I have been noticing lately that there is a very strong correlation of affinity in relation to a philosopher’s personal life/background, tastes, interests, political views, etc. and their philosophical arguments. I find that, with a startling degree of regularity and accuracy I can read a philosophical essay by an unknown author and deduce–without any other information–whether or not I share some particular degree of affinity with the author’s personal experiences, tastes, preferences, social views, etc. And the reverse has been tested with similar results–that I can expect to easily comprehend and support a philosophical viewpoint based on a brief examination of the biographical information about that particular author.
Thus, Nietzsche’s observation that philosophy is “the personal confession of its author and a kind of involuntary and unconscious memoir”.
Based on this observation, it would seem that ad hominem arguments are, in fact, quite useful empirical predictors (and of course, it is accepted that they are a particularly effective form of argument in terms of success). Thus, the designation as “fallacy” (which has negative connotation) is misplaced.
It seems to me that the main consideration of ad hominem as a fallacy is the representationalist idea that the linguistic statements (arguments) of an individual “actually” provide a reference to something other than (and apart from, outside, separate, etc.) the “nature” of the individual. If you don’t agree with this position (and I am disinclined to agree), then it seems that ad hominem is quite useful.