If you had a child to make yourself happy, as most people do, would that violate the Kantian imperative to avoid treating people as means?

Unfortunately, this is a tricky question for Kantian ethics to address.

On its face, it might appear that procreation (bringing a child into existence) in order to advance one’s own happiness treats the child merely as a means: One ‘uses’ the child to promote one’s own happiness.

But things get more complicated once we attend to exactly what this Kantian imperative says. The Kantian moral requirement you mention states that we are not to treat “humanity” merely as a means. There are debates as to exactly what Kant had in mind by “humanity” but the standard view is that “humanity” means the capacity for rational agency — the ability to choose our ‘ends’ (our goals or objectives) and the best means to those ends. But a newborn lacks “humanity” in this sense; it cannot choose ends for itself, etc. Nor can a fetus. All the more, a child who does not yet exist does not have humanity! Hence, it would appear that the apparent answer to your question is ‘no’: You cannot treat someone’s capacity for rational agency merely as a means if they simply do not have such a capacity.

This is (I suspect) the orthodox Kantian answer. But this answer has counterintuitive implications. For if wronging someone involves treating their humanity merely as a means, then it would seem impossible to wrong anyone without humanity. So it would be impossible to wrong a child at all – and that doesn’t seem correct. Surely it’s possible to wrong children by treating them merely as a means (having a child purely for the sake of, say, harvesting its organs to save others’ lives). So how might Kantians arrive at a more intuitively plausible answer?

The most promising route, in my estimation, is to cast doubt on an assumption on which the orthodox Kantian answer seems to depend. That assumption, very roughly, is that if it would be wrong to treat someone in a particular way because she has property F, then she must have property F at the very time the mistreatment occurs. This assumption is, pretty clearly, false: Jane is fully unconscious and unable to feel pain on Monday. On Monday, a sadistic doctor injects her with a drug that, when Jane awakes on Tuesday, will cause her great pain. In order for Jane to be wronged by this act, she must be capable of feeling pain. But when the injection is made (Monday) she is not able to feel pain. And yet it’s hard to deny that Jane is wronged by the injection. (Perhaps we should say she is wronged on Monday but suffers the wrong on Tuesday?)

Yet if that assumption is rejected, then it seems open to Kantians to argue that despite children not having humanity, they can never nevertheless be wronged by choices that will subsequently treat their humanity merely as a means. A child born so as to make her parents happy is wronged though perhaps not at the very moment she is born.

(There is one final wrinkle here: Does existing matter to whether one is wronged? Jane exists throughout the duration of Monday and Tuesday. A not-yet-conceived child does not exist. Is it possible for a wrong to be done by conceiving her that she only suffers later on?)

In any event, this isn’t a simple question for Kantian ethics to handle, but at least this response may help in discerning how Kantians might analyze it.

Read another response by Michael Cholbi
Read another response about Philosophers, Children