Content uploaded by David Bryce Yaden
Author content
All content in this area was uploaded by David Bryce Yaden on Sep 14, 2015
Content may be subject to copyright.
Psychology of Religion and Spirituality
The Language of Ineffability: Linguistic Analysis of
Mystical Experiences
David B. Yaden, Johannes C. Eichstaedt, H. Andrew Schwartz, Margaret L. Kern, Khoa D. Le
Nguyen, Nancy A. Wintering, Ralph W. Hood, Jr., and Andrew B. Newberg
Online First Publication, July 27, 2015. http://dx.doi.org/10.1037/rel0000043
CITATION
Yaden, D. B., Eichstaedt, J. C., Schwartz, H. A., Kern, M. L., Le Nguyen, K. D., Wintering, N. A.,
Hood, R. W., Jr., & Newberg, A. B. (2015, July 27). The Language of Ineffability: Linguistic
Analysis of Mystical Experiences. Psychology of Religion and Spirituality. Advance online
publication. http://dx.doi.org/10.1037/rel0000043
The Language of Ineffability: Linguistic Analysis of Mystical Experiences
David B. Yaden, Johannes C. Eichstaedt,
and H. Andrew Schwartz
University of Pennsylvania
Margaret L. Kern
University of Pennsylvania and University of Melbourne
Khoa D. Le Nguyen
University of Pennsylvania
Nancy A. Wintering
Thomas Jefferson University
Ralph W. Hood Jr.
University of Tennessee at Chattanooga
Andrew B. Newberg
Thomas Jefferson University
Mystical experiences are often described as “ineffable,” or beyond language. However, people readily
speak about their mystical experiences if asked about them. How do people describe what is supposedly
indescribable? In this study, we used quantitative linguistic analyses to interpret the writings of 777
participants (45.5% female, 51.0% male) who recounted their most significant spiritual or religious
experience as part of an online survey. High and low scorers on a measure of mystical experiences
differed in the language they used to describe their experiences. Participants who have had mystical
experiences used language that was more socially and spatially inclusive (e.g., “close,” “we,” “with”) and
used fewer overtly religious words (e.g., “prayed,” “Christ,” “church”) than participants without such
experiences. Results indicated that people can meaningfully communicate their mystical experiences, and
that quantitative language analyses provide a means for understanding aspects of such experiences.
Keywords: inclusion, linguistic analysis, mystical experiences, religion, spirituality
If that is ineffable which cannot be spoken, then that is not ineffable
which is called ineffable
—Augustine, 1958, pp. 10 –11
Mystical experiences are temporary mental states that include
profound feelings of unity, positive emotions, a noetic quality,
and most notably for the present study, alleged ineffability
(Austin, 2006; Griffiths, Richards, McCann, & Jesse, 2006;
Hood, Hill, & Spilka, 2009; James, 1902/1985; Pahnke, 1969;
Richards, 1975; Stace, 1960; Wulff, 2000). The criterion of
ineffability means that by definition, mystical experiences are
beyond verbal description. Whereas some scholars maintain
that language cannot adequately capture any aspect of mystical
experiences (a view called “apophatic mysticism”), others
maintain that the term “ineffability” qualifies one’s capacity to
convey the full breadth of these profound experiences (Proud-
foot, 1985). In general, people rarely discuss these deeply
meaningful and generally positive experiences with others on
their own accord (Hay, 1990; Tamminen, 1991). Yet when
asked about them, those who have had such experiences often
go on to describe them in detail, using language. This pattern
led Chinese poet and comic Po Chu-I to comment on the
mystical text, the Tao te Ching: “‘He who talks doesn’t know,
he who knows doesn’t talk’ that is what Lao-tzu told us, in a
book of five thousand words. If he was the one who knew, how
could he have been such a blabbermouth?” (Mitchell, 1988, p.
89).
Attempts have been made to examine verbal and written reports
of mystical experiences. James’s (1902/1985) Varieties of Reli-
gious Experience rests mostly on personal accounts of New Eng-
land Protestants gathered by one of the fathers of the psychology
of religion and spirituality, Edwin Starbuck. Durkheim (1912/
2012) undertook a study of religious experience in Aborigines, and
found that such experiences were central to these communities.
Maslow (1964) described the phenomenon of “peak experiences,”
which share a large degree of overlap with mystical experiences,
by consistently seeing them appear in transcripts of interviews
with highly successful, “self-actualized” individuals. Laski’s
David B. Yaden and Johannes C. Eichstaedt, Department of Psychology,
University of Pennsylvania; H. Andrew Schwartz, Computer & Informa-
tion Science, University of Pennsylvania; Margaret L. Kern, Department of
Psychology, University of Pennsylvania, and Melbourne Graduate School
of Education, University of Melbourne; Khoa D. Le Nguyen, Department
of Psychology, University of Pennsylvania; Nancy A. Wintering, Myrna
Brind Center of Integrative Medicine, Thomas Jefferson University; Ralph
W. Hood Jr., Department of Psychology, University of Tennessee at
Chattanooga; Andrew B. Newberg, Myrna Brind Center of Integrative
Medicine, Thomas Jefferson University.
Khoa D. Le Nguyen is now at Department of Psychology, University of
North Carolina at Chapel Hill.
Correspondence concerning this article should be addressed to David B.
Yaden, Department of Psychology, University of Pennsylvania, 3701 Mar-
ket Street Suite 200, Philadelphia, PA 19104. E-mail: dyaden@sas
.upenn.edu
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Psychology of Religion and Spirituality © 2015 American Psychological Association
2015, Vol. 7, No. 4, 000 1941-1022/15/$12.00 http://dx.doi.org/10.1037/rel0000043
1
(1961) work on “transcendent ecstasy,” found that these experi-
ences are rather common, and that when asked to describe them,
people can readily do so.
Adding a more systematic approach, the Alister Hardy founda-
tion collected thousands of accounts of experiences from people
who had responded to the questions, “Have you ever had the
feeling of being close to a powerful spiritual force that seemed to
lift you out of yourself?” or “Have you ever been aware of or
influenced by a presence or power, whether you call it God or not,
which is different from your everyday self?” (Hardy, 1979; Hay &
Morisy, 1985). These studies replicated previous findings that
religious, spiritual, and mystical experiences are relatively com-
mon, and not the rare, esoteric experiences that some mistakenly
suppose (Scharfstein, 1973). A synthesis of recent survey research
found that about 35% of respondents have had mystical experi-
ences (Hood et al., 2009). However, this figure is open to a large
degree of interpretation. For instance, Hay (1979) found that 65%
of respondents affirmed being aware of or influenced by a pres-
ence or power, but this number dropped to 29.4% affirmative
responses after follow-up interviews clarified the question and
responses. The authors further differentiated between numinous
experiences, which refer to a sense of presence (20.2%), and
mystical experiences, which are characterized more by the sense of
union (9.2%). More research is needed to develop an epidemiology
of religious, spiritual, and mystical experiences that includes well-
defined classifications to establish prevalence, triggers, descrip-
tions, underlying mechanisms, and outcomes of these often posi-
tively transformative experiences.
Analyzing the language of reports of mystical experiences may
help with this effort. However, linguistic descriptions of mystical
experiences defy easy categorization (Hay, 1979; Thomas & Coo-
per, 1978). Although rich qualitative data have been compiled
through a number of studies, the sheer volume of information—
and differing classification systems— has made it difficult to draw
clear conclusions (Hood et al., 2009). The lack of quantitative
analyses makes it difficult to provide evidence for or against
claims about the linguistic content of reports of mystical experi-
ences. For example, some claim that reports of mystical experi-
ences include universal aspects (Hood, 2006; Stace, 1960) whereas
others argue that there are no common elements (Katz, 1978;
Proudfoot, 1985). Still others claim that reports of spontaneous
mystical experiences are indistinguishable from drug-induced ex-
periences (Smith, 1964), whereas others argue that they are very
different (Zaehner, 1961). These are still open questions and the
subject of ongoing research (Wulff, 2000).
While quantitative studies have demonstrated that many people
have varieties of religious, spiritual, and mystical experiences
(e.g., Gallup, 1978; Hood et al., 2001), and quantitative linguistic
analyses has been used to examine religious and spiritual self-
identification (e.g., Keller et al., 2013), quantitative linguistic
analyses of descriptions of mystical experiences have generally not
occurred. One exception was a linguistic analysis that differenti-
ated mystical experiences from psychedelic experiences and
schizophrenic experiences (Oxman et al., 1988). However, this
study was limited by the data collection methods that were used.
Narratives were taken from people who had written publically
about their experience with the onset of schizophrenia, taking a
psychedelic drug, or having a mystical experience. Therefore, the
study identified more superficial words used to describe the expe-
rience, including contextual information such as how schizophren-
ics use the word “schizophrenic” and “illness” more often in their
descriptions. The study did demonstrate, however, the feasibility
of differentiating mental states through language.
Examining underlying features of narrative descriptions has
been limited largely by the lack of available methods for such
analyses. Newer tools from computer science are now making
quantitative analysis of qualitative information possible. In this
study, we used tools from computational linguistics to quantita-
tively analyze the language people used to describe their spiritual,
religious, and mystical experiences. To analyze the writing, we
used two different approaches. First, we used a common tool for
“top-down” linguistic analysis: Linguistic Inquiry and Word Count
(LIWC), a computer-based program that counts the frequency of
words used from a predetermined set of lists or lexicons (Tausczik
& Pennebaker, 2010). The program includes 64 categories, with
words that were rationally assigned based on psychological theory,
linguistic knowledge, and common sense. For example, the “ex-
clusive” language category contains words such as “without,”
“either,” “except,” and “versus.” Diaries, blogs, homework assign-
ments, and research abstracts have all been analyzed using this
closed-vocabulary method (Pennebaker & King, 1999). Intuitive
results have been found, such as a study that showed that people
scoring highly on neuroticism used more negative emotion words
(Pennebaker & King, 1999). Less obvious results have also been
generated, such as older individuals use more positive emotion and
fewer self-reference words (Pennebaker & Stone, 2003).
Second, we used an open vocabulary (“bottom-up”) approach,
called Differential Language Analysis (DLA; Schwartz et al.,
2013), which identifies words and topics most correlated with
scores on a given scale or outcome measure. Unlike the closed,
top-down, a priori approach of LIWC, DLA is not limited by
preestablished categories and dictionaries, but rather allows the
data to dictate relevant words and topics. Whereas LIWC identifies
abstract categories of language, DLA provides a transparent view
into fine-grained patterns that distinguish individuals on a given
outcome.
We applied both closed (top-down) and open (bottom-up) vo-
cabulary approaches to written narratives of religious, spiritual,
and mystical experiences. We focused particularly on mystical
experiences, while bearing in mind the common qualification that
such experiences are ineffable. Based on what is currently under-
stood about mystical experiences, we expected that those individ-
uals who scored highest on a measure of mystical experiences
would describe their experiences using language indicative of
unity, as this quality has been most emphasized in theoretical,
qualitative, and empirical studies (Austin, 2006; Griffiths et al.,
2006; Hood, 2003; James, 1902/1985; Stace, 1960). However, the
analyses were also exploratory in nature, as our primary intention
was to apply computational linguistic analyses to the corpus of
narratives about religious, spiritual, and mystical experiences,
identifying key patterns for further study.
Method
Participants
We used a convenience sample of participants who found and
completed our online survey between 2008 and 2013, which was
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
2YADEN ET AL.
hosted on a website by the University of Pennsylvania under the
supervision of one of the coauthors (ABN). The University of
Pennsylvania’s Institutional Review Board approved the collection
and analyses of the data. Of 2,718 respondents who began the
survey, 1,986 completed the questionnaire. Of those, 1,182 were
excluded for having fewer than 25 words about their experience.
An additional 27 respondents were excluded due to incomplete
data, leaving a final sample size of 777.
The word count threshold was chosen to exclude participants
who did not adequately responded to the prompt. Responses con-
tained a median word count of 227 words per written response. We
chose 25 words as a balance between adequately sampling a
person’s language and keeping our total number of participants as
high as possible. Responses falling below this level were essen-
tially nonresponses that were qualitatively different than responses
with more than 25 words (e.g., “none,” “I will update this later”).
Participants were generally middle class (75.42%), Caucasian
(82.37%), and relatively balanced between females (45.30%) and
males (51.22%). Slightly fewer of excluded individuals were mid-
dle class (72.37%), female (40.36%) and Caucasian (78.66%). In
terms of religious affiliation, our sample was disproportionately
“other” (31.27%) and atheist (26.38%), compared to the U.S.
national averages of 14.9% and 1.6%, respectively (Pew Research
Center, 2008). Christians represented 19.69% (compared with the
national average of 78.4%). Other religions included Jewish, Bud-
dhist, Muslim, Hindu, Pagan, Native American, agnostic, and
Unitarian. Over half of the sample (56.89%) reported that they had
taken psychedelic drugs at some point in their lives. Comparing
those included versus those excluded (when data were available),
the main sample included significantly more atheists,
2
(1) ⫽
3.87, p⫽.049 and “other,”
2
(1) ⫽11.04, p⬍.001, than the
excluded sample, with no significant difference for Christians,
2
(1) ⫽0.02, p⫽.89. The main sample scored higher on mysti-
cism, 3.22 vs. 3.06, t(925) ⫽3.03, p⬍.01, d⫽.18. There were
no significant differences between the main and excluded sample
for socioeconomic status,
2
(2) ⫽3.49, p⫽.17, gender,
2
(1) ⫽
2.41, p⫽.12, or ethnicity,
2
(7) ⫽10.67, p⫽.15.
Measures
Participants completed a questionnaire that included a mystical
experience measure, demographic information, and several other
scales.
1
Participants were then asked to write about any variety of
spiritual or religious experiences they have had.
Mystical experience. The Death Transcendence Scale was
developed by Hood and Morris (1983) and is based on the theo-
retical work of Lifton (1976, 1979), who identified five modes of
“death transcendence.” These modes, which are represented as
subscales, consist of ways in which one can conceptualize oneself
as living on after bodily death through identification with some-
thing more enduring. From a third-person perspective, one can
“live on” in the hearts and minds of family and friends (biosocial
subscale), through the work one has produced (creative subscale),
by becoming a part of the natural world (nature subscale), or in a
personal afterlife (religious subscale).
The fifth mode, “mystical,” refers to a mystical experience of
unity, and is the only subscale that measures the occurrence of an
experience rather than a conceptualization of the self (Hood,
1976). The items are based on the M-Scale (Hood, 1975), which is
the most widely used measure of mystical experiences and has
demonstrated validity and reliability across a number of cross-
cultural samples, as well as across samples of varying religious
commitments (Hood et al., 2001). The current study included the
five items from the mysticism subscale, answered on a 4-point
Likert scale (1 ⫽strongly disagree,4⫽strongly agree;␣⫽.90).
Narrative descriptions. Participants were asked to
describe in detail the various spiritual and/or religious experiences
that you have had and how they have affected you. If you have had a
specific religious or spiritual experience(s), please describe it in as
much as detail as possible—as long or as short as you wish.
Data Analyses
We used two forms of linguistic analyses to quantitatively
analyze the narrative descriptions: a closed vocabulary (top-down)
approach and an open vocabulary (bottom-up) approach. For the
closed approach, we used the Linguistic Inquiry and Word Count
(LIWC) program, which includes 64 dictionaries
2
(Pennebaker,
Mehl, Niederhoffer, 2003). For every participant’s narrative de-
scription, the program counted the number of word instances that
belong to each of the 64 categories. These counts were then
divided by the total number of words, yielding a percentage of total
words for each category. For example, for one narrative, 7.3% of
the words used were in the exclusion dictionary, whereas for
another narrative, exclusion words made up only 2.2% of the total
words. Using the Pearson correlation coefficient, we then corre-
lated the percentage of use of each of the 64 categories with the
mystical experience subscale of the Death Transcendence Scale.
Because of the high number of correlations, pvalues were cor-
rected with Simes (1986) multiple test procedure to adjust for
multiple comparisons.
For the open vocabulary approach, we used Differential Lan-
guage Analysis (DLA; Schwartz et al., 2013). DLA consists of
three steps: (a) open-vocabulary language feature extraction; (b)
correlational analysis; (c) result sorting and visualization. In the
first step, a tokenizer split the sentences into single words. In the
second step, words were correlated with the mystical experience
score from the Death Transcendence Scale. We limited analyses to
words mentioned in at least 5% of the writings, resulting in 2,745
different words included in the analysis. In the third step, corre-
lations were sorted by magnitude and direction, and then visual-
ized as word clouds, which display the 100 words most strongly
correlated with mystical experiences. In the word clouds, the size
of the word corresponds to the strength of the correlation coeffi-
cient with the mysticism scale score, and color indicates the
frequency the word occurs. The word clouds were used descrip-
tively, so no adjustments were made for multiple comparisons.
1
Scales included The Quest Scale (Batson & Schoenrade, 1991), The
Religiousness Measure (Sethi & Seligman, 1993), The Intrinsic Religious
Motivation Scale (Hoge, 1972), and The Index of Core Spiritual Experi-
ences (INSPIRIT; Kass, Friedman, Lesserman, Zuttermeister, & Benson,
1991). As the focus of the current study was on mystical experiences, these
measures were not analyzed in this study.
2
Note that categories are not mutually exclusive. For example, the
words in the “anger” dictionary are also in the “negative emotions”
category.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
3
LANGUAGE OF INEFFABILITY
Results
Religious affiliation influenced the reported degree of mystical
experience. Monotheistic individuals (Christian, Jewish, Islamic)
scored lowest on the mystical experience measure, with eastern/
other religions (Buddhist, Hindu, Pagan, Secular Philosophies)
scoring highest and atheists in the middle (see Figure 1).
To examine the language of mystical experiences, we first
computed the correlations between LIWC categories and mystical
experience scores. Results, summarized in Table 1, indicated a few
highly significant findings of modest effect size. Participants with
higher mystical experience scores were more likely to use inclu-
sive language, r⫽.20, p⬍.001, using words like “we,” “with,”
“into,” and “and.” High scorers on the mystical experience mea-
sure used fewer third person singular words like “he,” “her,”
“him,” and “his”, r⫽⫺.13, p⬍.001. Additionally, participants
with higher mystical experience scores used significantly fewer
overtly religious words, r⫽⫺.20, p⬍.001.
We then used the open vocabulary approach. As summarized in
Table 2, across the sample as a whole, religious language is
prominent, with “God” being one of the most commonly used
words. However, mirroring the closed approach results, those who
have had mystical experiences used more inclusive language and
fewer religious words. see Figure 2 shows the words that most
distinguish high and low levels of mystical experience by listing
the 100 words most positively and negatively correlated with
mystical experience. For example, those scoring higher on the
mystical experience measure were more likely to mention “one-
ness,” “and,” and “everything,” whereas those on the low end were
more likely to mention “Christ,” “prayed,” and “monk.”
To rule out the possibility that our finding that monotheists tend
to score lower on the mystical experience scale was driving the
language results, we conducted a supplemental analysis on lan-
guage use, separately across religious affiliations. Based on self-
reported religion, participants were classified as monotheists (Is-
lamic, Jewish, Christian, n⫽194), atheists (n⫽205), and eastern/
other (Buddhist, Taoist, Hindu, Pagan, and idiosyncratic responses
that were written in to a text box; n⫽327). We correlated the
LIWC categories and mystical experience scores separately for the
three groups. Inclusive and religious language showed consistent
patterns of correlation across the groups; inclusive and religious
language remained significantly correlated with similar effect sizes
at conventional significant thresholds (without Simes correction).
Inclusive language was positively correlated with mystical expe-
riences across all categories of religions, including monotheists,
r⫽.18, p⫽.011, atheists, r⫽.26, p⫽⬍.001, and eastern/other,
r⫽.18, p⬍.001. The negative correlation with religious lan-
guage also held across monotheists, r⫽⫺.16, p⫽.027, atheists,
r⫽⫺.22, p⫽.002, and eastern/other, r⫽⫺.14, p⫽.012.
Discussion
Applying closed and open vocabulary analytic approaches to
narrative descriptions about religious or spiritual experiences, we
found that people who have had mystical experiences tended to use
more inclusive language (i.e., “everything,” “and,” “with,” “one-
ness”) and less third person singular language (i.e., “he,” “she,”
“him,” “her”). We also found that participants who had mystical
experiences use less religious language (i.e., “Christ,” “religious,”
“holy,” “hell”; Table 1 and Figure 2).
We expected participants who had mystical experiences to use
more language indicative of unity, which was partially supported
by the correlation with “inclusive” language. The finding lends
support to the primacy of the perception of unity in mystical
experiences (Austin, 2006; Griffiths et al., 2006; Hood, 2003;
James, 1902/1985; Stace, 1960). Unlike “exclusive” language,
which is associated with distinction making and separation, inclu-
sive language reflects joining or union on spatial and social
dimensions.
In research using LIWC, the inclusive language lexicon has
previously been organized under the “spatial” category, because of
words reflecting closeness with objects in the environment (Pen-
nebaker, & King, 1999). It has also been used to measure “social”
factors such as interpersonal attachment style, acknowledging that
Figure 1. Mystical experience scores across the different religious groups with more than 10 participants. Bars
indicate 95% confidence intervals. See the online article for the color version of this figure.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
4YADEN ET AL.
inclusive words could refer not only to closer environmental space
but also to closer perceived social space, and thus a kind of social
connection (Stone, 2003). Because of its role in joining distinct
objects and concepts, the inclusive language lexicon is now cate-
gorized as a “cognitive process,” alongside other lexica such as
insight and causation (Tausczik & Pennebaker, 2010). Thus, the
current inclusive language lexicon contains conceptual joining
words such as “plus,” “come,” and “add;” spatial words such as
“around,” “inside,” and “close” as well as social words such as
“we,” “with,” and “both.” In this context, the negative correlation
with third person singular language also makes sense, as words
like “we” trump distinctions like “her” or “him” from the perspec-
tive of unity.
Inclusive language is related to social connection and engage-
ment. For example, the inclusive language lexicon is associated
with secure attachment styles rather than preoccupied or dismis-
sive styles (Cassidy, Sherman, & Jones, 2012). Inclusive language
is also predictive of more social interaction on social networks
(Mahmud, Chen, & Nichols, 2014). Use of the word “we” has been
specifically tied to a strong sense of community, and is often
evoked after collective upheaval or tragedies (Pennebaker & Lay,
2002).
This association between inclusive language and social connec-
tion may help explain some of the mechanisms through which
mystical experiences promote prosocial intentions (Griffiths et al.,
2006; Griffiths, Richards, Johnson, McCann, & Jesse, 2008). Self-
expansion theory (Aron & Aron, 1997) postulates that one’s sense
of self can include other individuals and groups. The sense of unity
with other people, which is felt to an extreme extent during
mystical experiences, has been shown to increase prosocial atti-
tudes and behavior (Waugh & Fredrickson, 2006). For example,
Aron et al. (1991) showed that subjects asked to anonymously
distribute money do so about equally between themselves and
others who they feel a sense of unity with, but unequally to those
with whom they feel less unity. The language findings indicate that
a primary feature of mystical experiences may derive from per-
ceived social connection.
Other research has found links between mystical experiences
and spatial unity. The theory of “allo-inclusive identity” suggests
that self/other overlap can extend beyond other individuals or
groups of people to one’s spatial environment, perhaps including
all of existence (Leary, Tipsord, & Tate, 2008). Thus the descrip-
tion of feeling “at one with all things.” In neuroimaging research,
mystical experiences are associated with alterations in brain areas
related to representing spatial boundaries. For example, practices
like meditation and prayer that produce states of unity correlate
with inhibition in the posterior superior parietal region of the brain
(Newberg et al., 2001; Newberg, Pourdehnad, Alavi, & d’Aquili,
2003). This brain area is associated with representing bodily spa-
tial boundaries, including the usual distinction that separates one’s
sense of self from the rest of existence (Newberg et al., 2001).
The finding that people who have had mystical experiences use
fewer religion-specific words is interesting, as “God” was one of
the most frequently used words in the overall corpus (see Table 2).
Therefore, there may be something about mystical experiences that
does not lend itself to religious language. This is relevant to the
theoretical debate between perennialists and constructivists. Peren-
nialists, the “lumpers” of this debate, believe that there are com-
mon, universal aspects of mystical experiences (Hood, 2006),
whereas constructivists, the “splitters,” believe that cultural differ-
ences prohibit any such relation (Katz, 1978). The perennialist
perspective, more formally described as the “common core theory”
or “unity thesis,” posits that there are a subset of features in
mystical experiences that are shared, even though descriptions of
these experiences are heavily influenced by the sociocultural factors
(Hood, 2003). From this perspective, mystical experiences should be
reported in ways that share similar features despite the vast differ-
ences in socioculture contexts (Forman, 1998). On the other hand, the
constructivist or “diversity theory” view holds that the experiences
themselves are influenced by sociocultural factors, not just the de-
Table 1
Correlation Between LIWC Categories and the Mystical Experience Subscale
Category Words r
Simes-corrected
Significance threshold
Inclusive and, with, we, into 0.20 ⬍.001
Cognitive processes all, always, every, ever 0.11 ⬍.01
Third person singular he, she, his, her ⫺0.13 ⬍.001
Religion religion, soul, Christ, pastor ⫺0.20 ⬍.001
Note. Correlations are limited to categories passing a Simes multi-test significance test at p⬍.05. See http://www.liwc
.net/descriptiontable1.php for category summaries and reliability information.
Table 2
The 25 Most Frequently Mentioned Words Across the Narrative Descriptions (N ⫽777; 351,806 Words Total)
Word Count Word Count Word Count Word Count Word Count
God 1,421 Felt 898 Years 562 Things 478 First 415
Experience 1,273 Like 844 Feel 559 Feeling 457 Body 413
Life 1,135 Experiences 800 People 532 Way 429 Love 392
One 1,115 More 776 Know 487 Mind 427 Sense 390
Time 1,036 Spiritual 730 Believe 480 World 420 Church 383
Note. Excludes function words (e.g., pronouns, prepositions, auxiliary verbs, conjunctions, particles).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
5
LANGUAGE OF INEFFABILITY
scriptions of them (Katz, 1978). According to this view, sociocultural
contexts are not merely a linguistic gloss over a universal mystical
experience, as the perennialists would have it, but rather these cultural
influences are vital elements of the experience itself, prohibiting
universal generalizability (Proudfoot, 1985).
From the perennialist perspective, the negative correlation be-
tween mystical experiences and religious language may lend sup-
port to Hood and Chen’s (2005) claim that religious language is
invoked reluctantly and only when it is the only cultural touch-
stone available to describe the experience. From the constructivist
(or diversity theory) perspective, the current study may be criti-
cized in that the mystical experience measure that we used may
reflect a bias toward a particular belief system. For example, an
item such as “I have had an experience in which I realized the
oneness of myself with all things” may represent a worldview of
immanence (there is a divine in and through physical reality;
Levine, 2002), or monism (there is an underlying unity to reality
despite seeming distinctions; Schaffer, 2010). Most monotheists,
however, subscribe to the logical opposites of these views, tran-
scendence and dualism, respectively (Bielfeldt, 2001; Engebret-
son, 1996). The language that suggests immanence and monism in
the scale may therefore be alienating to individuals who hold
monotheistic worldviews. For example, Zaehner (1961) argues
that monotheists are interested in theistic union, not union with
nature or an impersonal, monistic absolute. While studies that have
tested different religious and cultural populations have generally
found the M-scale to be neutral (Chen, Yang, Hood, & Watson,
2011; Hood, 2006; Hood & Chen, 2013; Hood & Francis, 2013;
Hood et al., 2001; Hood & Williamson, 2000), we found signifi-
cantly lower endorsement of mystical experience among monothe-
Figure 2. Top 100 words that were correlated most positively (top) and negatively (bottom) with the mystical
experience sub-scale scores. Larger words indicate stronger correlations with mystical experience. Color
indicates frequency of word use among participants (grey ⫽infrequently used, blue ⫽moderately used, red ⫽
frequently used). All words were independently significantly correlated with mystical experience at p⬍.05 (as
a descriptive summary, significance values were not corrected for multiple comparisons).
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
6YADEN ET AL.
ists. Either way, by evaluating the language itself, our approach
can potentially provide additional empirical insight to both the
perennialist and constructivist perspectives.
Another implication of our findings derives from the high per-
centage of respondents who did not report a religious or even
spiritual commitment, yet nonetheless reported having mystical
experiences. This may point to a certain type of person, a “mystical
but not religious or spiritual” type. Similarly, there is a “spiritual
but not religious” type that emphasizes the experiential elements of
spirituality rather than the more institutionally affiliated, dogmatic
emphasis of religion (Zinnbauer et al., 1997). The “mystical but
not religious or spiritual” type identified in this study may go one
step farther than the “spiritual but not religious” type by empha-
sizing experiences to an even greater extent while downplaying or
rejecting religious institutions in addition to spiritual beliefs. The
“mystical but not religious or spiritual” type is apparent in recent
books by avowed atheists like Alain d’Botton’s (2012) Religion
for Atheists, Barbara Ehrenreich’s (2014) Living With a Wild God,
and Sam Harris’s (2014) Waking Up. The authors of these books
relate mystical-type experiences without accompanying religious
or even spiritual language or endorsement of these beliefs. Along
these general lines, some have promoted a “perennial psychology”
that holds the mystical experience itself as of primary importance,
over and above interpretive beliefs (Forman, 1998), though this
has rarely extended to atheism and other secular philosophies in
popular usage. The distinctions between religious, spiritual, and
secular beliefs and varieties of mystical experiences require further
scholarship and empirical research, both of which can be comple-
mented by linguistic analyses.
Limitations and Future Directions
Several limitations must be acknowledged. Although the mys-
tical experience subsection of the death transcendence scale that
we used was derived from the widely used M-Scale and the alpha
was relatively high (␣⫽.90), this subscale has not received
extensive psychometric validity testing. There may also be a
perennialist bias to the scale, creating an interaction between the
religious beliefs of participants and their responses to the scale.
Future research on this scale should continue to explore response
characteristics from diverse samples and be compared with scales
capturing differing but related varieties of experiences. A value of
the data-driven linguistic analysis approach is that alternative
measures and concepts can be explored, in turn informing the
questionnaires and scales used to measure a given construct.
There was also a selection bias to our sample, as the individuals
who visited our survey website might have been more likely to report
more religious, spiritual, and mystical experiences, produce certain
types of narratives, or use specific terminology. The demographic that
we sampled was, in general, Caucasian and middle class, and around
half the sample was nonreligious. Therefore, care should be taken in
generalizing results to other populations. It is likely that a larger
sample recruited using different mechanisms might yield different
results. Although our supplemental analyses did show that our lan-
guage findings hold across all religious affiliations, the presence of so
many nonreligiously affiliated individuals limits generalizability. Al-
though results from the linguistic analysis techniques that we used
have been interesting and at times impressive (e.g., Eichstaedt et al.,
2015; Pennebaker et al., 2003; Schwartz et al., 2013; Tausczik &
Pennebaker, 2010), correlations with content categories such as those
found in this study are often quite modest. So although the findings in
this study include reliable trends and can usefully point to underlying
social and psychological processes, these results account for only a
rather small percentage of the total variance in the language used to
describe religious, spiritual, and mystical experiences.
Some scholars also critique the lack of appreciation for nuance,
ignorance of word sense, and computational complexity endemic
to these methods. Although these quantitative techniques find
replicable patterns in large amounts of data, they are not able to
produce the kind of nuance garnered from a grounded theory
approach or other in-depth qualitative methods. However, findings
from these quantitative approaches can provide a rich source of
hypothesis generation, which other qualitative and quantitative
approaches can explore in greater detail.
Linguistic analysis nonetheless offers an important tool for
studying religious, spiritual, and mystical experiences. In particu-
lar, these methods are capable of differentiating between mental states
based on language use. For example one linguistic analysis study
successfully identified the specific psychoactive drug that an individ-
ual had been administered before writing (Bedi et al., 2014). It is our
hope that similar approaches are applied to differentiate various re-
lated types of experiences (e.g., mystical, numinous, self-
transcendent) in large-scale, cross-cultural, and historical corpora, to
empirically derive the unique features of these varieties of experi-
ences. Furthermore, the successful identification of feelings of unity
through language provides a new tool to explore potential relation-
ships between perceived unity and other prosocial, well-being, and
health outcomes.
Conclusion
Mystical experiences are notoriously difficult to put into words,
leading people to rarely discuss them, and earning them the label
“ineffable” (Hay, 1990; James, 1902/1985). Perhaps it is as Melville’s
(1851) Ishmael claims, “For whatever is truly wondrous and fearful in
man, never yet was put into words” (p. 447). And yet linguistic
descriptions of these supposedly unspeakable experiences exist and
through their analysis we may discover they have some stable, un-
derlying features. The results of the present study, that people who
have mystical experiences describe them using more inclusive lan-
guage and fewer religious words, may be one such telling feature. The
profoundly positive significance that mystical experiences often hold
for those who have them underscores the scientific value of investi-
gating these allegedly indescribable experiences. This study repre-
sents an attempt, using linguistic analysis tools, to make headway into
“effing” the ineffable by increasing our empirical understanding of
mystical experiences.
References
Aron, A., & Aron, E. N. (1997). Self-expansion motivation and including
other in the self. In S. Duck (Ed.), Handbook of personal relationships:
Theory, research, and interventions (2nd ed., pp. 251–270). Chichester,
England: Wiley.
Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships
as including other in the self. Journal of Personality and Social Psy-
chology, 60, 241–253. http://dx.doi.org/10.1037/0022-3514.60.2.241
Augustine, S. (1958). On Christian doctrine (D. W. Robertson, Jr. Trans.).
Indianapolis, IN: Bobbs-Merrill.
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
7
LANGUAGE OF INEFFABILITY
Austin, J. H. (2006). Zen-brain reflections. Cambridge, MA: MIT Press.
Batson, C. D., & Schoenrade, P. (1991). Measuring religion as a quest: 2.
Reliability concerns. Journal for the Scientific Study of Religion, 30,
430 – 447. http://dx.doi.org/10.2307/1387278
Bedi, G., Cecchi, G. A., Slezak, D. F., Carrillo, F., Sigman, M., & de Wit,
H. (2014). A window into the intoxicated mind? Speech as an index of
psychoactive drug effects. Neuropsychopharmacology, 39, 2340 –2348.
http://dx.doi.org/10.1038/npp.2014.80
Bielfeldt, D. (2001). Can western monotheism avoid substance dualism?
Zygon, 36, 153–177. http://dx.doi.org/10.1111/0591-2385.00345
Cassidy, J., Sherman, L. J., & Jones, J. D. (2012). What’s in a word?
Linguistic characteristics of Adult Attachment Interviews. Attachment &
Human Development, 14, 11–32. http://dx.doi.org/10.1080/14616734
.2012.636649
Chen, Z., Yang, L., Hood, R. W., Jr., & Watson, P. J. (2011). Mystical
experience in Tibetan Buddhists: The common core thesis revisited,
Journal for the Scientific Study of Religion, 50, 328 –338.
D’Botton, A. (2012). Religion for atheists. New York, NY: Pantheon.
Durkheim, E. (2012). The elementary forms of the religious life. Mineola,
NY: Courier Dover Publications. (Original work published 1912)
Ehrenreich, B. (2014). Living with a wild God: A non-believer’s search for
the truth about everything. New York, NY: Hachette Book Group.
Eichstaedt, J. C., Schwartz, H. A., Kern, M. L., Park, G., Labarthe, D. R.,
Merchant, R. M.,...Seligman, M. E. P. (2015). Psychological language
on Twitter predicts county-level heart disease mortality. Psychological
Science, 26, 159 –169. http://dx.doi.org/10.1177/0956797614557867
Engebretson, J. (1996). Considerations in diagnosing in the spiritual do-
main. International Journal of Nursing Terminologies and Classifica-
tions, 7, 100 –107. http://dx.doi.org/10.1111/j.1744-618X.1996
.tb00301.x
Forman, R. K. C. (1998). The innate capacity: Mysticism, psychology, and
philosophy. New York, NY: Oxford University Press.
Gallup, G., Jr. (1978). The Gallup Poll: Public opinion 1972–1977. Wash-
ington, DC: Scholarly Resources.
Griffiths, R., Richards, W., Johnson, M., McCann, U., & Jesse, R. (2008).
Mystical-type experiences occasioned by psilocybin mediate the attri-
bution of personal meaning and spiritual significance 14 months later.
Journal of Psychopharmacology, 22, 621– 632. http://dx.doi.org/
10.1177/0269881108094300
Griffiths, R. R., Richards, W. A., McCann, U., & Jesse, R. (2006).
Psilocybin can occasion mystical-type experiences having substantial
and sustained personal meaning and spiritual significance. Psychophar-
macology, 187, 268 –283. http://dx.doi.org/10.1007/s00213-006-0457-5
Hardy, A. (1979). The spiritual nature of man: A study of contemporary
religious experience. Oxford, UK: Clarendon Press.
Harris, S. (2014). Waking up: A guide to spirituality without religion. New
York, NY: Simon & Schuster.
Hay, D. (1979). Religious experience amongst a group of postgraduate
students: A qualitative study. Journal for the Scientific Study of Religion,
18, 164 –182. http://dx.doi.org/10.2307/1385938
Hay, D. (1990). Religious experience today: Studying the facts. London,
England: Mowbray.
Hay, D., & Morisy, A. (1985). Secular society, religious meanings: A
contemporary paradox. Review of Religious Research, 26, 213–227.
http://dx.doi.org/10.2307/3511273
Hoge, D. R. (1972). A validated intrinsic religious motivation scale.
Journal for the Scientific Study of Religion, 11, 369 –376. http://dx.doi
.org/10.2307/1384677
Hood, R. W., Jr. (1975). The construction and preliminary validation of a
measure of reported mystical experience. Journal for the Scientific Study
of Religion, 14, 29 – 41. http://dx.doi.org/10.2307/1384454
Hood, R. W., Jr. (1976). Conceptual criticisms of regressive explanations
of mysticism. Review of Religious Research, 17, 179 –188. http://dx.doi
.org/10.2307/3510609
Hood, R. W., Jr. (2003). Conceptual and empirical consequences of the
unity thesis. International Series in the Psychology of Religion, 13,
17–54.
Hood, R. W., Jr. (2006). The common core thesis in the study of mysticism.
In P. McNamar (Ed.), Where God and science meet (Vol. 3, pp. 119 –
138). Westport, CT: Praeger.
Hood, R. W., Jr., & Chen, Z. (2005). Mystical, spiritual, and religious
experiences. In R. Paloutzian & C. Park (Eds.), Handbook of the psy-
chology of religion and spirituality (pp. 422– 440). New York, NY:
Guilford Press.
Hood, R. W., Jr., & Chen, Z. (2013). The social sciences and Christian
mysticism. In J. Laman (Ed.), The Oxford handbook of mysticism (pp.
577–591). New York, NY: Oxford University Press.
Hood, R. W., Jr., & Francis, L. J. (2013). Mystical experience: Concep-
tualizations, measurement, and correlates. In K. I. Pargament (Ed.),
Handbook of the psychology of religion and spirituality: Vol. 1.Context,
theory, & research (pp. 391– 405). Washington, DC: American Psycho-
logical Association.
Hood, R. W., Jr., Ghorbani, N., Watson, P. J., Ghramaleki, A. F., Bing,
M. N., Davison, H. K.,...Williamson, W. P. (2001). Dimensions of the
Mysticism Scale: Confirming the three-factor structure in the United
States and Iran. Journal for the Scientific Study of Religion, 40, 691–705.
http://dx.doi.org/10.1111/0021-8294.00085
Hood, R. W., Jr., Hill, P. C., & Spilka, B. (2009). The psychology of
religion: An empirical approach. New York, NY: Guilford Press.
Hood, R. W., Jr., & Morris, R. J. (1983). Toward a theory of death
transcendence. Journal for the Scientific Study of Religion, 22, 353–365.
http://dx.doi.org/10.2307/1385773
Hood, R. W., Jr., & Williamson, W. P. (2000). An empirical test of the
unity thesis: The structure of mystical descriptors in various faith sam-
ples. Journal of Psychology and Christianity, 19, 232–244.
James, W. (1985). The varieties of religious experience (Vol. 13). Cam-
bridge, MA: Harvard University Press. (Original work published 1902)
Kass, J. D., Friedman, R., Lesserman, J., Zuttermeister, P., & Benson, H.
(1991). Health outcomes and a new index of spiritual experience. Jour-
nal for the Scientific Study of Religion, 30, 203–211. http://dx.doi.org/
10.2307/1387214
Katz, S. (1978). Mysticism and philosophical analysis. New York, NY:
Oxford University Press.
Keller, B., Klein, C., Swhajor-Biesemann, A., Silver, C. F., Hood, R., &
Streib, H. (2013). The semantics of “spirituality” and related self iden-
tifications: A comparative study in Germany and the USA. Archiv für
Religionspsychologie / Archive for the Psychology of Religions, 35,
71–100. http://dx.doi.org/10.1163/15736121-12341254
Laski, M. (1961). Ecstasy in secular and religious experiences. Los An-
geles, CA: Tarcher.
Leary, M. R., Tipsord, J., & Tate, E. B. (2008). Allo-inclusive identity:
Incorporating the natural and social worlds into one’s sense of self. In H.
Wayment & J. Bauer (Eds.), Transcending self-interest: Psychological
explorations of the quiet ego (pp. 137–147). Washington, DC: American
Psychological Association. http://dx.doi.org/10.1037/11771-013
Levine, M. P. (2002). Pantheism: A non-theistic concept of deity. New
York, NY: Routledge.
Lifton, R. J. (1976). The life of the self: Toward a new psychology. New
York, NY: Simon & Schuster.
Lifton, R. J. (1979). The broken connection. New York, NY: Simon &
Schuster.
Mahmud, J., Chen, J., & Nichols, J. (2014). Why are you more engaged?
Predicting social engagement from word use. arXiv Preprint
(arXiv:1402.6690).
Maslow, A. H. (1964). Religions, values, and peak-experiences (Vol. 35).
Columbus, OH: Ohio State University Press.
Melville, H. (1988). Moby-Dick; or, The whale. Evanston and Chicago, IL:
Northwestern University Press. (Original work published 1851)
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
8YADEN ET AL.
Mitchell, S. (Trans.). (1988). Tao te Ching. New York, NY: Harper
Collins.
Newberg, A., Alavi, A., Baime, M., Pourdehnad, M., Santanna, J., &
d’Aquili, E. (2001). The measurement of regional cerebral blood flow
during the complex cognitive task of meditation: A preliminary SPECT
study. Psychiatry Research, 106, 113–122. http://dx.doi.org/10.1016/
S0925-4927(01)00074-9
Newberg, A., Pourdehnad, M., Alavi, A., & d’Aquili, E. G. (2003).
Cerebral blood flow during meditative prayer: Preliminary findings and
methodological issues. Perceptual and Motor Skills, 97, 625– 630. http://
dx.doi.org/10.2466/pms.2003.97.2.625
Oxman, T. E., Rosenberg, S. D., Schnurr, P. P., Tucker, G. J., & Gala, G.
(1988). The language of altered states. Journal of Nervous and Mental
Disease, 176, 401– 408. http://dx.doi.org/10.1097/00005053-
198807000-00002
Pahnke, W. N. (1969). Psychedelic drugs and mystical experience. Inter-
national Psychiatry Clinics, 5, 149 –162.
Pennebaker, J. W., & King, L. A. (1999). Linguistic styles: Language use
as an individual difference. Journal of Personality and Social Psychol-
ogy, 77, 1296 –1312. http://dx.doi.org/10.1037/0022-3514.77.6.1296
Pennebaker, J. W., & Lay, T. C. (2002). Language use and personality
during crises: Analyses of Mayor Rudolph Giuliani’s press conferences.
Journal of Research in Personality, 36, 271–282. http://dx.doi.org/
10.1006/jrpe.2002.2349
Pennebaker, J. W., Mehl, M. R., & Niederhoffer, K. G. (2003). Psycho-
logical aspects of natural language use: Our words, our selves. Annual
Review of Psychology, 54, 547–577. http://dx.doi.org/10.1146/annurev
.psych.54.101601.145041
Pennebaker, J. W., & Stone, L. D. (2003). Words of wisdom: Language use
over the life span. Journal of Personality and Social Psychology, 85,
291–301. http://dx.doi.org/10.1037/0022-3514.85.2.291
Pew Research Center. (2008). U.S. Religious Landscape Survey. Retrieved
from http://religions.pewforum.org/pdf/report-religious-landscape-
study-full.pdf
Proudfoot, W. (1985). Religious experience. Berkeley, CA: University of
California Press.
Richards, W. A. (1975). Counseling, peak experiences and the human
encounter with death: An empirical study of the efficacy of DPT-assisted
counseling in enhancing the quality of life of persons with terminal
cancer and their closest family members (Unpublished doctoral thesis).
Catholic University of America, Washington, DC.
Schaffer, J. (2010). Monism: The priority of the whole. The Philosophical
Review, 119, 31–76. http://dx.doi.org/10.1215/00318108-2009-025
Scharfstein, B. A. (1973). Mystical experience. Indianapolis, IN: Bobbs-
Merrill.
Schwartz, H. A., Eichstaedt, J. C., Kern, M. L., Dziurzynski, L., Ramones,
S. M., Agrawal, M.,...Ungar, L. H. (2013). Personality, gender, and
age in the language of social media: The open-vocabulary approach.
PLoS ONE, 8, e73791. http://dx.doi.org/10.1371/journal.pone.0073791
Sethi, S., & Seligman, M. E. P. (1993). Optimism and fundamentalism.
Psychological Science, 4, 256 –259. http://dx.doi.org/10.1111/j.1467-
9280.1993.tb00271.x
Simes, R. J. (1986). An improved Bonferroni procedure for multiple tests
of significance. Biometrika, 73, 751–754. http://dx.doi.org/10.1093/
biomet/73.3.751
Smith, H. (1964). Do drugs have religious import? The Journal of Philos-
ophy, 61, 517–530. http://dx.doi.org/10.2307/2023494
Stace, W. T. (1960). Mysticism and philosophy. Philadelphia, PA: Lippin-
cott.
Stone, L. (2003). Using language to detect and change attachment style
(Unpublished Masters dissertation). University of Texas, Austin, TX.
Tamminen, K. (1991). Religious development in childhood and youth: An
empirical study. Helsinki, Finland: Suomen Tiedeakatemia.
Tausczik, Y. R., & Pennebaker, J. W. (2010). The psychological meaning
of words: LIWC and computerized text analysis methods. Journal of
Language and Social Psychology, 29, 24 –54. http://dx.doi.org/10.1177/
0261927X09351676
Thomas, L. E., & Cooper, P. E. (1978). Measurement and incidence of
mystical experiences: An exploratory study. Journal for the Scientific
Study of Religion, 17, 433– 437. http://dx.doi.org/10.2307/1385407
Waugh, C. E., & Fredrickson, B. L. (2006). Nice to know you: Positive
emotions, self-other overlap, and complex understanding in the forma-
tion of a new relationship. The Journal of Positive Psychology, 1,
93–106. http://dx.doi.org/10.1080/17439760500510569
Wulff, D. M. (2000). Mystical experience. In E. E. Cardeña, S. J. E. Lynn,
& S. E. Krippner (Eds.), Varieties of anomalous experience: Examining
the scientific evidence (pp. 369 – 408). Washington, DC: American Psy-
chological Association. http://dx.doi.org/10.1037/10371-012
Zaehner, R. C. (1961). Mysticism sacred and profane: An inquiry into some
varieties of praeternatural experience (Vol. 56). New York, NY: Oxford
University Press.
Zinnbauer, B. J., Pargament, K. I., Cole, B., Rye, M. S., Butter, E. M.,
Belavich, T. G.,...Kadar, J. L. (1997). Religion and spirituality:
Unfuzzying the fuzzy. Journal for the Scientific Study of Religion, 36,
549 –564. http://dx.doi.org/10.2307/1387689
Received November 6, 2014
Revision received May 13, 2015
Accepted June 16, 2015 䡲
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
9
LANGUAGE OF INEFFABILITY
