1 Emotion and Music: Inherent Responses and the Importance of Empirical Cross-Cultural Research Cory McKay Faculty of Music McGill University 555 Sherbrooke Street W est Montreal, QC, Canada, H3A 1E3 ABSTRACT of leaned associations themselves. A number of theoretical arguments have been presented that Although it is generally accepted that encul- contend that any emotional meaning gleaned turation plays an important role in the associa- from music is due entirely to enculturation. tion of emotions with musical stimuli, there is still debate over whether inherent pre- Meyer (1956) argues that musical expressiv- programmed responses to psychophysical cues ity is linked primarily to the structural aspect of also play a role. This paper reviews and evalu- music. Expectation and deviation from expecta- ates existing theoretical arguments and empirical tion lead to the perception of emotion. Since results and proposes ideas for future research. An these expectations must be learned, the implica- emphasis is placed on the importance of cross- tion is that affective responses to music are also cultural empirical studies. learned. Sloboda and Lehmann (2001) present a more recent illustration of this perspective. Although these arguments likely do have It is well known that certain types of music some validity in relation to W estern music, they are able to evoke emotional responses in certain are of limited applicability in a broader context, listeners. The mechanisms that cause these re- as the music of most cultures tends to place sponses are not clearly understood, however. It is much less emphasis on strict structure and for- also uncertain whether these mechanisms can malisms than W estern music (Nettle et al., operate across cultural boundaries or if they are 1997). Meyer‘s perspective is cast into further culture-specific. doubt by a study by Tillman and Bigand (1996). The overwhelming body of psychological The authors broke several W estern musical ex- research to date has involved W estern subjects cerpts into very short segments, which were then listening to W estern tonal music. Although there played in the original order to some subjects and do not appear to be any immediately and easily in reverse order to others. Not only were the sub- available explanations of exactly how humans jects able to perceive emotion when the segments are able to express and interpret emotions in mu- were played backwards, but they did so in ways sic, valuable insights could be gained through the similar to the subjects who heard the segments study of a wide variety of cultures. Cross- played forward. This appears to indicate that the cultural studies in particular could prove valu- role of global structure may be overstated even able in understanding the relative importance of in W estern music. enculturation and biologically pre-programmed Sloboda (1985) argues that musical stimuli emotional responses to particular musical stim- acquire emotional meaning through association uli. An extension of the cultural breadth of re- with words that often accompany them. In order search and a refinement of experimental methods for two people to have similar affective re- could provide valuable insights into a potentially sponses to a musical stimulus, then, they must very fertile area of study. both have learned to associate the stimulus with There is little doubt expressed in the current words with the same semantic meaning. The literature that enculturation plays an important consistency and precision of inter-participant role in affective responses to music. There is still agreement arrived at in numerous recent experi- debate, however, over whether culturally univer- ments linking psychophysical stimuli with emo- sal psychophysical perceptual cues also play a tional responses (Balkwill and Thompson, 1999; role, perhaps acting in concert with learned asso- Behrens and Green, 1993; Campbell, Krysciak ciations, or perhaps influencing the development and Schellenberg, 2000; Crist, 2000; Gabrielson 2 and Juslin, 1996; Gabrielson and Laukka, 2000; sense without having learned about its music Juslin, 1997; Juslin and Madison, 1999; Kame- through experience. He contends that the music netsky, Hill and Trehub, 1997; Kaminska and of each culture is dependant entirely on its cir- W oolf, 2000) calls Sloboda‘s emphasis on se- cumstances, technology and ways of thinking, all mantic meanings into question. of which give rise to unique musical values that lead to music that cannot be translated between Kivy (1980) has suggested that musical ex- cultures. pression is linked to what he calls the —emotive life“ of a culture, which is to say expressive ges- W hile the above perspectives do present tures, postures and speech. Kivy argues that, some interesting ideas, they are unfortunately since the emotive life of each culture is pre- based primarily on theoretical arguments. This sumably distinct, the links made between various approach can certainly lead to important insights musical stimuli and particular emotions must and ideas, but one must rely on empirical evi- also be distinct. The music of cultures with un- dence in the end in order to resolve theoretical familiar emotive lives will therefore sound nei- dispute. Although it is conceivable that musical ther musical nor expressive to uninitiated listen- expression is a result solely of enculturation, ers. reliable and repeatable empirical evidence is needed before any definitive stance can be In a somewhat similar vein, Tolbert (2001) adopted. There is an important need for empiri- suggests that musical expression is learned at an cal cross-cultural research to test the above theo- early age through processes related to modelling retical models. of facial expressions, expressive gestures and body movements in general. Some indirect sup- Now that factors related to enculturation in a port for this position is offered by the research of broad sense have been discussed, it is appropri- Davidson (1993), who found that the visual as- ate to turn our attention to the effects of localized pect of a musical performance can be more in- psychophysical parameters. Although it was formative as to a performer‘s expressive inten- once argued that affective responses to music are tions than the music itself. due only to high-level elements, recent research has made it clear that isolated variations of spe- Although the influences outlined by Kirvy cific psychophysical parameters can elicit spe- and Tolbert may well play important roles in the cific emotional responses (Balkwill and Thomp- development of emotive responses to music, it son, 1999; Behrens and Green, 1993; Campbell, would be a mistake to say that other factors are Krysciak and Schellenberg, 2000; Crist, 2000; not significant as well. Musical traditions can Gabrielson and Juslin, 1996; Gabrielson and evolve separately from a culture‘s emotive life, Laukka, 2000; Juslin, 1997; Juslin and Madison, for example, yet still influence expressive prac- 1999; Kamenetsky, Hill and Trehub, 1997; tices. It is also important to consider the possible Kaminska and W oolf, 2000). Specifically, ma- existence of cross-cultural commonalities in nipulations of timbre (Behrens and Green, 1993; emotive life itself, perhaps with biologically pre- Gabrielsson and Juslin, 1996), tempo (Crist, programmed causes. For example, a review of 2000; Gabrielsson and Juslin, 1996), timing pat- cross-cultural studies (Frick, 1985) revealed that terns (Gabrielsson and Laukka, 2000; Juslin and many cultures use similar techniques to express Madison 1999), pitch (Campbell, Krysciak and emotion vocally. It would be interesting to per- Schellenberg, 2000; Kaminska and W oolf, 2000) form further cross-cultural research searching for and dynamics (Crist, 2000; Kamenetsky, Hill links between commonalities in musical expres- and Trehub, 1997) can all evoke emotional re- sion and commonalities in gestural and verbal sponses in listeners. These studies have also expression. shown that skilled performers are able to con- Feld and Keil (1994) contend that all mean- sciously control the affective content of their ing that we perceive in music is socially consti- music, although certain emotions are harder to tuted through experiences of sounds in our lis- evoke and perceive than others, and only a lim- tening histories. They present this as an explana- ited number of emotions have been consistently tion for the variety of ways in which different tested (primarily anger, fear, happiness and sad- individuals can interpret a given piece within the ness). same culture. Unfortunately, only W esterners listening to W alker (1996) argues that it is impossible to W estern music were studied in these experi- understand the music of another culture in any ments. This limits the meaningfulness of the spe- 3 cific results of each of these studies to the sphere consistent links between the ratings for type of of W estern music and listeners familiar with it. emotion and the ratings of the psychophysical These studies are valuable, however, in that they dimensions. show that manipulations of isolated and well- W hat is particularly interesting about the re- defined psychophysical parameters can cause sults is that the authors found a correspondence affective responses, even if it is possible that between the psychophysical dimensions of music these responses are due only to associations ar- that other researchers had previously mapped to rived at through enculturation. certain types of emotional responses in W estern The next step, then, is to carry out compara- music and the dimensions that the listeners in tive studies on the music of multiple cultures in this experiment matched with the same types of order to discover if similar psychophysical cues responses. This implies that Hindustani and evoke similar responses across cultural bounda- W estern musicians may use similar psycho- ries. If they do, then this could imply that there physical manipulations in order to evoke similar are inherent mechanisms at play, and that affec- emotions. This offers some limited support for tive responses may not be due only to encultur- the existence and importance of psychophysical ation.