>I think you missed the point of my post, since the argument was that "free
>access" to one choice (abortion) but not the other choice (funding to take
>care of a child) is hardly "free access" to a choice, but merely an
>(improved) resolution of a bad situation.
Nathan, some would merely remove the parenthetical (improved) and argue that under present conditions free access interacts with poverty to coerce women to choose abortions, thereby undermining the strength of the race in war of survival. It would then follow that free access should not be granted until poverty and racism were ended first in struggles to be led by, as Hattie Gossett once put it, gun-toting, reefer smoking revolutionary daddies. I know that *you* don't think that way. Was ony suggesting that some may read your comments that way. I am sorry for misreading you. Truly concerned about how your formulations may be vulnerable to misreading by some whose misinterpretations must therefore be more carefully guarded against.
To get back to that Chicago study--and your expert critical analysis as a sociologist would be appreciated--the relationship between abortion and low crime rates can only be stable if is grounded in the existence of a socio economic machine in which the poor surplus to capital's demand are tracked for criminalisation, instead of, say, shared work with others to reduce labor time for all or public employment. Moreover, the study not only implies the existence of such a machine but also that it is strongly resistant to externally induced change or evolutionary transformation; otherwise the relationship between abortion and low crime rates would not turn out to be relatively invariant or possess counterfactual import (if these children had been born, they would have been criminals): the machine must be ideologically frozen.
Moreover, since the relationship between the underlying causal structure/machine and the statistical relationship between abortion and lower crime rates to which that structure/machine putatively gives rise is not itself a causal relationship, the difficulty cannot be solved merely by adding a new variable (one indexing that particular machine) to the equations used to formulate specific causal models. The statistical technique thus used to demonstrate a causal relationship between abortion and low crime rates fails in its task as long as the 'discovered' relationship is understood to be relatively invariant or to possess counterfactual import.
I draw here from Vaughan MCKim's summary of Nancy Cartwright's conceptual critique of social science statistical analysis. His summary and Cartwright's essay are in Causality in Crisis? Statistical Methods and the Search for Causal Knowledge in the Social Sciences.
yours, rakesh