1

Theme 3: Culture, Self and Emotion

Theories of Culture and Self

How might we start to understand the impact of our cultural background on ourselves as
thinking, feeling beings? This is a very hard question to answer. It is hard from the point
of view of disentangling our personal selves from our cultural selves. Is there a ‘me’
that can be located and explored both separately from, as well as in combination with,
life in general as it goes on in my household, workplace, social group, sports club, or
neighbourhood? How might I cast the ‘me’ – the individual person – known as Graham
(or whoever you happen to be)? What words might I choose to describe all the bits that
go to make up who I am and what I'm like? Might those words reflect the essence of the
person who is searching for the essence of myself – the ‘I’ as scientist, subject, and doer
in the act of searching – and also the one who is the focus of my analysis – the ‘me’ as
known object of my musing? How might ‘I’ describe ‘me’ without making some
references to other beings or objects that inhabit my world? Is it possible for the ‘I’ to
describe ‘me’ without making reference to my social roles and relationships with
others? One way of trying to answer these questions is to look at ways in which we
describe ourselves. To guide you through your self-description, try the first exercise in
your resources for Theme 3. While as conscious human beings we are able to
contemplate such questions, we are nevertheless constrained in the ways we answer
them by our experience of the world. Consider the analogy of a goldfish – on which we
will confer consciousness for the purpose of this exercise – which lives all of its life in a
goldfish tank. It knows only its watery world, the other goldfish with which it shares its
tank, and what life holds for a goldfish in a goldfish tank. It can reflect on life outside of
the tank and how it receives its daily ration, but can it ever know that other life? Even if
it were to be relocated to a natural freshwater habitat, would it still view its existence as
if it were in the tank? Culture constrains us as human beings in a similar fashion to the
constraints that being a goldfish and being in a goldfish tank places on our marine
friend. It does this in subtle ways as you might come to understand – I myself have
difficulty appreciating what it would be like to grow up in another culture – when we
analyse the differences between individualistic and collectivistic cultural orientations.

When people define themselves in terms of how ‘I am, what ‘I like’, what ‘I do’, and
when their self-description clearly delineates between themselves and others or their
social affiliations, their view of self is described as an independent one. Such a self view
is a typically Western self-vies: "The Western conception of the person as a bounded,
unique, more or less integrated motivational and cognitive universe, a dynamic centre of
awareness, emotion, judgment and action organised into a distinctive whole and set
contrastively both against other such wholes and against its social and natural
background is, however incorrigible it may seem to us, a rather peculiar idea within the
world's cultures" (Geertz, 1975, p.48). It is a “peculiar” self-view, Geertz argued,
because many other cultures do not view the self in that way.

Another view of self is the interdependent view of self. What makes us "us" (not what
"I" am) are the distinctive sets of relationships into which we are placed with other
people. Within this interdependent construction, in order to preserve one's identity, or
sense of self, one must preserve the harmony of one's relationships with others, by
behaving appropriately within those relationships. Markus and Kitayama (1991) have
used the term connectedness to describe the essential quality of the interdependent self.

2

"A normative imperative of [connected] cultures is to maintain ...... interdependence
among individuals. Experiencing interdependence entails seeing oneself as part of an
encompassing social relationship and recognising that one's behaviour is determined,
contingent on, and, to a larger extent, organised by what the actor perceives to be the
thoughts, feelings, and actions of others in the relationship." (p. 225) Matsumoto (2000,
pp.54-58) discusses the independent-interdependent distinction as it pertains to self-
description and understanding.

Although Markus and Kitayama (1991) directed their analysis of self-understanding
primarily at Asian cultures, the independent-interdependent distinction has emerged
from research in other cultures as well. In Indigenous cultures in Australia, according to
Davidson and Reser (1996), "Aboriginal self-construction is in many ways collective or
interdependent ...... Who one is in an Aboriginal community is expressed as a complex
intersection of roles, relationships, mutual obligations and expectations, and common
connections. While country, clan membership, kinship location, and shared matrilines
and patrilines are primary intersects, other self-reference points include shared totemic
affiliations, ceremonies, and language. Those Aboriginal skeins of relatedness extend to
land, the natural world, and the spirit world, contrary to the western divisions of natural
versus man-made, myth versus reality, and past versus present and future. Self is thus
situationally and socially defined." (p.116-117) This view of Indigenous self not only
accounts for one’s relationships with other members of one’s community and family
group, but also for one’s relationships with other species and natural objects and
physical spaces. As we shall see, how the self is viewed has implications for how one
should feel about and behave toward others. For Westerners the depth of meaning in
those human and anthropomorphic connections is difficult to fathom.

Here a scientific analogy might aid our understanding of interdependence or
connectivity. “We are driven to the conclusion that there are two ways of advancing
from primitive truth. One such way was the way taken by some of the Greeks: to refine
the idea of causation in such a way that one finished up with a mechanical explanation
of the universe, such as Democritus did with his atoms. The other way is to systematise
the universe of things and events into a structural pattern which conditioned all the
mutual influences of its different parts” (Needham, 1978, p.166). Recognising that one
should be attending to one’s relationships with others, maintaining the emotional
valence of those relationships, and reciprocating to other’s behaviour and emotional
responses to oneself, is easier said than done when one’s own view of self is of the
independent kind. A full appreciation of the interdependent self is expressed in an
everyday sense in terms of what makes me ‘me’ and not who ‘I’ am. What makes us
‘us’ is the distinctive sets of relationships into which we are placed with other people,
and in which the actual identities of those others is important. It is not our personal
traits that make our interconnected selves distinctive, but those relationships with
others. Therefore, to preserve one’s interdependent identity – one’s sense of self – one
must preserve the harmony of one’s relationships with others who are a significant part
of one’s life. Furthermore, one will need to behave in ways that are appropriate within
those relationships and within the social contexts in which the relationships are found.
So one does not behave in the same way toward everyone, but instead one behaves
according to the obligations that are inherent in one’s established relationships with
others. This is a different way of thinking about one self from seeing one self as
straightforward, or caring, or tough, etc. Connected selves are strongly influenced also
by how others are behaving at the time. One has expectations about the situational

3

nature of others’ behaviour, and one seeks to normalise interactions in terms of those
expectations. In contrast, an independent, Western view of self holds that selves are
independent, self-contained, and autonomous. They are configured by internal attributes
such as personality traits, attitudes and beliefs, and values. People with independent
self-concepts tend to behave, and describe their behaviour, primarily as a function of
those internal attributes.

Inherent in this distinction between independent and interdependent views of self is the
assumption that these views of self reflect a cultural disposition toward holding these
self-views. Independent self-views are said, and indeed have been shown, to be the
predominant self-view in individualistic cultures, i.e., cultures that strongly reinforce
the value of individualism. Interdependent or connected self-views are said, and have
been shown, to be more prevalent in collectivistic cultures, i.e., cultures that strongly
reinforce the value of relationships between members of a community. The
measurement of individualism and collectivism is discussed further in Matsumoto
(2000, pp.41-47).

Research into the relationship between culture and self-views has addressed questions
about how we view and describe ourselves, how we view and describe others and,
interestingly but not surprisingly, how we insult others. Bond and Cheung (1983) asked
Japanese, Hong Kong Chinese and American students a set of 20 ‘Who am I’ questions.
American students gave more trait descriptions (friendly, happy, withdrawn) than
Japanese students. Hong Kong students fell in between the other two groups. The
authors explained the ordering of groups in terms of the educational and acculturation
influences on Hong Kong students. Triandis, McCusker and Hui (1990) asked the same
question of American, European and Chinese students, and found that Chinese students
gave more responses than the other two groups that signalled membership of a social
category. However, the scoring system used by Bond and Cheung did not distinguish
between statements such as ‘I am a student’ and ‘I am a member of the HKU
psychology class. It might be asked whether both statements are role statements or
whether the latter statement locates the self as a member of a specific collective of
relationships. Keeping in mind comments that I made earlier about the different facets
of the self and locking self-description into a particular way of self-reflection, the ‘I am
...’ question might predispose participants to adopt an independent and agency-oriented
strategy when answering it. Cousins (1989) asked Japanese and American students a
similar question. American respondents were more likely to offer trait labels while
Japanese students were more likely to offer contextually qualified descriptors. Cousins
then changed the task, so that the ‘I am ...’ question contained reference to specific
social contexts, e.g. ‘I am ..... at home’, ‘I am ..... at work’, ‘I am ..... with my friends’,
etc. Japanese respondents used more trait labels than they did when the context was
unspecified. American students used fewer trait labels than they did before and started
to qualify their trait descriptions of themselves, e.g. ‘I am often lazy at home.’ Cousins
suggested that even when Americans are required to act contextually they attempt to
maintain an independent, context-free self concept. Consider the two different ways of
responding to the ‘I am .....’ question. Ms Smith teaches English language to Japanese
students in a Japanese school. She describes herself as female, extraverted, honest, a
teacher, and a movie buff. Mr Kawai is one of her students who describes himself as a
member of the Kobe judo academy, one of three children in the Kawai household, a
pupil in the 11th grade at Motomachi Senior College, a son who respects his parents,
and a friend of Yoshi and Kazuo.

4

A similar difference emerges between independent and interdependent self-views when
respondents are made to describe emotions. Markus and Kitayama (1991) reported that
Japanese respondents made distinctions between emotions that reflected whether they
were ‘connected’ with the person to whom the behaviour was directed and those that
were not about interpersonal connections. Stipek, Weiner and Li (1989) asked Chinese
and American respondents to describe situations in which they became angry. American
respondents were more likely to talk about events that happened to them, but Chinese
respondents were more likely to talk about events that happened to others.

A similar pattern emerges when respondents are asked to describe other people.
Shweder and Bourne (1982) asked American and Indian students to describe their peers.
72% of American responses were context-free trait descriptions, e.g., he is generous.
Only 50% of Indian responses were context-free descriptions, and 50% were
contextualised descriptions, e.g., he takes care of his elderly parents. Korten (1974)
reported on American and Ethiopian descriptions of others. American respondents
accentuated others’ knowledge, abilities and personal characteristics. Ethiopian
respondents gave answers that focused on how a person interacted with, and thought
about, others.

Miller (1984) gave American and Indian respondents scenarios that required them to
make judgements about an actor’s behaviour. For example, a driver and passenger were
on a motor cycle whose back tyre bursts causing an accident. The passenger fell from
the cycle and struck his head. The driver, who was a lawyer on his way to court to
appear on behalf of a client and was uninjured, took his passenger by taxi to a hospital
where he left him, and continued on to the court without consulting the medical staff
about the seriousness of his friend’s injuries. Ultimately the passenger died from an
internal injury. Miller asked her respondents why the driver acted as he did (ignoring
the legality of his actions). American respondents answered in terms of the driver’s
personality characteristics and motives, e.g., he was untrustworthy, traumatised or
career oriented. Indian respondents answered mainly in terms of his assessment of the
situation and his other commitments, e.g., he had a duty of care to his client, or his
passenger did not look seriously injured.

Kearins (1988) reported on her attempts to investigate the nature of intelligent
behaviour among Indigenous Australian people living in the north of Western Australia.
Adults described intelligent children as helpful, being responsible for their own needs,
having good bush skills, and good sportspeople. Unintelligent children were those who
were lazy, dependent on others’ goodwill, silly – they tormented, bullied or destroyed –,
disrespectful and ill-mannered.

If asked to describe myself, I would say as one thing about me that I like reading Asian
literatures, whether they are in English or translated in English. This type of response
typically reflects an independent self. It is interesting in Chinese literature, as a
comparison to my way of describing myself and others, to study how Chinese writers
describe their characters. For example, young women who seek dalliances are not said
to be immoral, but instead are said to bring shame on their family by continuing to meet
with men who are not their chosen husbands.

5

So what is involved in knowing how to ‘sledge’ (to use a common Australian
expression) or, if you like, how to insult another person? Semin and Rubin (1990)
received the following types of responses to their inquiries about insults in individually
oriented northern Italy: ‘stupid’, ‘cretin’, sexual profanities, and religious profanities. In
collectivistically oriented southern Italy, insults focused on relatives, sisters, parents,
and incest relationships. Referring again to my penchant for Chinese literature
characters insult one another frequently by reference to sexual assault of ancestors,
mothers, wives, and other relatives, or by emphasising likeness to certain animal
species. In Japan, one might insult another in a more dignified fashion by using an
overly familiar or common form of address rather than a form of address that denotes
respect behaviour. Bond and Cheung found that Chinese students responded more
strongly to insults that were group directed rather than individually directed.

Finally, under this section on culture and self-views we will look at how one interprets
one’s performance outcomes. How does one view one’s success or failure on a task?
How does one differentially weight one’s contributions to success and failure of an
enterprise with the contributions of others to that same enterprise? Are there any
cultural differences in the manner in which individuals attribute success and failure, and
how might those cultural differences be explained? (See Matsumoto (2000, pp.421-431)
for a more in-depth discussion on attribution behaviour in different cultural settings.)

Two attribution biases have been the focus of cross-cultural research into how
individuals explain their success or failure on a task. One bias, known as the self-
serving bias (Nisbett & Ross, 1980) involves individuals attributing success to their
own skills and abilities, but contributing failure to the actions and influences of others,
the nature of the situation or task, or to other external factors. If I pass my assignment
it’s because I’m clever or because I’ve put in the hard work required. If I fail my
assignment it’s because the lecturer’s instructions to the class were unclear, the
questions were unfair, I was having difficulty with Web access, work commitments
prevented me from studying hard, or the marking was inaccurate. Does this sound
familiar? I know I occasionally engage in this type of thinking when things are not
going well for me at work. Another, though equally biased, way of judging the reasons
behind one’s successes and failures is to attribute one’s successes to the contributions of
others or to favourable circumstances and failures to one’s own shortcomings or lack of
ability. This attribution bias is known as a self-effacing, or modesty, bias. If I pass my
assignment, it is because the lecturer was very helpful, the question reflected material
that had been covered adequately in class, or my study group worked well together. If I
fail, it is because I am not clever enough to be at university or because I did not put in
the work required to succeed.

There has been a significant amount of research into performance attribution. It is
generally the case with Westerners that explanations of success and failure are biased in
a self-serving fashion. However, even Westerners differ in terms of the extent to which
they are prepared to take responsibility for, and control over, their own actions or events
that have occurred. Strangely, but understandably from the perspective of attribution
theory, abrogating all responsibility for failures and negative events is often an
unproductive way of dealing with unpleasant circumstances. It can cause us to become
angry about what has occurred, and this in turn can activate various psychological
distress symptoms. But does self-serving bias operate to the same extent in other
cultures and, if not, what effect does self-blame have on individuals from those

6

cultures? Kashima and Triandis (1986) gave Japanese and American students a difficult
task to do. American students were more likely to explain their success than their failure
in terms of their own ability. Japanese students conversely were more likely to explain
their failure rather than their success in terms of their own ability. Fry and Ghosh (1980)
studied Caucasian and Asian Indian Canadian children’s explanations for task
performance outcomes. The Caucasian children were more likely to attribute their
success than their failure to their own level of ability, and failure than success to
circumstances such as luck or poor explanations of the task. The Indian Canadian
children conversely were more likely to contribute success rather than failure to luck,
and failure to their own level of ability. These results and others suggest that self-
serving bias is more prevalent in individualistic cultural settings, and that self-effacing
bias is more prevalent in collectivistic cultural settings. Two qualifications need to be
added to this statement of the general trend. First, attribution of ability seems to work
somewhat differently from attribution of effort. Second, the self-effacing bias is more
pronounced in collectivistic cultural settings when the context of the performance is
clearly specified in the questions that are asked of the respondents.

The remainder of Theme 3 explores aspects of human emotion that may be universal –
we have previously labelled such aspects of behaviour etics – and compares and
contrasts these components with aspects of human emotion that are culturally
determined – or emics. In order to achieve that aim we will look in order at the
following component aspects of human emotion:
§ experiencing emotion
§ expressing emotion
§ perceiving others' emotional expression
§ labelling emotions
§ displaying emotions
§ distinguishing between emotion and other aspects of human experience and

behaviour

There is a very basic question that requires an answer before we can proceed further to
analyse questions about culture's effects on human emotion. The question is whether
emotional experience is part of the human condition? Matsumoto (2000, pp.306-307)
considers a number of theories that link emotion and physical arousal. Matsumoto
makes specific mention of the work of Schacter and Singer’s theory. In their theory of
emotion, the latter is reduced to a state of physiological arousal. The labelling of
emotion is determined by the circumstances that led to the arousal. For example,
heightened physiological responses associated with the successful birth of a baby might
be labelled happiness. The same physiological responses experienced at the end of a
bungee rope might be labelled fear by some and excitement by others. What changes is
not one’s physiological reactions but how one interprets them. Other theorists have
placed emotional responding within the confine of cortical arousal. The step from
cortical arousal to cognitive interpretation is not large. If these theories have substance,
then if one were to suggest that individuals in some cultures are unemotional, i.e.,
without emotion, then one would be suggesting that those individuals are "unwired".
Knowing what we do about physiological similarities between individuals of different
cultural backgrounds, it is reasonable to assume that all people in all cultures experience
emotions of one kind or another. In this respect emotional experience – the experiencing
of emotion – is a universal phenomenon.

7

Research into emotional expression has been concerned primarily with facial expression
of emotion (see Matsumoto, 2000, pp. 272-282). Research by Eckman and his
colleagues in over twelve countries whose residents differ markedly in terms of their
ethnicity, customs and traditions has suggested that at least 6 facial expressions of
emotion can be expressed and reliably discriminated between. They are: happiness,
sadness, anger, disgust, surprise, and fear. Eckman's research showed that even when
the person expressing the emotion and the person observing that behaviour were very
different in terms of their cultural background, e.g., New Guinean and Caucasian
Australian, individuals could still reliably produce and recognise the facial expressions
associated with those 6 emotions. Furthermore, the intensity of emotions being
expressed facially could be identified with reasonable accuracy. Therefore, emotional
expression, in the form of some facial expressions, also appears to be an etic or
universal phenomenon.

Eckman’s research findings, if taken at face value, suggest that the categories of
emotion listed above are also universals, in the sense that they are perceived and
labelled in the same fashion in all of the cultures in which the research was conducted.
However, when making the judgement that emotional experience and expression are
universals, it should be borne in mind that the research studies on which that judgement
is based used a common procedure which required participants to view facial
expressions and to select from a set of labels the label which best described the
expression being viewed. That is, the task required research participants to put one label
chosen from a set of labels to one face chosen from a set of faces. It is reasonable, and
necessary, to ask the following questions and to search for answers to them before
assumptions about universality of emotional expression can be sustained. How might
people in their own words describe the expression of emotion by others? Will
individuals from similar cultural backgrounds use similar descriptive terms for
emotional expressions when asked to do so? Matsumoto (2000, pp.282-290) reviews in
some detail the evidence for and against the argument for universals in human
emotional expression and perception. To summarise the findings, it seems that many
more descriptions emerge when participants are required to supply their own labels to
those facial expressions. However, some emotions like happiness and surprise are
consistently labelled, while others like interest and shame are less consistently
identifiable. Markus & Kitayama (1991) listed 20 different emotional terms in Japanese.
Each word was then paired with each other word and each pair of words was rated in
terms of whether the emotions they described were similar to or different from one
another. Some emotions were easily distinguishable from other emotions, but others
were not. For these Japanese participants, emotions that related to maintenance or
severance of social relations were not easily distinguishable from one another. This
suggests that interpretation and labelling of emotions are not as universal as Eckman’s
original research had indicated.

It is important to draw a distinction between experiencing, expressing and displaying
emotions. One might feel a sense of elation – happiness – following receipt of an
excellent test mark. One would normally smile broadly when one is feeling so happy.
However, one may not wish to appear so happy in the presence of a classmate whose
test grade was disappointing. So the difference between emotional expression and
emotional display is the difference between how we show emotion and whether we
show emotion. Much of the research knowledge that has been gathered about emotional
display has come from the labelling studies of Markus and Kitayama (1991) and

8

Matsumoto (1989). Matsumoto found that happiness was more readily identifiable in
individualistic cultures and sadness was more readily identifiable in collectivistic
cultures. American and Japanese respondents were given the task of identifying 6 (real)
emotions from facial expressions of males and females. American respondents were
better able than Japanese respondents to identify anger, disgust, fear and sadness, but
both groups were equally adept at identifying happiness and surprise. Japanese
judgements were more accurate for male than for female models. Japanese respondents
were less adept than the American respondents in judging emotions and situations that
were socially unacceptable or undesirable. Scherer and other researchers (see Scherer &
Wallbott, 1994) asked people in 37 different countries to report on naturally occurring
emotions. Happiness and anger were consistently reported. Reports on other emotions
suggested that there was considerable variation for the latter in terms of their frequency,
intensity and duration. Guddykunst, Ting-Toomey and Chua (1988) argued that
different situations trigger different emotions in different cultures, e.g., in high
masculine/high uncertainty avoidance cultures novel situations will frequently elicit a
fear response. Guddykunst et al. (1988) in their research also elaborated on the
relationship between individualism-collectivism and emotional perception and display.

An example of that difference can be found in the juxtaposition
of our Ms Smith’s and Mr Kawai’s feelings and conduct after a
classroom disagreement. During the disagreement Mr Kawai
was very guarded in letting Ms Smith know what he was feeling.
However, he vented some of his frustration to his friend, but was
still careful about showing too much of his embarrassment about
Ms Smith’s behaviour toward him. Ms Smith had raised her
voice when they talked and her face revealed her sense of
annoyance. When she later spoke with the school principal she
was very critical of Mr Kawai for his failure to accept her point
of view. Both acted as they would with other members of their
own cultural group, and neither realised just how annoyed the
other party was. Mr Kawai deep down was ashamed about the
disagreement with his teacher. Ms Smith was angry about the
intransigence of her pupil.

(adapted from Smith & Bond, 1993, p. 63)

As with understanding the importance of individualistic or collectivistic values for how
we think about ourselves, it is difficult for us to break out of our cultural domain when
we analyse our feelings. If our upbringing is essentially Western, in addition to our
heightened sensitivity to positive rather than negative emotional expression and display,
we grow up clearly distinguishing between how we act – or should act, what we think
and how we feel. There is pressure on us to separate out our thoughts from our feelings.
Have you ever been confronted with the statement, ‘Let’s look at this rationally’, i.e.,
‘Let’s not let our emotions get in the way of our making a sound decision’? Or have you
heard someone say, ‘If I’d stopped to analyse what I was feeling, I would never have
done what I did’? The delineation between thought, emotion and action is also evident
in general psychology. Emotion is seen and studied separately from other aspects of
human experience such as thought, motivation, perception and action. In some other
cultures all action, all thought, is accompanied by an emotional response. The system of
social relationships prescribes the emotion one should display when doing a task, when
interacting with others, etc. The relationship is as much dependent on the emotional

9

display as it is on the other aspects of behaviour in those situations. An example of this
can be found in the importance that appropriate emotional conduct has for Indigenous
Australians:

‘Feelings come first’ in this indigenous psychology. They are
the most meaningful touchstones with respect to ongoing
transactions with one’s physical and social worlds, and are also
central to communication and meaning. “Not to show proper
feeling in one’s interactions with others is to question the
relationship, not just to violate an expectancy, but to threaten a
severing of connectedness, critical to a sense of self and well
being.”

(Davidson & Reser, 1996)

Matsumoto (2000, pp.307-309) also provides some specific examples of culture-specific
constructions of emotion. Some emotional labels specific to particular cultures not only
divide the spectrum of human experience differently from the division that we find in
scientific psychology, i.e., a model that clearly separates feeling from thought and
action, but also divide the emotional spectrum differently. An understandable, but less

appreciable, culture-specific label taken from the work of Lutz (1980) is

fago

, which is

a combination of compassion, love and sadness. Some of the other examples that
Matsumoto discusses, e.g., itoshii and amae, appear to represent a complex pattern of
feelings associated with being connected (engaged) or disconnected (disengaged) with
other members of one’s family and community. These culture-specific concepts provide
evidence for the existence of emic emotional constructs, suggesting in different cultures
that there may be unique patterns of human feeling.