Abstract
The purpose of the present study is to gain clarity into the definition of religious fundamentalism. Specifically, the goal of the current research is to engage in a multi-method study defining the various aspects of fundamentalist movements across cultures and religions. Social scientists from around the world with recent peer-reviewed journal publications concerning religion and fundamentalism (n = 125) complete a survey where they define the construct and rate how representative 29 items distilled from the existing literature represent the term. Quantitative analyses reveal the scientists find Williamson et al.’s intratextual religious beliefs as the most important aspects of religious fundamentalism. Qualitative analyses reveal no theme is prevalent among the majority of scientists, with one category noted by 40 percent of participants, six by 20 percent to 39 percent of respondents, seven by 10 percent to 19 percent, and four reported in 7 percent to 9 percent of the sample. I integrate findings within existing religious fundamentalism theories and provide research suggestions with the purpose of creating a more comprehensive measure of fundamentalism that could be tailored to different cultural and religious contexts and used in future investigations to help applied practitioners better understand and interact with fundamentalists.
Religious fundamentalism is a relevant construct in today’s world. Recently, the Taliban retook Afghanistan (Maizland 2023), Buddhist fundamentalists assaulted Muslims in East Asia (Ramakrishna 2021), and terrorist groups such as al-Qaeda and the Islamic State of Iraq and Syria continue to pose a threat (Anna 2023). Citizens of Saudi Arabia, Egypt, Pakistan, Iran, and Malaysia reported elevated levels of religious fundamentalism (Ellis 2017; Moaddel and Karabenick 2021; Rothschild, Abdollahi, and Pyszczynski 2009). Religious fundamentalism was one of the primary predictors in identifying Americans who voted for President Trump in 2016 (Smith and Hanley 2018).
Thus, it was no surprise that many scientists studied religious fundamentalism. However, a limitation in the field was the variety of definitions (see Phillips and Kitchens 2021, for a recent review), which led to different forms of measurement, which, if context was not considered, could lead to the inappropriate use of scales within some religions and cultures. For example, Altemeyer and Hunsberger’s (1992, 2004) religious fundamentalism scales, though intended to be relevant across religions and cultures, were criticized (Hood, Hill, and Williamson 2005; Liht et al. 2011; Moaddel and Karabenick 2021) as best suited for Christians and Western cultures yet were used in Eastern and non-Christian religions.
Further, religious fundamentalism definitions and scales were critiqued as over-sampling some aspects and under-sampling other aspects of the construct. Take Altemeyer and Hunsberger’s scales once again; they were accused of oversampling fundamentalist features of authoritarianism, militance (Williamson et al. 2010), and the literal interpretation and inerrancy of religious teachings (Moaddel and Karabenick 2021). However, even those who critiqued Altemeyer and Hunsberger and created purportedly context-free, cross-cultural religious fundamentalism definitions, had themselves been critiqued for focusing on a limited number of fundamentalist features (Phillips and Kitchens 2021).
In an attempt to deal with these issues, a recent literature review integrated the variety of fundamentalism definitions and found 16 major components and further subcomponents (Phillips and Kitchens 2021). The authors worked within the assumption of a family resemblances framework, as had other social scientists who took a cross-cultural approach to studying fundamentalism (Almond, Appleby, and Sivan 2003; Herriot 2007). Just as a family contained similar traits—but no members had all the same characteristics—fundamentalisms from different contexts around the world may have had an overarching set of elements, but no two fundamentalisms would have had all the same components.
The present study also worked within a family resemblances model, with the goal of discovering the various commonly occurring elements of fundamentalism across different contexts (religions and cultures). The Phillips and Kitchens (2021) literature review served as an important first step in construct definition (Sireci 1998; Zamanzadeh et al. 2015). The next step would be to utilize multi-method procedures to empirically examine the definitions of fundamentalism in those who studied the construct (Koller, Levenson, and Glück 2017; Phillips and Kitchens 2021; Sireci 1998; Zamanzadeh et al. 2015)—the purpose of the current study. Thus, the present research asked social scientists who recently published peer-reviewed academic journal articles on religion and fundamentalism how they defined the term, as well as how much they thought each of 29 aspects identified by Phillips and Kitchens (2021) were representative of the construct. Fundamentalism appeared differently in various contexts (religions and cultures); however, the question was, how much do social scientists who studied the construct think each of the 29 features was relevant to fundamentalisms across the world? The goal was to move one step closer to a comprehensive measure of religious fundamentalism, making sure to not over-sample or under-sample certain features, as had been done in prior definitions. A comprehensive measure could then be tweaked within each contextual situation (e.g., factor analyses could eliminate irrelevant items within particular contexts), but an overarching understanding of the aspects of fundamentalism would remain. Such a comprehensive, cross-cultural measure could be used by applied practitioners such as religious leaders, grant funders, service providers, and others who interact with fundamentalists to better understand this important group across the world.
Multiple Conceptualizations of Religious Fundamentalism
The present research began with a review of the definitions of religious fundamentalism in the social science of religion. Some scientists took a contextual approach, applying their definition of fundamentalism only to certain groups (e.g., North American Protestants, Abrahamic religions), whereas other researchers took a universal approach within a family resemblances framework, examining fundamentalist traits that existed across cultures, while recognizing that no one feature was essential, and no two fundamentalisms had the exact same traits. In either case, the different features of the construct were presented below, and, in the process, provided evidence for the family resemblances framework from which the current paper operated.
Perhaps the most-cited definition of religious fundamentalism was provided by Altemeyer and Hunsberger (2004:48): “the belief that there is one set of religious teachings that clearly contains the fundamental, basic, intrinsic, essential, inerrant truth about humanity and deity; that this essential truth is fundamentally opposed by forces of evil which must be vigorously fought; that this truth must be followed today according to the fundamental, unchangeable practices of the past; and that those who believe and follow these fundamental teachings have a special relationship with the deity.” Altemeyer and Hunsberger also described religious fundamentalism as a right-wing authoritarian response to the religious impulse, defining authoritarianism as (1) submission to authority, (2) authoritarian aggression (strict, punitive social control), and (3) endorsement of traditional norms.
Hood et al. (2005) published a book describing a new theory concerning fundamentalism—intratextualism—meaning that fundamentalists identified truth only through the worldview of their sacred text, in contrast to intertextual approaches, where other sources (other religions’ texts and scientific scholarship) offered insight into truth even if they differed from the sacred text. Hood et al. noted limitations to Altemeyer and Hunsberger’s definition—for example, too much focus on militance, and item content biased toward Christians. Hood et al. identified six aspects of intratextualism within the Abrahamic faiths, the first four of which were also in Altemeyer and Hunsberger’s definition (although the latter addressed religious teachings and traditions more generally)—the sacred text was inerrant, unchanging (eternally relevant), privileged over other religions’ texts, authoritative as compared to other sources of knowledge (e.g., science), of divine origin, and self-interpretive (i.e., anyone could understand the text without need for hermeneutics). Williamson et al. (2010) created the five-item Intratextual Fundamentalism Scale (the self-interpretive item did not load on factor analyses) to assess intratextualism and presented evidence of scale reliability and validity in U.S. Christians and Pakistani Muslims. To round out the intratextual assessment in the Abrahamic faiths, Hammer and Lazar (2019) provided psychometric data for the scale in Israeli Jews. Nonetheless, the Intratextual Fundamentalism Scale was strongly correlated with Altemeyer and Hunsberger’s measure (correlations between .70 and .83 in Williamson et al. 2010), indicating that though the former assessed components of religious fundamentalism that were different (e.g., less militant) than the latter, both scales appeared to be measuring the same construct. Perhaps these scales tapped into some similar—and some different—aspects of fundamentalism.
Similar to Williamson et al., Liht et al. (2011) attempted to define religious fundamentalism in the Abrahamic faiths, and included the privileged, authoritative, and eternal features of Williamson et al. (though more broadly tied to religious traditions and rituals than sacred texts in particular), as well as aspects important to Altemeyer and Hunsberger (e.g., submission to authority). Liht et al. provided additional fundamentalist features, such as the prevalence of religion across an individual’s life, the integration of religion into public society (including government), viewing secular culture as a threat, and belief in end-of-days prophecy with a coming messiah. Illustrating the cross-cultural, family resemblances model of fundamentalism, end-of-days prophecy was also prevalent in Shiite Muslim, Jew, and Sikh fundamentalists, but rarely mentioned in their Sunni Muslim and Buddhist counterparts (Almond et al. 2003; Phillips and Kitchens 2021).
Like Liht, Williamson, and colleagues, Moaddel and Karabenick (2021) defined fundamentalisms within the Abrahamic faiths. Their definition included four components—three of which were found in prior definitions—authoritarian aggression, the inerrant and privileged status of the religion—and added a distinct feature—intolerance toward others.
In their definition of North American Protestant fundamentalism, Deal and Bartoszuk (2014) identified other fundamentalist features, such as evangelism and boundary-setting (delineating the true believers from everyone else). Illustrative of the cross-cultural family resemblances approach, evangelism was also commonly noted in fundamentalist Israeli Jews, Malaysian Muslims, Indian Hindus, and Buddhist Dhammakaya, but less described in Indian Sikh and Middle Eastern Shiite fundamentalists (Phillips and Kitchens 2021).
The variety of religious fundamentalism definitions, each with components that sometimes overlapped and sometimes did not, lent itself to the need for an overarching definition (Sireci 1998; Zamanzadeh et al. 2015). Phillips and Kitchens (2021) recently provided a comprehensive definition in their literature review. They reviewed definitions and theories concerning fundamentalism, conducted a content analysis of 10 religious fundamentalism measures, and identified 16 components of the construct. The authors suggested that future research empirically test their model by asking those who study fundamentalism the meaning of the construct. Such an approach was noted by other researchers working toward construct definition (Koller et al. 2017; Sireci 1998; Zamanzadeh et al. 2015). The goal of the current research was to carry out such a study.
In summary, there was a lack of agreement in peer-reviewed social scientific research concerning construct definitions of religious fundamentalism. The present study, like other cross-cultural research concerning fundamentalism (Almond et al. 2003; Phillips and Kitchens 2021), worked from a family resemblances approach, and assumed fundamentalisms across religions and cultures shared certain features. However, no two fundamentalisms had the same traits, given the unique context of each. Thus, the current study was the next step toward domain definition, a multi-method research project that explored the various aspects of religious fundamentalism.
The Present Study
The goal of the present study was to empirically examine how social scientists who studied religious fundamentalism defined the construct, through qualitative and quantitative methods. Quantitatively, participants were asked how well each of 29 different components (16 domains and 13 subdomains) identified in the social scientific literature by Phillips and Kitchens (2021) represented how they defined fundamentalism. Qualitatively, thematic and content analysis were performed on respondents’ definitions of religious fundamentalism.
Methods
Participants
Participants included 125 social scientists, of whom 113 held a doctorate, 6 a master’s, 1 a bachelor’s degree. (Unless otherwise indicated, missing data included 5 cases.) Most (105) participants described their field of study as some form of psychology; the most common subdisciplines were social (44), clinical/counseling (22), psychology of religion (9), personality (5), and neuroscience (4). Twelve noted their field of study as psychology in general. Ten other fields were noted by one or two participants (some participants listed more than one subfield): sociology (9), philosophy (3), history (2), psychiatry, public health, mass media, demography, religious studies, social work, criminology, and management (one each).
Most participants (89) were male, with 99 of European ancestry, 11 Asian ancestry, 4 Middle Eastern, 2 African, 2 mixed race, and 1 Hispanic (6 cases of missing data). Region of residence included the U.S. (87 participants), Europe (19), Israel/Middle East (7), Southeast Asia (4), and Australia (3). Ages ranged from 25 to 92 (M = 51.1, SD = 15.0). Forty-six identified as Christian, 8 as Jewish, 7 as Muslim, 5 with other religions, and 49 reported they were atheist, agnostic, or had no religious affiliation (10 did not answer this item). Seven identified as fundamentalist.
Measures
Participants completed a demographic questionnaire that asked their age, gender, race, religion, level of education, field of study, and whether they were a religious fundamentalist. Participants were then asked to define religious fundamentalism and clarify how it is different from being religious.
Next, participants were asked to rate how representative 29 items (presented in Table 1) were of their definition of religious fundamentalism (1 = Not at All; 5 = Extremely). Two researchers familiar with religious fundamentalism literature (having 1–2 publications concerning fundamentalism within the past five years) created the items, based on Phillips and Kitchens (2021) review of fundamentalism definitions and scales. Items were revised after discussions with two self-identified lay fundamentalists.
Scientist (n = 117–120) Views of Items’ Representation of Religious Fundamentalism.
Procedures
After the author’s university Institutional Review Board deemed the study exempt from review (PRO18080606), items were placed in an online survey (Qualtrics). To identify scientists who studied fundamentalism, a literature search was conducted on August 1, 2020, and again on July 17, 2022, in PsycInfo—the largest psychological repository, containing over 5 million records (American Psychological Association [APA] 2024). The search entailed fundamentalism as a keyword, subject, or within the “Tests and Measures” specifier along with the limiters “English language” and “peer-reviewed journal” published since 01/01/2010 and resulted in 405 articles; 41 were removed because they were a book review or unrelated to fundamentalism. The 364 remaining articles contained 760 authors.
To qualify for the study, authors had to meet one of two conditions: (1) published at least two peer-reviewed academic journal articles with the word fundamentalism anywhere in the PsycInfo record since 01/01/2010 (roughly 10 years from the time of the present study’s original literature search); or (2) published at least one peer-reviewed article meeting the aforementioned standard and at least one additional peer-reviewed academic journal article with the words religion/religiosity/religious/spiritual anywhere in the PsycInfo record since 01/01/2015. The peer-reviewed criterion ensured high quality science. The criterion of two recent fundamentalism publications provided a minimum pattern of research activity to suggest participants were knowledgeable about the construct. The second criterion was created because the current study wished to assess neither the small number of foremost religious fundamentalism experts (e.g., 10–20) nor all social scientists of religion (who might have only cursory familiarity with the construct), but a “concentric circle” between these two—researchers of religion with significant knowledge of religious fundamentalism (at least one recent peer-reviewed academic article on the construct and another in the general religion field). The second criterion also ensured a large enough sample size to obtain a confidence interval of the population mean for the quantitative items in the survey.
In all, 386 authors met the above criteria. Secondary references within the 364 articles resulted in an additional eight peer-reviewed journal articles with a fundamentalism measure since 1/1/2010, containing a further nine authors who met the above criteria. Nineteen more potential participants were added based on (1) prominent work concerning religious fundamentalism prior to 2010 (Altemeyer, Hunsberger, Lee Kirkpatrick, Sam McFarland), (2) prominent books published on religious fundamentalism (Gabriel Almond, Scott Appleby, Lawrence Davidson, Peter Herriot, Martin Marty, Dimitrios Methenitis, Emmanuel Sivan), and (3) a single peer-reviewed journal article published since 01/01/2010 that analyzed a new fundamentalism scale (Liht et al. 2011; Rouse et al. 2019; Williamson et al. 2010). Overall, 414 potential participants were identified, with three known deceased and no contact information for 5, leaving 406 contacted via email (383) or other means (23 via ResearchGate, Academia.edu). Of those contacted, the average number of peer-reviewed journal articles concerning fundamentalism between 01/01/2010 and 07/17/2022 was 1.69 (SD = 1.24), with 157 having at least two fundamentalism publications in that timespan, and another 232 meeting the second criterion—1 fundamentalism publication since 2010 and an average of 4.18 (SD = 5.29) publications concerning religion/spirituality since 01/01/2015. Authors were contacted twice over four months between August 2020 and December 2022. The response rate was 30.8 percent (125 completed the survey of the 406 contacted), above the minimum 20 percent to 25 percent required for reliable estimates in sample sizes below 500 (Fosnacht et al. 2017).
Concerning qualitative analyses, three coders reviewed the data for themes, using the principles of grounded theory (Charmaz 2006). One of the coders (the author) has a Ph.D. in psychology and was knowledgeable of research on fundamentalism, which was used to help guide the creation of themes, while remaining open to new definitions. The other coders were undergraduate psychology majors without prior knowledge of religious fundamentalism research.
After reviewing the data, the coders discussed and agreed on themes, defined categories, and identified example quotes. Content analysis was conducted, with the unit of measurement the number of participants who provided a portion of their definition (a word, phrase, or sentence) as labeled within each category. Each word within a participant definition could only belong to one code. Seven different coders (the current author and undergraduate psychology majors) were trained and coded the data, with three coders for each participant’s answer. Fleiss’ Kappa was .87; any coding disagreements were discussed until unanimous agreement was reached.
Results and Discussion
Quantitative Analysis
Table 1 provided statistics concerning scientists’ ratings of the 29 items. A commonly reported statistic in studies assessing participant ratings of construct features is the Item-Content Validity Index (I-CVI; Koller et al. 2017; Zamanzadeh et al. 2015), which entailed dividing the number of respondents who deemed the item “content valid” by the total number of individuals rating the item. Items with I-CVI over 0.80 are considered valid construct indicators (Koller et al. 2017; Zamanzadeh et al. 2015). Two I-CVI statistics were used in the current study—one more stringent (I-CVI-S; the numerator entailed those rating the item as “4-very” or “5-extremely” on the five-point scale of how representative the item is of fundamentalism), one more moderate (I-CVI-M; numerator included those rating the item “3-moderately”, “4”, or “5”). It was common to include more than just the extreme response as an indicator that the item was valid (Rodrigues et al. 2017; Zamanzadeh et al. 2015). Further, it appeared reasonable to include moderate ratings as valid given the variety of fundamentalism definitions; indeed, some researchers (e.g., Koller et al. 2017) went so far as to recommend a cutoff below the midpoint if the construct was quite broad. A third statistic used to determine if an item was a key feature of fundamentalism was whether at least 50 percent of respondents rated the statement with the highest response choice (“extremely” representative of fundamentalism). The fourth and final statistic used to determine item relevance was the item confidence interval for the population mean. One advantage to the present study was its large sample size, as the typical study assessing scientists’ knowledge of a construct contained 5 to 10 raters (Koller et al. 2017; Rodrigues et al. 2017). Thus, a confidence interval could be an indicator of the population mean concerning how much scientists viewed the item as representative of fundamentalism, and allowed for a comparison between items (i.e., if some were viewed as significantly more indicative of fundamentalism than others). Indeed, a one-way within-subjects MANOVA revealed a statistically significant difference between the 29 items on how well they represented the fundamentalist construct, F(29,85) = 9.28, p < .001, Wilks’Λ = .240, η2 = .76. Given the number of item comparisons, a stringent confidence interval (99.9%) was used. If the confidence interval between two items did not overlap, it suggested the population mean was significantly higher in the item with the higher confidence interval. For example, in Table 1, the highest-rated item, item 1, had a confidence interval lower bound of 4.31, which was higher than the upper bound confidence interval in the 16 least-endorsed items. Table 1 listed the items from most to least endorsed (sample mean), with the exception of items 8 to 12 and 16 to 22; items 8 and 9 were listed above items 10 to 12 because the former and not the latter were rated by at least 50 percent of participants as “5-extremely representative of fundamentalism”, and items 16 to 19 were listed above items 20 to 22 because the former and not the latter met the 0.80 threshold for the I-CVI-M.
Considering the various statistics in Table 1, the first six items appeared most relevant to the construct. These were the only items to meet the 0.80 threshold for the I-CVI-S statistic and have a lower limit confidence interval above 4.0 on the 5-point scale, significantly higher than at least 14 other items. Further, over half the sample rated these items with the highest response choice (5-extremely representative of fundamentalism). These included all five items (items 1–5 in Table 1) in Williamson et al.’s (2010) intratextual scale. (Note: item 1 was altered in two ways from the original scale, after a review of the literature and discussion with lay fundamentalists—(1) the item focused on the religion generally, rather than the sacred text; (2) the item was of a more moderate tone than the original.) The only intratextual item not in the top 6 was item 14 concerning self-interpretation of the sacred text, which was not included in Williamson et al.’s (2010) final scale because of poor fit within factor analyses. The sixth item with an I-CVI-S above 0.80 (item 6) measured a form of authoritarian submission known as religious humility (Phillips and Kitchens 2021).
Items 7 through 13 in Table 1 had moderate support for inclusion in a religious fundamentalism scale. All 7 items met the 0.80 threshold for the I-CVI-M that included a rating of “3” out of 5 as content-valid. Further, these 7 items were not significantly lower than items 1 to 6 but were significantly higher than some of the least-endorsed items. However, these items did not meet the 0.80 threshold for the I-CVI-S, and their lower limit confidence interval was lower than 4.0 (3.5) on the 5-point scale. Items 7 to 9 were rated by more than 50 percent of respondents as “extremely representative” of fundamentalism. Item 7 addressed the idea of returning to the pure, original, fundamental aspects of the faith (Altemeyer and Hunsberger 2004; Phillips and Kitchens 2021). Items 8 and 9 were further specifications of Hood et al.’s (2005) privileged concept—how one’s own religion was the surest path to meaning and transcendence (Altemeyer and Hunsberger 2004; Moaddel and Karabenick 2021; Phillips and Kitchens 2021). Items 10 and 11 dealt with aspects of authoritarian submission (Altemeyer and Hunsberger 2004; Liht et al. 2011; Moaddel and Karabenick 2021; Phillips and Kitchens 2021). Items 12 and 13 addressed very strong religiosity (Altemeyer and Hunsberger 2004; Liht et al. 2011; Phillips and Kitchens 2021).
Though items 14 to 19 met the 0.80 threshold for the I-CVI-M (items 20–29 did not), a significant amount of evidence suggested items 14 to 29 were not adequately representative of the construct in this sample. None of these items met the 0.80 threshold for the I-CVI-S. The confidence interval for these items were significantly below at least one other item, and less than 50 percent of scientists viewed the item as “extremely representative” of fundamentalism. Item 14 involved self-interpretation of the sacred text (Williamson et al. 2010). Items 15 to 19 included one item involving intolerance, another belief in prophecy, and three assessed dualistic thinking (concerns with apostates, boundary setting, and protecting oneself from society), all of which were components of religious fundamentalism in several definitions (Deal and Bartoszuk 2014; Liht et al. 2011; Moaddel and Karabenick 2021; Phillips and Kitchens 2021). The least endorsed items, 20 to 29, included items that addressed other aspects of dualistic thinking, intolerance, and prophecy (millennialism and messianism). Three of the least-endorsed items assessed authoritarian aggression—punitiveness and verbal and physical militance, components within some fundamentalism definitions (Almond et al. 2003; Altemeyer and Hunsberger 2004; Phillips and Kitchens 2021) but not all (Williamson et al. 2010). Other least-endorsed items addressed evangelism (Deal and Bartoszuk 2014; Phillips and Kitchens 2021), believing religion should pervade society (e.g., school and government—Liht et al. 2011; Moaddel and Karabenick 2021; Phillips and Kitchens 2021), and holding traditional values (Altemeyer and Hunsberger 2004; Liht et al. 2011; Phillips and Kitchens 2021).
Overall, there was strong quantitative evidence that the current sample viewed six items as highly representative of religious fundamentalism—Williamson et al.’s (2010) intratextual items and an aspect of authoritarian submission—one must accept all of the religion’s teachings (Phillips and Kitchens 2021). There was moderate empirical support that scientists considered seven additional items as relevant to their definition of fundamentalism, concerning the domains of extreme religiosity (Liht et al. 2011), returning to the fundamentals of the faith, further facets of authoritarian submission (Altemeyer and Hunsberger 2004) and more specific elements of the intratextual privileged component (Williamson et al. 2010). The current sample viewed the remaining 16 items as less representative of religious fundamentalism.
Qualitative Analyses
Content analysis of scientists’ definitions of fundamentalism revealed 18 themes or elements of the construct. Table 2 defined and gave example quotes for each. No theme was mentioned by the majority of participants; 1 theme was noted by 40 percent, 3 by 30 percent to 39 percent, 3 by 20 percent to 29 percent, 7 by 10 percent to 19 percent, and 4 by 7 percent to 9 percent. Such variance in definitions provided support for a family resemblances approach to the construct in that no feature of religious fundamentalism was essential (Almond et al. 2003, Phillips and Kitchens 2021). Further, the large number of themes supported Phillips and Kitchens’ model, with 15 of their 16 fundamentalism elements noted in the qualitative data (all but millennialism), and 17 of the 18 themes identified in their work (all but Certainty/Security; two themes—Rigid Beliefs and Unquestionable—were subsumed within one of their elements—e.g., avoiding doubts/questions).
Themes and Frequency Count of Scientists’ (n = 121) Definitions of Religious Fundamentalism.
The most common theme, noted by 42.1 percent of participants, was Rigid Beliefs—fundamentalists were dogmatic, inflexible, and not open to new ideas. Though not specifically mentioned in most definitions of fundamentalism, the theme shared similarities with the commonly mentioned fundamentalist features of the belief in the unchanging nature of the sacred text (Williamson et al. 2010) and traditional values (Liht et al. 2011). Further, though Rigid Beliefs was not directly examined in the present study’s quantitative items, it was indirectly assessed through the unchanging intratextual item 5 and traditional values item 22 in Table 1. Indeed, the fourth-most common theme in the data was Unchanging—believing that the sacred text and the religion’s traditions, norms, and values should remain unaltered, whereas the Rigid Beliefs theme was broader, addressing how the person’s entire belief system was unmalleable. Altemeyer and Hunsberger (2004) found correlations so strong between dogmatism and religious fundamentalism (i.e., .70 < r < .76), that one could argue they were measuring the same thing. After data collection in the present study was completed, Saroglou et al. (2022) published empirical evidence that dogmatism was one of the main features of fundamentalism. Measures of fundamentalism should assess general rigidity and calculate its relationship to other fundamentalist items, including belief in traditional values and the unchanging nature of the sacred text.
The second most prevalent theme, noted by 36.4 percent of respondents, was Privileged, or viewing one’s sacred text or religion above others. This domain was noted within many fundamentalism definitions (Phillips and Kitchens 2021), particularly Williamson et al.’s (2010) intratextual theory, with one caveat. Only 4 participants reported the sacred text, specifically, as privileged; whereas another 4 noted it was more than the sacred text that was privileged (religious traditions, practices, etc.). The majority (35 of 43) used more general phrasing (religious teachings, beliefs, etc.). Future research could directly ask respondents if they believed the privileged component is specific to the sacred text. The other elements of the intratextual theory were also themes in the qualitative data, albeit at different levels of prevalence—the fourth (Unchanging, noted by 30.6 percent of participants, 8th (Authoritative; 19.0%), 13th (Inerrant; 16.5%), and 15th (Divine; 9.9%) most common categories. In contrast, the quantitative data in the present study clearly indicated the five intratextual items were the most representative aspects of fundamentalism. As with the Privileged theme, most quotes within the Unchanging and Authoritative categories did not clarify it was the sacred text that held these features (thus not fully supporting intratextual theory). Most quotes within the Inerrant and Divine categories, however, did specify it was the sacred writings that held these qualities (15 of 20 and 8 of 12, respectively, with the rest giving ambiguous answers that may or may not be centered on the text). Overall, there was moderate support in the present study’s qualitative data for the intratextual theory, to supplement its very strong backing in the quantitative data. At the very least, scientists knowledgeable of religious fundamentalism thought the construct involved believing that some aspect of the religion (if not the sacred text in particular) was unchanging, inerrant, of divine inspiration, privileged above other faiths (texts), and authoritative compared to other knowledge sources.
The third most-common theme, noted by 33.9 percent of participants, was Literal Interpretation, or the belief that sacred texts should be interpreted exactly as they are written, without attending to symbolism or historical context. This fits some fundamentalist definitions (Wulff 1997), although others described literal interpretation as an inaccurate depiction of fundamentalism (Herriot 2007) or took a more nuanced approach (Williamson et al. 2010). For example, the latter’s intratextual scale originally included an item describing how the sacred text was interpretable without the help of historians, although that item did not fit in a factor analysis, nor was that item highly rated in the present study’s quantitative analysis (item 14 in Table 1). Fitting with the complexity surrounding literal interpretation as a religious fundamentalism definition, this category was one of four themes (the others being Militant, Privileged, and Unchanging) where one to two scientists noted that the element was not a defining feature. (Those noting an element was not a defining feature were not included in the frequency count in Table 2.) Perhaps Literal Interpretation was a popular category because it was often strongly linked with intratextual theory, which, as noted earlier, were the main components of fundamentalism in the present study’s quantitative data and fairly prevalent in the qualitative data. For instance, participant 22 linked literal interpretation with authoritativeness, describing fundamentalism as “following a very literal interpretation of the scripture in a way that disregards other major factors…while many non-fundamentalists…acknowledge that human beings developed from other animals…” Further, a common Biblical literalism operational definition was “the Bible is the actual word of God…to be taken literally, word for word” (Lytle and ten Bensel 2016:315), linking the divine nature of the sacred text with literal interpretation. To summarize, literal interpretation was a controversial aspect of fundamentalism, but often linked to other elements of the construct. Perhaps studies should separate each aspect of fundamentalism—for example, one item should address literal interpretation of a sacred text, another how the text is God’s word. Procedures such as factor analysis could then assess whether each element was a component of fundamentalism.
After the aforementioned Unchanging category, Simplistic Thinking was the fifth most common category, noted by 29.8 percent of respondents. Indeed, many fundamentalism definitions and scales described fundamentalists as dualistic in their thinking (see Phillips and Kitchens 2021, for a review), such as believing the world consists of “two kinds of people,” with one group favored by God (Altemeyer and Hunsberger 2004). Yet the quantitative data in the present study revealed simplistic thinking items were less representative of fundamentalism than other domains. Further, religious fundamentalism was often weakly or not at all related with integrative complexity (Hunsberger, Pratt, and Pancer 1994; Hunsberger et al. 1996; Liht et al. 2011; Weeks and Geisler 2019). As will be discussed in the limitations, further qualitative and quantitative research is needed to clarify the importance of fundamentalist domains such as simplistic thinking.
The sixth most common theme, mentioned by 25.6 percent of participants, was Individual Pervasiveness of Religion. This domain was included in few fundamentalism definitions (Liht et al. 2011; Phillips and Kitchens 2021), though many qualitative and historical studies noted how religion pervades all aspects of a fundamentalist’s life (Almond et al. 2003; Ammerman 1987; Peshkin 1986). Recently Saroglou et al. (2022) provided evidence that moral rigorism, or strict adherence to religious norms, was a major component of religious fundamentalism. Further, the quantitative data in the present study provided moderate support that punctilious religiosity was an important feature of religious fundamentalism. The pervasiveness of religion in a person’s life appeared to be an important aspect of the fundamentalist construct.
The seventh-most common theme—Intolerance—was mentioned by 24 percent of participants. Intolerance was mentioned in some fundamentalism definitions (Altemeyer and Hunsberger 1992; Deal and Bartoszuk 2014; Moaddel and Karabenick 2021; Phillips and Kitchens 2021). A large research literature has examined the consistent link between fundamentalism and prejudice (see McCleary et al. 2011, for a review). However, the quantitative data in the present study indicated that intolerance items were judged as less essential aspects of fundamentalism. Perhaps it was best to keep intolerance as a criterion variable rather than a component of fundamentalism, though including intolerance items in factor analyses with other fundamentalist items could clarify the matter.
The 8th through 13th most common themes were identified in 16 percent to 19 percent of scientists’ definitions, two of which (8th and 13th) were mentioned previously (the intratextual categories Authoritative and Inerrant). The 9th, 11th, and 12th most common themes were major components of Altemeyer and Hunsberger’s (2004) definition—Fundamentals, Militant, and Authoritarian Submission. Focusing on the faith’s fundamentals was noted by 19 percent of scientists in the present study. It was rarely mentioned in most definitions of fundamentalism, although the term was often used by fundamentalists themselves (Phillips and Kitchens 2021) and there was moderate support for the fundamentals item in the present study’s quantitative data. Militance and Authoritarian Submission—two of three aspects of authoritarianism—were found in 17.4 percent of scientists’ fundamentalism definitions. (The third element of authoritarianism—traditionalism—was part of the aforementioned Unchanging category.) The equal prevalence of these two categories was surprising, given that, in the present study, scientists rated the authoritarian submission items (as well as the unchanging item) as more representative of fundamentalism than the militance items. Further, although correlations between militance and religious fundamentalism ranged between small and large (see Phillips and Kitchens 2021, for a review), a growing number of researchers reported that most fundamentalists were not violent (Deal and Bartoszuk 2014; Herriot 2007; Hood et al. 2005; Sadowski, Endrass, and Zick 2019; Williamson et al. 2010). Indeed, two participants in the present study’s qualitative data reported militance was not a component of religious fundamentalism. A third participant stated: “There is a subset of fundamentalism that might better be classified as extremism that would advocate violence against secularism and other religions…most fundamentalists are concerned with upholding a worldview…rather than engaging in conflict.” Given how some fundamentalists caused major violence in the world (Phillips and Kitchens 2021), perhaps identifying the subset of fundamentalists who are militant is needed.
The 10th most common theme concerned Boundary Setting, noted by 18.2 percent of respondents. Yet the present study’s quantitative findings revealed participants did not view boundary setting as one of the most important features of fundamentalism. Further, boundary setting was a component of less-cited religious fundamentalism definitions (Deal and Bartoszuk 2014; Liht et al. 2011; McFarland 1989). In contrast, prior research has found a moderate to strong relationship between fundamentalism and boundary setting (Kellstedt and Smidt 1991; Phillips and Ano 2015). Though a component of fundamentalism in Phillips and Kitchens’ (2021) definition, Saroglou et al. (2022) provided empirical evidence that setting strong boundaries was a major component of fundamentalism in some fundamentalists, in contrast to dogmatism and high religiosity, which was a major domain in most fundamentalists.
The five least common categories—mentioned by 7 percent to 11 percent of participants—included Certainty/Security, Divine, Unquestionable, Evangelize, and Societal Pervasiveness of Religion. Certainty/Security was a category not mentioned in most fundamentalist definitions, nor was such an item included in the present study’s quantitative methodology. Further, research found only low to moderate correlations between need for closure (desire for certainty) and fundamentalism (Carlucci et al. 2021; Liht et al. 2011; Watson et al. 2003). Concerning Divine, its lack of popularity was surprising. It was the only aspect of intratextualism found in less than 16.5 percent of scientists’ definitions, and the present study’s quantitative data revealed the item was an important aspect of religious fundamentalism. Further, this element was in many fundamentalist definitions (Deal and Bartoszuk 2014; McFarland 1989; Phillips and Kitchens 2021; Watson et al. 2003; Williamson et al. 2010). Unquestionable was a component of some fundamentalist definitions (Liht et al. 2011; Moaddel and Karabenick 2021) and Phillips and Kitchens (2021) placed it as a subcomponent of the broader domain of authoritarian submission. The present study’s quantitative data provided moderate support that scientists perceived not questioning one’s religion as an important feature of fundamentalism. Fundamentalism and a questioning religious orientation generally had a moderate inverse association, though correlations range from low to high (Altemeyer and Hunsberger 1992, 2004; Deal and Bartoszuk 2014; Liht et al. 2011; Williamson et al. 2010). The categories of Evangelize and Societal Pervasiveness of Religion are included in few fundamentalism definitions (Phillips and Kitchens 2021) and corresponding items within the present study’s quantitative data were rated as less essential to participants’ description of religious fundamentalism.
Integrating the Qualitative and Quantitative Data
No aspect of religious fundamentalism was noted by a majority of scientists in the qualitative data, supporting the family resemblances assumption that there was no essential feature to the construct (Almond et al. 2003; Phillips and Kitchens 2021). Nonetheless, certain domains of religious fundamentalism were more prevalent across the present study’s qualitative and quantitative data, including five of Williamson et al.’s (2010) intratextual components; the sacred text (or some aspect of the religion) was unchanging, inerrant, divinely created, privileged above other faiths, and authoritative compared to other knowledge sources. Also popular across both sets of data was rigorous religious involvement (Liht et al. 2011), a focus on the fundamentals of the faith and submission to authority (Altemeyer and Hunsberger 2004). Some elements of religious fundamentalism were fairly popular in the qualitative but not the quantitative data—intolerance, simplistic thinking, boundary setting, and militance. Less central to the construct of religious fundamentalism across both sets of data were ideas of millennialism, messianism, prophecy, evangelism, and religion in public life.
Limitations, Implications, Suggestions for Future Research
Though strengths of the current research include the conservative p value and a sample size that was larger than the typical study that assessed scientists’ ratings of a construct’s features (Koller et al. 2017; Rodrigues et al. 2017), many quantitative comparisons were made with a sample of 125. Thus, the comparisons should be interpreted with caution. Further, the sample contained only English-speaking persons, mostly residing in the US, where Christianity was dominant, which could bias the definitions of religious fundamentalism. For example, multiple studies documented the importance of Sharia law in Islamic fundamentalism (see Phillips and Kitchens 2021, for a review), yet in the current sample, societal pervasiveness of religion was rarely mentioned as a key component of fundamentalism. Thus, research is needed with scientists who study fundamentalism and (1) do not speak English or (2) live in locations where non-Christian religions are prevalent. Other limitations concerning the sample were the criteria of one to two peer-reviewed journal publications concerning religious fundamentalism, and the lack of information concerning participant identity. Though anonymity is commonplace in survey design, it prevented the identification of the exact number of publications (and therefore the level of knowledge of fundamentalism) in the current study’s sample. Further, where to draw the line in terms of construct knowledge was a matter of debate and ambiguity; however, a more stringent criterion could ensure a more knowledgeable sample, though detracting from the statistical power of the current study, and the goal of assessing more than just the foremost experts on the topic.
Research methodologies of a similar design are needed with lay experts (Zamanzadeh et al. 2015)—persons who identify as fundamentalist. Watson et al. (2003) noted the different worldviews of social scientists in comparison to religious laypersons, and, particularly, fundamentalists. Features of fundamentalism such as simplistic thinking, for example, might not be viewed as relevant to lay fundamentalists.
There were also limitations concerning some of the quantitative items used in the present study. For example, there was a lack of items addressing unexpected themes in the qualitative findings—interpreting the sacred text literally, a dogmatic general belief system, and the need for certainty/security. The present study originally conceived Biblical literalism as a poor measure of religious fundamentalism given prior scholarship (Hood et al. 2005; Herriot 2007), though the intratextual item that addressed the self-interpretive nature of the sacred text was similar to literalism. Relatedly, the current research considered dogmatism more within the fundamentalist elements of the traditional values and unchanging nature of the sacred text. However, scientists in the present study appeared to view rigidity and literalism in a more general sense. The present study did not include an item addressing certainty as this was not found in existing definitions of religious fundamentalism (Phillips and Kitchens 2021). Future studies could ask questions concerning these matters to scientists or laypersons knowledgeable about fundamentalism. Further, future studies could quantitatively assess whether the intratextual principles are tied to the sacred text or to the religion more generally (i.e., a more intrareligious than intratextual theory). Finally, another limitation to the quantitative items was also a limitation to the family resemblances approach as a whole—are there truly a global set of traits that make up religious fundamentalism? Or are fundamentalist features contextual, with most components only existing in a select few religions and cultures? If the latter is more accurate, then asking scientists how much each item represented their definition of religious fundamentalism is less applicable. This was a matter that could be empirically tested. A scale could be developed from fundamentalist features delineated in the present study, and factor analyzed with participants across multiple religions and cultures, to determine if there was a core set of fundamentalism features that might vary slightly depending on the setting. For example, religious scrupulosity, authoritarian submission, and beliefs in the divine nature of the sacred text might be key to American Christian and Saudi Muslim fundamentalists, but the need for a societal pervasiveness of religion might not load in a factor analysis in the former because of cultural beliefs in the separation of church and state. In contrast, factor analyses might reveal that Sri Lankin Buddhist fundamentalists endorse all of the features just mentioned (including societal pervasiveness of religion), except belief in the divine nature of the sacred text, given the less theistic nature of their religion. Thus, empirical testing could determine if a core set of fundamentalism features exist, but vary slightly depending on the context, as the family resemblances model assumes.
Another means to test the family resemblances approach is to conduct factor analytic studies across religions and cultures, examining items from a broad array of existing fundamentalism measures (e.g., Altemeyer and Hunsberger 2004; Deal and Bartoszuk 2014; Liht et al. 2011; Moaddel and Karabenick 2021; Williamson et al. 2010). Baer et al. (2006) conducted a similar study with the mindfulness construct, giving participants five validated scales of mindfulness and conducted factor analyses on the data and created a new measure. Certain items not in most fundamentalist measures (literal interpretation, dogmatism, and need for certainty) should be added. These items can then be factor analyzed to determine if they are indeed a component of fundamentalism in most religions and cultures around the world.
The goal of the aforementioned suggestions was to create a more comprehensive, accurate definition and scale of religious fundamentalism that can then be tested (e.g., factor analyzed) in various cultures and religions. Such a context-appropriate measure can be used in research examining the causes and consequences of fundamentalisms around the world. Research of this kind would be useful to religious and lay leaders (e.g., politicians), service providers and nongovernmental organizations as they attempt to understand and interact with fundamentalists. Data in this area is vital because fundamentalists exist in many religions and cultures, from Myanmar Buddhists (Ramakrishna 2021) to Hindu Indians (Madan 2010), Middle Eastern Muslims (Moaddel and Karabenick 2021), Israeli Jews (Hammer and Lazar 2019), and American Christians (Hood et al. 2005).
Conclusion and Implications
The present article contributed to the literature by using a mixed-methods approach that empirically examined how scientists who studied religious fundamentalism defined the term (Koller et al. 2017; Sireci 1998; Zamanzadeh et al. 2015). The data elucidated the most relevant features of fundamentalism, including multiple elements of intratextualism theory (Williamson et al. 2010), authoritarian submission, focusing on the fundamentals of the faith (Altemeyer and Hunsberger 2004), and the pervasiveness of religiosity in fundamentalists’ lives (Liht et al. 2011), while recognizing that no one component was identified by a majority of the sample, supportive of a family resemblances approach to the construct (Phillips and Kitchens 2021). The present study also identified fundamentalist domains that appear less relevant, such as millennialism, messianism, prophecy, evangelism, and requiring religion across all aspects of society. Finally, the present study identified new domains for item specification, as well as suggestions for further item modification. This information can lay the groundwork for future refinements to religious fundamentalism measures, which can then be used in applied research to better understand fundamentalists and how to interact with them, which is prudent given the prevalence of this phenomenon in religions across the world.
Footnotes
Acknowledgements
The author would like to thank Michael B. Kitchens, Bryan McCarthy, and the reviewers for their helpful suggestions on the manuscript; Chase Beezer, Nicole Cortino, Juliet Hoeth, Chloe O’Hara, Kristina Paladin, Stephanie Thomas, and Samantha Mahr for their help in coding the data.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Data Availability
The data in this study are available upon request to the author. The data are not publicly available due to privacy and confidentiality reasons.