Are Musical Emotions Invariant Across Cultures?

Are musical emotions invariant across cultures? This (apparently) simple question is the one I encounter most frequently when I present my work to lay audiences. But the question is difficult to answer in a simple way, because there exist both similarities and differences across cultures, depending on focus and level of analysis.

Higgins (2012) offers a balanced and useful survey of the terrain, highlighting the many issues that scholars will need to address. It is significant, however, that her review does not result in any definitive conclusion. In this commentary, I will suggest that to determine to what extent musical emotions are invariant across cultures, we must break down the problem into smaller constituents based on certain distinctions. First, we should distinguish between perceived and aroused emotions. The respective contributions made by culture and biology will be different depending on which process we are concerned with.

Starting with perception of emotions in music, I think one could easily obtain evidence of either cross-cultural invariance or diversity, simply depending on how one is selecting the music in a study. By selecting pieces that are perceived as expressive of an emotion through simple cues such as tempo, loudness, and timbre (what I refer to as “performance cues” since they often vary according to the performer’s interpretation), you could find a fair level of cross-cultural agreement. (This was the case in the Juslin and Laukka [2003] meta- analysis, which focused largely on performance.) Conversely, by selecting pieces that are expressive mainly through cues that are part of the composition (e.g., melody, harmony), you could presumably find a larger in-group advantage. Thus, without a representative sample of pieces (a tall order indeed!), you could easily support either position.

The relationship between biological and cultural influences becomes even more intricate when it comes to felt emotions aroused by music, because emotion induction is more complex and multifaceted than is emotion perception, with greater diversity in response even within the same culture, or subculture. This partly reflects that emotional reactions to music are strongly influenced by factors in the listener, the music and the context. Because all of these are likely to differ across cultures, what is surprising is not that there are cross-cultural differences, but that they are not even larger! How can this be explained?

Higgins’s initial overview of universal and culture-specific aspects focuses on surface features of the music (Higgins, 2012, pp. xxx–xxx). However, it is important to note that diversity on the level of surface features of the music does not necessarily mean that there is diversity on the level of underlying mechanisms; for instance, although music that arouses sadness in listeners in one culture might sound different from music that arouses sadness in listeners in another culture, this does not rule out that the emotions were aroused for the same reasons in both cases.

Juslin and colleagues have outlined a new multilevel theory of emotion-causation that assumes that music can evoke “basic,” “complex,” or “refined” emotions (e.g. Juslin & Västfjäll, 2008) via eight mechanisms that music shares with other emotional stimuli in everyday life: brain stem reflexes, rhythmic entrainment, evaluative conditioning, contagion, visual imagery, episodic memory, musical expectancy, and appraisal. Each mechanism has unique features (e.g., in terms of information focus, mental representation, brain regions, development, cultural impact, induced affect, and degree of volitional influence), precluding general conclusions about musical emotions. This applies to the present issue also: The relationship between biology and culture will differ depending on the mechanism involved in a specific episode (e.g., less cross-cultural diversity for brain stem reflexes than for statistically learned musical expectancies).

The combination and interaction of the mentioned mechanisms can, perhaps, partly account for the “playful attention” in musical experience discussed by Higgins (2012)—how the listener may “switch focus” between content (e.g., contagion) and form (e.g., musical expectancy); present (e.g., brain stem reflex) and past (e.g., episodic memory). The theoretical framework has also implications for the connection between perceived and aroused emotion: Whether a piece that expresses a specific emotion will arouse the same emotion or a different emotion depends on the precise mechanism involved. Emotional contagion, for example, by definition involves a similar emotion, whereas episodic memory could involve any emotion (e.g., a “happy” piece could evoke a “sad” episodic memory). Empirical data suggest that emotions aroused (if any) do not uniformly “match” emotions perceived (if any).

Higgins (2012) argues that music may encourage an affective sense of affiliation and security, thus facilitating feelings of transcultural solidarity. Again, breaking down the music– emotion process into discrete mechanisms is helpful, given that these may contribute differently to the sense of “attunement” with both self and others. For instance, both entrainment and contagion may produce a sense of affiliation; evaluative conditioning may promote security through the mere exposure effect (assuming that absence of aversive events constitutes the unconditioned stimulus); and episodic memories may serve to confirm one’s identity.

The main theme of Higgins’s (2012) analysis is that culture is implicated on every level of the music and emotion relationship. (I think it would be difficult to find anyone who would argue otherwise, particularly since music is by definition a cultural artifact.) However, the same can of course be said about biology: it is implicated on every level. The nature–nurture dichotomy is false. However, it has been proposed that what position a scholar takes on the universalism–relativism continuum should depend partly on the research issue in question. Thus, as regards mechanisms, it appears feasible to adopt a version of “moderate universalism”: An account of the arousal of emotions can be cross-culturally valid at the level of mechanisms despite cross-cultural diversity in musical surface features and evoked emotions. Ultimately, biological and social-constructionist accounts are truly meaningful only if they are integrated and consistent across the respective levels of analysis. Theories of psychological mechanisms could provide the necessary “bridge” between these approaches by helping to delimitate what is biologically constrained and what is socially constructed in particular instances of musical emotion.