Are religion and science in conflict? National surveys indicate that a majority of Americans think so. Surveys that measure this (Pew Research Center: Religion & Public Life; Public Religion Research Institute) ask respondents to indicate whether they believe science and religion are “often in conflict” or “mostly compatible.” The findings are consistent: slightly over half choose “often in conflict” (54 to 55 percent), 38 to 40 percent choose “mostly compatible”, and 5 to 7 percent are unsure what they believe.
On their face, these statistics seem to suggest that a little more than half of the public believe that science and religion are in opposing camps: one favors reason and empirical evidence, the other faith and divine revelation. The two are locked in conflict over fundamentally incompatible methods of knowledge production. While some who answer this question probably believe in the epistemological conflict model, it is a mistake to impute this meaning to everyone who adopts the conflict stance.
Why do I think this? To begin, variations on this question produce quite different results. If, instead of two possible positions, the respondent is asked to pick from three possible positions – that science and religion “generally agree”, “generally conflict”, or are “not related to each other in any meaningful way”— then only between one quarter and one third pick the conflict answer (Inter-university Consortium for Political and Social Research; Gallup; Science Daily.)
This additional option – that the two are independent from one another – maps closely on to the late Stephen Jay Gould’s concept of Non-Overlapping Magisteria (NOMA). This view holds that science’s proper domain is the factual description of the world, while religion’s is human purpose and value.
For Gould this was a way to avoid conflict between religion and science, but this is not the perception of many in the general public. Although the addition of a third option reduces respondents from both of the other positions, the reductions from the “conflict” position are greater. The association between the NOMA position and the conflict position is clearer in the National Study of Religion and Human Origins. This survey asks whether respondents agree or disagree with the following statement: “Science is about facts and religion is about faith. The two do not overlap.” One-third of the population agrees with this statement, quite close to the proportion affirming the independence position in other surveys. Those who agree with this statement are actually more likely to agree with another statement: “Sometimes I feel like the findings of science and the teachings of [my] religion conflict with each other” (68 percent versus 48 percent for those that disagreed with the NOMA position). And, they were less likely to agree with the following statement: “Ultimately, I believe that the findings of science and the teachings of [my] religion are compatible” (27 percent versus 50 percent of those who disagreed with NOMA).
Many who hold to the conflict position still end up rejecting NOMA though. Do they believe that science and religion are fundamentally incompatible and competing ways of gaining knowledge about the world? Some probably do, but it’s unlikely that all do. The 2007 Baylor Religion Survey asked a slightly different question that helps us to see this. Instead of asking whether science and religion conflict, they asked whether science and religion were incompatible. Only 17 percent of Americans agreed with this statement. This is quite a bit less than the one-quarter to one-third who identify with the conflict position in the three-position question. In other words, a number of people believe there is legitimate conflict between science and religion but don’t hold that science and religion are inherently at odds with one another.
In many ways, this 17 percent fits most closely with the strong epistemological warfare position. Science and religion necessarily make competing claims about the world, therefore, one must choose sides. The evidence suggests that this group is split evenly between siding with religion and siding with science (additional evidence for an even split is available here).
Let’s return to those who agree that the two are in conflict but are not fundamentally incompatibility. What are they thinking about when they answer the conflict question? For many, the answer is likely evolution and creationism. A 2009 Pew survey on science asked respondents who felt that science conflicted with their own religious beliefs to report what the conflict was about. Evolution and Darwinism was cited more than four times as much as the next most common answer (which was stem cells, followed closely by abortion). For many, the relationship between religion and science is not one of general epistemological conflict, but conflict over very particular knowledge claims. The sociologist John Evans finds something similar among conservative Protestants. Compared to other religious and nonreligious groups, conservative Protestants were just as likely to seek out scientific knowledge, something that would be unusual if they believed in a general epistemological conflict. At the same time, Evans found that conservative Protestants were more likely to register opposition to evolution and the Big Bang. These conflicts are real for many Americans, but they don’t necessarily lead to a belief in the fundamental incompatibility of science and religion.
In sum, although a slight majority indicates they believe science and religion are “often in conflict”, the meaning behind this claim is not uniform. Roughly one-third of this group believes that science and religion are completely separate domains. They are likely to see prominent public debates about religious and scientific claims, but conclude these are ultimately unnecessary conflicts because one or the other is stepping out of its natural domain. Roughly another third believe that there are real conflicts between science and religion, but these conflicts are limited to very specific domains, not to a general epistemological conflict. The last third probably does view the practices of religion and science to be fundamentally incompatible. Within this group about half side with religion and half with science.
Frequently missing from discussions about religion and science is the moral nature of these disputes. Public skirmishes are rarely dry and dispassionate. For many, their own position is tied to moral ideas about human identity and societal progress. The National Study of Religion and Human Origins asked whether or not it was personally important to the respondent to have the correct beliefs on this topic. Forty-two percent said it was either extremely or very important to them. This group was asked to articulate why it was important in a sentence or two. For the religiously committed creationists, the most frequent responses involved the centrality of trusting in God or the Bible, as well as the importance of personal salvation and the afterlife. On the other end of the spectrum, those who believe that humans evolved without help from God gave equally morally-tinged responses to this question. Several of the most popular responses cite the importance of scientific progress and the personal and societal benefits of trusting in science. Some warn of the negative social and environmental consequences of wrong beliefs about this. For many in this group, placing trust in a rational science over an irrational faith is a moral imperative.
The historian Peter Harrison describes contemporary debates between religion and science as between proponents of different “natural theologies.” He argues that the long historical development of science from an interior virtue to an external body of knowledge was accompanied by a new notion of social progress. While it was once linked to a Christian teleology, science became entirely divorced from specific religious commitments in the nineteenth century. The secular notion of progress that developed was a progress that moves “beyond” religion to a more enlightened rationality. Competing notions of personal or social salvation are often central to contemporary debates that pit science and religion against each other.
All of this suggests that there is more than meets the eye in survey questions about the relationship between science and religion. Although the results appear to have a straightforward reading at first, we need to recognize that changes in the wording and framing of these questions are tapping into different conceptions of science, religion, and the boundaries between the two. Social scientists would do well to move beyond simplistic survey questions and begin to investigate public conceptions of the boundaries and content of religion and science more intentionally.
(1) To what extent does conflict between religion and science depend on how religion and science are conceptualized?
(2) Are there better ways to measure the public’s beliefs about the relationship between religion and science than the survey questions referenced in this article?
My essay, “Do Americans Believe Science and Religion Are in Conflict?” assessed the US survey data on the public’s beliefs about the relationship between religion and science. In the most basic version of the “conflict” question, a slight majority affirms that science and religion are in conflict. However, changes in wording and answer categories tend to produce very different results. By using multiple sources we can infer that this slight majority is actually a heterogeneous group. About one third of this group appear to believe that science and religion are entirely separate, non-overlapping domains, another third believe that science and religion conflict over very specific issues (e.g., human origins), and the last third appear to believe that science and religion are fundamentally opposing methods for gaining knowledge about the world.
In the online discussion, a reader brought up the concern that, with such a plethora of seemingly contradictory results, one should probably just ignore survey and poll results altogether. They are simply unreliable. I disagree with this position. Survey responses are frequently sensitive to word choice, ordering, and answer categories. But the very fact that questions about science and religion bend to some of these factors can be an invitation for further exploration. Respondents are clearly drawing on different conceptions and frameworks, and these need to be unpacked.
Let me suggest two fruitful avenues forward. One is the increased use of vignettes in survey questions. When the meanings imputed to ideas — such as religion and science — vary substantially, short vignettes that describe hypothetical behavior and beliefs, and ask the respondent to assess or react to the story, will often be more telling than an abstract, contextless question. There is a growing literature on using “anchoring vignettes.” to help ground survey responses in areas such as health, political efficacy, and job satisfaction. Perhaps extending this method to science and religion research would be a productive way to get “underneath” the ideas that the public draws upon when thinking about religion and science.
The second suggestion is to disaggregate complex beliefs into their component parts. I did something along these lines in the National Study of Religion and Human Origins. Instead of asking questions designed to categorize respondents as creationists, theistic evolutionists, or atheistic evolutionists (as the frequently cited Gallup poll question does), I asked a series of questions about evolution, God’s involvement in the emergence of humans, the historical existence of Adam and Eve, twenty-four hour days of creation, and geological time frame. These simpler questions become building blocks to more accurately construct complex beliefs and identities. Using this method, I found that very few people fit unambiguously into camps such as “Young Earth creationists” or “atheistic evolutionists”.
This brings up another concern from the discussion: namely, that public discourse on science and religion is dominated by relatively small groups of ideologues. Although the survey results suggest a little over half of the population see conflict between religion and science, it is only around 17 percent that view the two as inherently at odds. Within this group, the sides are roughly equally divided between those that “side with” science and those who “side with” religion.
This shouldn’t be too surprising. Controversial public issues are frequently dominated by groups whose voices are disproportionate to their numbers. The resulting perception is of a public who appear more divided than they probably actually are. We end up imputing “culture war” attitudes to large swaths of the public who probably harbor no such views.
Lastly, I think both of these concerns should cause us to reflect more carefully on who is presenting the findings. Certainly one of the reasons for the distrust in survey results lies with the assumption that people are wielding statistics on behalf of some agenda. The truth is, I don’t see this going on in the professional social science community. There is a robust commitment to careful analysis coupled with a curiosity and openness to following the data wherever they may lead. That said, “advocacy” groups (for lack of a better term) sometimes do selectively and carelessly use these findings to push their own agenda. I know most sociologists think that what people choose to do with their research is not their problem, but I think we need to recognize that this can damage the credibility of what we do and lead to the erroneous assumption that survey results can never be trustworthy.
New Big Questions:
1. What are the underlying narratives that the public primarily draws upon to shape their beliefs about the relationship between science and religion?
2. How can the social scientific study of religion and science be improved?