#namethetrait - transformations and impossible traits

[privkeav]

One of the ways that trait equalization within NTT was explained to me was as follows: imagine starting off with a human and a cow, and slowly transform the human into a cow by eliminating or switching traits. At what point in the transformation does the human lose "moral value"?

I think that this type of approach creates problems and will explain why.

Suppose that T1 is a trait that, if eliminated, would cause the human to lose moral value. The trait T1 might be the first trait that is eliminated during the transformation, the last trait eliminated, or it may be eliminated somewhere in the middle of the transformation. So, it should be clear that you would only be able to describe the point during the transformation that the human loses value if you know the order in which traits are being eliminated or switched.

The fact that analysis of a transformation may be dependent on the order in which the traits are analyzed is important because the order can produce what I will call impossible traits. These are traits that could be examined outside the transformation process, but become implicitly eliminated from analysis during the transformation. For example, suppose I have the following traits for a human:

T1 = has opposable thumbs
T2 = is a biped

We can combine the two traits to form a compound trait:

T = T1 and T2 (has both opposable thumbs and is a biped)

Now, suppose the order of the transformation is as follows: first examine and eliminate T1 - we now have a human that does not have opposable thumbs. At this point, the compound trait, T no longer applies to the "human" (because the "human" is now a biped without opposable thumbs). I'm not sure how it would make sense to examine a trait that can no longer applies to the "human" at a certain point in the transformation. Any compound trait can be eliminated in this way, so the order that traits are examined becomes very important.

It is possible to argue that what is important is actually identifying a trait. However, this would make the transformation process irrelevant, and in my opinion, a red herring.

Or, is there something I'm not understanding? I'm still trying to get an idea of why transforming a human to a cow would be relevant in an NTT discussion. I'm also struggling a bit to come up with a decent way of viewing "trait equalizable" would mean when considering hypothetical worlds.