'Abort it and try again. It would be immoral to bring it into the world if you have the choice.'
People have argued that it's morally permissible to abort in order to prevent having a DS child, but I haven't heard of anyone contending that it's actually obligatory. Dawkins is saying that any parents who knew their baby would have DS but who didn't abort are guilty of doing something immoral. They had a duty to abort.
Dawkins has since backpedalled, issuing this statement:
If your morality is based, as mine is, on a desire to increase the sum of happiness and reduce suffering, the decision to deliberately give birth to a Down baby, when you have the choice to abort it early in the pregnancy, might actually be immoral from the point of view of the child's own welfare.The locution 'might actually' is mysterious. Dawkins clearly doesn't want to say merely that, for all we know, it's possible that having a DS child will reduce the sum of happiness. Given the context of his earlier remark (on Twitter), he seems to think it's likely that the child's creation will have this negative impact.
The antecedent in the above, longer quotation suggests that Dawkins bases his view on a utility calculation for those who are affected by the birth of a DS child -- taking all of them into consideration, the sum of happiness is greater (and the amount of suffering lower) if we prevent the creation of a DS child. However, the final clause of the quotation suggests that Dawkins wants to assess the impact on only the DS individual's well-being -- it's for her sake that we shouldn't permit her to be born.
That's not even plausible. DS does not saddle one with such a low quality of life that one is better off not existing in the first place. It's hard to see how the DS individual is better off by being killed prenatally (or by having its fetal precursor killed -- if that's how one prefers to speak).
Turning to the former claim (that the creation of a DS child is most likely contrary to the project of increasing the sum of happiness for all who are affected by the act), evidence is called for. Dawkins is making a prediction about the net impact on happiness of having a DS child. Why does he give no empirical support for his prediction? After all, these are very complicated matters of fact, so we oughtn't to rely on armchair intuition-crunching.
Has Richard Dawkins, the great champion of empirically supported reason, proclaimed such a shocking conclusion and charge of immorality without first examining the evidence?
Update (August 21, 2014): Ari Kohen summarizes some of the relevant evidence.