FIGURE 1.5. The physical makeup types distinguished by Kretschmer. Note. From Körperbau und Charakter: Untersuchungen zum Konstitutionsproblem und zur Lehre von den Temperamenten [Physique and Character: Research concerning Problems of Constitution and Knowledge on Temperaments] (17th-18th ed., pp. 18, 23, 27), by E. Kretschmer, 1944, Berlin, Germany: Springer. Copyright 1944 by Springer Verlag. Reprinted with permission.
vidual cases the following findings emerged: among schizophrenics 50.3% are leptosomatics, 64% of manic-depressive individuals are pyknics, and epileptics are mostly found among athletics (28.9%) and dysplastics (29.5%), as presented in Figure 1.6. The dysplastic type comprises all kinds of departures from the normal physique (Kretschmer, 1944).
Body Types and Temperaments. Taking these findings as a point of departure, Kretschmer put forward a hypothesis stating that in healthy, normal people there is also a relationship between the physical makeup and temperaments, including the inclination to psychiatric disorders. Normal people with a given type of physique have temperaments that are somewhat like the psychiatric disorders typical for their physical makeup. Kretschmer distinguished the following three types of temperament: schizothymic, cyclothymic, and ixothymic.
The schizothymic temperament has a leptosomatic (asthenic) body type. In case of illness individuals representing this type are prone to schizophrenia, are autistic (withdrawn), their emotions oscillate between irritability and coldness,
FIGURE 1.6. The distribution ofphysical makeup types among individuals with different psychiatric disorders. Note. From Korperbau und Charakter: Untersuchungen zum Konstitutionsproblem und zur Lehre von den Temperamenten [Physique and Character: Research concerning Problems of Constitution and Knowledge on Temperaments] (17th-18th ed., p. 35), by E. Kretschmer, 1944, Berlin, Germany: Springer. Copyright 1944 by Springer Verlag. Adapted with permission.
FIGURE 1.6. The distribution ofphysical makeup types among individuals with different psychiatric disorders. Note. From Korperbau und Charakter: Untersuchungen zum Konstitutionsproblem und zur Lehre von den Temperamenten [Physique and Character: Research concerning Problems of Constitution and Knowledge on Temperaments] (17th-18th ed., p. 35), by E. Kretschmer, 1944, Berlin, Germany: Springer. Copyright 1944 by Springer Verlag. Adapted with permission.
they are rigid in habits and attitudes, they withdraw from reality, and they have difficulty adapting.
The cyclothymic temperament, named from the cyclical psychosis known as manic-depression, is mostly found among pyknics. Pyknics are prone to manic-depressive psychosis; their emotions vary from joy to sadness. They have easy contact with their environment and are realistic in their views.
The ixothymic (Gr. ixos—sticky) temperament has an athletic body type. When psychiatric disorders occur, individuals with this temperament have a tendency toward epilepsy, are quiet, have low sensitivity, are modest in gestures and mimicry, have low plasticity, and have difficulty adapting to their environment.
On the basis of his research and experiences Kretschmer described the characteristic qualities of temperament. The temperaments are present in the following four psychological domains: (1) psychasthesia (Psychaestesie) expressed in over- or undersensitivity to psychic stimuli, (2) mood, which may vary from pleasure (joy) to unpleasantness (sadness), (3) psychic tempo expressed in the acceleration or inhibition of psychic processes in general as well as in their specific rhythmicity, and (4) psychomotility (Psychomotolitaet), the general tempo of motions as well as specific modes of locomotion (Kretschmer, 1944, p. 298). It is clear that these temperament characteristics spotlight the formal aspect of human behavior.
In Europe Kretschmer's constitutional typology gained exceptional popularity, especially in the 1930s. Many psychologists and clinicians were under the influence of Kretschmer. During World War II a German psychiatrist, K. Conrad ( 1941), referring to Kretschmer's theory, published a constitutional typology based on a genetic approach. His view, according to which temperament characteristics as well as more complex psychic phenomena (e.g., attitudes, beliefs) are, like the physical makeup, genetically determined has been rejected by most psychologists. Such views were conducive to the development of racist attitudes.
To treat the physical makeup as an indicator ofhuman psyche, or at least temperament, was an attractive and easy route to psychological and psychiatric diagnostics. After World War II Kretschmer's typology lost its popularity. In recent decades few papers referring to his theory have been published.
The Constitutional Theory of Temperament Developed by Sheldon
The popularity that Kretschmer gained in Europe did not spread to American psychologists and physicians. Twenty years after Kretschmer's famous monograph W. H. Sheldon and coworkers (Sheldon & Stevens, 1942; Sheldon, Stevens, & Tucker, 1940) published a constitutional theory of temperament that earned popularity in the United States. Sheldon's theory of temperament has been described in almost all U.S. handbooks of personality; thus I limit the presentation ofhis concept to the most important issues.
Sheldon's constitutional theory, which developed under the influence of Kretschmer and to test his theory, was based on the assumption that body and temperament constitute two integral aspects of the same object: the human being. The body, viewed structurally, determines temperament which in turn is a function of the body. The interdependence of structure and function is masked by the complexity of the organism and the psyche. In order to grasp the relationship between the physical and psychological characteristics, the basic variables (physical and psychological) must be separated.
The Morphological Taxonomy. The starting point of Sheldon's theory was the description of basic morphological components by means of a standardized photographic technique and anthropometric studies (Sheldon et al., 1940). He distinguished 17 different measures (parameters expressed as ratios to stature) based on a 7-point rating scale. This procedure led him to the distinction of three primary components (dimensions) of the physical makeup. Using embryological terminology, he called the morphological components endomorphy, mesomorphy, and ec-tomorphy, referring to the three tissue layers.
The composition of the three components results in the somatotype which provides the basis for a morphological taxonomy. The strength of these components is expressed by means of a 7-point scale. Thus the endomorphic type, in which the digestive viscera are massive and highly developed, is symbolized by the numbers 7-1 -1. The mesomorphic type, with strongly developed bones, muscles, and connective tissue has the numbers 1-7-1. The numeral configuration 1 -1-7 represents the ectomorphic type, in whom the nervous system, sensory tissue, and skin are dominant. The configuration 4-4-4 is typical for an average (mixed) so-matotype.
In contrast to Kretschmer, Sheldon and his coworkers conducted studies on normal subjects. Most of his more than 4,000 subjects were males, normally nourished and in a broad age range (Sheldon et al., 1940).
The Temperament Components. Whereas the morphological components represent the static aspect of constitutional psychology, temperament refers to the dynamic aspect. By temperament Sheldon and Stevens (1942) meant "the level of personality just above physiological function and below acquired attitudes and beliefs" (p. 4). The starting point for the study of temperament, which lasted over 5 years, was a list of 650 alleged temperament trait terms. No information was given about the sources from which they were derived apart from indicating that most of them refer to extraversion or introversion. On the basis of prolonged observations and dozens of interviews conducted with each subject the authors distinguished 60 trait terms which represent three clusters, also called components or factors.
Borrowing the suffix- tonia from Eppinger and Hess (1910), who conducted studies on vagotonia, the authors labeled the three primary temperament components viscerotonia, somatotonia, and cerebrotonia. The names illustrate the functional dominance of given organs of the body. As with the morphological measures, the temperament components, as well as the separate traits constituting the components, were estimated by means of a 7-point scale. Viscerotonia comprises temperamental traits closely related to the functional predominance of the digestive viscera (endomorphic physique). Somatotonia is characterized by traits associated with the functional predominance of the somatic structures (mesomorphic physique) and cerebrotonia refers to the functional prepotency of the higher centers of the NS, especially to attentional consciousness (ectomorphic physique). Thus the numeral configuration 7-1-1 represents the viscerotonic temperament, the configuration 1-7-1, the somatotonic temperament, and the numbers 1-1-7 illustrate the cerebrotonic temperament. In Table l .2 each of the three primary temperament components has been described by means of 10 traits which constitute
TABLE 1.2. The Short Form of the Temperament Scale Developed by Sheldon and Stevens
Relaxation in posture and movement
Love of physical comfort
Love of polite ceremony
Evenness of emotional flow
The untempered characteristic
Smooth, easy communication of feeling, extraversion of viscerotonia
Assertiveness of posture and movement
Love of physical adventure
The energetic characteristic
Need and enjoyment of exercise
Love of risk and chance
Bold directness of manner
Physical courage for combat Competitive aggressiveness The unrestrained voice
Overmaturity of appearance
Restraint in posture and movement, tightness
Overly fast reactions
Mental overintensity, hyperattentionality, apprehensiveness
Secretiveness of feeling, emotional restraint
Self-conscious motility of the eyes and face
Inhibited social address
Vocal restraint, and general restraint of noise
Note. From The Varieties of Temperament: A Psychology of Constitutional Differences (p. 26). by W. H. Sheldon and S. S. Stevens, 1942, New York: Harper & Brothers. Copyright I942 by Harper & Brothers. Adapted with permission.
the short form of the Temperament Scale developed by Sheldon and Stevens (1942, p. 26).
Studies conducted by Sheldon and Stevens (1942) on a sample of 200 men aged from 17 to 31 showed that there is a highly significant and strong correlation between the physical makeup and temperament characteristics: for viscerotonia and endomorphy, .79; for somatotonia and mesomorphy, 82; and for cerebrotonia and ectomorphy, .83.
Sheldon and his coworkers also conducted studies in which the somatotypes and temperament components were related to psychiatric disorders. Wittman, Sheldon, and Katz (1948) showed that, in general, somatotypes, as well as temperament components, correlate with psychiatric disorders in the hypothesized direction.
Critical Remarks Regarding the Constitutional Approach
There is at least one common denominator for all constitutional approaches, explicitly present in both Kretschmer's and Sheldon's typologies. This is the assumption that human physique, being inherited, does not change during ontogenesis. The dualistic doctrine, typical for the constitutional approach, postulates that temperament (personality) is a parallel of the physical makeup, which means that temperament also does not change during ontogenesis. Ignorance of the environment (physical and social) as a factor codetermining temperament or personality has consequently led to the rejection of the constitutional typologies. This rejection was particularly expressed in the United States but was also present in Europe. Psychologists, physicians, and educators became more interested in changes occurring in human behavior; they concentrated on factors other than inheritance contributing to the development of human psyche. The static, purely descriptive approach and a fatalistic view of human temperament and personality as proposed by constitutional typologies were probably the main reasons for the loss of interest in the theories of Kretschmer and Sheldon, the most prominent investigators in this field of study.
There is also a more detailed critique regarding the constitutional theories, specifically referring to the concepts of Kretschmer and Sheldon. Most of the critical remarks, some of which are mentioned in the following sections, were made several decades ago.
Kretschmer's Typology. The typology of Kretschmer was developed with pathology as the starting point. It was based on the assumption, not commonly accepted, that psychiatric disorders differ from normal mental states quantitatively rather then qualitatively. Kretschmer's methods of data collection and the findings of him and his coworkers have been subject to criticism from several angles. For example, it has been shown that college students essentially differing in body types (pyknics versus leptosomes) do not differ in a series of psychological tests, in-
eluding temperament characteristics (Klineberg, Asch, & Block, 1934) which speaks against the temperamental specificity of the body types under study.
Brengelmann (1952), in a study conducted on 100 normal subjects showed that the tests used by Kretschmer (1944) and his students (Enke, 1930) as experimental markers for the schizothymic and cyclothymic temperament did not allow for differentiation of individuals according to the cyclothymia-schyzothymia hypothesis as proposed by Kretschmer. The tendency to mental diseases, as well as to given body types, is to some extent age specific-a finding not taken into account by Kretschmer. In general, schizophrenics and leptosomes are about 10 years younger than manic-depressive individuals and pyknics (e.g., Burchard, 1936). Kretschmer ignored the essential changes in the physical makeup that occur during and after puberty (Homburger, 1926).
The empirical data presented by Kretschmer, limited to descriptive statistics, are handicapped by lack of statistical sophistication (H. J. Eysenck, 1970) as well as by ignorance of other methodological requirements, especially the matching of samples for age and socioeconomic status (SES).
Sheldon's Typology. The fact that Sheldon distinguished three morphological components, the combination of which resulted in somatotypes, differentiates his typology from Kretschmer's, which is based on traditional typological thinking, where categorization referring to extreme characteristics leads to the distinction of separate types. Sheldon's typology refers to dimensions and different combinations of dimensions, thus allowing for quantitative gradations among types. His constitutional theory has been criticized for several reasons, some of which are given in the following discussion.
Factor analytic studies have shown that the three-dimensional system as proposed by Sheldon in the morphological and temperament domains can easily be reduced to two dimensions. Extreme ectomorphy may be regarded as the absence of either the endomorphic or the mesomorphic component (G. Ekman, 1951); ec-tomorphy and endomorphy are opposite manifestations of one factor (Howells, 1952). Humphreys (1957), criticizing the statistical and empirical bases of Sheldon's theory, pointed out that when proper statistics were used to analyze his data, measures of any two of the dimensions enabled the prediction of a third dimension.
The correlations between somatotypes and objective tests used as measures of the three temperament components are much lower than postulated by Sheldon, often not significant at all (Janoff, Beck, & Child, 1950; H. C. Smith, 1949). The estimations of the somatotypes and temperament components were often conducted by the same person, thus leading to a strong halo effect which in turn biased the objective relationship between the physique and temperament (Adcock, 1948; Tyler, 1965).
Definite opinions and stereotypes exist regarding the relationship between the physical makeup and behavioral characteristics. They are mainly based on folk wisdom and in line with Sheldon's findings. These kinds of stereotypes regarding the physique-temperament relationship influenced the collection of data (Gacsaly & Borges, 1979; Wells & Siegel, 1961). The fact that somatotypes correspond with temperament components does not necessary mean that the latter are determined by the body type. The stereotypes and opinions regarding the physique-temperament relationship may influence the educational treatment of a child in such a way as to favor the kind of activity that corresponds with the individual's physical makeup. Also, changes in morphology may occur due to special child-rearing events (Lindzey, 1967).
Sheldon, presenting the idea of interdependence of the structure (body) and the function (temperament), did not make use of the embryological hypothesis (H. J. Eysenck, 1970) that was his starting point for the three morphological dimensions. He limited his theory to a purely descriptive level.
There are hundreds of publications, starting from antiquity, in which authors have tried to define temperament. A few years ago a roundtable discussion by child-oriented temperament researchers aimed at answering the question "What Is Temperament?" was published in the journal Child Development. In four presentations four different views regarding the definition of temperament were delineated (Goldsmith etal., 1987). As demonstrated in the following sections, in spite of many differences in the understanding of this phenomenon, there are several characteristics on which most temperament scholars agree.
One of the main aspects on which temperament scholars differ is the domain of behavior characteristics and psychic functioning to which the concept temperament refers. To consider the extremes, according to some researchers temperament characteristics should be limited to emotions only, whereas to others behavioral expressions of temperament are present in all kinds of human functioning.
It is difficult, if not impossible, to determine the extent to which the contributions of former researchers have influenced our own ideas and inquiries. From my own reading on temperament over my 40 years of studies in this area, it is apparent that the conceptualization of temperament presented in the 1920s and 1930s by Gordon Allport is one of the most influential contributions in the attempt to determine the meaning of the concept of temperament, at least in Western countries. His definition of temperament constitutes the basis for all or almost all emotion-oriented temperament researchers.
Gordon Allport's Definition of Temperament
G. W. Allport (1897-1967) is regarded as the founder of trait-oriented personality psychology. This statement is not neutral for our considerations on temperament, since with but a few exceptions temperament is viewed as a structure consisting of traits.
Allport's starting point in his task of describing temperament was a deep penetration of the temperament literature, including the contributions of German and French researchers, and his creative understanding of personality. I contend that G. W. Allport (1937) regarded temperament as one of the components of personality. According to him:
Temperament refers to the characteristic phenomena of an individual's emotional nature, including his susceptibility to emotional stimulation, his customary strength and speed of response, the quality of his prevailing mood, and all peculiarities of fluctuation and intensity in mood; these phenomena being regarded as dependent upon constitutional make-up, and therefore largely hereditary in origin. (p. 54)
In an article written by G. W. Allport and Vernon (1930) we find a very similar definition of temperament, quoted, however, after Floyd H. Allport (1924), the brother of Gordon Allport.
G. W. Allport's understanding of temperament as referring to individual differences in emotions, especially in the formal characteristics of these phenomena, has its historical forerunners. As shown previously, Wundt, as well as many other authors, limited temperament to emotional characteristics. The ancient Greek ty-pologists also considered individual differences in emotions as the core of temperament when they claimed that "humors" were the physiological basis of temperament.
Further statements from Allport's classic monograph Personality: A Psychological Interpretation add to our knowledge of his understanding of temperament. For example, for Allport temperament, like intelligence and physique belongs to the class of raw material from which personality is fashioned. Temperament being dependent on the biochemical constitution should be regarded as psychobiological.
Temperament belongs to the category of dispositions that are almost unchanged from infancy throughout life. "The more anchored a disposition is in native constitutional soil the more likely it is to be spoken of as temperament" (G. W. Allport, 1937, p. 53). The behavioral expressions of temperament are present since early infancy.
According to Allport (1937) two aspects of temperament may be characterized by means of quantitative continua (dimensions): broad emotions-narrow emotions, which refers to the range of objects and situations to which an individual reacts emotionally, and strong emotions-weak emotions. The latter dimension, which pertains to the intensity of feeling evoked by objects and situations can be measured objectively by means of blood pressure, pulse rate, and psychogalvanic skin response (pp. 407-408).
Allport's writings were of utmost importance in further considerations and studies regarding temperament. As seen in the next sections of this chapter, many of Allport's ideas still hold today and modifications made by others usually go in the direction of less radical ascertainments and expanding the domain of temperament.
Emotion-Centered Definitions of Temperament in Contemporary Research
The emotion-oriented definitions of temperament, of which Allport's is regarded as the classic one, have also gained popularity in contemporary psychology. In his earlier publications H. J. Eysenck (1970), referring Allport, among others, defined temperament as a "more or less stable enduring system of affective behaviour ('emotion')'' (p. 2). However, Eysenck, whose temperament theory is described in Chapter 2, has seldom used the term "temperament."
The most popular emotion-oriented definitions of temperament may be found in Mehrabian's and Goldsmith and Campos's conceptualizations regarding this phenomenon. Since about the mid-1970s Albert Mehrabian (1978b, 1991; Mehra-bian & Falender, 1978; Mehrabian & O'Reilly, 1980) has presented an emotion-based theory of temperament (see Chapter 3) which refers to the following understanding of this concept: "Temperament is viewed as a characteristic emotional state" (Mehrabian & Falender, 1978, p. 1120). Probably to avoid misunderstanding in defining temperament in terms of states, Mehrabian (1991) supplemented this definition by stating that " 'Temperament' is defined here as a 'characteristic emotion state' or as an 'emotion trait' " (p. 77).
By using the term state the author wanted to underscore that temperamental characteristics are present only in states. The rather confusing term characteristic has been used by Mehrabian to stress that only those emotional states that are typical and unique for the given individual, that is, the emotional states in which individuals usually differ, are the subject of study in temperament research. The definition of temperament as presented by Mehrabian and his coworkers is rarely cited by other researchers, even when these others are centered on emotions as the core concept of temperament.
Whereas Mehrabian's concept of temperament was developed by studying adults, Goldsmith and Campos (1982, 1986, 1990; see Chapter 3) were centered on infants. The analysis of infant behavior led them to the conclusion that the motoric, facial, and vocal behavior of infants all are expressions of the affective systems (Goldsmith & Campos, 1982); temperament in infants is expressed mainly in emotional behavior. As the authors write:
We conceive of temperament as individual differences in emotionality. . . . This includes individual differences in the primary emotions — fear, anger, sadness, pleasure, interest, and so forth—and more generalized arousal, as expressed in the temporal and intensive parameters of behavioral response. We delineate temperament from individual differences in cognition, perception, and emotional states (as opposed to traits). (Goldsmith & Campos, 1986, p. 231)
Taking a more general view Goldsmith and Campos (1990) proposed to define infant temperament "as individual differences in tendencies to express the primary emotions" (p. 1945). Temperament dimensions form the emotional substrate of some later personality characteristics (Goldsmith et al., 1987). The authors make clear that in the understanding of temperament they follow a tradition dating to Hippocrates and Galen and resurrected by Allport.
To close the presentation of the emotion-centered conceptualizations of temperament one may conclude that the most conspicuous common denominator of the definitions of temperament as proposed by Allport, Eysenck (in his earlier writings), Mehrabian, and Goldsmith and Campos is the statement that temperament should be regarded as a construct referring exclusively to emotional behavior. Explicitly or implicitly these definitions say that more or less stable individual differences in emotions are the subject of temperament research. For these authors temperament is a synonym of the expression, "individual differences in emotional behavior."
One of the most popular definitions of temperament was formulated in the 1960s by Alexander Thomas and Stella Chess (1977; A. Thomas, Chess, & Birch, 1968), the founders of contemporary temperament research in children. Considering temperament as a behavioral style the authors described temperament thus:
Temperament may best be viewed as a general term referring to the how of behavior. It differs from ability, which is concerned with the what and how well of behaving, and from motivation, which accounts for why a person does what he is doing. Temperament, by contrast, concerns the way in which an individual behaves. (A. Thomas & Chess, 1977, p. 9)
Thomas and Chess considered temperament as a phenomenological term, with no implications as to etiology and immutability. Temperament is defined on a purely descriptive level without any inference to the determinants of individual differences in temperament. However, on different occasions the authors assumed that temperament has a genetic background (Chess & Thomas, 1989; A. Thomas &
Chess, 1977). There is a high consistency in the authors'understanding of temperament which is defined in recent publications of the two eminent temperament scholars as it was 25 years ago (Chess &Thomas, 1986, 1989, 1991).
Giving some more detailed explanation to their definition of temperament Thomas and Chess (Goldsmith et al., 1987, pp. 508-509) emphasized that temperament is an independent psychological attribute; must at all times be differentiated from motivations, abilities, and personality; is always expressed as a response to an external stimulus, opportunity, expectation, ordemand; and is an attribute ofthe child that modifies the influence ofthe environment.
Every behavior, irrespective of its content, may be characterized by means of the stylistic component; it might be concluded that temperament understood as a behavior style reveals itself in all kinds of behavior and it refers rather to the formal characteristics and not to the content of behavior. J. V. Lerner and Lerner (1983) paid attention to the fact that it is difficult to define the concept ofstyle. Bates (1987) argued that the concept ofstyle does not apply to all temperament dimensions, even those distinguished by Thomas and Chess (see Chapter 2).
Many temperament researchers, especially in the United States, take the stylistic definition of temperament as formulated by Thomas and Chess as a starting point for their studies, centered mostly on infants and older children (see, e.g., W. B. Carey, 1983; P. S. Klein, 1984; Maziade, 1988; McDevitt & Carey, 1978; Person-Blennow, McNeil, & Blennow, 1988; Rutter, 1982; Sameroff, Seifer, & Elias, 1982; Windle, Iwawaki, & Lerner, 1988).
Most researchers in the area of temperament take the fact that individual differences in temperament are determined or codetermined by some biological mechanisms as the basic or one of the basic criteria for defining temperament. This is not to say that the biology-oriented understandings of temperament compose a class of concordant or convergent definitions. They differ mostly in the number of criteria to be represented in the definition, such as heritability, stability, population, and age to which temperament refers. As illustrated in Chapters 2 and 3, they also differ in the understanding of biological mechanisms underlying temperament. Because most of the definitions refer to more than one criterion in defining temperament, it is not easy to present a clear-cut distinction between them. Even some ofthe definitions of temperament presented until now, especially the one by Allport, emphasize the importance of the constitutional factor as one of the criteria to be applied in order to distinguish temperament from other phenomena. Several groups of definitions representing the biological approach to temperament are given in the following sections.
Temperament as an Expression of the Type of Nervous System
Since Pavlov there has existed in the former Soviet Union, especially in Russia, a tradition according to which temperament is regarded as the psychological expression of the type of higher nervous activity (e.g., Golubeva & Rozhdestven-skaya, 1976; Leites, 1956, 1972) or as the dynamic characteristics of behavior as expressed in individual differences in speed and intensity of reaction (behavior) and determined by the type of nervous system (Ilin, 1978; Merlin, 1973; Rusalov, 1979; Teplov, 1985).
The fact that temperament is determined by the type of nervous system regarded as a given composition of the basic properties of the central nervous system was commonly accepted among Russian psychologists at least until the 1980s. Also established was the view that temperament refers to formal traits, such as the energetic and temporal characteristics of behavior.
The understanding of temperament as presented by Russian psychologists in the period from the 1950s to the 1980s influenced research on temperament in Eastern Europe to some extent (see, e.g., Halmiova & Sebova, 1986; Strelau, 1969; Zapan, 1974). However, such a comprehension of temperament did not gain much popularity in Western countries (see Mangan, 1982; Strelau, 1983).
Temperament as an Inherited Component of Personality
The inheritance of temperament traits had already been pointed out by G. W. Allport (1937). One of the most influential theoreticians on temperament in the 1950s, Solomon Diamond (1957), considered temperament as the inherited (though influenced by environment) component of personality. He pointed to the similarities between temperament characteristics in humans and other mammals, as well as to the evolutionarily determined adaptive function of temperament. When referring to the understanding of temperament by Allport, Diamond (1957) preferred to define temperament "in terms of the ease of arousal of unlearned [emphasis added] patterns of adaptive behavior, and to define its dimensions in terms of whole classes of adaptive response, rather than in terms of emotional expression'' (p. 95). Both Allport and Diamond, as well as Thomas and Chess, influenced to some extent the theory of temperament as developed by Buss and Plomin (1975, 1984). According to their theory (see Chapter 3) temperament must be defined by taking into account two basic criteria that distinguish this phenomenon from other personality traits: inheritance and presence in early childhood. Thus the authors define "temperaments as inherited personality traits present in early childhood" (1984, p, 84).
This definition of temperament emphasizes most consistently the fact that temperament is inherited. It also excludes from the domain of temperament all personality traits that originate solely in environmental events. The definition does not specify, however, the domain of behavior in which temperament is expressed, thus giving way to the search for temperament characteristics among a broad range of behaviors, assuming they fulfil the definitional criteria—inheritance and presence since early infancy. In most definitions of temperament the exposition of inheritance is absent, even when referring to biological bases of temperament.
A contemporary Russian psychologist, V. M. Rusalov (1985, 1989c), also makes inheritance one ofhis basic criteria when defining temperament. According to him, the basis of temperament consists in the general constitution of the human organism which should be considered a composite of physical and physiological properties of the individual. These composites are rooted in the inherited apparatus (Rusalov, 1985, p. 25).
In studies conducted during the past two decades the inheritance criterion has lost its strength. It has been shown in several studies that the heritability index does not differentiate between the traditionally recognized temperament traits and other personality characteristics (Holden, 1987; Loehlin & Nichols, 1976).
Temperament as Referring to the Formal Characteristics of Behavior
The fact that temperament refers not to the content of behavior but to formal characteristics ofhuman functioning has been emphasized by many authors. When temperament is limited to emotional behavior, such formal characteristics as intensity, duration, speed, fluctuation of emotions, and so on are the basic criteria for describing individual differences in temperament. The conceptualizations of temperament offered by Wundt, Allport, Goldsmith, and Campos may serve as examples here. The stylistic definition of temperament as presented by Thomas and Chess also refers mainly to the formal characteristics of behavior. The speed and intensity of behavior have been used as definitional criteria of temperament by Russian investigators as well.
In some biologically oriented definitions of temperament the formal characteristics of behavior are particularly exposed. To some extent the statement that temperament refers to formal features of behavior implies the biological roots of temperament, the fact that temperament refers not only to emotions but to all kinds of behavior, its specificity as compared with other personality traits. In definitions given by Strelau (1983, 1989b), Eliasz (1990), Rusalov (1985, 1989c), and partly also by Rothbart (1989b; Rothbart & Derryberry, 1981), the formal characteristics of behavior serve as a definitional criterion of temperament.
As mentioned previously, in most definitions of temperament more than one criterion is used for specifying the nature of temperament. For some authors the relative stability of temperamental traits has been regarded as an essential feature. The requirement of stability is explicitly exposed in definitions of temperament given by Kagan (1982a, 1989b), Hagekull (1989), Eliasz (1990), and Rusalov (1985). As Kagan (1982) wrote:
If there is a correlated cluster of qualities that seems intuitively to belong together
(for whatever reason), and the composite of the cluster is stable [emphasis added], that cluster is a reasonable candidate for temperamental status. (p. 24)
Concluding Remarks Regarding the Understanding of Temperament
In spite of the differences in defining temperament, there seems to be some agreement regarding the definitional criteria by means of which this phenomenon is characterized. At least there is more consistency in the understanding of this concept as compared with such notions as "personality" (see, e.g., Hall & Lindzey, 1978; Pervin, 1990, 1996; Wiggins, Renner, Clore, & Rose, 1971) or "intelligence" (e.g., Sternberg, 1982; Vernon, 1979).
The Most Common Features Determining the Understanding of Temperament
Most temperament researchers, whether emotion-oriented or expanding this phenomenon to other kinds ofbehavior as well, agree, explicitly or implicitly, with the view that temperament is a phenomenon that may be characterized by the following features:
(a) Temperament refers to behavior characteristics in which individuals differ. These differences are described by such concepts as disposition (e.g., Betz & Thomas, 1979; Diamond, 1957), trait (e.g., Allport, 1937; Buss & Plomin, 1984; Strelau, 1983), quality (e.g.,Kagan&Reznick, 1986;Roback, 1931), attribute (e.g., Stevenson & Graham, 1982; Windle, 1989b), factor (e.g., Cattell, 1934-1935; J. P. Guilford, 1975), dimension (e.g., H. J. Eysenck, 1990b, 1991a; J. P. Guilford & Zimmerman, 1950), type (e.g., Kagan, 1989b; Zapan, 1974), and category (e.g., Chess & Thomas, 1989; Kagan, 1989b); these concepts are used interchangeably by many researchers.
(b) Temperament is relatively stable as compared with other phenomena and is also characterized by considerable cross-situational consistency. The terms "stability" and "consistency," however, should not be understood literally. In the context of temperament research they underline only that temperament, as compared with other behavior characteristics, belongs among the most stable and cross-sit-uationally consistent phenomena.
(c) Temperament has a biological basis; however, as shown in Chapters 2 and 3, a variety of views regarding the kind and quantity of anatomic structures and physiological mechanisms underlying temperament can be found.
(d) Temperament refers mainly to formal characteristics of behavior or reactions, such as intensity, energy, strength, speed, tempo, fluctuation, and mobility. In some conceptualizations regarding the nature of temperament these character istics are limited to emotions only (e.g., G. W. Allport, Goldsmith, Campos); in others they spread to all kinds of behavior (e.g., Rothbart, Strelau).
The Status of Temperamental Traits as Exemplified by a Methaphor
For a better understanding of the nature of temperament I refer to an analogy with the characteristics of the automobile which I have applied in my lectures on temperament since the end of the 1960s. In order to explain the nature of temperament some authors have used the functioning of a steam engine or a car as a metaphor of temperament features in man (e.g., Buss & Plomin, 1975; Eliasz, 1990; Ewald, 1924).
Acceleration ofthe car seems to illustrate the concept oftemperament better than any other feature. As we know, each type of car has given acceleration characteristics, which are expressed in time units (seconds), that is, the time needed to move the car from standstill for a given distance under optimal and fully controlled conditions. The acceleration characteristic is one of the most important features of the car and may be found in every car catalog. No statistics are needed to state that there are differences in acceleration between cars.
By using the term "acceleratability," treated here as an analogy to temperament traits, I underscore the fact that cars differ from each other in acceleration and that these differences, being relatively stable, have the status of a feature resembling traits or dimensions in psychology. Acceleratability can be measured in time units, suggesting that we are dealing with a property that objectively exists. However, it does not have the ontological status of a feature as does the color or shape of a car. The latter are permanently observable features of the car, whether the car is moving, has stopped, is operating, or is damaged. In contrast, acceler-atability reveals itself under specific conditions and can be measured only when the car moves (behaves), similar to temperament traits which are expressed exclusively in behavior and reactions. Thus acceleratability, like temperament traits, has the status of a latent property, which is activated and expressed under specific conditions, such as movement (car), or behavior and reactions (man and animals). Ac-celeratability, like temperament traits, should be understood as the tendency of a car to move (behave) with a given speed.
The mechanisms determining acceleratability of a car throw some light on our understanding of the determinants of temperamental traits. They also exemplify how to understand cross-situational stability and temporal consistency of temperament traits. Acceleratability of the car depends on many of its aggregates and elements, the most important of which seem to be the following: type of engine, capacity of cylinders, ignition, carburetor, shape, size, and weight of the body, and type of wheels. Other conditions being equal, the interaction ofthese aggregates and the "makeup" elements of the car determine its acceleratability. One may also say that these composite features of the car ensure stability and cross-sit-
uational consistency of the acceleratability characteristics, as the biochemical and physiological mechanisms underlying temperament do.
Drivers know, however, that such factors as a dirty spark plug, disordered ignition, choked carburetor, leaky cylinder, and so on, directly influence the acceleratability of a car by lowering its capacity. Even one element in the system of aggregates determining acceleratability, for example, leaky cylinders, may essentially decrease this feature. This gives the erroneous impression that acceleratability is determined by the one factor only. Such a state of affairs is often found in temperament research. A disorder or change in a given physiological mechanism, causing changes in temperament characteristics, was used by some researchers (e.g., Kretschmer, 1944) as an argument proving that a given temperament trait is predominantly or only determined by this mechanism. The number of aggregates and elements of the car the interaction of which determines such a simple trait as acceleratability suggests that the biochemical and physiological bases of any temperament trait in man and animals must be much more complex and cannot be reduced to single biological mechanisms.
When assessing acceleratability of the same car but under different conditions we easily observe differences in scores expressing this "trait." We then find that acceleratability also depends on such factors as air pressure of the wheels, quality of gasoline, temperature of the engine, kind and quality of the road (highway), weather (e.g., rain, wind, snow). Only when these environmental conditions are comparable for the two cars may we expect similar scores on the accelerata-bility dimension. Under constant conditions the scores expressing this feature are stable, that is, they are predictable.
The fact that the actually measured acceleration depends not only on the anatomy and functional capacity of the aggregates (the interaction of which determines acceleratability) but also on other (environmental) conditions shows distinctly that the expression of this trait is a result of interaction between the composite mechanisms of the car and a variety of environmental conditions. This is all the more true when referring to temperament characteristics. The more constant the environmental variables, the stronger their influence on the stability of a given characteristic. A car that always runs on a dust track will always show lower acceleration characteristics as compared with the same car tested on free highways. This is also the case with temperament characteristics that may be expressed differently depending on the particular environment.
Some external conditions, especially permanent ones, may cause changes in the elements and aggregates of the car that in turn lead to changes in its acceleratability. For example, the way the new car has been run in, the mode of exploitation, the quality of gasoline usually applied, and so on, are factors that codetermine the acceleratability of the car. Again, the analogy with temperament is obvious. Educational treatment, especially in the first period of life (the analogy to the running in of a new car), and permanently acting environment (social and physical)
cause changes in the biological bases of temperament and therefore also lead to changes in temperament characteristics.
Acceleratability has the status of a formal trait, as most temperament characteristics do. Each car may be characterized by means of this property, measured in time units (in analogy to tempo, duration, or mobility in temperament).
Finally, acceleratability understood as the car's property is not the speed with which the car passes a given distance from starting point, yet it expresses itself in speed units. Neither can acceleratability be reduced to the mechanisms and factors underlying this property. Acceleratability is the latent property of the car which results from the interaction among the different aggregates and elements. In my understanding temperament traits have a similar status.
Temperament traits are expressed in behavior characteristics but cannot be reduced to these characteristics. They are determined by internal (inborn and acquired) mechanisms but, again, cannot be reduced to these mechanisms. Temperament traits are the result of a given interaction among a variety of internal mechanisms; they have a specific status expressed in the tendency to behave (react) in a given way. This tendency, because it is more or less consistent and stable, may be modified by external conditions. As shown in Chapter 4 and 5 the biological bases determining temperament traits are far from familiar and identified, thus temperament traits have the status of hypothetical constructs as some authors have mentioned (e.g., Bornstein, Gaughran, & Homel, 1986; Kagan & Reznick, 1986; Eliasz, 1990). Several years ago I emphasized that temperament "does not exist as such; it is a type of a theoretical construct, referring to existing phenomena, just like the theoretical construct of intelligence" (Strelau, 1986, p. 62).
Much less agreement occurs, however, regarding the relationship between temperament and personality. This is mainly due to the ambiguity of the concept of "personality" itself, already shown most explicitly by G. W. Allport (1937) who distinguished 50 definitions of this concept (see Fig 1.7).
The tendency to distinguish between temperament and personality has a longstanding tradition. Whereas Galen used the term temperament to describe behavior characteristics that have their roots in the individual's organism (endogenous factors), others have tried to explain human behavior characteristics by means of external conditions. The monograph Characters by Theophrastus (4th-3rd century B.C.), who explained individual differences in character4 mainly in terms of environmental settings (exogenous factors) exemplifies this view.
4In more contemporary writings the term character has been substituted by the term personality or temperament and character have been treated as two separate components of personality (see G. W. Allport, 1937; Roback, 1931).
Was this article helpful?
This guide Don't Panic has tips and additional information on what you should do when you are experiencing an anxiety or panic attack. With so much going on in the world today with taking care of your family, working full time, dealing with office politics and other things, you could experience a serious meltdown. All of these things could at one point cause you to stress out and snap.