In 1991, when my wife Janet was pregnant with our second child, we were told by her obstetrician that we might want to consider an amniocentesis - just to make sure.

"Just to make sure of what?" we asked. "We had a sonogram, and we imagine that any serious problems would have shown up there."

An amniocentesis would take genetic material from the fetus itself, the doctor explained, thus affording us a crystal-clear indication of any anomalies that might lead us to terminate the pregnancy.

Janet and I discussed the possibility that day, but we didn’t discuss it for long. Janet pointed out that we wouldn’t get the results until 16 to 18 weeks into the pregnancy and, though we both supported abortion rights, we did not want to wait until the second trimester before deciding to terminate - unless the fetus was very severely damaged. And something very severe, such as anencephaly, would have been spotted on the sonogram.

I added (because I’d just learned it from reading the brochure) that amniocentesis induces miscarriage in one of every 200 attempts. Curiously, at age 36, those were the same odds against Janet’s having a child with Down syndrome.

"And what if we had a child with Down syndrome?" she asked. "Well, then," I said, uttering one of the more prescient and consequential sentences of my life, "we’d just love the baby all the more."

I guess it was a sweet thing to say, though in retrospect it was incredibly naive. Jamie Bérubé was born that September. He has Down syndrome. We love him dearly, and he delights us daily - not merely by being cute, but by being clever and sly and witty and charming and a good kid to his brother Nick.

But I didn’t know anything about Down syndrome in 1991. I thought that people who had it were kind of flabby and passive and had a life expectancy in the 20s. I knew just enough — about disability and about my own heart - to make what I think was the right decision about prenatal screening.

I would make the same decision today, even more emphatically, but I would bring to that decision a deeper skepticism, born of both experience and study, about whether prenatal screening tells us everything we think we need to know.

Since 1991, people with Down syndrome have made - as the cliché has it —great strides. So has biomedicine. It is now possible, as I suggested in the book about Jamie I published in 1996, to do a definitive genetic analysis of a developing fetus as early as eight weeks. And it’s possible to do so without sticking a great big fat needle into the uterus, as one does in amniocentesis.

Prospective parents who encounter the medical apparatus of prenatal screening today, then, are facing a world utterly unlike the one Janet and I inhabited 16 years ago: On the one hand, they can see happy and healthy children with Down syndrome all around them, being integrated into public life as never before; on the other hand, they can choose to screen for "genetic anomalies" (of which Down syndrome is by far the most prevalent) without fear of miscarriage and without the moral murk of a second-trimester abortion.

Many opponents of abortion rights, of course, do not understand why the moral murk should be any murkier in week 18 than in week eight. For them, life ("defects" or not) begins at conception and childbirth is mandatory. Yet even some opponents of abortion rights, together with most supporters, see moral shades of grey at every point in a pregnancy. They believe that some fetuses are more developed than others, and some fetuses are more disabled than others. Recognition of moral complexity in childbearing, in fact, can make for some strange alliances.

For example, as Rayna Rapp points out in Testing Women, Testing the Fetus, the ultra-orthodox Hasidim in New York are strenuous promotors of prenatal genetic screening because Tay-Sachs disease - a genetic disability so excruciatingly debilitating that it sometimes seems as if it were invented by bioethicists as an extreme limit case - occurs disproportionately often in Ashkenazi Jews