Bringing Up Children in Conlangs: Language Death

One detail about bringing children up in your own conlang that might not have occurred to anyone is language death. Now, the usual argument with regards to language death and conlanging is the trite and fairly stupid why create languages when there's already hundreds of them out there dying? That is most emphatically not the argument I will present, and to make this very clear, I'll present a rather strong argument against that argument - despite the fact that it is entirely beside the point I am trying to make, and despite the fact that I am not arguing for conlanging as such, since I figure there is no great need to argue for it.

So, what is the problem with the bad rhetorical question why create languages when there's already hundreds of them out there dying? Due to globalization, lots of cultures are slowly vanishing, their songs, their ethical traditions, their philosophical inquiries, their stories being replaced by those of dominant cultures. Why would anyone write new stories when old ones are vanishing? Why would anyone compose music, when music is vanishing?

Since that argument is clearly preposterous, we probably can reject any argument of a parallel structure unless it has some other qualities that make them not be fallacies.

I only bring that argument up in order to show that I do not hold it. I have already seen people argue against positions I do not hold sufficiently often in discussing this that I want to be very clear on where I stand.

Now, let's get to the details of my argument: we know from extensive research into dying languages that the last speakers generally are very unhappy about the state of their language - no longer able to speak a language they grew up with to anyone who would understand it or be able to respond in it. It's not just unhappiness, generally, but a genuine anguish. 

For an example of this, David Crystal wrote a play that is a character study of a last speaker. Granted, it describes a situation with other important characteristics, such as the almost complete extinction of a tribe. However, we know from people who have been isolated from their language too that this anguish can happen even without the actual extinction of the language.

Further, how likely does anyone think a conlang will ever be at convincing a large enough community to speak it for there to be a likely transmission into a second generation beyond the first few native speakers it might manage to get? Seriously?

This argument is thus not why create languages when there's already hundreds of them out there dying - it is why create moribund language communities, whose first native speakers will also be first-hand experiencers of language death?