al. (Eds.), Close relationships (p. 265–314). New York: W. H. Freeman.
Lee, J. A. (1973). The colors of love. Englewood Cliffs, NJ: Prentice-Hall, Inc. Mervis, C., & Rosch, E.H. (1981). “Categorization of Natural Objects,” Annual Review of Psychology, 32, 89–113.
Ossorio, P. G. (2006). The behavior of persons. Ann Arbor, MI: Descriptive Psychology Press.
Rempel, J. K., & Burris, C. T. (2005). Let me count the ways: An integrative theory of love and hate. Personal Relationships, 12, 297–313.
Rosch, E. H. (1973). Natural categories. Cognitive Psychology, 4, 328–350.
Singer, I. (1984). The nature of love,Vol.1:From Plato to Luther (2nd ed.). Chicago, IL: University of Chicago Press. Sternberg, R. J. (1988). The triangle of love: Intimacy, passion, commitment. New York: Basic Books.
Stemberg, R. (2006). A duplex theory of love. In R. Sternberg & K. Weis (Eds.), The new psychology of love. New Haven, CT: Yale University Press. Wittgenstein, L. (1953). Philosophical investigations. New York: MacMillan.