We can conceive of a lot of things.

One thing I have recently conceived of is what I call the Worst Imaginable Scenario (in the world we live in). The Worst Imaginable Scenario (WIS) goes like this:

Worst Imaginable Scenario (WIS): Every single person who existed, exists, and will exist has already been in existence since the beginning of time, residing in a Platonic realm of unimaginable suffering. Birth into the world we live in currently is the only sweet release any one of us ever gets to experience. Not everyone gets this chance, either; some are left behind. When we die, we get transported right back into this transcendental Hell for all eternity.

Ignore the metaphysical problems this arises and the fact that it is definitively the product of a neurotic mind, and focus on the ethical problems this raises. If WIS were the case, then it seems like we would have an obligation to have as many children as possible to minimize how much torture ultimately happens. Torture is inevitable; our only choice would be to minimize how much happens in the long run.

Compare this to a wholly different scenario, the Drunk-Driving Dilemma (DDD):

Drunk-Driving Dilemma (DDD): You live in a rural mountain town, and are heavily intoxicated but want to drive home from a party. It is very late at night and although you can conceive of getting into an accident, you brush this aside as a very low probability given the remoteness and small population size of your community. You decide to drive home drunk and in the process hit a car with a family on a vacation, severely injuring every single passenger.

DDD seems to be a case of clear irresponsibility and wrong-doing.

Now, what if the WIS is not known to be the case, but simply conceivable (which it is, I am capable of imagining a great many things including WIS). Do we still have an ethical obligation to have as many children as possible, simply out of the conceivability of WIS?

The tension here arises when we see conceivability as an important measure for ethical obligation in situations like DDD but not in situations like WIS.

Probably most of us, myself included (when I'm not obsessive), would scoff at WIS. Just because we can imagine something doesn't mean it exists! In fact most of us would probably argue that it's just obvious that people don't exist before they are born. Like, duh...

Yet just because we can't imagine ourselves getting into a car accident due to intoxication, doesn't mean it won't happen!

In both cases, ignoring the conceivability of something has the potential of harming others.

This seems to lead to the conclusion that the way we see ourselves ethically has to, in part, depend upon probability and likelihood.

Of course, it could be argued that the sheer conceivability of unconceivable pain disqualifies any talk of likelihood.

But we can also conceive of a lot of other things as well. Perhaps by typing the letter "s" into my keyboard, a trillion people suddenly come into existence and are tortured. Conceivable? Yes. Likely? No. Therefore I shouldn't worry about pressing the letter "s".

However, what this also means is that, without further considerations, we shouldn't see your decisions in DDD as morally problematic, as the probability of getting into a car accident were presumably quite low. Which, without any other considerations, seems wrong. If you don't have a good reason to go out drunk driving, then it doesn't seem like you should go drunk driving!

Another example clarifies this point: what about our decisions regarding those whom we are not sure are capable of being sentient? For example, it might be doubted that insects are capable of feeling anything. How are we to assess this uncertainty? If we just ignore that they might be able to suffer, we come across as insensitive and cruel. But if we focus on mitigating encounters with insects as to prevent (conceivable) suffering, we fall back into the same problem as before; we'll treat insects with respect because they might feel pain, but we won't usher people into worldly existence simply because we doubt they actually exist before they are born? What's the cut-off here?

In this case, it seems that our doubt of insects' ability to suffer is more a product of prejudice and bias than reason. The jury is still out on the possibility of insect suffering, but it seems like a pretty good idea to be precautionary and treat them as if they can suffer.

But once again we're back at the same problem: why not be precautionary and take seriously the WIS?

There does seem to be a difference between these two cases, in that insect suffering is something we are not sure of (agnosticism), yet the suffering of unborn people is something we are pretty sure is non-existent. Thus the difference is once again one of probability.

But how sure are we actually? Say, for example, we end up with pretty conclusive results that lead us to believe that insects cannot, in fact, feel anything and that our treatment of insects are thus morally unimportant. Little do we know that insects actually can feel pain and can suffer.

Oops. Ockham's Razor isn't always reliable.

To attempt to solve this issue, I will present a principle that I call the Ethical Qualification of Investigative Capability:

Ethical Qualification of Investigative Capability (EQIC): that which cannot be conceivably investigated is not morally important.

The key word here is conceivably. For without it, it would mean that if just because were aren't physically able to investigate the consciousness of insects, we then are not obligated to treat them fairly. Which is quite short-sighted. The conceivability of investigation condition limits the scope of our ethics, then. Just because we can conceive of something doesn't mean it is automatically ethically important - we must additionally be able to conceive of a way of investigate this conceivability.

Taken to the extreme, the only things that fail to meet the standard of EQIC would be those things that are metaphysically impossible for us to investigate. Such things are thus like black boxes whose contents can only be pie-in-the-sky speculated upon and cannot be accessed by any means whatsoever, whether that be logical syllogism, empirical observation, or whatever. Insects, therefore, are not black boxes in that we can see how they might be able to be investigated.

Thus, the conclusion is that, from a welfare-centered view, we ought to see the value of a something as seen from the point of view of the universe as additionally seen through the eyes of value-beings. What reality is actually like is important, but only insofar as observers can actually conceivably know about it. The focus goes from an universal objectivity to an inclusive, yet limited, objectivity.

A weaker version of EQIC can be formulated, the:

Ethical Qualification of Pragmatic Investigative Capability (EQPIC): that which cannot be conceivably investigated, or that which cannot reasonably be investigated without disproportionate risk or effort on our part is not morally important.

Thus the addition of a pragmatic element has us consider the impact such an investigation would have on us in classic consequentialist input/output terms.

Our treatment of things that do not fulfill this requirement is thus indeterminate: we do not know how to treat these things. There may be a ghost next to me who is horribly tortured every time I play a certain song - then again, they might be horribly tortured every time I play a different certain song. I just don't know, and if I can't find out one way or another, then I can't be expected to worry about it.