Content uploaded by John Cromby
Author content
All content in this area was uploaded by John Cromby on Jun 25, 2014
Content may be subject to copyright.
1
BEYOND BELIEF
Cromby, J. (2012)
Journal of Health Psychology:
DOI 10.1177/1359105312448866
Numbers in square brackets [p.xx] refer to page numbers in the published version.
This paper is based on a keynote speech given at the International Society for
Critical Health Psychology conference in Lausanne, Switzerland in 2009. It was
published together with a series of commentaries from peers: Tim Corcoran,
Rachel Falmagne, Joseph Forgas, Antonia Lyons & Kerry Chamberlain,
Crysdara Park, Marie Santiago and Louise Sundarajan. I was invited to respond
to these commentaries, and my response is also available:
Cromby, J. (2012). Response to commentaries on ‘Beyond Belief’.
Journal of Health Psychology: DOI: 10.1177/1359105312457144
Dr. John Cromby
Psychology, SSEHS
Loughborough University
Loughborough, Leics LE11 3TU
England UK
J.Cromby@lboro.ac.uk
2
[p.1] BEYOND BELIEF
Abstract
Psychology, including health psychology, frequently invokes the concept of belief but
almost never defines it. Drawing upon scholarship associated with the ‘affective turn’,
this paper argues that belief might usefully be understood as a structure of socialised
feeling, contingently allied to discursive practices and positions. This
conceptualisation is explained, and its implications for health psychology discussed
with respect to research on religiosity and spirituality and debates about the value of
social cognition models such as the Theory of Planned Behaviour.
Introduction
Belief figures prominently in health psychology, both as substantive topic and
theoretical construct. Taking perhaps the most obvious example, as substantive topic
belief is integral to religiosity (“a multi-dimensional variable, which refers to the
personal beliefs and experience connected to religion, and includes overt behaviours,
beliefs, values and goals, and subjective experiences” (Karademas, 2010 p.240) and
spirituality, a more amorphous concept not necessarily associated with organised
religion but typically including “belief in a higher being, the search for meaning, and
a sense of purpose or connectedness” (Aukst-Margetic & Margetic, 2005 p.366).
Evidence for possible connections between health and religiosity/spirituality (R/S) has
been accumulating for some time. A review of 91 studies by Chida, Steptoe, & Powell
(2009) concluded that R/S was associated with reduced mortality in healthy (but not
diseased) populations, and that this effect was partially independent of behavioural
factors (e.g. smoking, drinking alcohol) and of structural variables such as SES.
McCullough, Hoyt, Larson, Koenig, & Thoresen (2000) reviewed 42 studies and
found that religious involvement was significantly associated with reduced mortality,
particularly where measures of public involvement (e.g. church attendance) were
used. The reviews by Aukst-Margetic & Margetic (2005) and Chida et al.(2009) both
suggest that R/S beliefs are associated with reduced cardiovascular disease. Lower
rates of cancer, better overall health and increased life expectancy are found amongst
some religious groups (Koenig, McCullough, & Larson, 2000), outcomes linked
primarily to strict dietary [p.2] and health regimes. With respect to mental health,
3
Aukst-Margetic & Margetic (2005) distinguish between intrinsic religiosity (i.e. for its
own sake) and extrinsic religiosity (for pragmatic or instrumental reasons); they find
that intrinsic religiosity is associated with a reduced likelihood of being diagnosed
with depression, and lower levels of anxiety. There is also evidence that physiological
processes are modulated by belief: Lissoni et al. (2001) review literature suggesting
that ‘psycho-spiritual status’ might enhance immune efficacy, and Koenig et al.,
(1997) found that religious attendance reduced pro-inflammatory cytokines by small
but significant amounts.
Nevertheless, some studies find little or no evidence for positive associations between
R/S and health (e.g. Meisenhelder & Chandler, 2002). Where correlations do emerge,
Sloan, Bagiella, & Powell (1999) highlight poorly-designed research, failures to
control for multiple comparisons, confounds (e.g. healthier people are more likely to
attend religious services) and ethical concerns. Rew & Wong (2006) note a
preponderance of cross-sectional research and frequent use of unvalidated, single-item
measures or indexical indicators. Reviewing 266 medical articles citing religion,
Sloan & Bagiella (2002) concluded only 17% were relevant to positive health
outcomes, and that the majority of studies were methodologically flawed. The
relationships between R/S and health might be mediated along various biological
pathways by diet, smoking, alcohol consumption, sexual behaviour, better sleep,
sympathetic nervous system activity or reduced cortisol (Chida et al., 2009; Rew &
Wong, 2006) but there is no consensus regarding which are most influential, and
Seeman, Dubin, & Seeman’s, (2003) review concludes that the strongest evidence is
for positive effects of meditation upon immune and cardio-vascular function. In short,
despite evidence for a small positive association between R/S belief and health, many
questions remain.
Taking another obvious example, as theoretical construct belief is fundamental to the
theory of planned behaviour (TPB: Azjen, 1985). Each of the TPB’s three core
constructs, attitude, subjective norm and perceived behaviour control, are
“underpinned by belief” (Darker & French, 2009 p.862): beliefs about the
consequences and corollaries of changing sedimented activity patterns, normative
beliefs about self and others, and beliefs about self-efficacy and environmental
factors. TPB research utilises belief elicitation studies, within which population
4
beliefs relevant to the activity of interest are sampled using open-ended questions.
The TPB and other social cognition models continue to be widely used in health
psychology (Hagger & Chatzisarantis, 2009), although they have been questioned in
recent years (e.g. Marks, 2008; Mielewczyk & Willig, 2007).
Belief in the TPB and in R/S is different in content but similar in conceptualization: in
both, belief is understood as durable and implicit; as associated with practices, choices
and activities; and as bearing personal significance and import. Strikingly, however,
in psychology belief is almost never defined. It is simply omitted from most
psychological dictionaries, and used without further specification in the
overwhelming majority of theories and studies. Related terms (‘schema’,
‘disposition’, [p.3] ‘predisposition’) are sometimes used in place of or alongside
belief, but these too are rarely elaborated. Indeed, this lack of clarity is itself seldom
acknowledged, although Jervis (2006) provides some explicit discussion with respect
to political psychology. He links belief to evaluations, attitudes and opinion, and
observes that many beliefs, even those of a non-religious character, are nevertheless
imbued with powerful elements of commitment, or faith. He notes that although
different kinds of belief appear to have different functions (to index ‘inner’ states,
urge others toward action, or specify causal relations) each appears to point toward
“the inextricable role of emotion in sensible thought” (Jervis 2006, p.642).
Philosophy is where we usually turn when conceptual clarity is required. Many
philosophers have engaged with the phenomena that belief indexes, providing
extended discussions from within particular philosophical systems (e.g. Stich, 1983;
Wittgenstein, 1982). Nevertheless, to the extent that there is a consensus it is an
unsatisfactory one. One well-regarded summary of key philosophical concepts defines
belief as “a propositional attitude” (Honderich, 1995 p.82), described more fully as a
“mental state, representational in character, taking a proposition (true or false) as its
content and involved .. in the direction and control of voluntary behaviour”. A belief,
then, is an attitude toward a statement, claim or proposition. This seems reasonable
until we encounter the definition of ‘attitude’ in the same volume: “any mental state
with propositional content. Attitudes, in this sense, include beliefs, desires, hopes and
wishes” (Honderich, 1995 p.64). This circularity sabotages any apparent consensus,
suggesting that further investigation is desirable.
5
In what follows, I argue that health psychology might usefully adopt a more
theoretically elaborated notion of belief. I propose that belief be reconceptualised as a
structure of socialised feeling, contingently allied to discursive practices and
positions. I show how this conceptualisation accords with relevant uses of belief: non-
trivial claims about causal relations, values, preferences and complex states of affairs,
rather than trivial claims about perception (e.g. ‘I believe this object is a table’). I also
demonstrate how it is consonant with the view that belief is associated with emotion
(Jervis, 2006) and connected to desires, hopes and wishes (Honderich, 1995). The
argument proceeds through a consideration of thinking, then of feeling, then of how
thinking and feeling conjoin in ‘felt thinking’. A further discussion of religiosity
illustrates how belief might be acquired; finally, some possible implications for health
psychology of this reconceptualisation are discussed.
Thinking
Since thinking is already a social process, treating belief solely as a psychological
state may obscure how it is already embedded in “meaning, experience, emotion,
order, individuality, thought, action, identity, sociality, rationality, symbolism and
power” (Day, 2010 p.10). In this regard, Baerveldt & Voestermans (2005) argue that
understandings of belief informed by discursive psychology (DP) are advantageous
because they demonstrate some of the ways that beliefs are already social. DP studies
psychological phenomena as they get enacted in everyday life, taking naturally-
occurring talk as its medium. Rather than treat this talk as a window into a separate
realm of individual cognition, DP treats the talk itself as the primary arena of
psychological activity: for DP, conversation is where cognition gets played out and
can be studied. DP shows how beliefs get constructed in discourse, where they are
tailored to specific contexts and used flexibly to manage dilemmas, construct
identities, impute responsibility, and so on (Edwards, 1997; Potter, 1996). Beliefs are
also rhetorically structured: their form is already argumentative or dialogical,
anticipating possible challenges or refutations (Billig, 1987). DP shows that beliefs
“do not exist in a social vacuum or in individual human minds”, but are “advanced
and responded to in the course of ongoing conversational activity, of claims and [p.4]
counterclaims” (Baerveldt & Voestermans 2005, p.454).
6
However, Baerveldt & Voestermans also argue that DP is incomplete because it
cannot adequately explain how people come to invest or ‘have faith’ in some
constructions of the world rather than others. Discursive explanations of the
variability and situatedness of belief in everyday life come at the cost of obscuring the
ways in which beliefs can compel and organise activity, and of how they persist
across social and material contexts. DP’s constructionist focus on beliefs’ discursive
deployment largely conceals their enduring, normative aspects, and does not
adequately consider their emotional, felt or affective dimensions. Other scholars have
pointed to (versions of) this problem, noting how discursive analyses problematise our
understanding of the way that people valorise some identities rather than others
(Wetherell, 1995), and how the embodied desires that produce the dilemmas analysed
using DP notions of stake and accountability are paradoxically excluded from its
analyses (Willig, 2001). Despite the advantages of DP, then, it shares with cognitive
psychology a tendency to treat belief solely “as something that is claimed or stated,
rather than lived” (Baerveldt & Voestermans, 2005 p.453).
We can begin treating belief as something that is lived by supplementing DP with an
account of how discourse inhabits thought. Billig (1987) described how thinking itself
is argumentative or rhetorical in structure, deriving its content and form from
successive prior conversations. We think largely as we talk: thinking is a discussion
with one’s self that shares features and positions with spoken arguments in everyday
life. What we call belief gets worked up rhetorically in discussions with others that
provide both its core content and its nuances. Consequently we do not have rigid
belief systems (p.254), but can and do finesse belief – in both interaction and thought
– according to occasion, context and situation.
In a similar fashion, Vygotsky (1962) described how we learn to think when two lines
of development – the biological or maturational, and the social – come together, and
that language, in the form of inner speech (the unspoken commentary we experience
on our own and others’ activities), provides many of the structures and tools of
thought. Vygotsky argued that children’s speaking aloud to themselves – which for
Piaget was irrelevant ‘egocentric speech’, produced when children fail to ‘de-centre’
and recognise that such talk is socially inappropriate – is in fact a transitional form,
7
halfway between the conversation it was derived from and the inner speech it will
become.
So inner speech, the most readily apparent component of thought, has social origins.
More than this, inner speech is continuously social because it does not simply run
along according to its own dynamics: it is a ‘boundary phenomenon’ (Shotter, 1993a),
always on the edge between ‘inner’ and ‘outer’, always being interpellated (‘called
out’) by people, activities, objects and events. Like Billig, then, Vygotsky understands
that thinking has a conversational aspect, and is shaped and maintained in social
relations. Consequently, belief tends to reproduce cultural norms, the precepts,
expectations and values of particular times and places (Day & Coleman, 2010).
Simultaneously, within such broad cultural patternings, the belief of any given
individual is produced through the mediation of that person’s particular history of
social relations – with parents, carers, teachers, significant others – with which these
acquired norms get inflected.
However, inner speech is not the whole of thinking: famously, for Vygotsky (1962)
language completes thought: which obviously means that thought itself contains more
than fragments of unverbalised conversation. This content that language ‘completes’
is provided by the lived, phenomenological body: “Thought itself is engendered by
motivation, i.e. by our desires and needs, our interests and emotions. Behind every
thought there is an affective-volitional tendency, which holds the answer to the last
‘why’ in the analysis of thinking.” (Vygotsky, 1962 p.150). In this way, Vygotsky
again directs us back to the embodied, emotional, felt dimension of belief, the
dimension associated with [p.5] values, desires, hopes and wishes. Belief has the form
of fragments of and positions within conversations: socially produced and mediated,
nuanced according to occasion and context, and flexibly deployed at the imprecise
boundary between self and other according to the contingencies of the present
moment. At the same time, belief resides in more than just the present moment and its
words. Belief also resides in emotions and desires, in an implicit, embodied sense of
how the world is or should be. It indexes not just words, but feelings: it works through
conversation, and also through an embodied sense of how conversation should
proceed (Shotter, 1993b). Belief is more than just language, more than just inner
speech: it is also, simultaneously, a structure of socialised feeling.
8
Feeling
The related terms feeling, affect and emotion are notoriously ill-defined: although
they are frequently used interchangeably, they can also index particular meanings or
theoretical allegiances (6, Squire, Treacher, & Radstone, 2007). In the humanities and
social sciences, where an ‘affective turn’ is said to be occurring (e.g. Athanasiou,
Hantzaroula, & Yannakopoulos, 2008; Clough & Halley, 2007), affect is typically
associated with psychoanalysis and primary process activity, or alternately with the
work of Deleuze and Guattari where it is understood as a force or intensity that flows
within and between individuals. In either case it is not necessarily conscious: we
know affect as a feeling when it enters awareness. Emotions, then, are normative
configurations of affect and feeling, enculturated body-brain responses tied to cultural
precepts and local moral orders (Blackman & Cromby, 2007). However, Greco &
Stenner (2008) observe that clear distinctions between emotion and affect cannot
always be sustained, and that when analysts invoke affects it is almost always
emotions to which they refer. Similarly, distinguishing ‘feelings’ as purely
phenomenological can be problematic, particularly in psychology which frequently
identifies feeling states that function just like emotions, despite not appearing in any
taxonomies (Cromby & Harper, 2009).
Here, feelings are defined as “experiences reflective of the momentary state of our
body-brain system as it mediates and enables the situated, relational flow of our being
in the world” (Cromby, 2007 p.99). It is necessary to emphasise the double aspect of
this definition: both these embodied states, and the experience of them, can be
influential. We do not always need to notice an embodied state for it to influence our
being: ingesting alcohol, for example, influences levels of inhibition whether or not
the drinker notices this occurring. Acknowledging this within his influential
neuroscientific account of consciousness, Damasio (1999) proposed that we
distinguish between states of feeling, sets of biological processes that include neural
patterns available to awareness, and states of feeling made conscious - such as when
we realise suddenly “that we feel anxious or uncomfortable, pleased or relaxed, and it
is apparent that the particular state of feeling we know then has not begun on the
moment of knowing” (p.36).
9
Damasio recognises that this distinction challenges everyday uses of feeling that
implicate conscious awareness, but thinks it necessary to capture some of its
complexity. He further acknowledges other embodied states that can be “triggered and
executed non-consciously”: but although in social science these non-conscious states
might be called affects, Damasio calls them emotions. Nevertheless, although
Damasio’s neuroscience sometimes informs the affective turn in social science and
the humanities, there is wariness about such borrowings (Papoulias & Callard, 2010)
and a general concern about ‘neuro-hype’ (Tallis,2009). Moreover, an emphasis on
feeling perhaps belies claims that affect, emotion and feeling ‘follow different logics’
(Massumi, 2002) or that affect can be influential entirely without feeling (Deleuze, 2005).
Pursuing these issues would re-activate definitional debates about affect, emotion and
feeling (Blackman & Venn, 2010) and, ultimately, lead to much wider [p.6] discussions
about both the character of experience and the adequacy of notions of subjectivity (see,
for example, Gregg & Seigworth 2010).
Rather than attempt to resolve these issues, I have explained that feelings are
influential both as embodied states and as the experience of those states, and will
consistently emphasise that notion, in preference to affect or emotion. This is partly
because, unlike these alternatives, feeling is a linguistic prime: all human cultures, so
far as we know, have a word or concept that denotes feeling (Shweder, 2004;
Wierzbicka, 1999). The term also places a psychological emphasis on the experiential
realm, whilst sidestepping some of the uncertainty surrounding definitions of emotion
and affect (Cromby, 2011). It may be an especially valuable focus for health
psychology because, alongside the emotional or affective, feeling includes the
embodied sensations associated with pain, fatigue, medication usage and other
relevant embodied states.
Analytically (although not necessarily experientially) feelings can be separated into
three classes (Cromby, 2007). In order of explanatory convenience, first, there are
emotional feelings, the embodied components of the enculturated complexes of
discourse, intention and action that we call emotions. Second, there are extra-
emotional feelings: the feelings associated with urges such as hunger and thirst, and
with bodily responses such as pain. In psychology these feelings are sometimes
relegated to the status of mere sensations and treated as primitive or insignificant.
10
Here, their import, diversity, complexity, motivational force and relations with other
feelings are recognised, and they are described as ‘extra-emotional’ only because they
typically have emotions bound up with them.
The third class is feelings of knowing: these were first discussed in psychology by
William James, and English speakers sometimes call them ‘gut feelings’ or intuition.
Although their somatic characteristics (location, intensity, duration, character) are
typically vague, these feelings both situate us within social relations and contribute to
processes of reasoning and deliberation. Emphasising their relational function,
(Shotter, 1993b, 1993c) theorises these feelings as ‘knowing of the third kind’ and
describes how they provide embodied, sensual, practical-moral knowledge about the
current relational moment. Emphasising their epistemic function, Johnson (2007)
argues that even the most abstract forms of reasoning, indexed by seemingly content-
free words such as ‘but’, ‘and’ or ‘not’, rely upon feelings of (for example)
obstruction, connection, disjunction or direction. These feelings help organise and
direct our thinking so that when, for instance, we want to say ‘but’ in response to a
claim, the urge to do so is not solely a discursive move. It is simultaneously the verbal
enactment of a felt sense of obstruction to the smooth movement of thought, a subtle
embodied feeling that something is extraneous, missing or wrong.
Like inner speech, all three kinds of feeling are both continuously social (bound up
with the social and material circumstances of the present moment) and socialised
(marked with the impress of prior experience). Feelings of knowing are intrinsically
social, thoroughly bound up with the ongoing flow of social interaction and the
deliberations of the lived moment. Emotional feelings are also social: even where
emotion is defined biologically, as in Ekman’s (1992) notion of hardwired basic
emotions, it is acknowledged that there are culture-specific display rules that regulate
how and when the emotion should be enacted. But experiencing the emotion and then
choosing a display rule before applying it is simply too slow: the spontaneous
performance of an emotion must already include the appropriate display rule. Because
how an emotion feels includes how it is bodily enacted (contrast anger which is tight-
lipped and contained with anger which is openly, blazingly furious), this means that
even on Ekman’s account emotional feelings are socialised. Finally, extra-emotional
feelings are socialised, sometimes explicitly and intentionally, sometimes implicitly
11
and contingently: consider feelings of pain as they get responded to within [p.7]
different parenting regimes, in military or combat training, in practices of sado-
masochism, in hypnosis, or in the work of pain clinics.
Although they can be analytically distinguished in this way, in lived experience these
distinctions are frequently obscure. Feelings typically blend and flow one into the
other, so that “Being overly hungry can lead people to be short-tempered and treat
those around them as irritating and unhelpful; being emotionally aroused can increase
sensitivity to pain; being anxious can lead us to see (project, as psychoanalysts have
it) anxiety in others, and so on.” (Cromby, 2007 p.102). Feelings are easily
communicated, sometimes against our wishes, and this pre-reflective ‘emotional
contagion’ and ‘affective attunement’ leaves us continuously responsive to the
feelings of others (Blackman, 2010). Feelings can therefore mingle in relational
circuits that intensify, extend and generalise them (Scheff, 2003) or that alternately
cause them to become mixed, vacillating or confused (Sullivan & Strongman, 2003).
Many feelings are produced by neural systems inaccessible to consciousness
(Damasio, 1999), but even where their source is available most are relatively subtle,
vague and fleeting (Langer, 1967), so we do not always notice what prompted them.
Moreover, we frequently have reason to disavow our feelings: to avoid hurting others,
manage conflicted or unpleasant situations, or handle information difficult to
acknowledge.
Feelings also have specific qualities that influence how they become meaningful.
First, they are a-representational, known immediately through the flesh (Langer,
1967): consequently, their meaning is not asocial, but neither is it solely decreed by
social convention. Second, because they are embodied and therefore somewhat
ineffable (Shilling, 2003), and because they blend and flow, their meanings are less
rigidly demarcated than those of words. Third, feelings operate at different speeds and
rates of acceleration to words (Connolly, 2002): frequently slower to depart, they are
sometimes slower to arise, although, conversely, they are sometimes all-but
instantaneous: both tendencies giving them the constant potential to be de-
synchronous with other aspects of experience. And fourth, feelings defy willed effort
in ways that words don’t: we can speak much as we wish (discourses legitimate the
12
meanings we articulate rather than the words that carry them), but can’t anywhere
near so readily change how we feel.
These qualities begin to show how the meaning of feelings is necessarily always
interpreted, produced contingently along with other meanings: and hence, how a focus
on feeling does not re-instate the Cartesian subject. Ruthrof (1997) explains how all
meaning is the product of multiple, dynamically contingent sign systems. Many of
these sign systems are linguistic, wherein meaning is decreed by cultural convention,
dependent upon networks of difference and deferral, and organised in regimes of
power. But experience is always embodied experience, so these linguistic signs are
always accompanied by others, generated by embodied systems: haptic, tactile,
thermal, kinaesthetic, olfactory, gustatory, visceral etc. These embodied signs
continuously contribute to meaning, which is consequently always embodied, and
always contingent: embodied signs are already meaningful, but their full meaning
must be interpreted.
Just now, revising the draft paper on my desktop feels absolutely impossible. I might
understand this feeling as despair engendered by the sheer quantity of other tasks I
must complete; I might understand it as resentment of the reviewers’ criticisms,
tinged with anxiety that I might not be able to address them; or perhaps, in a
contemporary version of accidie (Harre & Finlay-Jones, 1986), I might understand it
as boredom and self-disgust. It is nevertheless vital to realise that embodied signs are
not meaningful only by virtue of their interpretation, because their intensities,
textures, valences and affordances resist some meanings and favour others: regardless
of any other signs it coincides with, my feeling about this paper is not one of joy.
[p.8] Felt thinking
Thinking consists of multiple embodied processes, including the sensory processes of
seeing, hearing, smelling and tasting, and the complex processes of memory that help
render experience coherent and sensible (Middleton & Brown, 2005). The interplay of
discourse and feeling is therefore decisively not the whole of thinking. Nevertheless,
these other processes and contents are always enchained within flows of socialised
feeling, always contingently associated with acquired fragments of language. Thought
is continuously constituted from a flux of embodied valences, textures, affordances
13
and intensities: an ongoing, corporeal sense of our own being and place in the world
that shades thinking with value and desire. Simultaneously, it is constituted from
acquired fragments of conversation that ‘complete’ these embodied feelings, allowing
us to ‘fix’ or represent them to ourselves and others, to cut into and ‘grasp’ this flux of
embodied feeling, rendering its moments more thoroughly meaningful. In the
embodied processes of thinking, feelings are always interpreted - but these
interpretations are already influenced by the feelings they strive to understand, and
may quickly modify the feelings that prompted them. Consequently, all thinking is felt
thinking (Ratner, 2000): no reasoning is entirely or simply ‘logical’, and every
rationality implicates processes of feeling (Damasio, 1994; Jagger, 1989; Johnson,
2007).
Despite the strengths of his analysis, Vygotsky consistently maintains a somewhat
dualistic separation between the biological and the social (van der Veer & Valsiner,
1991). This produces a relatively static view of the relation between language and
feeling: language ‘completes’ thought (which is felt, motivational, affective,
volitional) but for Vygotsky the two remain largely encapsulated within distinct lines
of development. Contemporary evidence, for example from neuroscience, challenges
this separation: whilst the systems that enable language are for the most part not those
that enable feeling (Damasio, 2003), social influence gets built into the brain from the
earliest moments of life (Rose, 1997; Schore, 2001). Scholarship associated with the
affective turn similarly emphasises how language and feeling are jointly enrolled
within social practice, with neither operating in a foundational manner (Gregg &
Seigworth, 2010).Vygotsky’s tendency toward dualism occludes the ways that
feelings are socialised, confines sociality primarily to language, and makes it difficult
to explain how feelings and (inner) speech are so frequently intertwined.
If this separation is not presumed, Vygotsky’s analysis can be elaborated in a way that
has implications for the notion of belief. We can extend his account of the acquisition
of inner speech, in a course from conversation through outer speech, by adding
another stage where inner speech gets converted into a structure of feeling. As
fragments of conversation and elements of feeling get rehearsed together they
interpenetrate and become interchangeable: corporeal analogues of meanings, initially
borne of conversation, can then get enacted through the body. An example: although I
14
can articulate reasoned critiques of current UK economic and social policy, I can also
simply experience feelings of anger and mistrust toward the voices of government
ministers, feelings that may arise even before I attend to what they say. Here,
discourse has inculcated a structure of feeling, but this transformational potential is
bi-directional: for example, my feeling of unease on first meeting someone can
subsequently get ‘fixed’ and interpreted, in inner speech, conversation, or both: this
stabilises and elaborates my felt sense of disquiet, letting me represent it to myself
and to others.
The value of recognising this potential for further transformation becomes clear if we
return to Johnson’s (2007) examples, because they suggest how we are able to
appropriately feel a sense of obstruction, or “but’ – rather than say a sense of
connection, or “and” - in response to a proposition. Johnson (p.151) acknowledges
that his account does not adequately incorporate social influence. This means that
although he successfully identifies the capacity [p.9] to experience different feelings
in response to an argument, he cannot explain how we might come to have one feeling
rather than the other, nor explain how the ‘correct’ feeling might ever arise.
Recognising that a feeling of “but” is a transmuted version of a prior discursive
position, a structure of socialised feeling that we previously knew through discourse
and narrative, supplies this explanation. With particular respect to belief, it also helps
show how belief can be just as much about desires, wishes and dreams as about
reasoning, evidence and logic: in its everyday living, belief gets enacted primarily as
feeling or primarily as argument, depending on circumstance.
Believing
Belief does not uniquely combine discourse and feeling: it is a particular form of felt
thinking, characterised by durability and personal significance. Since the origins of
these qualities are in the social and material circumstances where belief becomes
relevant, we can elucidate belief by examining those circumstances. To illustrate I
will again consider religiosity, specifically Western Christianity. The account is very
general, intended to illustrate typical processes of belief formation rather than make
specific, grounded claims. It follows others (e.g. Durkheim, 1995 [1912]; Marshall,
2002) that focus on ritual and discursive practice, and echoes Smith (2007) in
emphasising the role of feeling in the workings of Christian belief.
15
Speaking very broadly, then, Western Christian belief consists of structures of feeling
and associated discourses and narratives developed through various ritualised
practices, including: contemplation, prayer, confession, penitence, abstinence,
meditation, clapping, dancing, drumming, swaying, singing and chanting. In varying
degrees, these practices discursively invoke hierarchy, ritual and tradition, and may
include reading, clothing, costumes, incense, touch, icons, statues, engravings, films,
food, drink, immersion in water, and other multi-sensorial components. They may
also include more mundane (for Durkheim, profane) practices associated with
“‘potluck dinners’, bingo nights, sports, youth groups, reading groups and ‘gossip
with friends’” (Smith, 2007 p.176). They are enacted in variable settings (home,
classroom, church, church hall), some with dedicated architectures and iconographies;
associated with both momentous life events (births, deaths, marriages) and with the
regular passage of time (Christmas, Easter); and involve variously-sized collectives of
differentially-significant others.
Whilst these multiple, interlocking practices necessarily recruit many different
feelings, in the Western Christian tradition particular emphasis is often placed on
feelings of happiness, love, pride (Tsai, Miao, & Seppala, 2007) and guilt (Albertsen,
O'Connor, & Berry, 2006). Western Christian practices simultaneously marshal
feelings of knowing, such as being accepted by others; having confidence in the
future; and having arcane or privileged knowledge about reality. They may also
recruit extra-emotional feelings of hunger, thirst, satiation, bodily discomfort, touch
and sexual desire, and emotional feelings of loneliness, affection, companionship and
trust. All these feelings get stabilised by their repetition within rituals imbued with
tradition, authority and power; by their relational interpellation in complex mixtures
(Scheff, 2003); by effortful engagement with the practices that generated them; by the
frequently-rhythmic character of those practices; and through the mediation of
embodied and social processes including heightened arousal, entrainment, social
facilitation and de-individuation (Marshall, 2002).
Thus, Western Christian practices are mediated and organised by multiple objects,
discourses and narratives, with which, through repetition and rehearsal, the feeling
structures they interpellate get contingently associated. Moreover, these associations
16
are not worked up in exclusively Christian sites since they may also be functional
elsewhere - for example, in everyday moral enactments (Smith, 2007). Consequently,
they produce somewhat heterogeneous structures of feeling that nevertheless [p.10]
imbue discourse and narrative with degrees of personal significance; feeling structures
that valorise some discursive constructions and narrative positions rather than others,
albeit in variable ways. They may associate particular structures of feeling with
behavioural injunctions (‘don’t eat fish on Fridays’, ‘attend church on Sundays’); with
discursive regimes of ethics and value that reproduce wider societal structures of
difference and power (adultery is a sin; charity is a virtue); and with more abstract or
universal discursive constructions (‘god’,’ divinity’, ‘soul’).
This account in no sense denies the significance of Western Christian belief, which is
not diminished by a better understanding of its ontogenesis. It shows that what we call
beliefs are not singular cognitive entities: they are complex, variegated habits of felt
thinking: far more contingent upon the flows of social practice, and far more rooted in
our bodies, than psychology usually allows (although see Barsalou, Barbey, Kyle-
Simmons, & Santos, 2005). Belief acquired in this way will persist across social
settings where different discursive registers may be relevant, and have motivational
force to compel or guide activity. It is simultaneously both discursive and felt or
affective: this will become particularly obvious if it is challenged. Indeed, belief
constituted thus may be intrinsically resistant to change: recall that feelings defy
willed effort, operate on timescales disjunctive with other aspects of experience, and
convey a-representational, embodied meanings. These characteristics may make all
non-trivial belief resistant to ‘rational’ argument, because the structure of feeling
interpellated by attempts to disprove belief by logic can generate pre-reflective
incentives to reject what is said, almost regardless of content.
Implications
This account has numerous implications for health psychology. Research into R/S and
health has assessed belief using behavioural indicators (e.g. church attendance) and
questionnaires. This account demonstrates that both are restricted because what are
superficially ‘the same’ beliefs and activities actually get constituted differently, and
have different meanings, according to the particular history of practices, the specific
discursive positions and structures of feeling, that they implicate. Potter & Wetherell
17
(1987) observed that forced-choice questionnaires artificially suppress response
variation, eliminating ‘it depends’ and decontextualising responses onto a grid of pre-
defined possibilities. With respect to religiosity questionnaires, this restriction extends
beyond an inability to reflect nuance and context. Whilst different respondents might
all ‘strongly agree’ that ‘God helps me to lead a better life’ (Francis Scale of Attitudes
Towards Christianity - Francis, 1993) the lived (rather than claimed or stated)
meaning of their agreement could nevertheless differ, in accord with the different
structures of feeling each respondent associates with concepts such as ‘God’. Within
spirituality, rituals are typically less formalised so variation will tend to be
correspondingly greater; conversely, even those discourses and structures of feeling
enacted during regular church attendances will differ - from person to person,
occasion to occasion. We cannot be surprised, therefore, that research using either
behavioural indicators or questionnaire indices of R/S generates somewhat variable
findings.
To the extent that reviews (e.g. Chida et al., 2009) suggest a small effect of R/S that
cannot be accounted for by differential patterns of diet, alcohol consumption, social
support and other variables, this account suggests explanations. First, the feelings
integral to belief may have direct consequences: emotions, for example, are
implicated in coronary and cardio-vascular diseases (Everson-Rose & Lewis, 2005;
Ruiz, Hutchinson, & Terrill, 2008) and can modulate immune function (Kiecolt-
Glaser, McGuire, Robles, & Glaser, 2002). Religiosity is associated with acceptance
of chronic pain (Gerberhagen, Trojan, Kuhn, Limroth, & Bewermeyer, 2008), and
pain is a socialised feeling with an emotional component (Rhudy, Williams, McCabe,
Rambo, & Russell, 2006). Park (2007) proposes that the associations between R/S and
health might be interrogated from a ‘meaning systems’ perspective where different
aspects of belief map [p.11] onto specific health practices and pathways: this account
extends her suggestion by showing how the meanings enacted in belief already
implicate embodied processes. Second, many measures used in this research already
implicate feeling. Religiosity is positively associated with health-related quality of life
(Gerberhagen et al., 2008), a measure that typically includes ‘emotional’ and
‘subjective’ wellbeing subscales. Evidence suggests that positive affect is associated
with good health (although this research suffers from similar problems to research
into R/S - Pressman & Cohen, 2005): to the extent that the structures of feeling
18
interpellated by religious/spiritual belief are also hedonically positive, this might be
largely the same effect viewed from a different perspective.
This account also has wider implications, with regard to the typical conceptualisation
of belief - in the TPB, and elsewhere - as a discrete, cognitive variable. Belief is
discursive before it is cognitive: its cognitive aspects are secondary to the social,
discursive practices that engendered them. Moreover, this influence does not end once
belief is acquired: belief remains continuously social, an element of the ‘boundary
phenomena’ that (Shotter, 1993a) identifies. Far from being rigidly demarcated and
cognitive, belief is enacted and influential in ways that continuously reflect the
contingency, multiplicity and variability of everyday life - even when research
procedures artificially minimise this variability (i.e. experiments are social situations).
Mielewczyk & Willig (2007) review evidence indicating that this causes particular
problems for two of the TPB’s three component variables, subjective norms (SN) and
perceived behavioural control (PBC). Even when measures of SN and PBC are
supplemented with or replaced by others, to increase their explanatory power, studies
using the TPB typically account for no more than 50% of the variance in intended
outcomes. Mielewczyk & Willig (2007) suggest that this is in part because the ‘health
behaviours’ targeted by social cognition models do not actually exist as discrete
categories of activity: the meaning of (for example) not using a condom is
relationship and circumstance dependent, and not calculable simply in terms of health
risks. This account complements their analysis by showing how the beliefs within the
core components of the TPB (and other social cognition models) are similarly open to
relational and circumstantial influences, upon which their salience and enacted
character is continuously predicated.
Simultaneously, this account shows that to treat belief as solely discursive or
cognitive would indeed be to erroneously reduce it to something claimed or stated,
rather than something lived. Alongside its discursive-cognitive aspect belief is a
structure of feeling that enacts and reproduces personally held, socially-obtained
values. Belief therefore tends to reproduce wider social divisions and their associated
power relations; recognising this allows us to investigate its character in relation to
socio-economic, cultural and political conditions (Marks, 2008). We might then
understand ‘health beliefs’ and their relation to ‘health behaviour’ as moments within
19
ongoing social-embodied processes, mediated by the organisation of bodies,
discourses, resources, objects and places. Models like the TPB instantiate a
psychology that isolates and abstracts individuals from the concrete particulars of
their lives, makes subjectivity into a problem rather than a resource, and generates
findings at such a level of generality that their practical utility is severely restricted
(Tolman, 1994). The concepts for a psychology that recognises how health and illness
are socially, materially and bodily constituted within experience and subjectivity are
already available (Brown & Stenner, 2009), and we have appropriate research
methods (Cromby, 2011): it is time, perhaps, to deploy such resources more widely.
Conclusion
Belief arises when social practice works up structures of feeling in contingent
association with discourse and narrative. Consequently, [p.12] beliefs are enduring,
yet variable and flexible; they have largely predictable content, yet are contingent
upon the actions and talk of others; they are social, yet can be endowed with deep
personal significance. Believing is not merely information-processing activity, and
belief is not an individual cognitive entity. Belief is the somewhat contingent, socially
co-constituted outcome of repeated articulations between activities, discourses,
narratives and feelings. In a second order process, a work of social construction
parasitic upon this co-constitution, belief is also a category of both lay and academic
psychological discourse. Health psychology has little to lose, and something to gain,
by engaging with these processes of co-constitution.
6, P., Squire, C., Treacher, A., & Radstone, S. (2007). Introduction. In P. 6, S.
Radstone, C. Squire & A. Treacher (Eds.), Public Emotions (pp. 1-34).
London: Palgrave Macmillan.
Albertsen, E., O'Connor, L., & Berry, J. (2006). Religion and interpersonal guilt:
variations across ethnicity and spirituality. Mental Health, Religion and
Culture, 9(1), 67-84.
Athanasiou, A., Hantzaroula, P., & Yannakopoulos, Y. (2008). Towards a New
Epistemology: the "affective turn". Historein, 8, 5-16.
Aukst-Margetic, B., & Margetic, B. (2005). Religiosity and Health Outcomes: review
of literature. Collegium Anthropologicum, 29(1), 365-371.
Azjen, I. (1985). From intentions to actions: a theory of planned behaviour. In J. Kuhl
& J. Beckman (Eds.), Action Control: from cognition to behaviour (pp. 11-
39). Heidelberg: Springer-Verlag.
20
Baerveldt, C., & Voestermans, P. (2005). Culture, Emotion and the Normative
Structure of Reality. Theory and Psychology, 15(4), 449-474.
Barsalou, L., Barbey, A., Kyle-Simmons, W., & Santos, A. (2005). Embodiment in
Religious Knowledge. Journal of Cognition and Culture, 5, 14-57.
Billig, M. (1987). Arguing and Thinking: a rhetorical approach to social psychology.
Cambridge: Cambridge University Press.
Blackman, L. (2010). Embodying Affect: voice-hearing, telepathy, suggestion and
modelling the non-conscious. Body and Society, 16(1), 163-192.
Blackman, L., & Cromby, J. (2007). Affect and Feeling. International Journal of
Critical Psychology, 21, 5-22.
Blackman, L., & Venn, C. (2010). Affect. Body and Society, 16(1), 7-28.
Brown, S. D., & Stenner, P. (2009). Psychology Without Foundations: history,
philosophy and psychosocial theory. London: Sage Publications.
Chida, Y., Steptoe, A., & Powell, L. (2009). Religiosity/Spirituality and Mortality.
Psychotherapy and Psychosomatics, 78, 81-90.
Clough, P., & Halley, J. (Eds.). (2007). The Affective Turn: theorising the social.
Durham NC: Duke University Press.
Connolly, W. (2002). Neuropolitics: thinking, culture, speed. Minneapolis: University
of Minnesota Press.
Cromby, J. (2007). Toward a Psychology of Feeling [Electronic Version].
International Journal of Critical Psychology, 21, 94-118 from http://www-
staff.lboro.ac.uk/~hujc4/.
Cromby, J. (2011). Affecting Qualitative Health Psychology. Health Psychology
Review, 5(1), 79-96.
Cromby, J., & Harper, D. (2009). Paranoia: a social account. Theory and Psychology,
19(3), 335-361.
Damasio, A. R. (1994). Descartes Error: emotion, reason and the human brain.
London: Picador.
Damasio, A. R. (1999). The Feeling of What Happens: body, emotion and the making
of consciousness. London: William Heinemann.
Damasio, A. R. (2003). Looking for Spinoza: joy, sorrow and the feeling brain.
Orlando: Harvest.
Darker, C., & French, D. (2009). What sense do people make of a theory of planned
behaviour questionnaire? A think-aloud study. Journal of Health Psychology,
14, 861-871.
Day, A. (2010). Propositions and performativity: relocating belief to the social.
Culture and Religion, 11(1), 9-30.
Day, A., & Coleman, S. (2010). Broadening Boundaries: creating inter-disciplinary
dialogue on belief. Culture and Religion, 11(1), 1-8.
Deleuze, G. (2005). Francis Bacon: the logic of sensation (D. W.Smith, Trans.).
Minneapolis: University of Minnesota Press.
Durkheim, E. (1995 [1912]). The Elementary Forms of Religious Life. New York:
Free Press.
Edwards, D. (1997). Discourse and Cognition. London: Sage Publications.
Ekman, P. (1992). Are There Basic Emotions? Psychological Review, 99(3), 550-553.
Everson-Rose, S., & Lewis, T. (2005). Psychosocial factors and cardiovascular
diseases. Annual Review of Public Health, 26, 469-500.
Francis, L. J. (1993). Reliability and validity of a short scale of attitude towards
Christianity among adults. Psychological Reports, 72, 615-618.
21
Gerberhagen, K., Trojan, M., Kuhn, J., Limroth, V., & Bewermeyer, H. (2008).
Significance of health-related quality of life and religiosity for the acceptance
of chronic pain. Schmerz, 22(5), 586-593.
Greco, M., & Stenner, P. (2008). Introduction: Emotion and Social Science. In M.
Greco & P. Stenner (Eds.), Emotions: a social science reader (pp. 1-21).
London: Routledge.
Gregg, M., & Seigworth, G. (Eds.). (2010). The Affect Theory Reader. Durham and
London: Duke University Press.
Hagger, M., & Chatzisarantis, N. (2009). Integrating the theory of planned behaviour
and self-determination theory in health behaviour: A meta-analysis. British
Journal of Health Psychology, 14(2), 275-302.
Harre, R., & Finlay-Jones, R. (1986). Emotion Talk across Times. In R. Harre (Ed.),
The Social Construction of Emotions (pp. 220-233). Oxford: Basil Blackwell.
Honderich, T. (Ed.). (1995). The Oxford Companion to Philosophy. Oxford: Oxford
University Press.
Jagger, A. (1989). Love and knowledge: emotion in feminist epistemology. Inquiry,
32, 151-172.
Jervis, R. (2006). Understanding Beliefs. Political Psychology, 27(5), 641-663.
Johnson, M. (2007). The Meaning of the Body: aesthetics of human understanding.
Chicago: University of Chicago Press.
Karademas, E. (2010). Illness cognitions as a pathway between religiousness and
subjective health in chronic cardiac patients. Journal of Health Psychology,
15, 239-247.
Kiecolt-Glaser, J., McGuire, L., Robles, T., & Glaser, R. (2002). Psychoneuro-
immunology: Psychological influences on immune function and health.
Journal of Consulting and Clinical Psychology, 70, 537-547.
Koenig, H. G., Cohen, H., George, L., Hays, J., Larson, D. B., & Blazer, D. (1997).
Attendance at Religious Services, Interleukin-6, and Other Biological
Parameters of Immune Function in Older Adults. International Journal of
Psychiatry in Medicine, 27, 233-250.
Koenig, H. G., McCullough, M. E., & Larson, D. B. (2000). Handbook of Religion
and Health. New York: Oxford University Press.
Langer, S. (1967). Mind: an essay on human feeling (Vol. 1). Baltimore: The Johns
Hopkins University Press.
Lissoni, P., Cangemi, P., Pirato, D., Grazia-Roselli, M., Rovelli, F., Brivio, F., et al.
(2001). A review on cancer-psychospiritual status interactions.
Neuroendocrinology Lettters, 22, 175-180.
Marks, D. (2008). The quest for meaningful theory in health psychology. Journal of
Health Psychology, 13, 977-981.
Marshall, D. (2002). Behaviour, Belief and Belonging: a theory of ritual practice.
Sociological Theory, 20(3), 360-380.
Massumi, B. (2002). Parables for the Virtual: movement, affect, sensation. Durham:
Duke University Press.
McCullough, M. E., Hoyt, W., Larson, D. B., Koenig, H. G., & Thoresen, C. (2000).
Religious Involvement and Mortality: a meta-analytic review. Health
Psychology, 19(3), 211-222.
Meisenhelder, J., & Chandler, E. (2002). Spirituality and health outcomes in the
elderly. Journal of Religion and Health, 41, 243-252.
Middleton, D., & Brown, S. D. (2005). The Social Psychology of Experience: studies
in remembering and forgetting. London: Sage Publications.
22
Mielewczyk, F., & Willig, C. (2007). Old Clothes an an Older Look: the case for a
radical makeover in health behaviour. Theory and Psychology, 17(6), 811-837.
Papoulias, C., & Callard, F. (2010). Biology's Gift: interrogating the turn to affect.
Body and Society, 16(1), 29-56.
Park, C. (2007). Religiousness/Spirituality and Health: a meaning-systems
perspective. Journal of Behavioural Medicine, 30, 319-328.
Potter, J. (1996). Representing Reality:discourse, rhetoric and social construction.
London: Sage Publications.
Potter, J., & Wetherell, M. (1987). Discourse and Social Psychology: beyond
attitudes and behaviour. London: Sage Publications.
Pressman, S., & Cohen, S. (2005). Does Positive Affect Influence Health?
Psychological Bulletin, 131(5), 925-971.
Ratner, C. (2000). A cultural-psychological analysis of emotions. Culture and
Psychology, 6, 5-39.
Rew, L., & Wong, Y. (2006). A systematic review of associations between
religiosity/spirituality and adolescent health attitudes and behaviours. Journal
of Adolescent Health, 38, 433-442.
Rhudy, J., Williams, A., McCabe, K., Rambo, P., & Russell, J. (2006). Emotional
modulation of spinal nociception and pain: The impact of predictable noxious
stimulation. Pain, 126(1-3), 221-233.
Rose, S. (1997). Lifelines: life beyond the gene. Oxford: Oxford University Press.
Ruiz, J., Hutchinson, J., & Terrill, A. (2008). For better and worse: Social influences
on coronary heart disease risk. Social and Personality Psychology Compass,
2(3), 1400-1414.
Ruthrof, H. (1997). Semantics and the Body. Toronto: University of Toronto Press.
Scheff, T. (2003). Male emotions/relations and violence: a case study. Human
Relations, 56(6), 727-749.
Schore, A. (2001). The effects of early relational trauma on right brain development,
affect regulation, and infant mental health. Infant Mental Health Journal,
22(1), 201-269.
Seeman, T., Dubin, L., & Seeman, M. (2003). Religiosity/Spirituality and Health: a
critical review of the evidence for biological pathways. American
Psychologist, 58(1), 53-63.
Shilling, C. (2003). The Body and Social Theory (2nd ed.). London: Sage
Publications.
Shotter, J. (1993a). Bakhtin and Vygotsky: internalisation as a boundary phenomenon.
New Ideas in Psychology, 11, 379-390.
Shotter, J. (1993b). Conversational Realities: constructing life through language.
London: Sage Publications.
Shotter, J. (1993c). Cultural Politics of Everyday Life. Buckingham: Open University
Press.
Shweder, R. A. (2004). Deconstructing the emotions for the sake of comparative
research. In A. Manstead, N. Frijda & A. Fischer (Eds.), Feelings and
Emotions: the Amsterdam Symposium. Cambridge: Cambridge University
Press.
Sloan, R., & Bagiella, E. (2002). Claims about religious involvement and health
outcome. Annals of Behavioural Medicine, 24, 14-21.
Sloan, R., Bagiella, E., & Powell, T. (1999). Religion, Spirituality and Medicine. The
Lancet, 353, 664-667.
23
Smith, C. (2007). Why Christianity Works: an emotions-focused phenomenological
account. Sociology of Religion, 68(2), 165-178.
Stich, S. (1983). From Folk Psychology to Cognitive Science: the case against belief.
Cambridge, Mass.: MIT Press.
Sullivan, G., & Strongman, K. (2003). Vacillating and Mixed Emotions: A
Conceptual-Discursive Perspective on Contemporary Emotion and Cognitive
Appraisal Theories Through Examples of Pride. Journal for the Theory of
Social Behaviour, 33, 203-226.
Tallis, R. (2009). Neurotrash. New Humanist, 124(6). Retrieved 11/8/2011 from
http://newhumanist.org.uk/2172/neurotrash
Tolman, C. (1994). Psychology, Society, Subjectivity: an introduction to German
Critical Psychology. London: Routledge.
Tsai, J., Miao, F., & Seppala, E. (2007). Good Feelings in Christianity and Buddhism:
religious differences in ideal affect. Personality and Social Psychology
Bulletin, 33(3), 409-421.
van der Veer, R., & Valsiner, J. (1991). Understanding Vygotsky: a quest for
synthesis. Oxford: Blackwells.
Vygotsky, L. S. (1962). Thought and Language (E. Hanfmann & G. Vakar, Trans.).
Cambridge, Mass.: M.I.T. Press.
Wetherell, M. (1995). Romantic Discourse and Feminist Analysis: interrogating
investment, power and desire. In C. Kitzinger & S. Wilkinson (Eds.),
Feminism and Discourse: psychological perspectives (pp. 128-144). London:
Sage Publications.
Wierzbicka, A. (1999). Emotional Universals. Language Design, 2, 23-69.
Willig, C. (2001). Introducing Qualitative Research in Psychology. Buckingham:
Open University Press.
Wittgenstein, L. (1982). Last Writings on the Philosophy of Psychology vol 1. Oxford:
Blackwells.
