In 2001 Regina McKnight, a 24-year-old South Carolina woman, was convicted of "homicide by child abuse" because she was a cocaine user whose baby was stillborn. She received the minimum sentence for that crime, 20 years, with eight years suspended. The conviction, the first of its kind, was outrageous for several reasons:

1) If McKnight had deliberately killed her baby by obtaining an illegal abortion, the maximum sentence she could have received would have been two years.

2) There was no solid evidence that McKnight's baby died as a result of her cocaine use, and even the general connection between cocaine and stillbirth had been called into question.

3) Although tobacco use is also associated with stillbirth, women who smoke during pregnancy are not prosecuted for homicide when their babies die (unless they smoke crack).

This week the South Carolina Supreme Court unanimously overturned McKnight's conviction, finding that her attorney had not made an adequate effort to present evidence on the latter two points and had failed to challenge confusing jury instructions regarding intent. The court said McKnight's lawyer should have introduced into evidence the autopsy report (which mentioned two factors in addition to cocaine as contributing to the baby's death), called an expert witness to question the state's conclusion about the cause of death, and questioned "the adverse and apparently outdated scientific studies propounded by the State's witnesses to find additional support for the State's experts' conclusions that cocaine caused the death of the fetus." The court said the defense should have presented testimony regarding "recent studies showing that cocaine is no more harmful to a fetus than nicotine use, poor nutrition, lack of prenatal care, or other conditions commonly associated with the urban poor."

Even if we accept the legal theory behind this prosecution, the government would have to show that cocaine use during pregnancy is independently associated with stillbirth, once confounding variables such as prenatal nutrition, health care, and use of other drugs are taken into account (the sort of problem that pervades the literature on the effects of prenatal cocaine exposure). Even if a general risk has been established, it is very difficult, if not impossible, to show beyond a reasonable doubt that a particular factor caused a particular stillbirth. Finally, the magnitude of the risk matters. If the risk to the fetus posed by the mother's cocaine use is comparable to that posed by various habits for which pregnant women whose babies die are not prosecuted, it's clear the government is irrationally discriminating against cocaine users.

Forget about cocaine for a moment. Do we want to accept, as a general principle, the idea that women who unintentionally cause harm to their fetuses by doing (or failing to do) certain things during pregnancy should not only be held criminally liable but prosecuted for homicide if the fetus does not survive? Such a rule is plainly inconsistent with the law regarding abortion, and it probably would deter many pregnant women from seeking medical care, thereby further endangering them and their babies. More to the point, a stillbirth is not a murder, and a woman who experiences this tragedy, even if her own behavior may have inadvertently contributed to it, deserves sympathy, not prosecution and prison. The fact that anyone needs to state this explicitly speaks volumes about the extent to which anti-drug hysteria has warped our system of justice.

The South Carolina Supreme Court ruling is here. The Drug Policy Alliance's press release on the case is here.