The Skeptics Society & Skeptic magazine

Jacques-Louis David [Public domain], via Wikimedia Commons

The Three Shades of Atheism:
How Atheists Differ in Their Views on God

When we think of prominent atheists, we may conjure up an image of one of the “Four Horsemen” of the New Atheism—Christopher Hitchens, Sam Harris, Richard Dawkins, and Daniel Dennett—authors famous for their steadfast rejection of any form of deity and their willingness to confront the world’s religions. Ironically, however, when we see them in debates and interviews, the confidence with which they make their case and discount the opposition may at times seem indistinguishable from the offputting dogmatism of the hyper-religious. How typical of atheists are the Four Horsemen?

Our research, based on a sample of hundreds of respondents to a survey distributed through social media, indicates that they probably represent a common form of atheism but not the majority view. Most atheists express some degree of tentativeness in their beliefs and would be prepared to consider contrary evidence and arguments. In other words, they are skeptical in their orientation rather than dogmatic. However, the prevalence of dogmatic atheism may come as a surprise to some observers, including Richard Dawkins,1 who stated that he “would be surprised to meet many people” who would say “I know there is no God.” Many respondents in our survey said this.

Distinguishing Between Categories of Atheistic Belief

To categorize the various forms of atheism, it is necessary to distinguish among several closely related concepts.

Formal v. informal meanings of atheism. The term atheism literally means an absence of belief in a deity, as in a theismwithout theism. This formal usage broadly encompasses both nonbelief and the explicit rejection of a deity. Nonbelief without any inclination to reject a deity is similar to, but distinguishable from, agnosticism, a term introduced by Thomas Henry Huxley in 1876 at a meeting of Britain’s Metaphysical Society, many of whose members were clergymen, and elaborated upon at a symposium published in 1884 by The Agnostic Annual. Huxley defined agnosticism as … the absence of belief one way or the other and the absence of a claim to having any scientific knowledge on the issue:

Agnosticism is of the essence of science, whether ancient or modern. It simply means that a man shall not say he knows or believes that which he has no scientific grounds for professing to know or believe.

Huxley described how he arrived at this position:

When I reached intellectual maturity and began to ask myself whether I was an atheist, a theist, or a pantheist…I found that the more I learned and reflected, the less ready was the answer. They [believers] were quite sure they had attained a certain ‘gnosis,’—had, more or less successfully, solved the problem of existence; while I was quite sure I had not, and had a pretty strong conviction that the problem was insoluble.2

In informal usage atheism usually refers to an explicit belief—or at least an inclination toward the view—that no deity exists. Within this category a further distinction can made between atheists who claim to have knowledge or proof that no deity exists (gnostic position) and atheists who claim that no such knowledge is available and may never be attained (agnostic position).

Agnostic-atheism v. gnostic-atheism. Robert Flint, an influential Scottish philosopher and theologian, advanced a concept of agnostic-atheism in his Croall Lecture of 1887–1888.3 He suggested that the two terms were synonymous and should not be differentiated, arguing that it was possible for an individual both to believe that there is no God and that the question of God’s existence is unanswerable. However, agnostic-atheism clearly differs from Huxley’s agnosticism in that the former includes the belief that no deity exists whereas the latter is noncommittal on the issue. Agnostic-atheism, in turn, implies a concept of gnostic-atheism in which the atheist claims to have proof or knowledge that no deity exists.

From this discussion we may identify three basic categories of atheistic belief:

  1. nonbelief in a deity without taking any position on the issue;
  2. agnostic-atheism, expressing disbelief to some degree but without a claim of knowledge (skepticism); and
  3. gnostic-atheism, firmly rejecting the existence of a deity and claiming to have knowledge or proof that no deity exists (dogmatism). We describe now our research showing that these categories are empirically distinguishable in survey respondents’ explanations of why they do not believe in any form of God.
Survey Methods

Data were collected from several questions included within a larger survey that was developed by the first author for a Master’s Thesis at California State University, Fullerton, on the topic of patterns of moral values among atheists, deists, and theists.

Identifying atheists. A total of 666 participants (a purely coincidental number!), gathered through Facebook and other social media, was categorized according to their responses to six “belief” questions that began with, “Do you believe in a God that…” and then stated a particular trait. Five traits were theistic (e.g., “monitors your behavior,” “intervenes in human affairs”) and one trait was deistic (“created the universe, but refrains from interacting with it”).

An atheist was operationally defined as a respondent who answered “No” to all six belief-in- God questions, thereby meeting the requirements of the formal definition of an atheist. As a check on the validity of this definition, the page that followed asked respondents to indicate if they had a “religious affiliation,” with 11 affiliative options arranged alphabetically from “Buddhist” to “Seventh Day Adventist” (with “Other” at the end) plus two rejectionist options, “None-Atheist” and “None- Agnostic.” It was expected that virtually all of the operationally defined atheists would identify as atheist or agnostic.

Exploring the thoughts of self-identified atheists. A total of 233 respondents selected “Atheist” instead of a religious affiliation and were transferred to another page that presented the question, “Since you selected atheist, would you please elaborate on why ‘atheist’ is a more appropriate characterization of your beliefs than ‘agnostic’? How do you differentiate between these two terms?” Respondents could answer by typing in an open-ended text response (219 total).

Development of coding criteria. The authors initially created tentative criteria for assigning responses to categories based on respondents’ flexibility of belief and whether or not they made a distinction between belief and knowledge. There were four categories, which in current terms were: gnostic-atheism, agnostic-atheism, ambivalentatheism, and “other” (not classifiable).

The authors independently applied the criteria to the first 50 responses and the level of agreement was assessed statistically (using Cohen’s kappa, maximum value =1) to see if it was greater than the level expected by chance. The value found was .746, which was above chance and would generally be considered to show “good agreement” (.60–.80= “good”, .80–1.00=“very good”).4 After discussion of cases of disagreement, the authors independently applied the criteria to the remaining 169 respondents. The kappa value for all 219 cases was .700. The criteria were as follows:

Gnostic-Atheism: Any explicit or implied characterization of the participant’s position as certain or definite.

Agnostic-Atheism: Any effort made to distinguish between a “belief” and “knowledge” position; or participants who indicate that they are open to evidence: they describe their belief as malleable and open to changing based on new information, evidence, or “proof.”

Ambivalent-Atheism: Any use of the phrase “I don’t know” or “I am not sure,” or similar characterizations of belief, without further explanation.

Other: Any statement that does not fit the criteria of the other categories.

Table 1 shows examples of statements that were assigned by both coders to each category.

Survey Results I: Nonbelievers Who Do Not Reject a Deity

Nonbelief and religious affiliation. Of the 666 social media respondents, 366 (55%) met our operational definition of an atheist (which is also the formal definition) by responding “No” to all six belief-in- God questions. Of those who were operationally defined as atheist, responses to the religious affiliation question were as follows: 306 (83%) responded “None”; 232 (63%) identifying as “Atheist”; and 74 (20%) as “Agnostic”.5

Nonbelievers are thus highly likely to reject a religious affiliation (95% confidence interval: 79% to 87%) but a substantial percentage of nonbelievers (at least 13%) do not reject it. It would seem that one could affiliate with a religion for a variety of social and psychological reasons other than belief in a deity, for example, secular Jews who attend religious services for social or emotional reasons, a family that practices the religion, identification with the religion’s moral values, or the absence of a deity in the religion’s ideology, such as Buddhism.

Defining nonbelief: agnosticism + formal atheism. The lightest shade of atheism in our model is nonbelief without taking a position on whether a deity exists. Respondents who identified as agnostic rather than as atheist would meet this requirement provided that they also responded “No” to all belief-in-God questions.

As noted above, there were 74 such respondents. However, there were 34 additional agnostics who could not be classified as nonbelievers because they did not answer No to all the belief-in-God questions.6 One could not simply go with the agnostic label. Combining it with the belief questions was essential.

In contrast, if a respondent self-identified as atheist, in virtually every case she or he answered “No” to all the belief questions. Beyond the 232 validated atheists, one additional respondent was classified as a deist.

Survey Results II: Nonbelievers Who Reject a Deity

Distribution of agnostic- and gnostic-atheists. The majority of self-identified atheists were gnosticatheists, an orientation that we have characterized as dogmatic rather than skeptical. Table 2 shows the distribution of respondents across these two categories.

Most of the respondents were classified as either gnostic- or as agnostic-atheists (206 or 196 depending upon the coder). Both coders classified 117 respondents as gnostic-atheists (53.4%; confidence interval: 47% to 60%) and classified an average of 84 respondents as agnostic-atheists (38.4% of respondents; confidence interval: 32% to 45%).

The Majority of Atheists are Skeptical, Not Dogmatic

The survey data indicate that most atheists in the sample maintained a skeptical orientation toward their own position and were open to considering evidence and arguments favoring a theistic position. The numbers of respondents in each belief category were as follows:

Gnostic-Atheist, 117;
Agnostic-Atheist, 84;
Nonbeliever (uncommitted), 74.

As percentages of the total (275) the distribution was:

Gnostic-Atheist, 43%;
Agnostic-Atheist, 31%;
Nonbeliever (uncommitted), 27%.

Combining the last two categories, 58% (95% confidence interval: 51% to 63%), acknowledged a distinction between what they believed and what they thought they knew, a precondition for critical thinking and reasoned debate.

Conclusions: Faith-Based Atheism

In demonstrating a sizable category of gnostic atheists, our data reveal a way of thinking among many atheists that is fundamentally religious in nature.

Do atheists accept atheism on faith? In The God Delusion, Dawkins proposed a “spectrum of probabilities7 to represent the range of judgments that people could make on the question of God’s existence. It is a continuous scale highlighted by seven landmarks: (1) strong theist, (2) de facto theist, (3) leaning towards theism, (4) completely impartial, (5) leaning towards atheism, (6) de facto atheist, and (7) strong atheist. Dawkins characterizes his own position as (6) and “leaning towards” (7). He states that it is not (7) only because, in principle, one cannot prove that something does not exist. It would have to be accepted on faith, and in contrast to believers in God, “Atheists do not have faith…”

However, when we look at the data we find that more than half of atheists who take a belief position express certainty in the non-existence of God, with statements such as “Atheist means that you are certain there is no such thing as god,” “I’m certain there are no gods,” and “There is no God or other deity and I don’t entertain the notion that there might be.” As Dawkins states, “reason alone could not propel one to total conviction that anything definitely does not exist.” What fills the gap here is faith. At the extreme ends of Dawkins’ scale we essentially have two opposing religions.

Skeptic 22.2 (cover)

This article appeared in Skeptic magazine 22.2 (2017)
Buy this issue

The two shades of theism. The dogmatic and skeptical shades of atheism seem likely to have counterparts on the theistic side of the issue, so that with appropriate defining criteria the methodology we have described here should also reveal gnostic and agnostic forms of theism.

Gnostic-theists would be individuals who equate their beliefs with facts, dogmatically insisting that they have positive knowledge of God’s existence. Agnostic-theists would be individuals who accept the distinction between belief and knowledge, thereby demonstrating a degree of skepticism about their own position, and would indicate that their belief is based on faith, intuition, or an interpretation of natural phenomena. A 5-level, bipolar scale relating theistic and atheistic beliefs would be:

  1. Gnostic-Atheism
  2. Agnostic-Atheism
  3. Nonbelief
  4. Agnostic-Theism
  5. Gnostic-Theism

The scale represents maximum darkness at both ends, the domains of dogmatic thinking. Maintaining a skeptical attitude toward one’s own beliefs can be a challenge but, as the achievements of science have shown, it is a better route to enlightenment. END

About the Authors

Brittany Page earned her BA in Psychology from California State University, Fullerton (CSUF). She is currently in her third and final year to obtain her Master of Science in Clinical Psychology at CSUF. Page’s research interests focus on issues related to morality, political psychology, and anti-atheist prejudice. Her master’s thesis is entitled, “Are atheists immoral? Patterns of values of atheists, deists, and theists on moral foundations.” Brittany also hosts the podcast, “I Doubt It with Dollemore” a twice-weekly news and comment show dedicated to all things news, politics, and religion.

Dr. Douglas J. Navarick is an experimental psychologist and Professor of Psychology at California State University, Fullerton. He regularly teaches courses in Introductory Psychology and Learning and Memory. Since the 1970s, Navarick has published research articles on choice behavior in pigeons and humans and is currently investigating how we make intuitive moral judgments.

  1. Dawkins, Richard. 2006. The God Delusion. New York: Houghton Mifflin Harcourt, 73.
  2. Huxley, Thomas H. 1894. Collected Essays, New York: D. Appleton and Co., Vol. 5, 237–238.
  3. Flint, Robert. 1903. Agnosticism: The Croall Lecture for 1887–88. Edinburgh, Scotland: William Blackwood and Sons, 49–51:

    If a man has failed to find any good reason for believing that there is a God, it is perfectly natural and rational that he should not believe that there is a God; and if so, he is an atheist… if he goes farther, and, after an investigation into the nature and reach of human knowledge, ending in the conclusion that the existence of God is incapable of proof, cease to believe in it on the ground that he cannot know it to be true, he is an agnostic and also an atheist—an agnostic-atheist— an atheist because an agnostic… while, then, it is erroneous to identify agnosticism and atheism, it is equally erroneous so to separate them as if the one were exclusive of the other.

  4. (retrieved 3/4/2017)
  5. Of the 28 nonbelievers who did not identify as atheist or agnostic, and who responded to the religious affiliation question, the affiliations were as follows: 6 Buddhist, 4 Christian – Catholic, 2 Christian – Other, 4 Christian – Protestant, 3 Jewish, 3 Seventh Day Adventist, 6 unaffiliated (“other”). An additional 32 nonbelievers did not respond to the religious affiliation question.
  6. 15 were deists (responded “Yes” to deistic trait, not to the five theistic traits), 9 theists (“Yes’ to at least 1 theistic trait), and 10 unclassifiable because they did not complete the belief questions.
  7. Dawkins, op. cit., 72–73.
For those seeking a sound scientific viewpoint


Be in the know!

Subscribe to eSkeptic: our free email newsletter and get great podcasts, videos, reviews and articles from Skeptic magazine, announcements, and more in your inbox once or twice a week.

Sign me up!

Copyright © 1992–2022. All rights reserved. | P.O. Box 338 | Altadena, CA, 91001 | 1-626-794-3119. The Skeptics Society is a non-profit, member-supported 501(c)(3) organization (ID # 95-4550781) whose mission is to promote science & reason. As an Amazon Associate, we earn from qualifying purchases. Privacy Policy.