Scarlet & Grey
Ohio State University
School of Music


Three Studies of Culture and Emotion

Notes by Sato Ashida


Music 829
May 29, 2001

Notes on articles by Anna Wierzbicka (1994), Sandra Trehub (2000) and Anna Unyk et al (1992).

Anna Wierzbicka (1994)

Anna Wierzbicka. Emotion, language, and cultural scripts. In: Shinobu Kitayama & Hazel Rose Markus (eds.) Emotion and Culture: Empirical Studies of Mutual Influence. Washington: American Psychological Association, 1994. pp.133-196.

-different cultures encourage different attitudes toward emotion and that is reflected in the lexicon and the grammar of the language associated with the culture

Part I. Theoretical Issues

Universalism -- there is a set of discrete and universal basic human emotions that can be identified by English words (happiness, anger, fear, surprise, disgust, or shame -- Ekman) -Cross-cultural researches --> emotion terms available in any given language are unique and reflect a culture's unique perspective on people's way of feeling.

  1. Language specific character of emotion concepts: some words for emotional expressions have no exact equivalent in English and cannot be adequately translated. Cross-linguistic variation -- there is no reason to assume that some English words have a privileged status and offer access to psychological states of universal significance. * Languages differ in the size and character of their affective lexicon (hundreds in English / Chepang language of Malasia has less than 10). Ifaluk language of Micronesia [Lutz 1982, 1985; Wierzbicka, 1992] does not have a word for `anger' -- `song' meaning something like `justified anger' (less aggressive feeling) Polish word `gniew' roughly means a dignified kind of anger *emotion terms cannot be matched across language boundaries
  2. The need for lexical universals as conceptual and descriptive tools. Researchers need to try to understand concepts from a native’s point of view. Lexical universals -- concepts encoded in distinct words in any human languages (such as good or bad, know or want) -- tentative list. Ifaluk does not distinguish lexically between feel and think.
  3. Basic emotions and the issue of discreteness. Problem of the relationship between language and actual feelings -- difference between `jealousy' and `envy'? --> differential use of these words depends on assumptions about rights, duties, values [Harré, 1986] (problem: theory that there are some discrete basic emotions -- e.g., Sadness and distress, frustration and anger) *Feelings tend to be nebulous, fluid, and difficult to compare without language / assumptions / interpretations

    Ekman (1992): a nondiscrete view -- stated that each of the basic emotions stands not for "a single affective state but a family of related states"; "for each emotion more than one universal expression has been identified"

  4. The relationships among emotions, sensations, and feelings -difference between emotion and sensation may be artificial and may be due to unconscious absolutization of Anglo folk dichotomy opposing body to mind. [Coulter, 1986] Misassimilation of emotions to sensations (feels tired, feels toothache, feels hungry, felt thirsty, feel pain) as a serious error -- "we cannot even begin to identify the emotion unless we take into account how a person is appraising an object or situation"

Universal point of view: it is the undifferentiated feeling that is a truly fundamental human concept, not the more elaborated, more culture-dependent, and theory-laden emotion Emotion in universal semantic primitives view: Jamesian semantic formula --English lexical category encourages English speakers to interpret their experience in terms of this model --> it is not true universally Phrases like `psychology of emotion', `psychobiological theory of emotion' or `operational definition of emotion (galvanic skin response)' create the impression that emotion is an objectively existing category delimited from other categories by nature itself.

E.g.,: English-language introductory psychology textbook: "emotion .... As a departure from his or her normal state of composure; at the same time there are physical changes that can be detected objectively" -- assumption that a person's normal state is a state of composure --- "emotions such as joy, worry, sadness, delight constitute most people's normal state = absence of emotions indicates a deadening of heart or soul"

E.g., Ekman's suggestion that "emotions are typically a matter of seconds, not minutes or hours" -- true only for Anglo males trained to control and suppress their feelings

**Whereas the concept of feeling is universal and can be used safely in the investigation of human experience and human nature, the concept of emotion is culture bound and cannot be similarly relied on. [It is better to define emotion with reference to feeling]

[Ellsworth, 1994] "the term `feel' (in English) was not generally used to refer to emotional states until the nineteenth century; before that one simply was sad or ashamed, and "to feel something" typically meant to touch it." --> a heightened cultural attention to feelings. [Russel, 1991] "It is an illusion to think that one can speak about emotions without a reference to internal states (feelings) because the concept of emotion is built upon the concept of feeling."

Wierzbicka -- it is important to realize the complex and culture-specific character of emotion and not to absolutize it, bearing in mind its dependence on the concept of feel and on other conceptual primitives. "with a coherent semantic theory and a well-developed semantic methodology, the meaning of words such as feelings and emotions can be stated in a nonarbitrary way open to intersubjective assessment; and it is the absence of serious investigation of the semantics of emotions, rather than its exaggerated pursuit, that has long hindered the progress of research in this area."

Part II. Cultural Scripts: Examples from Anglo and Polish Cultures

Different cultures take different attitudes toward emotions, and these attitudes influence the way in which people speak. -emotional cultures (Russian, Polish, Jewish, Black American cultures) -nonemotional culture (Anglo and Japanese from Slavic perspective) ** Need individualized culture scripts instead of global labels or schematic dichotomies

  1. Expressing good feelings toward the addressee
    "My mother says I'm becoming "English". This hurts me, because I know she means I'm becoming cold ...." [Eva Hoffman]. Polish culture: behavior that shows feeling is seen as the norm, not as departure from the norm (emotional in English word). Anglo culture: "cold speech" is associated with ideals such as self-control and the ability to be dispassionate.

  2. My feeling versus someone else's feelings (p.163)
    Polish culture: speaker is not discouraged from saying bad things about others and thus hurting their feelings. Anglo: Don't want others to feel bad. Japanese: Discourage people from saying what they think and urge them to pay attention to other people's feelings -- "consideration for other people's feeling"

    *Anglo conversational routine of saying "thank you" in response to a range of positive remarks. [Wolfson, 1981] "In American English, compliments occur in a very wide variety of situations and the frequency of compliments is often remarked by foreigners." -- reinforces a feeling of solidarity between speakers Polish word "komplement" is rather like "flattery" in English

  3. Spontaneity and Self-Control
    Australian: rich in swearing resources (cultural norm to express bad feelings). Polish: encourage uninhibited expression of emotion -- "storminess of emotion". American culture: important to constantly scan one's feelings. Anglo culture: ability to analyze own feelings rationally is important because self-analysis enables people to gain some distance from emotions -- emotional self-control <--> Polish: no need to know "what I feel" but "want others to know how I feel" (need to express feelings without analyzing it)

    [Lutz, 1988] Interviews on emotion with 15 Americans "the rhetoric of control" -- frequently occurring theme: someone or something controls, handles their emotion. "under control" -- linguistic evidence *Anglo tendency to suppress spontaneous expression of feeling is a by-product of a different norm: encouraging controlled behavior "talk about emotion rather than showing it"

  4. Good Feelings in American Culture
    American culture fosters and encourages "cheerfulness". "social smile" -- conscious or semiconscious effort to get rid of bad feelings <--> Japanese "social smile" may be used as a cover for bad and painful feelings. American culture: great emphasis on being liked and approved, on being perceived as friendly and cheerful, and as someone who is competent and in control of oneself and of the surrounding world. [Stewart, 1972]; Americans need to be liked ... "Obsession with making a good impression". Expectation of positive reply in greeting routine [Renwick, 1980]; "American friendliness" [Wanning, 1991] -- often a source of culture shock for others

Conclusion

All cultures evolve different attitudes toward feelings, different communication strategies associated with feelings, and different norms governing the handling of feelings. Norms are shared by a given speech community on an unconscious level -- may be made explicit in the form of cultural scripts formulated exclusively in terms of conceptual and lexical universals.

Anglo-American culture scripts encourage people to feel something good all the time, to be aware of what they feel at any given moment, to be able to analyze and verbalize their feelings, to control their feelings and thus to prevent themselves from feeling something very bad for a long time, to think before saying something to someone when the thing one wants to say could cause the addressee to feel something bad, to separate the expression of one's opinions from any expression of feelings, to behave as though one felt something good toward everyone and as though one felt something good all the time.

Polish culture scripts encourage people to express their feelings freely; to act upon their feelings; to be guided in their actions by their feelings; to express their thoughts and the feelings caused by these thoughts at the same time; to feel and to express very good feelings toward individual people and to do so spontaneously, on the spur of the moment; to express both good and bad feelings spontaneously and fully without inhibitions, delays, and self-censorship.

Lexical, grammatical, conversational routines, and culture-specific speech acts -- evidence for cultural norms [e.g., "Thank you" response to praise and compliments]

Emotion and culture are inextricably interwined. To study the role of emotion in cultural patterns and the role of culture in the shaping and conceptualization of emotions, we need to pay close attention to language, in all its aspects -- lexical, grammatical, and pragmatic.

Sandra Trehub (2000)

Sandra Trehub (2000). Human processing predispositions and musical universals. In: Wallin, Merker and Brown (eds.) The Origins of Music. Cambridge, MA: Massachusetts Institute of Technology Press. pp.427-448.

Trehub considers the possibility of human predispositions for processing music, and speculates about the broader question of musical universals. --> Study of human infants to explore the possibility of such processing predispositions.
>Similarities in musical pattern perception between adults with extensive exposure to music and infants with minimal exposure suggest a biological basis for several aspects of music processing.

Similarities:

>Infants remember the component tones of scales more readily when the scale steps are of unequal size, as in the major scale, rather than of equal size. They also encode more details of a melody when its rhythmic arrangement is conventional rather than unconventional

Music of the world reveals greater emphasis on global structure than on local details and on smaller-integer frequency ratios than on large ratios; similarities from the ubiquity of unequal steps in scales, preferred rhythms, and a special genre of music for infants.

[Trehub & Trainor, 1993; Trehub, Schellenberg, & Hill, 1997] Studied infants' perception of music or music-like patterns (6 to 9-month olds). Used sequence of pure tones (sine waves) -- estimated infants' ability to detect specific changes in a repeating melody. Methods: presented with repetitions of a melody and rewarded for responding to specified changes in the melody. (repetitions were presented at different pitch levels or tempos, forcing infants to solve the detection tasks on the basis of relational cues rather than absolute cues. Conclusion: Infants' perception of music-like patterns is remarkably similar to that of adults.

Relational Processing of Auditory Patterns

Infants treat the transposition as equivalent to the original melody. Treat altered melody (altered pitch relations but unchanged contour) as familiar rather than new. <--> a change in contour resulting from the substitution of a single tone or the reordering of tones leads infants to consider the altered melody as unfamiliar (like adults) *Infants can detect contour changes even when comparison patterns are presented at pitch levels different from the original. *Infants consider faster or slower version of a tone sequence as functionally equivalent, provided the rhythm or temporal pattern remain unchanged. *Infants group or chunk components of tone sequences on the basis of similar pitch, timbre, or loudness [Demany, 1989 Thorpe et al. 1988] in much the same way as adults Universals: 1) the priority of contour over interval processing; 2) the priority of temporal patterning over specific timing cues; 3) the relevance of gestalt principles of grouping --> priority for global, relational cues over precise, absolute cues.

Interval Processing: Frequency Ratios

[Ancient and Medieval] claimed that tones related by small-integer ratios are pleasant, or consonant, and that those related by large-integer ratios are unpleasant, or dissonant. [8ve -- 2:1, perfect fifth -- 3:2, perfect fourth -- 4:3, and the tritone -- 45:32] --> People retain more information from sequences whose component tones are related by small-integer ratios [Trehub et al., 1986; Cohen, Thorpe, & Trehub, 1987] --> infants and adults tend to categorize intervals on the basis of consonance or dissonance rather than size. **Studies suggest physiological concomitants of consonance and dissonance. *Infants also exhibit preferences for consonance over dissonance -- facilitate the encoding and retention of melodies. --> processing universal

Scale Structure

Scales: -Similarities across cultures: the typical division of the octave into five to seven different pitches likely originates in cognitive constraints [Dowling and Harwood, 1986] -Assumed processing advantages of unequal-step scales: 1) allow different tones to assume distinctive functions; 2) facilitate the perception of tension and resolution; 3) and provide the listener with a sense of location within a melody. Experiment: presented infants and adults with transposed repetitions of three ascending-descending scales (an equal-step scale, the major scale, and a unequal-step scale) and subjects were expected to detect semitone changes. Result: Adults performed better on the familiar major scale than the other two, when no difference between equal and unequal-step scale. Infants performed significantly better on both unequal-step scales than on the equal-step scale, but major scale had no advantage over unequal-interval scale -- rules out exposure effect. *Consistent with the view that unequal-step scales have their origin in perceptual processing predispositions. Also indicates the potency of culture-specific exposure

Rhythmic Structure

Diversity of rhythmic structures across cultures -- suggests the notion that musical rhythms have their foundation in culture rather than in nature. Experiment: "good" and "bad" rhythms (as identified by untrained adults) were presented to 6 to 9-month old infants. Infants detected pitch and rhythm change better in the context of adults' preferred rhythm. ["good" rhythm exemplified Gestalt grouping principles and tone durations related by small-integer ratios.

Lateral Asymmetries in Processing

Infants generally exhibit the characteristic right-ear (left hemisphere) advantage for speech and left-ear (right h.) advantage for music [Best, Hoffman, & Glanville, 1982] Music processing in adults and infants' left-ear advantage for contour processing and a right-ear advantage for interval processing.

Implications of Adult-Infant Similarities

Musical systems across cultures: diversity exists, but it is not unlimited Musical culture is built on perceptual processing predispositions -- exposure and training often lead to progressive improvement in the skills that are favored by nature

Speech and Sign for Infants

>Infants' response: positive affect to approving rather than disapproving utterances, and to infant-directed speech rather than to adult-directed speech. >Infant-directed speech: effectively recruits and maintains infant attention -- biological makeup predisposes infants to attend selectively to distinctive pitch contours, whose primitive emotional meanings can be decoded in the absence of language

Music for Infants

Lullaby is perceptually distinct from other song genres. [simple structure, repetitiveness, preponderance of falling pitch contours]--identifiable by naïve adult listeners >Infant's presence alters the caregiver's emotional state and further affects the vocal musculature. [higher pitch, slower tempo, distinctive timbre, and perturbations in fundamental frequency and intensity = reflect heightened emotional expressiveness] *Mothers subtly alter their performance (soothing or playful manner) of the same song.

Infants' Responsiveness to Infant-Directed Music

Infants `prefer' consonant harmonizations of melodies to dissonant; show enhanced attention to recording of a woman singing to her infant relative to a comparable performance with no infant audience. [Dissanayake, 1992] *Songs could be considered embellishments of human vocal communication or ritualized expressions of love, hope, or complaint.

Consequences for Musical Structure

Universals derived from infants' perceptual abilities have their counterparts in universals or near-universals of musical structure. *Different types of music reveal greater relative emphasis on global features (contour, rhythm) than on local details (specific pitch levels and durations); the prevalence of small-integer frequency ratios; unequal scale steps; and preferred rhythms; and existence of a special genre of music for infants. [Terhardt, 1987] Composers intuitively create patterns that build on universal principles of pattern perception.

Sumerian love song from approximately 1400 B.C. that was decoded from clay tablets found in the Middle East [Kilmer, Crocker, & Brown, 1976] -- sounds like ordinary lullaby, hymn, or folk song. -- feeling of familiarity may have originated in the diatonicity of the underlying scale.

Ivan Turk's discovery of a Neanderthal "flute" (approximately 44,000 years old) -- distance between the second and third of four visible holes (two complete holes and two partial holes) is twice that between the third and fourth holes, which would be consistent with whole steps and half steps in a diatonic scale).

**These evidences make an intriguing case for the biological basis of at least some musical principles.

Anna Unyk, et al (1992)

Unyk, A. M., Trehub, S. E., Trainor, L. J., & Schellenberg, E. G. (1992). Lullabies and simplicity: A cross-cultural perspective. Psychology of Music, 20(1), 15-28.

  1. Whether there are structural features that differentiate lullabies from other songs.
  2. Identify features that underlie adults' assignment of songs (correctly or incorrectly) to the category of lullabies.

Experiment 1: Whether lullabies are perceived as simpler than other songs. Methods: 20 university students (45% w/o musical training) were asked to identify simpler melodies in each pair (30 lullabies and matching songs). [African, Asian, European, and North American Indian]. Matching songs from same culture with similar tempo/style/orchestration. Results: Adults rated the lullabies in cross-cultural samples as simpler than the comparison songs. Perceived simplicity played a role in lullaby identification. (some non-lullaby songs had been consistently misidentified = atypical)

Experiment 2: Whether lullabies would be perceived as simple independent of their lyrics. Methods: 30 pairs of songs used in Experiment 1 were filtered to eliminate the verbal content. 20 university students (30% w/o musical training) were asked to select the simpler song of each pair. Results: Lullabies in the sample were still perceived as simpler than the comparison songs. Perceived simplicity of lullabies is influenced, in part, by features that are unaffected by the lyrics --> melodic form and voice quality(?).

Experiment 3: Whether lullabies would be perceived as simple independent of their vocal quality. Methods: The melodies of the sample songs were played in a uniform timbre using a synthesizer. Results: Listeners continued to rate the lullabies as simpler even when lyrics and voice quality were controlled. --> other features of melodies such as pitch range, contour and interval size must contribute to perceived simplicity]

Experiment 4: Identify specific musical features contributing to lullaby identity. Methods: Analyze the structure of 56 songs, 28 lullabies and matching songs from Experiment 3, after transcribing them into standard Western musical notation. [median pitch, pitch range, rate of contour change, descending intervals, rate(average number of tones per minute), phrase length, average interval size, and pitch variety] Results: accuracy of lullaby identification was positively related to the proportion of descending intervals, negatively related to the median pitch of the matching adult song, and negatively related to the number of contour changes per minute in the lullaby. Discussion: lullabies are not characterized by the features that distinguish infant-directed (ID) from adult-directed (AD) speech.

*Adult listeners perceived lullabies or infant songs from different cultures as simpler than other judgements of simplicity. Their classification of songs as lullabies seemed to be influenced by melodic features that parallel some prosodic features of infant-directed speech.



This document is available at http://csml.som.ohio-state.edu/Music829D/Notes/Culture.html