We voted unanimously against the placebo proposal for the following reasons:
1. The factor of "risk" wasn't being controlled for. The experiment would need to include another control group that takes the placebos without being informed of the non-existent risk.
2. Something is wrong in principle about deceiving pregnant women and intentionally denying them treatment for morning sickness, if there is indeed a treatment.
3. We think the women should at least be told there is a possibility of receiving a placebo (even though all of them are). This is common practice in clinical trials, and as long as it's kept double-blind (i.e., neither the patient nor the clinician knows that the drugs are indeed placebos) we don't see how this would significantly alter the results.
4. In general, there are too many confounding factors-- not all of the women will experience the same degree of morning sickness, if at all, and their hormones are already fluctuating in response to the pregnancy. Furthermore, if the perceived risk does create a physiological effect, there's a possibility that the mother's and fetus' health may be affected. This risk may even outweigh the perceived "nonexistent" risk.
We realize that the women signing up for such a study would be committing themselves to more of a risk than they actually are being exposed to. However, we feel the fundamental problem is that the women are not providing fully informed consent because they do not know the full scope of the study. To do so may render the experiment impossible, but perhaps this it is reason enough not to see it through.