Argument

Stellar Formation Refutation

Intro

Astronomers say stars form when a cold cloud of gas and dust collapses under its own gravity, heats up in the middle, and finally starts nuclear fusion. Some young-earth writers have argued that this cannot work: hot gas pushes outward, spinning clouds should fly apart, a collapsing cloud should break into pieces instead of making one star, and one famous old star, nicknamed Methuselah, once seemed to be older than the universe itself.

This page takes that case seriously. It states each move in its strongest form, then gives the astrophysical answer, then says plainly how the move holds up.

The honest result is this. The strong claim, that stars cannot form naturally, does not survive. Cold clouds really are unstable, the heating and the breaking into pieces are part of how stars form, and the Methuselah problem went away when the star was measured better. Some real puzzles remain, especially how a forming star gets rid of its spin. Danny Faulkner, the leading young-earth astronomer, reviewed the science in 2021 and concluded that some star formation today appears unavoidable. Christians do not need this refutation. Scripture says God made the stars. It does not say they cannot form.

◎ See this page's graph and discover linked topics →

In full

The refutation case is a set of attempted defeaters against the standard model of star formation, the gravitational collapse of dense cores within cold molecular clouds into protostars with accretion disks. It runs: (1) by the Jeans criterion, thermal pressure in interstellar gas resists gravitational collapse; (2) collapse heats the gas, which should halt contraction; (3) conservation of angular momentum should spin a collapsing cloud up until it can contract no further, and the Sun holds very little of the solar system's angular momentum; (4) collapse should fragment a cloud into many small pieces rather than produce an individual star; and (5) the metal-poor subgiant HD 140283, the "Methuselah star", was dated older than the universe. A sixth framing claim is that no star has been observed forming from start to finish.

Each move is answered from the professional literature. (1) Molecular clouds cool to about 10 K by radiating through molecules and dust, and at those temperatures and densities the Jeans mass of a dense core is a few solar masses, so cores are unstable. (2) While the gas is optically thin, the collapse is nearly isothermal, so heating does not stop it until the core becomes opaque, which is when a protostar forms. (3) The angular-momentum problem is real and openly stated in textbooks; the proposed solutions (magnetic braking, transport through disks, winds and jets) are physically grounded and partly observed, though their details remain under active study. (4) Fragmentation is a prediction of the model, not a problem for it: most stars form in groups and clusters. (5) Later work on HD 140283 puts its age at about 12 billion years, inside the age of the universe.

The assessment is that the refutation fails as a refutation. What survives is narrower and still useful: open problems in angular-momentum transport and in what sets the masses of stars, and the observation that the formation of any single star takes far longer than human history. Those support humility about details, not the conclusion that stars cannot form. The young-earth movement is itself divided here, and the codex sides with the evidence. The design question is better asked one level up, about the laws that make star formation possible, which is where Fine-Tuning Argument and Argument from the Suns Origin take it.

Argument structure

Cheatsheet

The 30-second reply (when a young-earth friend uses this argument):

I would not use that one. Cold gas clouds really can collapse. They cool down to about ten degrees above absolute zero, and at that point gravity wins. The cloud breaking into pieces is how clusters of stars form. The Methuselah star was re-measured and it fits inside the age of the universe. There is one real puzzle, how a new star gets rid of its spin, and scientists are still working on the details. Even Danny Faulkner, a young-earth astronomer at Answers in Genesis, says some star formation today looks unavoidable. The Bible says God made the stars. It never says they cannot form.

The 30-second reply (when a sceptic asks whether Christians deny star formation):

Some young-earth writers have, but the Bible does not require it, and the best-informed young-earth astronomer does not hold it. Genesis says "he made the stars also." That is a claim about who, not a ban on how. The deeper question is why the laws of physics let gas clouds become long-lived stars at all, and that is a question naturalism has to answer too.

The 5 fast facts:

1. Cold clouds. In giant molecular clouds and dense cores the gas is around 10 K, and a dense core's Jeans mass is a few solar masses (University of Maryland ASTR 406 notes).
2. Cooling. Clouds shed heat through molecules such as H2 and CO and, at high density, through dust radiating in the infrared.
3. The spin problem. A 0.1 parsec cloud core has about a million times the Sun's angular momentum per unit mass; the excess must be carried away, and there is still a lot of discussion about how.
4. Clusters. Lada and Lada (2003): 90 percent of stars that form in embedded clusters form in rich clusters of 100 or more members.
5. Methuselah. Bond and colleagues (2013) found 14.46 billion years with an uncertainty of about 0.8; Tang and Joyce (2021) found 12 plus or minus 0.5; Guillaume and colleagues (2024) found 12.3.

What NOT to defend:

• Do not say gas pressure makes collapse impossible. The Jeans criterion says when collapse happens, and cold cores meet it.
• Do not say star formation violates the second law of thermodynamics. Collapse radiates energy away; Faulkner rejected this argument as early as 2001.
• Do not cite the Methuselah star as a problem for the Big Bang. Its revised age fits, and even the old figure overlapped within the error.
• Do not say astronomers have no evidence. Protostars, disks and jets are observed directly.

The closing line:

God does not need a gap in astrophysics to be the Maker of the stars. He counts them and names them, and the laws that let them shine are His too.

M1, Gas pressure beats gravity

The refutation case

1. The Jeans criterion. Sir James Jeans showed that a cloud collapses only if its mass exceeds a critical value set by its temperature and density. Below that mass, thermal pressure wins and the cloud disperses.
2. Interstellar gas is too hot and too thin. For typical interstellar conditions (about 100 K and very low density) the Jeans mass is around ten thousand solar masses, far more than any single star.
3. Most gas is not collapsing. At any given time most of the gas in a galaxy is not forming stars, which suggests that collapse is not the natural default.

The mainstream answer

1. Star formation happens in cold, dense gas, not average gas. The Jeans mass scales as temperature to the 3/2 power and density to the minus 1/2. Lecture notes for the University of Maryland's ASTR 406 give about 10 solar masses (thermal) for giant molecular clouds at about 10 K and a few solar masses for dense cores at the same temperature.
2. Clouds cool themselves. The same notes explain that cooling proceeds through molecular radiation (H2 and CO) and, at low temperature and high density, through dust grains radiating in the infrared; once a cloud is optically thick it is shielded from outside radiation and "the interior portions of the gas can cool in peace," with the equilibrium temperature dropping to about 10 K.
3. The measured clouds are unstable. Faulkner's review for young-earth readers concludes that calculations based on the measured conditions within giant molecular clouds indicate that they are unstable against collapse (Answers Research Journal 14, 2021).
4. The "most gas is quiet" point is granted and already in the model. Turbulence and magnetic fields support much of the gas; McKee and Ostriker describe turbulence as both creating the overdensities that start collapse and countering gravity elsewhere (ARA&A 45, 2007).

Verdict

Fails. The Jeans criterion is the tool astronomers use to show which regions collapse. Applied to cold dense cores, it predicts collapse.

M2, Collapse heats the gas and stops itself

The refutation case

1. Compression heats gas. Squeezing a gas raises its temperature, and higher temperature raises the Jeans mass.
2. So collapse should stall. For an ideal gas that cannot lose heat, the collapse would heat itself to a standstill.

The mainstream answer

1. While the gas can radiate, it stays cool. Course notes at Case Western Reserve University explain that early in the collapse the cloud is optically thin, so "the energy of collapse is immediately radiated away" and the collapse is isothermal.
2. Isothermal collapse runs away. As density rises at constant temperature the Jeans mass falls, so collapse speeds up instead of stalling. The Maryland notes put the condition precisely: if the effective polytropic index is below 4/3, the collapse runs away.
3. Heating is the next stage, not the end. When the density and opacity rise high enough, the energy of collapse is trapped and the core heats. That is the formation of a protostar, which then continues to contract and accrete.

Verdict

Fails. The objection describes a gas that cannot cool. Molecular clouds can, until the point where heating is exactly what the model needs.

M3, Spin should halt collapse

The refutation case

1. Angular momentum is conserved. As a rotating cloud shrinks it spins faster. The Maryland notes work the example: a solar-mass parcel sharing the Galaxy's rotation would have to spin about a hundred trillion times faster to shrink to the Sun's size, a rotation period of about a minute against roughly 3 hours for break-up.
2. The numbers are enormous. The same notes give the angular momentum per unit mass of a 0.1 parsec dense core as about 10^21 in cgs units, against about 10^15 for the Sun, and conclude: "This is a serious problem!"
3. The Sun is backwards. It has about 99.8 percent of the solar system's mass but rotates slowly, and Jupiter carries far more angular momentum than the Sun does.

The mainstream answer

1. The excess is carried away, and the mechanisms are physical. The Maryland notes list magnetic braking (a field threading the cloud enforces rotation and moves angular momentum outward, before and during collapse), collapse to a disk with internal stresses such as the magnetorotational instability moving angular momentum outward and mass inward, star-disk magnetic coupling, and winds and jets.
2. The machinery is observed. Disks around young stars are imaged directly. ALMA's 2014 image of HL Tauri, a system no more than a million years old, shows a disk with concentric bright rings separated by gaps (ESO, eso1436).
3. The theory has its own known difficulties, and they are being worked on. Zhao and colleagues review the "magnetic braking catastrophe": realistic magnetic braking in simulations removes so much angular momentum that disks become hard to form, the opposite of the original problem. They discuss non-ideal magnetic effects, field misalignment and turbulence as resolutions (Space Science Reviews, 2020).

Verdict

The strongest move, but not a defeater. The angular-momentum problem is real, and the literature says so openly: "There is still a lot of discussion about how this happens" (Maryland ASTR 406 notes). But an unsettled mechanism for shedding spin is not evidence that spin cannot be shed, and observed disks, jets and slowly rotating young stars show that it is. A Christian can cite this as an area of honest uncertainty; it cannot carry the conclusion that stars cannot form. The Sun's own angular-momentum split is treated as a design datum, not an impossibility, on Argument from the Suns Origin.

M4, Collapse fragments the cloud

The refutation case

1. Falling Jeans mass means breaking up. If the Jeans mass drops as the cloud collapses, smaller and smaller pieces become unstable on their own.
2. So the cloud shatters. A collapsing cloud should break into fragments rather than build a single star of the right mass.

The mainstream answer

1. Yes, and that is the point. The Case Western notes put it plainly: instead of one giant cloud undergoing a monolithic collapse, the cloud fragments into small collapsing pieces. Each piece becomes a star or a multiple system.
2. Fragmentation has an end point. As each fragment grows denser its opacity rises until the energy of collapse is trapped and it begins to heat (Case Western notes), which ends the cool, isothermal phase that drives the breaking up.
3. Stars form in groups. Lada and Lada's review of embedded clusters reports that 90 percent of stars that form in embedded clusters form in rich clusters of 100 or more members, and that fewer than 4 to 7 percent of embedded clusters survive emergence from their clouds as bound clusters (ARA&A 41, 2003). Most clusters disperse, so stars born together do not stay together.
4. What sets each star's mass is still an open question. McKee and Ostriker list "what determines the initial mass function" among the important problems of the field (2007). That is a question about the distribution of masses, not about whether stars form.

Verdict

Fails. Fragmentation is a prediction of the model and matches what is seen: stars born in groups. The real open question is the origin of the mass distribution, which is a research problem, not a refutation.

M5, The Methuselah star

The refutation case

1. A star older than the universe. HD 140283, a nearby metal-poor subgiant, was dated in 2013 to 14.46 billion years, older than the 13.77 billion-year age of the universe.
2. So the timeline is broken. If stellar dating and cosmic dating disagree, one of them is wrong, and the model of star formation built on them is suspect.

The mainstream answer

1. The 2013 paper did not claim a conflict. Bond and colleagues gave 14.46 plus or minus 0.31 billion years from the measurement alone, but noted that uncertainties in composition, especially oxygen, raise the total uncertainty to about plus or minus 0.8 billion years, and concluded that within the errors the age "does not conflict with the age of the Universe," though the star "must have formed soon after the big bang" (ApJ Letters 765, 2013).
2. Better measurements lowered the age. Tang and Joyce, using interferometric radius measurements and the MESA stellar-evolution code, found 12 plus or minus 0.5 billion years and stated that the best-fitting age "is not in conflict with the currently accepted age of the universe" (Research Notes of the AAS 5, 2021).
3. Composition was the key. Guillaume and colleagues found 12.3 billion years using the star's own measured abundances, and showed that assuming a solar-scaled mixture instead pushes the age to about 14 billion, in tension with the universe's age (A&A 692, 2024). The original high figure came from the inputs, not the physics of formation.
4. It was never a star-formation argument. An apparently over-old star would challenge the cosmic timeline, not the collapse of gas clouds. And every proposed age, old or revised, is billions of years, which does not help a young-universe reading either.

Verdict

Fails; resolved. The anomaly was a matter of error bars and inputs, and later work removed it.

F, Nobody has watched a star form

The refutation case

No astronomer has observed a single star go from diffuse cloud to ignited star. The sequence is assembled from many different objects, each caught at a different stage.

The mainstream answer

1. The timescales forbid watching one star. Even unopposed free-fall in typical cloud gas takes millions of years (the Maryland notes give about 5 million years at 50 to 100 particles per cubic centimetre).
2. Every stage is seen somewhere. Cold dense cores, embedded protostars, young stars with disks and jets, and newly visible young stars are all observed. Faulkner's review notes that T Tauri stars show what a protostar should: excess brightness, strong winds, bipolar outflows, and closeness to the clouds where star formation is thought to occur.
3. This is how historical sciences work. Geology does not watch a canyon form either. Reconstructing a sequence from many snapshots is a testable method; see the discussion of historical science on Argument from the Suns Origin.

Verdict

True but weak. It is a fair reminder that star formation is a reconstruction, which is a reason for care about details. It is not evidence against the reconstruction.

Honest assessment

1. The young-earth movement is divided on this. Faulkner's 2021 review for Answers Research Journal lists young-earth writers who argued that star formation from gas clouds would violate thermodynamics (Mulfinger, Williams, Henry) and one who argued that stars cannot form today because God made them all on day four (Thomas). Faulkner disagreed with the thermodynamic case and concluded, after reviewing the theory for fellow creationists, that from the physical standpoint "the conclusion that some star formation may occur today appears unavoidable."
2. Scripture does not settle the mechanism. In a 2014 article for Answers in Genesis, Faulkner wrote, "I don't see that the Bible absolutely precludes star formation today, nor do I see that the Bible demands it." He compared stars to animals and mountains: God's creating them does not stop new ones from being born or built.
3. The codex's own position. Elsewhere the codex argues that stellar lifetimes and the cosmic record point to a universe billions of years old (see 20 Arguments for Old Earth, argument 4). On that view the formation of stars over long ages is simply part of how God made the heavens, and nothing in the faith depends on its being impossible.
4. What a Christian can rightly say. Star formation depends on gravity, the behaviour of molecules and dust, magnetic fields and nuclear physics all working together. Why the laws are such that cold clouds become long-lived, life-supporting stars is a fair question for the naturalist, and it is the question Fine-Tuning Argument presses. That is a far stronger place to stand than a claim that the physics fails.

Master objections to this assessment

MO1: "Conceding star formation concedes naturalism."

• No. Describing how a process works does not explain why the process exists or why its laws hold. Scripture regularly ascribes to God what also has a describable process, such as rain and the growth of crops.

MO2: "Genesis says the stars were made on day four."

• "He made the stars also" (Genesis 1:16, ASV) says that God made them. Whether new stars can form within the order He made is a separate question, and Faulkner's point stands: Genesis neither forbids nor requires it.

MO3: "Open problems are open. Why not use them?"

• Because an argument that rests on an unsolved detail fails the moment the detail is solved, and it takes the credibility of the gospel with it. See God of the Gaps.

Connection to Scripture

• Genesis 1:16, "he made the stars also" (ASV), the Maker named without the method described.
• Psalm 147:4, "He counteth the number of the stars; He calleth them all by their names" (ASV).
• Isaiah 40:26, "Lift up your eyes on high, and see who hath created these, that bringeth out their host by number; he calleth them all by name" (ASV).
• Psalm 19:1, "The heavens declare the glory of God; And the firmament showeth his handiwork" (ASV).
• Job 38:31, "Canst thou bind the cluster of the Pleiades, Or loose the bands of Orion?" (ASV), a star cluster named as God's work.

Live-cite kit

• Scripture: Genesis 1:16; Psalm 147:4; Isaiah 40:26; Job 38:31.
• Young-earth review: Danny R. Faulkner, "A Review of Stellar Formation Theory," Answers Research Journal 14 (2021): 417-426.
• Mainstream review: Christopher F. McKee and Eve C. Ostriker, "Theory of Star Formation," Annual Review of Astronomy and Astrophysics 45 (2007): 565-687.
• Methuselah: Bond et al., ApJ Letters 765 (2013); Tang and Joyce, RNAAS 5 (2021): 117; Guillaume et al., A&A 692 (2024): L3.
• Aphorism: "God made the stars. He did not promise that astronomers would fail to understand how."

Tactical notes

• Do not run this argument in a debate. If an opponent raises it to caricature Christians, agree that the strong form fails and move to fine-tuning.
• If a fellow believer raises it, start with Faulkner. A young-earth astronomer's verdict carries weight that an old-earth critic's would not.
• Keep the one real puzzle in view. Angular momentum is genuinely open in its details; saying so shows you know the field.
• Never cite Methuselah as a live anomaly. It was resolved in 2021.

See also

• Argument from the Suns Origin, the design reading of the Sun's formation, which grants the model and argues from its specifics
• Fine-Tuning Argument, where the better question about stars belongs
• Argument from Thermodynamics, the cosmological argument from entropy, distinct from the failed second-law objection to star formation
• Second Law of Thermodynamics, why local order with energy loss does not violate it
• Stellar Nucleosynthesis, what stars do once they form
• Age of the Universe, the cosmic age the Methuselah star was measured against
• 20 Arguments for Old Earth, including stellar evolution timescales
• Young Earth Creationism and Old Earth Creationism, the two frames in which this argument is weighed
• God of the Gaps, the risk this argument runs
• Anthropic Principle, the selection-effect reply to stellar fine-tuning