Popular imagination casts this as a four-century war. But the first person to propose that the universe had a beginning was a Belgian priest — and when a pope described that theory as scientific proof of creation, the objection came from the priest himself: a hypothesis not yet confirmed by observation has no business serving as evidence for a doctrine. Historians stopped writing this history as a war decades ago. The real relationship is not one battle line but at least four ways of coexisting.
"The intention of the Holy Spirit is to teach us how one goes to heaven, not how heaven goes."— Cardinal Baronio, quoted by Galileo in his Letter to the Grand Duchess Christina (1615) to argue that scripture and astronomy have separate jurisdictions
Start with what the warfare story hides. Medieval Latin Europe built its universities inside church institutions; most practitioners of natural philosophy were clerics, and the church was the single largest patron of natural inquiry of its day. The Islamic scientific flourishing of the 9th–13th centuries was likewise embedded in religious institutions: Ibn al-Haytham's Book of Optics established the practice of settling hypotheses by experiment, and part of the drive behind astronomy and mathematics was severely practical — fixing the direction of prayer and the calendar.
Copernicus published On the Revolutions of the Heavenly Spheres in 1543; he was himself a canon of the church. The rupture came when Rome placed heliocentrism on the Index in 1616 and when Galileo was found "vehemently suspect of heresy" and confined for life in 1633. But that case is tangled with a collapsed patronage relationship, a turf fight over the standing of Aristotelian natural philosophy, and one awkward technical fact: stellar parallax could not yet be measured, so the empirical case for heliocentrism was genuinely incomplete. When On the Origin of Species appeared in 1859, supporters and opponents again cut across denominational lines.
The word "warfare" arrived remarkably late: Draper's History of the Conflict between Religion and Science (1874) and White's A History of the Warfare of Science with Theology in Christendom (1896). Both were written inside a fight over the secularization of American higher education — polemics, not historical findings. The 1925 Scopes trial then turned "may evolution be taught?" into public theatre and hardened the impression further.
From the mid-20th century onward, historians of science such as Lindberg, Numbers and Brooke dismantled the warfare narrative through archival work and replaced it with the complexity thesis: within the same period and the same tradition, support, opposition and indifference typically coexist. From 1966, the physicist-theologian Ian Barbour built "science and religion" into a field of its own, and in his 1989–91 Gifford Lectures set out the fourfold typology still in use.
Start where the argument is most concentrated. On human origins, American adults split (Gallup, 2024): 37% say God created humans in their present form within roughly the last 10,000 years, 34% say humans evolved with God guiding the process, and 24% say humans evolved with no divine involvement. The first figure is the lowest in a forty-year trend; the last is the highest.
More telling is that the wording decides the answer. In Pew's 2015 survey, 59% of American adults said science and religion are "often in conflict" in general — but only about three in ten said science conflicts with their own beliefs. Among weekly worship attenders, 50% saw a general conflict; among those who seldom or never attend, 73% did. The further people stand from religious practice, the more certain they are that a war is going on somewhere else.
Now the scientists. Pew's 2009 survey of members of the American Association for the Advancement of Science: 33% said they believe in God and another 18% in some higher power — about half in total, against 83% of the public at the time. The gap is real, but "scientist equals atheist" does not hold.
And the gap depends heavily on country. Elaine Ecklund's team surveyed 9,422 biologists and physicists across eight regions (France, Hong Kong, India, Italy, Taiwan, Turkey, the UK, the US): 94% of Indian respondents reported some religious affiliation; 54% of Taiwanese scientists identified as religious, above the 44% of the local public; in Hong Kong the figures were 39% against 20%. In those places, scientists are more likely than ordinary people to have a religious affiliation.
Muslim societies are just as uneven (Pew, 2013): the median share accepting human evolution was 53%, rising to 79% in Kazakhstan, while 49% in Turkey and 55% in Indonesia said humans and other living things have existed in their present form since the beginning of time. The plain summary: conflict is regional and issue-specific, not a global constant, concentrated in a handful of societies around a handful of questions.
Barbour's four models are not four camps but four settings for the relationship; one person may use different ones on different questions.
The premise is that both sides are contesting the same kind of claim — how the world actually is. Both ends of the spectrum stand here: the literal creationist who dates the earth at 6,000 years and the person who holds natural science to be the only legitimate source of knowledge share that premise. The strongest version runs: if a tradition really asserts a 6,000-year-old world and the resurrection of the dead, it is making empirical claims, and empirical claims should be tested. Its weakness is assuming every religion has empirical propositions at its centre, when most traditions are weighted toward ritual and community.
The best-known version is NOMA (non-overlapping magisteria), proposed by Stephen Jay Gould in 1997: science holds the licence for fact and theory, religion for meaning and value, and the two do not overlap. As a truce it works well. The criticism (Dawkins and others) is equally direct: actual religions do make claims about the material world — miracles, the efficacy of prayer, the soul — and zoning those out amounts to rewriting believers' faith for them.
No merger, but an admission that each side's questions spill over. Typical topics: why is the universe intelligible in mathematical terms? How should the values of the physical constants be explained? It also covers methodological analogies — scientific models and religious metaphors are both referring tools that admit their own incompleteness, a position known as critical realism.
Natural theology, process theology, and "multi-level explanation" (one event described completely and without competition at the physical, biological and intentional levels). The highest payoff and the highest risk: hang a theology on a specific scientific conclusion and it falls when the conclusion moves — which is exactly why that Belgian priest objected to a pope treating the Big Bang as proof of creation.
The four models were formulated inside Christian theology; move traditions and the coordinates shift. Islam calls the natural world ayat (signs of God), so studying nature carries a flavour of religious duty. Hindu and Jain cosmologies already run on enormous timescales, leaving no purchase for a "6,000-year-old earth" dispute. In East Asian traditions, "Heaven" is not a legislator issuing physical laws, and creationism controversies barely exist. Before asking whether the conflict model applies, ask where a tradition places the divine.
In Kitzmiller v. Dover Area School District (2005), a US federal court held that "intelligent design" is not science but a religious claim, and that teaching it in public high-school biology violates the Establishment Clause. This was a ruling about a public curriculum, not about whether anyone's beliefs are true.
For the overwhelming majority of believers, science enters life as bioethics: brain-death criteria and organ transplantation, assisted reproduction, gene editing, end-of-life decisions, vaccines and public health. Each tradition digests new technology with the legal tools it already has — Islamic fatwa (juristic opinion), Jewish halakha rulings, Catholic magisterial documents, the opinion traditions of Hindu and Buddhist scholars. This is where "science and religion" actually happens every day.
The most common approach among religious scientists is neither an either/or nor a grand synthesis, but using two modes of questioning in different settings: in the lab, "what is the mechanism?"; elsewhere, "is this worth doing?"
Misconception: science and religion have been fighting throughout history.
That frame was set by two polemical books of 1874 and 1896, products of a fight over the secularization of American higher education. The mainstream position among historians of science today is the complexity thesis.
Misconception: the Galileo affair was a pure clash of science against faith.
It is tangled with a broken patronage relationship, a fight over the standing of a philosophical school, and a genuine shortage of decisive evidence — stellar parallax was not measured until 1838. But complicated does not mean blameless: in 1992 the Vatican itself acknowledged that the theologians' judgment had been wrong.
Misconception: scientists are basically atheists.
About half of AAAS members report belief in God or a higher power; in the eight-region study, 94% of Indian respondents reported a religious affiliation, and scientists in Taiwan and Hong Kong were more likely to be religious than their general publics.
Misconception: the creation-evolution fight is a universal religious battlefront.
It is highly localized, mainly to parts of American evangelicalism and some Muslim societies. Catholicism, mainline Protestantism, most Jewish movements and East Asian traditions have essentially no such dispute.
The switch is hermeneutical (see Day 23). How sharp the conflict gets is usually decided not by science but by how a text is read: read Genesis as literal chronology and a collision with geology is guaranteed; read it as a theological text and there is none. Which reading counts as "faithful" is an argument traditions have had internally for centuries, and science cannot settle it for them.
Where the divine is placed decides whether science can touch it (see Day 21). A God who intervenes continuously within natural law and a God who lights the fuse and then stands back expose wholly different surface areas to empirical testing.
Science is not an automatic engine of secularization (see Day 25). The relationship between education, scientific literacy and religious participation does not point the same way in every country — the higher-than-public affiliation rates among scientists in Taiwan and Hong Kong are the counterexample.
The live frontier has moved. The sharpest negotiations today are not about cosmic origins but about governing technology: climate, vaccines, reproduction and gene editing, artificial intelligence and the boundary of "consciousness" and personhood. On these questions religious bodies can be a source of doubt and resistance or a source of mass mobilization — and often both roles appear inside the same tradition.
This article presents multiple models of the science–religion relationship together with survey evidence, from a neutral and comparative standpoint. It does not endorse any one model as correct and passes no judgment on any tradition or individual. Survey figures are sample estimates for particular years and populations, and vary with question wording and definitions.