Sibilants, according to one source, is (treble) information out-side of the time envelope.
First, lets get our terms straight. "Sibilants" is the plural of "sibilant", which is a word for the sound shared by the letters "s", "sh", "z" and "zh" (in English), plus a few others in other languages. Producing or reproducing sibilants without stridency, distortion or harshness may be challenging, but sibilants don't necessarily include any of those problems. They're just a type of vocal sound.
"Sibilance" is an adjective meaning, "sounding like a sibilant". It too does not necessarily imply unpleasantness, though its more often used so. Your source probably meant to write, Sibilance [distortion] is treble information out-side of the time domain.
With that I mostly agree. Time-domain errors are the major cause of distorted or harsh sibilants. (Another is exaggerated amplitudes, which leads to stridency.) Weve discussed speed control of the turntable, slewing distortions in the phono stage (or anywhere in the amplification chain), cartridge overhang, zenith angle and VTA/SRA. All these directly affect reproduction of waveforms in the time-domain.
I also agree that capacitors (in the crossover or elsewhere) affect time-coherency. Anyone whos done a cap upgrade has probably heard this. Its impossible for capacitance to occur without a time delay, but the better the cap the less delay therell be. Best of all is to eliminate capacitors completely, when feasible.
Another source of high frequency time-domain errors is the insulation of wires in the signal path. If the insulator in contact with a conductor has a significant dielectric quotient it can act as a micro-capacitor for the frequencies its in contact with. Now in any electrical conductor its the high frequencies (like sibilants in the 8-10kHz region) that travel at or near the surface, ie, near the insulator. Since teflon and other typical insulators are poor conductors (obviously), they induce spurious time delays in those frequencies.
The cure is to use wire that does not put materials in contact with the conductor. Interconnects and speaker cables with an air dielectric (or, theoretically, a vacuum) produce audibly clearer, less distorted highs. (If that sounds like a plug for Nordost it is, because it works.)
The speed and linearity of the tweeter also influences sibilant reproduction. My old B&W Nautilus aluminum tweeters were unforgiving. Everything else in the system had to be perfect or they could split your ears. The new diamond tweeters are faster, more linear and more revealing, but paradoxically more forgiving. They distort so much less that distortions coming to them from elsewhere in the signal path are not exaggerated. A good ribbon tweeter is even better.
The opposite approach, soft dome fabric tweeters, tames sibilants by smoothing each waveform and reducing amplitudes. This is a different kind of time domain distortion, similar to what Koetsu does in cartridges, but if its well done it can produce kinder, gentler highs. Not my cup of tea, because at the end of the day it conceals musical information, but its an approach used by many very popular speakers.