One of the better hi-fi theorists (perhaps Peter Moncrieff in IAR) opined that the only way for the 901 to work as Dr. Bose intended was with a recording made in an anechoic chamber (or outdoors, I suppose). The sound captured by the recording mics would be only the direct sound of the instruments and voices, no reflections or room sound.
Such a recording then played back on the 901 could at least have a chance of working, though as Al correctly points out, the arrival time of reflections off the walls and ceiling in a small listening room are much closer in time to the direct sound than are the reflections in a concert hall.
J. Gordon Holt was of the opinion that the real answer was to capture the direct sound with one set of mics, the hall ambiance with another, the two on different sets of recorder channels. The direct sound channels would then be played back on the front loudspeakers, the ambiance channels on rear speakers.