Hubble sights a galaxy with 'forbidden' light
phys.org
phys.org
Non-expert readers could be forgiven for reading that and thinking this is some sort of "discovery" ("Hubble observes something new and unusual!"), that the forbidden emission lines only appear in Type 2 Seyfert galaxies ("The spectral lines that Type-2 Seyfert galaxies emit are associated with specific 'forbidden' emission lines."), and that they only come from the active nucleus ("in the midst of an incredibly energetic galactic core"). But none of that is true.
Forbidden emission lines are ubiquitous in astronomy. Any time you see an object with emission lines, some of the emission will be forbidden. So we see forbidden emission lines from Seyfert 2 nuclei, from Seyfert 1 nuclei, from quasars, from non-Seyfert nuclei, from the nebula of star-forming regions, from planetary nebulae, from supernova remnants...
You could pick literally any spiral galaxy that Hubble has observed and title the image "Hubble sights a galaxy with 'forbidden light'", and the title would be just as correct. Or swap "nebula" for "galaxy" and use any Hubble image of a nebula in our own galaxy. (There's even forbidden emission coming from the solar corona.)
I have a physics degree and I still don't understand what "forbidden spectra" meant according to this article.
It has nothing to do with black holes, except that accretion disks around black holes can, due to their high temperature, produce the necessary energetic photons. (But so can massive stars, for example.) The rotational energy of the black hole is not involved.
In a way, the tautology captures the generally squishy quantum mechanical definition of "forbidden".
according to usual approximations (such as the electric dipole approximation for the interaction with light), the process cannot happen, but at a higher level of approximation (e.g. magnetic dipole, or electric quadrupole) the process is allowed but at a low rate.
The way out is contracting the edge of the forbidden intermediate between the interactions you'd have to decompose the forbidden transition into, until it's so short in time that you hit Heisenberg or similar and basically quantum-tunnel across the sequence of decomposed interactions.
Which, just like quantum tunneling, can and does happen, but it requires multiple low-occurrence events to randomly happen so quickly in succession that the quantum coherence time is essentially long enough to span across these individual events that have to occur in sequence.
Same here, and probably true of most readers.