Absolutely - it offers a very good baseline for one reason and one reason alone - it removes confounding variables and allows studies to cross reference each other without those variables confounding results. From a purely scientific perspective, it makes sense as a baseline.
If you read what I actually said, the dubious part is assuming that the music presentation in an anechoic chamber is at all "natural", "neutral", or "desirable".
Harman's research starts from the position that sound reproduction inside a anehoic chamber is the very definition of neutral. That is the aspect that is dubious.
And if you consider the conclusions of the Harman research itself, in which no one preferred a room equalized frequency response that mimiced an anechoic chamber, you'll understand how Harman's own research confirms this.
In fact, I'm not aware of any study which correlates the absolutely flat acoustics of an anechoic chamber with anyone's preference for ideal sound reproduction.
All of these things support my original claim that assuming an anechoic chamber is natural/neutral/desirable is dubious. When proposing an ideal target for sound reproduction defined as neutral/natural/desirable - the acoustics in an anechoic chamber are not a good definition for that.
That is where the hand waving from Harman comes in. They want to say that listener preferences magically correct for optimal ideal frequency response. The fact that 87% of listeners preferred the Harman curve, they argue, is evidence that it is natural/neutral/"correct". There is scientific evidence that 87% of listeners prefer the Harman curve, there is zero scientific evidence for the sweeping claim that listener preference implies neutral/natural sound reproduction.
So again, my point is simply that by choosing Harman has an objective standard, you are making a subjective determination about what is important to you (which, in turn, is based on the subjective determinations of the research subjects involved in the study). 87% of listeners agreeing with a certain sound signature is, subjectively to you, enough to propose an objective standard. Your objective standard target is fundamentally based on your own subjective value decision.
You have used a lot of words to dance around this point, and stopped short of admitting this in all of our exchanges. I long ago said that the Harman curve is one objective standard of many, and that you can make objective decisions regarding equipment that deviates from it. The manner in which you are continuing to reply to me, you continually want to weasel out of admiting that by even choosing the Harman standard in the first place, you are making a subjective decision, and you don't want to admit that it is "subjectivity all the way down".
I'll address your last paragraph first. I don't dance and you have not said that “the Harman curve is one objective standard of many” until now, so I'll take that as a concession of sorts. you have said this, however:
"I think part of the blame is the rise of measurements in the first place. It gives the illusion of an objective standard by which to measure all other headphones. The problem is, it is just an illusion.”
“Strange, isn't it? People parade around these "reference curves" which are based entirely on people's preferences, then turn around and say that all preferences are invalid because accuracy is king. It's a certain ironic level of cognitive dissonance that is totally lost on them.”
“It's all over the top absurd. People can live and die by the accuracy illusion, because that's all it is, an illusion. A ghost. A false prophet. Chasing the dragon. The rest of us can get back to the music.”
you’re a dyed in the wool subjectivist and that’s your prerogative. If your goal all along has been for me to admit that by adopting the Harman headphone target response curve I would be making a subjective decision then I am happy to oblige – yes, of course it is. So what of it?
Is that really what you mean when you say “it is subjectivity all the way down’? if so, then based on your logic every choice I make renders the objective subjective and that is absurd. If I choose to use objective data, modelling, free-field, diffuse field or Harman target curves, census results etc., then I am making a subjective choice but what I have chosen
is still objective.
with regard to the use of an anechoic chamber, you appear to be confusing it with a frequency response curve, which is actually the baseline or target. the anechoic chamber is used by audio engineers to measure the transducer against that target frequency response curve.
the measurements can be taken at a substantial distance away from the transducer in a large anechoic chamber. this can also be done outside in a suitably quiet environment. both methods are taking measurements in the “free field”. and as I said in my previous post, the objective is to obtain accurate loudspeaker measurements by eliminating sonic artifacts caused by room reflections.
and as I also said, Harman international isn’t the only loudspeaker manufacturer that does this. it’s common practice in the industry and has been for years. headphone manufacturers have also adopted it. so while you regard it as dubious, they clearly don’t.
with regard to your remaining points, well you’re just repeating yourself frankly, and I’ve already addressed them in my previous responses. the argument has become circular as these arguments usually do, so I think we should agree to disagree as they say.
here's a thought - let’s say that I audition a number of headphones and find myself preferring the pair that most closely aligns with the Harman target response curve, without actually knowing that it does. could that be compelling evidence for the accuracy of the Harman target response curve as a predictor of people’s sonic preferences and could that make it the new “objective” standard?
some interesting thoughts on "hard-core subjectivists" and neutrality in audio from a notorious headphone reviewer:
https://www.innerfidelity.com/content/response-cacophony-tyll-stance-reviewing