8
Religious Conversion and
Modes of Religiosity
Ilkka Pyysiäinen
Whitehouse's theory of the modes of religiosity is a good analytical tool. But
it is also easy to misunderstand the point. When I, for example, use the theoiy
in my teaching, a frequent comment is that no known religion seems to
be imagistic or doctrinal; the features rather mix and mingle. It is then difficult
to tiy to argue that the modes theoiy actually is meant only as an analytical
distinction, not a typology. It may thus be helpful to forget the labels
of "imagistic" and "doctrinal" altogether for a while and only focus on the
way the differing contents of the twelve modal variables (Whitehouse 2002b)
may "excite" or "inhibit" each other. We must remember that "imagistic" and
"doctrinal" only refer to ideal types, that is, to purely abstract clusters of variable
contents. A single ritual, belief, or any other single feature cannot be
imagistic or doctrinal, because being imagistic or doctrinal depends on all 12
variables simultaneously. It is only a larger combination of features that can
be more or less imagistic or doctrinal. But also these are seldom simply imagistic
or doctrinal.
Let us substitute the letters from a to l for the imagistic contents of the 12
variables and the letters from m to x for the doctrinal contents of the same variables
(the actual contents are discussed below one by one). The theoiy now
seems to presuppose that in practice {a . . . l} are not the singly necessary and
jointly sufficient characteristics of imagistic religiosity, and that {m . . . x} are not
the singly necessary and jointly sufficient characteristics of doctrinal religiosity.
As the distinction is only an analytical one, we can always find some of the
{a . . . l} in otherwise doctrinal constellations and some of the {m . . . x} in imagistic
ones.
The question now follows how many of the features {a . . . l} can we find
in an otherwise doctrinal constellation and still be able to recognize/regard
149
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
ISO Ilkka Pyysi]inen
it as doctrinal? Or, how many of the features {m . . . x} can we find in an otherwise
imagistic constellation and still be able to recognize/regard it as imagistic?
Also, if the contents of the variables in the sets of {a . . . l} and of [m
. . . x} mix in practice, what might be the "evolutionarily stable strategy/' that
is, the most fit combination of {a . . . .x}? In other words, which combination
of features from the two sets results in a religious tradition that has the best
potential for cultural survival? This would be the real attractor position; pure
imagism and pure doctrinality are not attractor positions since they do not
canalize actual religiosity. Pure imagism and pure doctrinality, rather, are
ideal types.
Still another question is how many of the features {a . . . .x} must be present
in order for the transmission to count as modal (assuming that also nonmodal
transmission exists)? Or, perhaps, the question cannot be framed in
quantitative terms and we should rather look for some specific contents of
the twelve variables that are necessary for a transmission process to count as
modal? It also seems that the distinction between modal versus nonmodal
transmission is more important in Whitehouse's theory than the distinction
between imagistic and doctrinal itself (see Whitehouse 2000, 2002a, 2000b).
C O N V E R S I O N
I shall here discuss the case of religious conversion in the light of the above
questions. I mean by religious conversion a sudden change in one's religious
convictions, behaviors, and affiliations. These may then also have repercussions
other than for religious beliefs and behaviors. "Sudden" here means
that the person in question is able to point out a specific moment when
"everything changed" in his or her life (see Miller and C'de Baca 2001). Even
if the conversion experience is a doctrinal construct, one can point out a specific
moment in his or her life when the import of the doctrine supposedly
was personally realized. To the extent that it is a real event, its suddenness
does not exclude the possibility that the change may have been preceded by
a longer period of restlessness, doubt, seeking, and so forth. Yet I am primarily
interested in the moment at which all this culminates in an experience
of a dramatic change (real or imagined). By "change" I mean the fact that
some beliefs, concepts, and practices that one previously either has ignored
or actively opposed suddenly begin to seem both true and relevant (see
Miller and C'de Baca 2001). "Relevant" here means that one is able to perceive
some positive connection between the target information and the
knowledge one already possesses; one is able to combine it with one's existing
knowledge and to form new premises on that basis. This, then, allows
for new inferences (Sperber and Wilson 1986, 46-50, 118-71). By "religious"
I refer to anything that involves counterintuitive representations, is an object
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 151
of belieť, is shared by a group, and is used in management of life (see
Pyysiäinen 2001, 9-23; 2002a; Boyer 2001, 51-91, Atran 2002). A conviction
is a metarepresentational attitude that takes certain concepts and beliefs to
be more important than others as premises in reasoning.
u
Conversion, ,
is a predominantly Christian and Protestant theological concept,
although its application has been extended to other religious traditions
as well (see Taves 1999; Malony and Southard 1992). It has been studied from
varying angles, yet the phenomenon seems to escape definition as well as explanation.
The actual process of conversion, as well as the effects of conversion
processes on the self, have so far not been properly specified; neither
have different types of conversion processes been adequately differentiated
(Zinnbauer and Pargament 1998, 161). Conversion has been explored mainly
from sociological, psychological, and theological perspectives, identifying the
phenomenon mainly by its effects, that is, transformations in the contents of
beliefs, behaviors, and religious affiliations. The neurocognitive mechanisms
of conversion have been by and large ignored in research (see Brown and
Caetano 1992; Miller and C d e Baca 2001, xi-xii, 3-7, 156). As lar as I know,
there is currently no evidence indicating the existence of a separate neurocognitive
(or socioneurocognitive) mechanism responsible for the sudden
transformations taking place in a conversion (cf. Miller and Cde Baca 2001).
The concept of "conversion' has mostly received its coherence from underlying
theological assumptions (see, e.g., Rambo 1993, 1998).
Atran (2002, 165-169) summarizes some of the work on religious conversion,
arguing that linked feelings of guilt, anxiety, and social alienation are
good predictors of a religious conversion; in large-scale industrial societies,
socially marginal individuals, especially adolescents, whose stress levels are
high and levels of self-esteem low are likely to pursue a religious identity.
Kirkpatrick (1997, 1998, 1999), for example, argues on the basis of empirical
studies that insecure-anxious and insecure-avoidant adult social-attachment
styles were more likely predictors of finding a new relationship with God
than a secure attachment style. Women who were initially anxious were significantly
more likely than either avoidant or secure women to report having
had a "religious experience or conversion" during the four-year period that
the study covered. (The attachment-theoretical approach, however, is elsewhere
[71-78] criticized by Atran.)
Pargament (2002a) argues from a coping perspective that religious converts
have experienced great difficulty in their lives, so much so that existence
itself has become a problem. In realizing that something is wrong, the
would-be convert sees his or her "old self" and way of life as inadequate
and regards a radical change that would transform his or her entire life as
the only possible solution (see Miller and C d e Baca 2001). Zinnbauer and
Pargament (1998) found that both a sudden spiritual conversion and a gradual
conversion markedly changed the self-definition of the converts. When
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
152 Ilkka Pyysi]inen
compared to a group of religious subjects who had not experienced religious
change, the sudden-conversion group showed more preconversion
perceived stress, a greater sense of personal inadequacy and limitation prior
to the conversion, and also greater pre-post improvement, together with a
greater increase in postconversion spiritual experiences. Paloutzian and
colleagues (1999), however, suggest that spiritually transforming experiences
can result in profound changes in life mostly at the level of goals, feelings,
attitudes, and behaviors. Also, such self-defining personality functions
as identity and meaning of life change dramatically after a spiritual experience.
Yet the elemental functions of personality (whether one is open to
experience, conscientious, extroverted, agreeable, or neurotic) change only
minimally. These are also the dimensions of personality that are strongly
heritable; in a typical population, 40 to 50 percent of the variation in them
seems to be tied to genes (Pinker 2002, 50).
Atran (2002, 1*76-79) suggests a mechanism based on the finding that
posttraumatic stress disorder (PTSD) is accompanied by heightened amygdala
activation in response to fearful stimuli. He speculates that the overly
stimulated amygdala goes into undirected hyperactivity, being unable to
process the emotional significance of individual stimuli. The hypothalamus
then receives a confounding flood of information and relays it to the autonomic
nervous system; this then provokes increased discharges in the sympathetic
and parasympathetic nervous systems (see also Pyysiainen 2001,
114-15). PTSD is characterized by uncontrollable arousal mediated by
adrenergic clysregulation. However, adrenergic blockers, such as beta
blockers and possibly antidepressants, can reduce terror and arousal and interrupt
the neuronal imprinting leading to long-term symptoms, if administered
in the first hours after the terrifying incident. Patients also may benefit
from cognitive-behavioral therapy (see MacReady 2002). Atran then
argues for the possibility that heightened expressions of religiosity can serve
a similar blocking function. This is supported by Atran and Norenzayans (in
press) result that the strength of belief in God and in the efficacy of supernatural
intervention are reliably stronger after exposure to death primes
than to a neutral or religious prime.
Atran*s (2002, 169) tentative conclusion about conversion (based on a
somewhat different sample of studies than here) is that "the more traditionally
and continuously religious the person, the less likely to suffer depression
and anxiety in the long run." However, religion provides not only comfort
but also is a source of fear. Many religious concepts, images, and beliefs can
be interpreted either as fearful or as comforting, depending on howT
one relates
to them. Gods, for example, have counterintuitive potency to punish
and to reward alike; the course of action is determined by the nature of the
relationship. (See Pyysiainen 2001, 136-37.) Thus religion can both increase
fears and anxieties and resolve them. The mechanisms work in both ways: a
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 153
very difficult life situation may trigger the fear of extinction and lead to a religious
solution, or religious stimuli may trigger the emotions of fear, anxiety,
and hope; that is, religion may be both "the sickness and the cure" (see Thorson
1998; Guthrie 1993).
This entails that the emotions of fear and hope are closely allied both cognitively
and religiously, just as Atran (2002, 68) argues. They may be cognitively
(but not physiologically) composed of identical elements, differing
only in valence. There is, for example, some empirical evidence to support
the hypothesis that an increase in tearfulness is accompanied by heightened
faith in God as well as hope for a providential outcome in a difficult situation.
Hope is here understood as the subjective sense of having a meaningful
future, despite obstacles (Post 1998, 24).
Fear is also related to anxiety, which is an internal state that generates multiple
fast scenes in the mind, as though they were occurring in the external
world. The turnover of neural constellations at the basts of anxiety is as high
as in fear, although the mental contents are as internalized as in depression
(Greenfield 2000, 152-53). Anxiety has no clear object; it is dominated by internal
reasoning processes, not by sensation; yet it also involves emotions. It
may also not be easy in practice to differentiate between anxiety and depression;
some common cognitive and behavioral features may underlie the
apparent differences (see Craig and Dobson 1995). There is, for example,
some empirical evidence for anxiety as a sequel to depression (MacDougall
and Brown 1984). Information may also he misattributeci to substitute objects,
the information about which is derived from affective states. This happens
when an affective state lacks a salient object, as is the case in anxiety.
An anxious mood is turned into an emotion when an object is provided by
focusing on whatever happens to be most salient to the experience!' (Gore
and Gasper 2000, 10-11.) Thus, for example, indefinite feelings of guilt can
be turned into religious emotions by connecting them to religious objects.
This may make it easier to handle them (see Thorson 1998).
But, as Emmons and Paloutzian (2003, 394) observe, almost all studies of
conversion suffer from various kinds of methodological shortcomings, such
as for example near total reliance on measures of self-perceived change. This
is no minor issue in the study of religion (cf. Barrett and Keil 1996). Many extensive
literature reviews have showr
n that results from studies on religion and
mental health are mixed and even contradictory. Bergin (1983), for example,
found that in 23 percent of the reviewed studies there was a negative relationship
between religion and mental health, in 47 percent of the studies the
relation was positive, and in 30 percent there was no relationship. This is
close to what one would expect by chance. Another alternative is that the results
are skewed because of methodological difficulties. For example, Gartner
(2002, 187-88), who is suspicious about the existence of such difficulties, \vi
acknowledges the fact that the very concept of "religious concept" has no
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
154 Ilkka Pyysi]inen
generally accepted definition. Krymkowski and Martin (1998), for instance,
found that in papers published in the Journalfor the Scientific Study of Religion,
beginning from 1986, religion was prominently taken to be an independent
causal factor, affecting such things as abortion attitudes and alcohol
consumption. They claim that such explanations are highly problematic because
no sufficient attention has been paid to the mechanisms by which "religion"
supposedly exercises influence, the direction of causality is not always
clearly established, and controls are not always used. Often it is not too clear
what is meant by religion (see Pyysiainen, in press).
Gartner (2002, 200), for his part, claims that much of the discrepancy in the
findings may be explained by differences in the ways mental health is measured.
Thus we actually are dealing with variables too vaguely defined; it is
therefore veiy difficult to find unequivocal causal relationships. Gartner
(2002, 201) argues that the studies reporting a negative relationship between
religion and mental health typically employed personality tests with only
limited reliability and validity, whereas the studies that found a positive correlation
were based on real-life observations concerning such things as drug
abuse, delinquency, and so forth. It should, however, be specified what it is
in religion that contributes to mental health: professing certain counterintuitive
beliefs, performing rituals, the social relationships among believers, or
what (see Levin and Chatters 1998)? Thus George and colleagues (2002) conclude
that "we are far from understanding the mechanisms by which religious
involvement promotes health," and Pargament (2002b, 169) says that,
even when significant results are obtained, they provide only little insight
into how religion works. Moreover, many of the relevant studies have been
made in the United States, and the results cannot necessarily be generalized
to other cultural contexts (see Pyysiainen, in press).
Thus, although traditionally and continuously religious individuals may report
less stress and anxiety than nonreligious or unconventionally religious
individuals, this may be because they have more enduring social relationships
and because they have conceptual and other cognitive-affective means
to deal with their negative experiences. They do not necessarily lack these
negative experiences; they only are better equipped to deal with them.
In the following section, I present the 12 modal variables systematically
according to Whitehouse (2002b; cf. Whitehouse 1995), tiying to evaluate
their relevance for explaining conversion phenomena.
C O N V E R S I O N A N D THE M O D E S
1. Frequency of Transmission
In the imagistic ideal type, the transmission of a given ritual takes place
once every year at the most. More frequent transmission means a decisive
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 155
move towacl the doctrinal position, although we cannot say that increased
frequency nakes the ritual doctrinal (doctrinality depending on all the other
variables a? well). It is also problematic to measure only the frequency of
one single itual, as some rituals may be embedded in other rituals; an individual
ritUcl always works in that wider context (Lawson and McCauley
1990; McCailey and Lawson 2002; cf. Lévi-Strauss 1979). Moreover, what
counts as participation in a ritual is not always clear (McCauley and Lawson
2002, 55, 1 7 - 3 0 ; Atran 2002, l6()-6l). Third, what is learned in a ritual may
also be learned from other sources; myths and doctrines are recounted not
only in ritutls.
With the^e caveats in mind, we can say that religious conversion may take
place in both frequently and infrequently performed rituals. Sometimes a
single even may trigger a conversion; sometimes it happens only after participation
01 a regular basis. It all depends on so many other things.
2. Level of arousal
The imagistic ideal type is characterized by high emotional arousal, the
doctrinal one by low. Whitehouse does not suggest any exact method of
measurement, but he seems to mean the type of arousal that can, in principle,
be registered by physiological measures (saliva samples, galvanic skin
responses, etc.). The scale to be used should be a universal one: what is high
among American Protestants is also high among the Haitian practitioners of
voodoo, as well as everywhere. We are not talking only about the external
expression of emotion and its cultural interpretation but rather about basic
physiological facts. The argument is clearly different from that advanced by
McCauley and Lawson (2002), which deals with relative levels of arousal
across the repertoire of rituals found in any single religious tradition (rather
than across traditions).
Conversions typically involve high emotions by both standards. It could
even be said that this is one of the defining characteristics of conversion as
here understood.
3. Memory
One of the central characteristics of the imagistic ideal type is episodic
memories for personally experienced events, the doctrinal ideal type relying
more heavily on semantic memory for teachings codified in language. While
the cognitive contents of doctrinal religions are largely based on semantic
knowledge organized into abstract scripts and schemata, conversions are
personal experiences encoded in episodic memory. They take place at a
specified time and place and are remembered as such: it happened to me
then and there. For Tulving, episodic encoding actually depends on prior semantic
encoding and vice versa, episodic and semantic memory working in
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
156 Ilkka Pyysiâinen
a parallel fashion in storing information: it is only retrieval of information
from these two stores that happens independently (Tulving 1995; Frith and
Dolan 2000; Brown and Craik 2000; see Atran 2002, 159-61). Semantic memory,
based on the lateral temporal cortex, and episodic memory, based on
the medial temporal lobe, need to cooperate for us to be able to recognize
surprising and exceptional events and phenomena: semantic memory represents
what is common across situations, while episodic memory represents
the exceptions. This means that situations requiring more conscious control
tend to give rise to stronger episodic memories (Lieberman et al. 2002, 228,
233-34; see Zola and Squire 2000; Schacter et al. 2000; Pyysiâinen 2004c).
Klein, Cosmides, Tooby, and Chance (2002), for example, present strong
empirical evidence to support the hypothesis that decision mles designed to
render judgments about the self and others recruit information both from
episodic and semantic memory stores. Episodic memories are about specific
events and behaviors involving a target, while semantic memories consist of
summaries of the target person's personality traits abstracted from a set of
particular events. Once a "trait summary" has been formed, judgments about
a person's traits are made on this basis. However, if there are episodic memories
that are inconsistent with the trait summary, they will always be activated
when a piece of summary information is retrieved. Thus, we might
speculate that also in their religious judgments people draw information
both from episodic and semantic memory stores, with semantic information
consisting of summaries of what has been acquired in cultural transmission.
It is the interaction between episodic memories and summaries that needs to
be considered in developing a theory of religious conversion.
4. Ritual Meaning
In the imagistic ideal type, ritual meanings are internally generated in contrast
to the doctrinal position typified by learned or acquired meanings. In
the doctrinal position, persons can interpret the rituals using concepts and
beliefs that can be inferred from cultural communication without much effort,
whereas in the imagistic mode, the interpretative concepts are arrived at
through a lengthy and laborious process of "spontaneous exegetical reflection"
(see Sperber and Wilson 1986; Barrett and Keil 1996).
In a conversion, what Barrett and Keil (1996, 612) call "rehearsed, nonintegrated
attributes that have no causal efficacy" might suddenly be invested
with a very personal meaning; what previously seemed like unconnected
bits and pieces that could not be connected to one's database without very
costly cognitive processing, suddenly appear as a coherent whole that can
be combined with what one already has in mind. Thus, although one is processing
familiar (partly at least) and "rehearsed" concepts, they acquire a
new meaning that is somehow internally generated, doctrinal summaries
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 157
and episodic memories being simultaneously activated. The difference between
imagistic and doctrinal is that in learning meanings one merely uses
some commonly accepted concepts without much reflective thought or
feeling of personal relevance, whereas in the case of internal generation
one consciously ponders over the meanings and arrives at an interpretation
that seems to have personal relevance. Thus conversion is typified by internally
generated meanings, although they can be arrived at by reinterpreting
learned concepts.
5. Revelation Techniques
Whitehouse's way of differentiating between two kinds of "techniques of
revelation" may be problematic. He argues that whereas imagistic religions
are typified by iconicity, multivocal imagery, and analogic reasoning, doctrinal
religiosity is characterized by "logical integration" and rhetoric. This
might lead a psychologist to think that Whitehouse is actually distinguishing
between intuitive/spontaneous/automatic processing and systematic/
rational/controlled processing. This, however, is not his point; the two types
of cognitive processing certainly belong to both imagistic and doctrinal religion,
although in varying degrees (Bargh 1994; Pyysiäinen 2003, 2004d).
Atran (2002, 155-57) correctly remarks that there is no evidence to support
the claim that the doctrinal mode tends toward some "logically integrated
and coherent ideology" connected by implicational logic. This is only true of
theology as an ideational artifact that exists in the collective book-mincls interaction
(Pyysiainen 2004a, 2004d). If, on the other hand, we step back from
this argument and define logical integration more loosely, we risk blurring
the difference between imagistic and doctrinal (see Pyysiainen 2004c).
6. Social Cohesion
The imagistic ideal type is characterized by intense social cohesion and
the doctrinal one by diffuse cohesion. It seems, on the basis of relevant literature,
that conversion takes place from a diffuse group to a more cohesive
one (see Malony and Southard 1992).
7. Leadership
In imagistic religion, leaders are either passive or totally absent; doctrinal
religiosity is characterized by dynamic leadership. Conversion seems to necessitate,
somewhat paradoxically, both intense group cohesion and dynamic
leadership. The leader in a way represents the whole group, which
gets personified in the leader. However, it might be that these two variables
ire complementary in the sense that a conversion requires either a coherent
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
158 Ilkka Pyysiäinen
group or a dynamic leader (of a diffuse group). If both are absent, there can
be no concrete doctrinal expectations to trigger the conversion.
8. Inclusivity/Excfusivity
Imagistic religions are exclusive, doctrinal religions inclusive. Yet it seems
that there are important differences in inclusiveness/exclusiveness between
various types of doctrinal movements: conversions are more typical in doctrinal
movements such as Pentecostalism than in the more hierarchical religions
of Lutheranism or Catholicism. As Weber (1965) argued, in institutionalized
religions, an "official grace" (Anstaltsgnade) replaces personal piety;
personal piety can even be considered a threat to the institution's capacity to
bestow salvation to anyone it chooses. This, however, only excludes conversion
to inclusive religions; conversion within an inclusive religion but to
a more exclusive one seems to be common.
9. Spread
The spread of imagistic religions is slow and inefficient in contrast to doctrinal
religions that spread fast and efficiently. As conversion is a means of
entering a religion, rapid spread also means great potential for conversions.
Conversions cannot take place in stable movements that can only be entered
by being born into the group.
10. Scale
Imagistic religions are small-scale phenomena, whereas doctrinal religions
are large-scale phenomena. It seems that conversions most typically
happen within large-scale movements but lead to smaller-scale movements.
Although for example Pentecostalism is a large-scale movement
worldwide, its noncentralizecl nature may make it appear locally as a smallscale
movement.
11. Degree of Uniformity
The degree of uniformity is supposed to be low in imagistic religions and
high in doctrinal religions. In the imagistic ideal type there is no binding
doctrine, whereas in doctrinal religions everybody is supposed to subscribe
to the same doctrine in principle. Yet at the same time, the way of life in
imagistic religions may be at least as uniform as in doctrinal religions. It
seems to be specifically the ideal of a doctrinal unity that is the catalyst for
conversions.
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 159
12. Structure
The structure is noncentralized in imagistic religions and centralized in
doctrinal religions. Such centralization seems to be a catalyst for conversions
that, however, happen toward noncentralized groups (see below).
C O N V E R S I O N IN T H E M O D E S DYNAMICS
There are now four possibilities:
• Conversion within imagistic religiosity (rare)
• Conversion from imagistic to doctrinal religiosity (tedium follows)
• Conversion within doctrinal religiosity (typically happens)
• Conversion from doctrinal to imagistic religiosity (might happen?)
The first option seems like a rather unlikely event; it is veiy difficult to
"convert" from a local cult to another such cult. Or, if it happens, one usually
does not substitute one cult for another but simply joins one more cult (as in
the Greek mystery cults, for example; see Martin 2004).
As to the second option, conversion from an imagistic religion to a doctrinal
religion is often a reaction to missionary efforts. Koskinen (1953, 91-94)
argues that when the Pacific Islanders were converted to Christianity en
masse, they still did not discard their old beliefs and practices. The influence
of Christianity "remained superficial"; the natives "had none of the ambition
of real Christians." At first Christianity was popular, but "disappointment and
tedium soon made their appearance." The natives felt that Christianity had
promised more than it had been able to give and that they did not gain the
expected this-worlclly advantages from "white man's magic" (see Whitehouse
2000 44-45, 142-43). Thus, in this type of conversion, one adopts
something of a doctrinal tradition, trying to combine it with one's imagistic
religion with varying degrees of success.
The most typical conversion seems to be a conversion within the doctrinal
ideal type. It is possible, for instance, to have a Christian belief environment
and yet be relatively ignorant about Christian doctrine. Conversion then revitalizes
the known (if only vaguely) doctrinal beliefs.
The fourth option might be realized when someone for example converts
from Christianity to an imagistic cult. It is, however, doubtful that one could
really totally abandon one's doctrinal way of thinking about religion. The result
may rather be some kind of hybrid religion.
Summarizing the analysis, we see that conversions seem to belong to religiosity
that is more typically imagistic on the basis of the variables of 2, 3, and
i. In the variables 1 and 4, both imagistic and doctrinal content allows for
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
160 Ilkka Pyysiäinen
conversions. The fifth variable is undetermined because it fails to differentiate
between the two modes. In the variables 6, 7, 10, 11, and 12, it is doctrinal
contents that are more likely to lead to conversions. In the variables 8 and
9, something like a mixture of imagistic and doctrinal contents leads to conversions.
It thus seems that although conversion is a phenomenon that typically
happens in a doctrinal context, it is nevertheless a phenomenon that
does not quite fit with Whitehouse's description of the doctrinal ideal type. It
is a soil of imagism within cloctrinality: the psychological variables (of the 12)
have an imagistic content, while the social variables have a doctrinal content.
Pyysiäinen (2001, 2004c) and Tremlin (2002) have emphasized on the basis
of Boyer's (1994, 2001) and Barrett's (Barrett and Keil 1996; Barrett 1998)
studies that because theological traditions are an epiphenomenal overlay on
natural religiosity (also McCauley 2000), doctrinal religiosity thus is constantly
threatened by the fact that its concepts seem irrelevant and are difficult to use
in everyday reasoning. Yet only doctrinal religions have the potential to
spread beyond the boundaries of the local community and unite large masses
of people. Tremlin (2002) thus argues that imagistic-like phenomena provide
individual motivation, while doctrinal-type phenomena offer systems-level
tools for the preservation of stable traditions (see also Pyysiäinen 2004a).
In this perspective, conversion might be said to be a reaction to a situation
where doctrinal development has produced such abstract religious concepts
and beliefs that they no longer can motivate people. Their abstractness makes
them difficult to process and also makes it difficult to see their relevance: what
for example concretely follows from the view that God is "The Ground of Being?"
(see Boyer 2001, 320-22). It seems that when theology gets really abstract,
religious institutions also tend to become all-encompassing but "empty"
(see Pyysiäinen 2004a). Weber (1965) described such a process in his argument
about new movements gradually getting institutionalized and then giving
birth to revivalist movements that aim to restore the lost original spontaneity.
At first the religious virtuosity of leaders may be important, but as
institutionalization proceeds they become suspect because they seem not to
trust the institution's capacity to bestow salvation freely to all. Weber takes as
an example the Scholastic distinction between fides implicit a and fides explícita:
ordinary people were not supposed to know the whole doctrine; it was
enough for them to accept it on the grounds that the specialists know it and
have rational reasons to believe it. The specialists had to know the doctrine explicitly.
Ordinary people thus have only implicit belief, whereas specialists
have explicit belief.
This kind of development is evident in the Reformation that broke the
doctrine of implicit faith (although similar phenomena, like the "New Devotion,"
were known also before). Christianity now was understood to be an
intimate matter between God and the individual believer. Luther's emphasis
on an individual believer's right to judge matters in the light of his personal
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 161
understanding of the Bible is well known. Yet Luther did not want radically
to depart from the Catholic Church or the scholastic tradition, the presuppositions
and methods of which he shared to a certain extent. Some of his followers,
however, went much further in their individualism, rejecting hierarchical
authority both on institutional and doctrinal levels (Williams 1962;
Saarinen 1988; Lohse 1999).
But, as the Weberian hypothesis predicts, the Reformation itself was in clue
course institutionalized; this led the so-called Lutheran orthodoxy and its
rigid dogmatism. A new kind of revival then took place in the Pietistic movem
e n t s ) that had their earliest beginnings in the early seventeenth century;
the word Pietism, however, came to be known only in 1677, two years after
the publication of Philipp Jakob Spener's Pia desideria; Spener's followers
were now called "Pietists." In the various forms of Pietism, great emphasis
was put on the personal life of piety (Brecht 1996).
In Finland, for example, the first appearance of radical Pietism is personified
in the mysterious figure of Lars Ulstaclius. He was a Lutheran minister and
a schoolteacher who came to be tormented by religious doubt, guilt, and general
anxiety. He fell ill, and for about two years he neither washed himself nor
had his hair or beard cut. In his agony he turned to the local vicar, asking for
public absolution for his sins. The vicar explained to him that such scruples
were merely the work of the devil and he should not pay attention to them.
On July 22, in 1688, Ulstaclius then in clue course appeared in the Dome of
Turku in his rags, with his hair hanging long and with a huge matted beard,
interrupting the service by starting to read aloud the radical theses he had
written down. When two men grabbed him to throw him out of the Dome,
what was left of his humble dress fell off, and poor Ulstaclius stood there
naked, only covered by his long hair and beard. Like some Old Testament
prophet, he proclaimed that the Lutheran doctrine was to be doomed, that
prayer books and postillas were a bunch of lies, and that the ministers were
not endowed with the Holy Spirit (Akiander 1857, I 3-30; Oclenvik 1940). The
revivalist movements that then emerged in Finland in the eighteenth and
nineteenth centuries all protested against a too-liberal and too-abstract interpretation
of the doctrine, emphasizing both correct doctrine and an individual
believer's responsibility of his or her religious choice: baptized members
of the Lutheran Church were now expected to make a personal decision in
favor of faith (see Pentikainen 1975; Pyysiainen 2004c).
Ulstadius's dramatic appearance in the Dome might have marked the beginning
of a new imagistic movement, but that did not happen; the revivals
took a very doctrinal form despite the emphasis on emotional experience.
We may thus make the prediction that doctrinal religiosity develops slowly
but has strong staying power: revivalism does not undermine doctrinal religiosity
but supports it. Conversions are reactions to inherent problems in
doctrinal religiosity; they derive their motivational power from imagistic-like
162 Ilkka Pyysidinen
phenomena but combine them with elements of doctrinality. The important
thing is that one experiences an intimate connection with some counterintuitive
agent. Episodic memories of conversion experiences are then always
activated when a doctrinal summary is activated, as they contradict what one
previously thought about the doctrine. Yet the episodic memories may themselves
in due course develop into schemata in semantic memory, due to the
doctrinal context of interpretation and numerous retellings of the episodes.
Thus the intimacy may also be lost to an extent.
In my opinion, the amount of emotional arousal in religious rituals and
other religious situations depends precisely on whether such intimacy is experienced
or not (Pyysiainen 2001, 80-97). Ritual form (McCauley and Lawson
2002) is one way of producing it, ritual frequency is another (Whitehouse
1995, 2000, 2001, 2002), and there are others still (see also Atran 2002,
290-91 )• Let me emphasize, however, that I am here speaking of the actual
experiences of real people, not of rituals' potential to arouse emotions—
ceteris paribus—in an ideal participant. Thus, from the modes point of view,
the strong emotional component in conversions is explained by the fact that
a sudden conversion is a unique and personal event that creates strong
episodic memories that seem to fulfill the criteria of flashbulb memory (see
Whitehouse 2000, 7-9, 119-22; Conway 1995). The "spontaneous exegetical
reflection" that supposedly follows is constrained by the doctrine, however.
Therefore it may gradually develop into fixed schemata.
Conversion also is a way of establishing a personal relationship with the
counterintuitive agent(s) as presented in the doctrine. Thus, from the point
of view of the ritual form hypothesis (McCauley and Lawson 2002), the emotional
component is explained by the logical form of the event: when one
experiences a conversion, one supposedly is the direct object of action of
some counterintuitive agent; this creates the intimacy that triggers the emotions.
Such conversion also often leads to frequent participation in various
kinds of services and devotions, and thus revitalizes doctrinal religiosity. Occasionally
conversions may also form "epidemics" and spread like wildfire,
as in the case of the Great Awakenings in the United States (see Taves 1999;
Slone 2004, 85-102).
C O N C L U S I O N
To the extent that the modes theory is meant as an empirical theory, its worth
can only be judged by the range of phenomena it helps explain. We must ask
how and for what purpose this theory can be used in the empirical sftidy of religions.
In the case of conversions, the modes theory helps draw attention to
the interplay between psychological and social factors in canalizing religious
transmission into differing feedback systems. They are not independent from
each other. This, then, helps explain, for example, why there can be no conEdited
by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
Religious Conversion 163
versions in the imagistic ideal type. In imagism there is no need for conversions
because the problem is not motivation but the coherence and durability of the
tradition. In doctrinal religiosity, the problem is motivation, not preservation.
The attractor position thus cannot be imagistic or doctrinal, but precisely in between
the two ideal types. Living religiosity always seems to combine features
of the two modes; pure imagism and pure doctrinalism are abstract ideals only.
Conversion is one example of how religious phenomena tend to oscillate in the
middle position; magic might be another (see Pyysiainen 2004b, 90-112). Conversions
thus represent an attempt at an evolutionarily stable strategy in cultural
transmission. Although a conversion is a psychological phenomenon, that
is, something that happens in the mind, an increase or a decrease in conversions
at some time or in some sociocultural context is a social phenomenon.
The modes theory, albeit in a somewhat modified form, can help analyze the
mutual interdependence of the psychological and social variables here.
REFERENCES
Akiander, Matth(ias). 1857. Religiose/ rorelserna i Finland i alder och senate tidei\
vol. 1. Helsingfors: J . Simelii arfvigar.
Atran, Scott. 2002. /// Gods We Trust: The Evolutionary Landscape of Religion. Oxford:
Oxford University Press.
Atran, Scott, and Ara Norenzayan. In press. Religion's Evolutionary Landscape: Counterintuition,
Commitment, Compassion, Communion. Behavioral and Brain Sci-
ences.
Bargh, J o h n A. 1984. The Four Horsemen o f Automaticity: Awareness, Intention, Efficiency,
and Control in Social Cognition. In Handbook of Social Cognition, ed.
Robert S. Wyer and Thomas K. Srull. Hillsdale, NJ: Erlbaum.
Barrett, Justin L. 1998. Cognitive Constraints on Hindu Concepts of the Divine, journal
for the Scientific Study of Religion 37: 608-19.
Barrett. Justin L., and Frank Keil. 1996. Conceptualizing a Nonnatural Entity: Anthropomorphism
in Cod Concepts. Cognitive Psychology 3 1 : 219-47.
Bergin, A. E. 1983. Religiosity and Mental Health: A Critical Revaluation and MetaAnalysis.
Professional Psychology: Research and Practice 14: 170-84.
Boyer, Pascal. 1994. The Naturalness of Religious Ideas: A Cognitive Theory of Religion.
Berkeley: University of California Press.
. 2001. Religion Explained: The Evolutionary Origins of Religious Thought.
New York: Basic Books.
Brecht, Martin. 1996. Pietismus. Theologische Realenzyklopcidie 26: 6 0 6 - 3 1 . Berlin:
Walter cle Gruyter.
Brown, Scott C , and Fergus I. M. Craik. 2000. Encoding and Retrieval of Information.
In The Oxford Handbook of Memory, ed. Enclel Tulving and Fergus I. M. Craik.
Oxford: Oxford University Press.
Brown, Warren S., and Carla Caetano. 1992. Conversion, Cognition, and Neuropsychology.
In Handbook of Religious Conversion, ed. H. N. Malony and S. Southard,
147-58. Birmingham, AL: Religious Education Press.
Edited by Foxit Reader
Copyright(C) by Foxit Software Company,2005-2007
For Evaluation Only.
164 llkka Pyysiäinen
Clore, Gerald L., and Kasper Gasper. 2000. Feeling is Believing: Some Affective Influences
on Belief. In Emotions and Beliefs: How Feelings Influence Thoughts, eel.
Nico H. Frijda, Anthony S. R. Manstead, and Sacha Bern. Paris: Editions de la maison
des sciences de l'homme & Cambridge: Cambridge University Press.
Conway, Martin A. 1995. Flashbulb Memories. Hove, UK: Lawrence Erlbaum.
Craig, Kenneth D., and Keith S. Dobson, eds. 1995. Anxiety and Depression in Adults
and Children. Thousand Oaks, CA: Sage.
Emmons, Robert A., and Raymond F. Paloutzian 2003. T h e Psychology o f Religion.
Annual Review of Psychology 54: 3 7 7 - 4 0 2 .
Frith, Chris, and Raymond J . Dolan. 2000. T h e Role of Memory in the Delusions Associated
with Schizophrenia. In Memory), Brain, and Belief ed. Daniel L. Schacter
and Elaine Scarry. Cambridge, MA: Harvard University Press.
Gartner, John. 2002. Religious Commitment, Mental Health, and Prosocial Behavior:
A Review^ o f the Empirical Literature. In Religion and the Practice of Clinical Psychology,
ed. E. Shafranske. Washington, DC: American Psychological Association.
George, Linda K., Christopher G. Ellison, and David B. Larson. 2002. Explaining the
Relationships between Religious Involvement and Health. Psychological Inquiry
13: 190-200.
Greenfield, Susan. 2000. The Private Life of the Brain: Emotions, Consciousness, and
the Secret of the Self. New York: John Wiley.
Guthrie, S. 1993. Faces in the Clouds. New York: Oxford University Press.
Kirkpatrick, Lee A. 1997. A Longitudinal Study of Changes in Religious Beliefs and
Behavior. Journal for the Scientific Study of Religion 36: 207-17.
. 1998. G o d as a Substitute Attachment Figure: A Longitudinal Study o f Adult Attachment
Style and Religious Change in College Students. Personality and Social
Psychology Bulletin 24: 9 6 1 - 7 3 .
. 1999. Attachment and Religious Representations and Behavior. In Handbook
of Attachment Theory: Research, and Clinical Applications, ed. J. Cassidy and P. R.
Shaver, 803-22. New York: T h e Guilford Press.
Klein, Stanley B., Leda Cosmides, John Tooby, and Sarah Chance. 2002. Decisions
and the Evolution of Memory: Multiple Systems, Multiple Functions. Psychological
Review 109: 306-29.
Koskinen, Aarne A. 1953. Missionary Influence as a Political Factor in the Pacific Isla)
ids. Helsinki: Suomalainen tiecleakatemia.
Krymkowski, Daniel H., and Luther H. Martin. 1998. Religion as an Independent Variable:
Revisiting the Weberian Hypothesis. Method & Theory in the Study of Religion
10: 187-98.
Lawson, E. Thomas, and Robert N. McCauley. 1990. Rethinking Religion: Connecting
Cognition and Culture. Cambridge: Cambridge University Press.
Levi-Strauss, Claude. (1958/1979). Structural Anthropology. Trans. C. Jacobson and
G. Schoepf. Harmondsworth: Penguin.
Levin, Jeffrey S., and Linda M. Chatters. 1998. Research on Religion and Mental
Health: An Overview of Empirical Findings and Theoretical Issues. In Handbook
of Religion and Mental Health, ed. Harold G. Koenig. San Diego: Academic Press.
Lieberman, Matthew D., Ruth Gaunt, Daniel T. Gilbert, and Yaacov Trope. 2002. Reflexion
and Reflection: A Social Cognitive Neuroscience Approach to Attributional
Inference. Advances in Experimental Social Psychology 34: 199-249.
Religio is Conversion 165
Lohse, Bernhard. 1999. Martin Luthers Theology: Its Historical and Systematic Development.
Ed. and trans. Roy A. Harrisville. Edinburgh: T&T Clark.
MacDougall, Mary Ann, and Robert S. 3rown. 1984. The Temporal Order of Depression
and Anxiety. In Tide Self in Anxtity, Stress, and Depression, ed. Ralf Schwarzer,
341-51. Amsterdam: Elsevier Science Publishers.
MacReady, Norm. 2002. Adrenergic Bbckers Shortly after Trauma Can Block PTSD.
Clinical Psychiatry Sews 30: 9.
Malony, H. Newton, and Samuel Soutiard, eds. 1992. Handbook of Religious Conversion.
Birmingham, AL: Religious Education Press.
Martin, Luther. 2004. Cognitive Science, Historiography and the Study of So-Called
Folk Religions: The Case of the Eleisinian Mysteries of Demeter. Temenos 39/40:
81-100.
McCauley, Robert N. 2000. The Naturahess of Religion and the Unnaturalness of Science.
In Explanation and Cognition, ed. Frank C. Keil and Robert A. Wilson. Cambridge,
MA: MIT Press.
McCauley. Robert N., and E. Thomas Lavvson. 2002. Bringing Ritual to Mind: Psychological
Foundations of Cultural Forms. Cambridge: Cambridge University Press.
Miller, William R., and Janet C'de Baca. 2001. Quantum Change: When Epiphanies
and Sudden Insights Transform Ordinary lives. New York: The Guilford Press.
Odenvik, Nathan. 1940. Lars Ulstadius. Stockholm: Förlaget Philadelphia.
Paloutzian, Raymond F , James T. Richardson; and Lewis R. Rambo. 1999. Religious
Conversion and Personality Change. Journal of Personality 67: 1047-79.
Parlament, Kenneth I. (1996/2002a). Religious Methods of Coping: Resources for the
Conservation and Transformation of Significance. In Religion and the Practice of
Clinical Psychology, ed. E. Shafranske. Washington, DC: American Psychological
Assc )ciation.
. 2002b. The Bitter and the Sweet: An Evaluation of the Costs and Benefits of
Religiousness. Psychological Inquiry 3: 168-81.
Pentikäinen, Julia. 1975. Revivalist Movements and Religious Contracultures in Finland.
In New Religions (Scripta InsUtuti Donneriani Ahoensis, 7), ed. Haralds
Biezais. Abo: The Donner Institute fcr Research in Religious and Cultural History.
Pinker, Steven. 2002. The Blank Slate: The Modern Denial of Human Nature. New
York: Penguin.
Post, Stephen G. 1998. Ethics, Religion, and Mental Health. In Handbook of Religion
and Mental Health, ed. Harold G. Koenig. San Diego: Academic Press.
Pyysiäinen, Ilkka. 2001. How Religion Works: Towards a New Cognitive Science of
Religion. Leiden: Brill.
. 2003. Dual-Process Theories and Hybrid Systems: A Commentary on Anderson
& Lebiere. Behavioral and Brain Sciences 26: 617-18.
. 2004a. Folk Religion and Theological Correctness. Temenos 39/40: 151-65.
. 2004b. Magic. Miracles and Religion: A Scientist's Perspective. Walnut Creek,
CA: AltaMira Press.
. 2004c. Corrupt Doctrine and Doctrinal Revival: On the Nature and Limits of the
Modes Theory. In Theorizing Religions Past: Archaeology, History, and Cognition,
ed. Harvey Whitehouse and Luther H. Martin. Walnut Creek, CA: AltaMira Press.
. 2004d. Intuitive and Explicit in Religious Thought. Journal of Cognition and
Culture 4: 123-50.
166 Ilkka Pyysiäinen
. In press. Religion is neither Costly, nor Beneficial: A Commentary on Atran
and Norenzayan. Behavioral and Brain Sciences 27, no. 6.
Rambo, Lewis. 1993. Understanding Religious Conversion. New Haven, CT: Yale University
Press.
. 1998. The Psychology of Conversion. Paper delivered at the International
Coalition for Religions Freedom Conference on "Religions Freedom and the New
Millennium," Berlin, Germany, May 2 9 - 3 1 . http://uww.religioiisfreedom.com/
conference/Germany/rambo htm (accessed 22 August 2002).
Saarinen, Risto. 1988. Gottes Wirken auf uns. Die transzendentale Deutung des
Gegen wart-Christi-Motivs in der Lutherforschung. Reports from the Department of
Systematic Theology, University of Helsinki, IX. Helsinki: University of Helsinki.
Schacter, Daniel L., Anthony D. Wagner, and Randy L. Buckner. 2000. Memory Systems.
In The Oxford Handbook of Memory, ed. Endel Tulving and F. I. M. Craik.
Oxford: Oxford University Press.
Slone, Donald Jason. 2004. Theological Incorrectness: Why Religious People Believe
What They Shouldn't. New York: Oxford University Press.
Sperber, Dan, and Deirdre Wilson. 1986. Relevance: Communication and Cognition.
Cambridge, MA: Harvard University Press.
Taves, Ann. 1999. Fits, Trances, & Visions: Experiencing Religion and Explaining Experience
from Wesley to fames. Princeton, NJ: Princeton University Press.
Thorson, James A. 1998. Religion and Anxiety: Which Anxiety? Which Religion? In
Handbook of Religion and Mental Health, ed. Harold G. Koenig. San Diego: Academic
Press.
Tremlin, Todd. 2002. A Theory o f Religious Modulation: Reconciling Religious Modes
and Ritual Arrangements. Journal of Cognition and Culture 2: 309-48.
Tulving, Endel. 1995. Organization of Memory: Q u o Vadis? In The Cognitive Neurosciences,
ed. Michael S. Gazzaniga et al. Cambridge, MA: MIT Press.
Weber, Max. (1922/1965). Sociology of Religion. Ed. and trans. E. Fischoff. London:
Methuen.
Whitehouse, Harvey. 1995. Inside the Cult: Religious Innovation and Transmission
in Papua New Guinea. Oxford: Clarendon Press.
. 2000. Arguments and Icons: Divergent Modes of Religiosity. Oxford: Oxford
University Press.
. 2001. Transmissive Frequency, Ritual, and Exegesis. Journal of Cognition and
Culture 1: 167-81.
. 2002a. Conjectures, Refutations, and Verification: Towards a Testable Theory
of "Modes of Religiosity." Jou mal of Ritual Studies 16: 4 4 - 5 9 .
. 2002b. Modes of Religiosity: Towards a Cognitive Explanation of the Sociopolitical
Dynamics of Religion. Method & Theoiy in the Study of Religion 14: 293-315.
Williams, George Huntston. 1962. The Radical Reformation. Philadelphia: T h e Westminster
Press.
Zinnbauer, Brian J., and Kenneth I. Pargament. 1998. Spiritual Conversion: A Study of
Religious Change among College Students. Journal for the Scientific Study of Religion
37: 161-80.
Zola, Stuart, and Larry R. Squire. 2000. T h e Medial Temporal Lobe and the Hippocampus.
In The Oxford Handbook of Memory, ed. Endel Tulving and F. I. M.
Craik. Oxford: Oxford University Press.