small-press poetry stats

nice to see john latta back in blog world with isola di rifiuti (which he's apparently been running since may, thus showing how current i am). repeated trips through his earlier hotel point will be well rewarded, trust me.

recently latta offered his take (the second section in a three-part post) on ron silliman's take on the attention span project of steve evans. each year since 2003 evans has invited readers to contribute lists of up to 11 recent poetry titles (or general books or recordings or films or performances or events) that have been most significant to them, with or without brief statements of why. evans posts the lists, and also compiles some statistics on which titles and publishers get cited most frequently. (i contributed a list, with comments, for the first time this year.)

the results are always interesting, especially the individual lists. but as i mentioned to david buuck and ca conrad in a private email yesterday, such things have about as much sociological significance as a survey on dailykos that says "click here if you hate bush." as most of us are aware, results from self-selection sampling can rarely be extrapolated to a general population with any confidence or accuracy. first, you can't identify or assess the full population that actually could have contributed a list. second, since self-selectors may have specific motivations for responding, you cannot correct for biases introduced into the sample on account of those motivations or take into account the biases of those who choose not to respond. and so on. (in fact, i am a perfect case in point: i contributed a list because increasingly i'm finding the books of poetry most important to me are being written by women and i wanted that fact to be registered.)

in any case, latta is spot on when he refers to silliman's "bean-counting" and calls this "a dumbshow algebra of proof." let me be clear: i do not disparage bean-counting. in fact i find it to be a very useful activity when the context calls for it, and silliman can be among the very best there is in this regard. but 46 responses from an estimated field of 10,000 isn't just one of "some methodological limitations" as ron calls them, it's tantamount to no methodology at all -- if not entirely "meaningless," according to latta, "as a data-set for statistical analysis," then at the very least pretty near insignificant. the resulting analyses in this case makes for a terribly poor showing of poets trying to be sociologists while misapprehending basic principles of statistical analysis.

i also find the competitiveness such attempts at numbercrunching fosters -- a competitiveness i might add that lies at the heart of global free market capitalism and that i would like to think small-press poetry opposes on more than a nominal level -- utterly disheartening. lisa robertson appears on these lists so frequently because she is a fantastic poet, pure and simple. must we then pit this fact against rodrigo toscano's only having appeared once a year? he's also a fantastic poet, such lists completely notwithstanding. sure, they give us some insight into reading habits from which perhaps some general inferences can tenatively be made. beyond that -- and to the extent that they foster the competitiveness, branding and who's-hot-who's-not faddism of larger marketplace banalities -- pseudo-sociological analyses derived from the attention span stats ultimately do small-press poetry a disservice.