Viability1 as a criterion for personhood is plagued with even more serious difficulties. To be viable is to be able to survive. In this context, it means to be able to survive (in theory) outside of the mother’s womb. Of course, this is all about shades of probability, and even very young fetuses do not die instantly when exposed to air. So viability is already an ambiguous concept, poorly suited to provide the crisp boundary we seek.
Even if a non-arbitrary and unambiguous definition of viability were established, worse problems remain. For example, viability is every bit as much a function of society’s technological capabilities as it is of the fetus’s innate properties. The gestational age at which a prematurely born baby has a shot at long-term survival has been steadily regressing for decades as medical technology has advanced. Since personhood is about intrinsic value, it should not be dependent upon these extrinsic conditions. Predicating personhood on viability would mean that a 26-week-old fetus is a person today but would not have been two hundred years ago; similarly, personhood would begin earlier in the developed world than in sub-Saharan Africa. Most people would agree that this is absurd.
Finally, a sick, elderly, or disabled person who reaches a level of dependency roughly analogous to that of a pre-viability fetus would cease to be a person under this schema. Again, few pro-choicers would be willing to embrace this conclusion.