Module 12: Cosmology today · Lesson 12.4 · about 16 min
JWST and the surprisingly early galaxies
The James Webb Space Telescope has seen galaxies from under 300 million years after the Big Bang, and more bright early galaxies than expected. How they are found, the records, and why ΛCDM survives.
Module 12 · about 16 minutes. Read first: 4.5 Cosmological redshift and 10.5 The first stars and reionisation. Since 2022 the James Webb Space Telescope has pushed our view back to galaxies shining less than 300 million years after the Big Bang. It also found more of them, and brighter, than many astronomers expected, and headlines asked whether cosmology was broken. By the end you will be able to explain how the earliest galaxies are found, read the record chart, and judge the “too early” claims for yourself.
1. A telescope built for stretched light
JWST, built by NASA with the European and Canadian space agencies, was launched on 25 December 2021 from French Guiana. Its 6.5-metre mirror of 18 gold-coated segments sits 1.5 million km from Earth, behind a sunshield the size of a tennis court that keeps it at about −230 °C. It sees from 0.6 to 28 micrometres: red light and infrared (the astronomy course explains these wavelengths).
Why infrared? A galaxy at redshift z = 14 sent out its brightest light in the ultraviolet. The expansion has stretched every wavelength by 1 + z = 15 times on the way (Lesson 4.5), so that ultraviolet now arrives as infrared. Hubble’s infrared camera stops at about 1.7 µm; JWST goes much further, with a mirror more than five times larger in area.
2. Finding them: the vanishing trick
Neutral hydrogen gas filled the young universe, and it swallows ultraviolet light bluer than the Lyman alpha line at 121.6 nm. So an early galaxy’s light has a cliff: bright redward of the line, almost nothing blueward. Redshift moves the cliff to 0.1216 × (1 + z) µm. Take pictures through several filters, and a very distant galaxy is visible in the red ones but drops out of the blue ones.
A dropout is only a candidate. In 2022 a galaxy called CEERS-93316 looked like a record-breaker at z ≈ 16. A spectrum in 2023 put it at z = 4.9: a dusty galaxy whose bright emission lines had faked the colours (Arrabal Haro et al. 2023). Only a spectrum, splitting the light to find the break or known lines exactly, makes a redshift sure. Galaxy records below are all spectroscopic.
3. The record keeps falling
In 2024 the JADES team confirmed JADES-GS-z14-0 at z = 14.32 (Carniani et al. 2024), and in 2025 MoM-z14 took the record at z = 14.44 (Naidu et al. 2025): light from about 280 million years after the Big Bang. You can read about a fainter neighbour, JADES-GS-z14-1 at z = 13.9, in fibub’s discoveries blog, and turn any redshift into an age with the redshift calculator and cosmology calculator. As of late 2025, candidates beyond z = 15 exist from colours alone, but none had a confirmed spectrum.
4. Too many, too bright, too soon?
Two surprises arrived quickly.
Too massive? In 2023 Labbé and colleagues reported six red galaxies at z ≈ 7 to 9 that seemed to hold up to 10 to 100 billion Suns’ worth of stars, only 500 to 700 million years after the Big Bang. Mike Boylan-Kolchin checked them against ΛCDM. A galaxy cannot have more stars than the ordinary matter in its dark matter halo:
- M★
- the mass in stars
- Mhalo
- the mass of the dark matter halo the galaxy lives in; ΛCDM predicts how many halos of each mass exist at each time (Lesson 10.3)
- Ωb/Ωm
- the share of matter that is ordinary, about 16%; the rest is dark matter, which cannot make stars
- ε
- the fraction of that gas actually turned into stars, usually 10 to 20% or less in galaxies today
- Physical meaning
- Even with ε = 1 (every atom made into a star), there are only so many big halos early on. The most extreme 2023 masses needed ε close to 1, which would have been a real problem.
Spectra then changed the story. Several of these red galaxies turned out to have broad emission lines, the sign of gas whirling around a growing black hole. Light from the black hole had been mistaken for starlight, so their star masses were overestimated, sometimes by a lot. They belong to a new population nicknamed little red dots: small, very red, very common between z ≈ 4 and 8, and still being debated (as of 2025) as a mix of hidden black holes and dense young stars.
Too bright, too often? This surprise has lasted. At z above 10, JWST finds several times more ultraviolet-bright galaxies than many pre-launch models predicted. But those models were guesses about star formation, tuned to later times. Ideas that fit within ΛCDM include: star formation in short, intense bursts (catching galaxies at their brightest); very efficient star formation in the dense early universe, before supernovae could blow the gas away; less dust; stars heavier on average than today’s; and extra light from black holes. Watching how galaxies form (Lesson 10.6) has become one of the busiest fields in astronomy.
Deeper and earlier.
- 1995 The Hubble Deep Field: thousands of galaxies in a dark patch of sky.
- 2004 The Hubble Ultra Deep Field.
- 2016 GN-z11 at z = 11.1, with Hubble (JWST later measured 10.6).
- 2021 JWST launches on 25 December.
- 2022 First images in July; first galaxies beyond z = 13 confirmed in December.
- 2023 "Too massive" candidates; the z ≈ 16 candidate turns out to be at z = 4.9; little red dots.
- 2024-2025 Records at z = 14.32 and 14.44.
“JWST has disproved the Big Bang.” Some articles claimed this in 2022-2023. The galaxies JWST sees fit comfortably in a 13.8-billion-year-old universe; their light even shows the stretch and time dilation that expansion predicts. The surprise is about how quickly gas turned into stars inside dark matter halos. Several of the most dramatic claims were later revised with better data, which is how science is meant to work.
- JWST (launched 25 December 2021) sees in the infrared, where the expansion has moved early galaxies’ ultraviolet light.
- Distant galaxies are picked out by the Lyman break dropping out of bluer filters; only a spectrum confirms the redshift.
- Records: JADES-GS-z14-0 at z = 14.32 (2024) and MoM-z14 at z = 14.44 (2025), about 280 million years after the Big Bang.
- The 2023 “too massive” galaxies were partly black holes (“little red dots”) with overestimated star masses.
- More bright galaxies than expected at z above 10 is real, but it points to fast, bursty star formation, not to a broken ΛCDM.
Check yourself
Sources
- S. Carniani et al. (2024), Spectroscopic confirmation of two luminous galaxies at a redshift of 14, Nature 633, 318 https://doi.org/10.1038/s41586-024-07860-9
- R. P. Naidu et al. (2025), A cosmic miracle: a remarkably luminous galaxy at z = 14.44 confirmed with JWST, arXiv:2505.11263 https://arxiv.org/abs/2505.11263
- I. Labbé et al. (2023), A population of red candidate massive galaxies ~600 Myr after the Big Bang, Nature 616, 266 https://doi.org/10.1038/s41586-023-05786-2
- P. Arrabal Haro et al. (2023), Confirmation and refutation of very luminous galaxies in the early universe, Nature 622, 707 https://doi.org/10.1038/s41586-023-06521-7
- M. Boylan-Kolchin (2023), Stress testing ΛCDM with high-redshift galaxy candidates, Nature Astronomy 7, 731 https://doi.org/10.1038/s41550-023-01937-7
- E. Curtis-Lake et al. (2023), Spectroscopic confirmation of four metal-poor galaxies at z = 10.3-13.2, Nature Astronomy 7, 622 https://doi.org/10.1038/s41550-023-01918-w
- P. A. Oesch et al. (2016), A remarkably luminous galaxy at z = 11.1 measured with Hubble Space Telescope grism spectroscopy, ApJ 819, 129 https://doi.org/10.3847/0004-637X/819/2/129
- J. Matthee et al. (2024), Little red dots: an abundant population of faint active galactic nuclei at z ~ 5 revealed by the EIGER and FRESCO JWST surveys, ApJ 963, 129
- NASA, James Webb Space Telescope mission pages https://science.nasa.gov/mission/webb/
Your ticks stay in this browser. Save them as a file on the course home.
Updated 2026-11-22. Numbers are rounded for teaching; measured values come from the sources listed in each lesson. Simulations run in your browser; nothing you type is sent anywhere.