Preprint Presents Sharper Spectroscopic Case for a Recoiling Black Hole in Quasar 3C 186

A revised preprint on arXiv says astronomers may have their clearest spectroscopic case yet for a recoiling supermassive black hole in the quasar 3C 186 — a striking claim that, if it holds up, would offer rare direct evidence that giant black holes can merge and then be kicked away from a galaxy’s center. But the result is still a preprint, not a settled finding, and the recoil interpretation remains the authors’ own.

In the paper, “A recoiling supermassive black hole in a powerful quasar,” posted on arXiv as arXiv:2501.18730, Marco Chiaberge and colleagues report a new measurement in 3C 186, a powerful radio-loud quasar at a redshift of about 1.068. The arXiv record shows the latest revision, v4, dated June 17, 2026, and notes it was revised after referees’ comments, though no final journal reference is listed.

The paper’s central result is that both the quasar’s accretion disk — the hot material spiraling into the black hole — and its broad-line region, the fast-moving gas close to the black hole, appear blueshifted by the same amount relative to the host galaxy. The authors give a line-of-sight velocity of minus 1,310, plus or minus 21, kilometers per second.

As Chiaberge and colleagues write in the abstract, “Here we present new and conclusive spectroscopic evidence that both the accretion disk and the broad line region in the spatially offset quasar 3C 186 are blue-shifted by the same velocity relative to the host galaxy, with a line of sight velocity of (-1310 +- 21) km/s. This is best explained by the GW recoil super-kick scenario.”

Gravitational-wave recoil refers to a kick delivered when two black holes merge and emit gravitational waves more strongly in one direction than another, sending the merged remnant moving away at speeds that can reach hundreds or even thousands of kilometers per second. Astronomers care because a convincing example would be unusually direct evidence that supermassive black holes do coalesce in nature, bearing on the long-standing “final-parsec problem” — the question of how black hole pairs get close enough for gravitational waves to finish the merger.

3C 186 has been on astronomers’ radar for years. In a 2017 study based on Hubble Space Telescope observations, Chiaberge and colleagues reported that the quasar’s active nucleus was not centered in its host galaxy. As that paper’s abstract put it, “HST imaging shows that the active nucleus is offset by 1.3 ± 0.1 arcsec (i.e. ~11 kpc) with respect to the center of the host galaxy. Spectroscopic data show that the broad emission lines are offset by −2140 ± 390 km/s with respect to the narrow lines.”

A 2022 study led by Takahiro Morishita, also using deep Hubble imaging, measured a projected offset of 11.1 plus or minus 0.1 kiloparsecs between the quasar point source and the center of the host galaxy, and concluded that the recoiling-black-hole scenario remained plausible. That same year, Castignani and colleagues used NOEMA observations of carbon monoxide gas, specifically CO(4-3), to measure the host galaxy’s molecular gas at z = 1.06811 plus or minus 0.00005. That helped independently anchor the host galaxy’s systemic redshift and supported a substantial velocity offset between the quasar’s broad-line region and the host.

The new number is notably smaller than earlier estimates of roughly 2,000 kilometers per second, so this is not simply a repeat of the older claim. It is a revised and more precise spectroscopic result. It also comes with an important caveat: broad-line shifts and off-center active galactic nuclei can also be discussed in terms of outflows, binary black holes or merger-related host dynamics, which is why independent confirmation matters.

That leaves 3C 186 in a familiar but important place: still one of the strongest candidates for a recoiling supermassive black hole, and now the subject of a sharper new claim. Whether the wider community accepts that interpretation will depend on peer review, follow-up work and independent reanalysis.