We've been having a slight discussion regarding this topic in the subwoofer thread and it might be a little more fun and on topic to discuss it further here. I believe that speakers should not have a flat SPL, but follow more closely to how an average human ear hears. Flat speakers to me sound hard, bright, and unnatural. I find B&W speakers to sound very pleasing tonally and if any measurements are to be looked at, they follow the equal loudness scale very accurately. B&W being a multi-million dollar company who owns two anechoic chambers and uses computer software to model crossovers could build a flat SPL speaker in a matter of minutes.
If a recording was made with a flat calibrated microphone, then it should be played back on a non flat speaker system in order to sound flat to an average ear. Playing the recording back on a flat system will actually make it sound incorrect, because at that point one hears what the microphone did and not what a person would have during the event.
Here is the equal loudness chart:
This is what Linkwitz had to say:
Electro-acoustic models
"H - Psycho-acoustic 3 kHz dip
Our perception of loudness is slightly different for sounds arriving frontally versus sounds arriving from random directions at our ears. The difference between equal-loudness-level contours in frontal free-fields and diffuse sound fields is documented, for example, in ISO Recommendation 454 and in E. Zwicker, H. Fastl, Psycho-acoustics, p. 205.
Diffuse field equalization of dummy-head recordings is discussed in J. Blauert, Spatial Hearing, pp. 363, and headphone diffuse field equalization by G. Theile in JAES, Vol. 34, No. 12.
Reference to a slight dip in the 1 to 3 kHz region for loudspeaker equalization is made in H. D. Harwood (BBC Research Department), Some factors in loudspeaker quality, Wireless World, May 1976, p.48.
Around 3 kHz our hearing is less sensitive to diffuse fields. Recording microphones, though, are usually flat in frequency response even under diffuse field conditions. When such recordings are played back over loudspeakers, there is more energy in the 3 kHz region than we would have perceived if present at the recording venue and a degree of unnaturalness is introduced.
This applies primarily to recordings of large orchestral pieces in concert halls where the microphones are much closer to the instruments than any listener. At most listening positions in the hall the sound field has strong diffuse components.
I use a dip of 4 dB (x1.gif, 2760NF) to equalize for this. The circuit consists of R, C and L in series, forming a frequency dependent ladder attenuator in conjunction with the 5.11k ohm source resistor. You may choose to make the notch filter selectable with a switch for different types of recordings.
I have found through my own head-related recordings of symphonic music that the dip adds greater realism, especially to large chorus and to soprano voice and allows for higher playback levels. "
And this is what BBC determined after doing a $1 million (nearly $5 million in today's dollar) psychoacoustic research some 40 years ago: