Reply to comment

“What difference can we define between that vague, (("floofy")) thing we call love and the HARD SCIENCE thing we call brain states?”

I'd imagine that a good approach is to clearly determine what “hard science” concepts one is attempting to study (i.e. pair bonding) and relate your findings to “love” insofar as a given instance of love includes that concept. I couldn't promise that this will “explain everything”, of course, but it can pick off the most familiar and empirically-accessible elements of otherwise vague and complex ideas. It might be possible to determine what brain states are common to love given certain cultural and social constraints, or what the consequences of a given state are with regard to pair bonding behavior, and so on, and in this way winnow the “floofiness” of the subject. At some point, you'd probably need a new way to tackle remaining uncertainty, which I would imagine involves finding the conditions that make “love” or another floofy thing so variable or simply holding a certain amount of variation as “given”.

This is still kind of difficult. When a good study - such as the pair bonding one - is careful to describe exactly what it is studying and takes care not to generalize too much or draw conclusions about "love" that are not supported, it is still important for the reader to recognize these limits before deciding what the study teaches them. A parallel that comes to mind is the somewhat strained attempt to produce a model of creativity that could connect a patient's art to their cognitive capabilities - even if this were done perfectly, it's tempting to interpret a finding as answering a “big question” that it might not really apply to. I can see this being a problem as new attempts to define “pieces” of things like love are passed from researcher to researcher in one long, overconfident game of telephone.

In the end, I think that it might be important to remember that, should the scientific community find that certain aspects of love are highly variable in a way that makes general explanations impossible despite functional parallels, that is still an explanation.