As I've mentioned before the 'omnibenevolence' leg of the stool has no real existence prior to the nineteenth century; nobody attributes the term "omnibenevolence" to God prior to that. Indeed, even in the nineteenth century, the term 'omnibenevolence' is usually used in ways that are too weak to formulate any sort of problem of evil, since the term originally seems to have meant merely, 'wishing everyone well' with all the vagueness and flexibility that can have, and which is, in any case, hardly a distinctively divine attribute. The first Christian I've found using it in definitely something like the sense in which it is usually used is Robert Browning, who has Guido argue, in The Ring and the Book:
Let the law stand: the letter kills, what then?
The spirit saves as unmistakably.
Omniscience sees, Omnipotence could stop,
Omnibenevolence pardons, — it must be,
Frown law its fiercest, there’s a wink somewhere.
(I have found a few prior to Browning, like William Penn; but in each case the force of the term is either ambiguous or obscure.) But, of course, it's still functioning in a different way: Guido's point is that "there's a wink somewhere", the claim being that Christianity is just paganism on the sly -- a higher moral tone, yes, but it's all a mask to allow sin loopholes. That is, we don't have the problem of evil here; we just have an argument (not Browning's own) that Christian doctrine is a breeding ground for hypocrisy. It's really only in the twentieth century that one finds any Christians using the term in a positive context and in a way that suggests its use in the argument from evil; and these are clearly backformations. The real source of all this seems to be Hugh MacColl shortly after the turn of the twentieth century, who, while not really read today, seems to be the only source who was widely read enough and who actually uses the formulation for the argument from evil. (MacColl also is the first to respond to the problem by arguing that evil is necessary, in particular for the progress of the universe; which, of course, was the whole point of his putting the argument this way, since progress as an 'omnibenevolent' objective would have been a relatively easy sell, even if the necessity of evil for it wasn't.) Even the freethinkers don't use the term for the argument from evil much earlier than the 1890's.
In any case, even if this isn't so, as it's not a traditional term, and as the root word, 'benevolence', does not generally convey any particular necessity or specific duty, the question does arise as to why one would think that anyone is committed to holding that God is omnibenevolent, at least in any sense that would cause a problem with regard to theodicy. Indeed, it's not even clear what the term means; it could mean 'wanting good for people in everything' or 'wanting what is best for each and every person' or 'wanting what is best overall for everyone' or 'always acting with a view to another's good', or any number of other things. In order to make the argument from evil work it has to be strong enough to establish a duty or obligation; but 'benevolence', as I've said, is a weak word as it is used colloquially, so it's not helpful for extrapolating any meaning. Why then does it persist? Because of the rhetorical parallelism: it makes it sound more distinctively divine, and gives the illusion that it is structurally parallel either to omniscience or to omnipotence -- an illusion because it is never explicated in a way that would actually make it so. The argument from evil is really the claim that, given the existence of any evil, every good is in some key way useless or ineffective; the parallelism conveniently hides the fact that the crucial issue in the argument is not one's account of God but one's account of good. It wraps it all up in what looks like a tidy and memorable package. Never mind that it isn't actually tidy when you start asking critical questions; packages that are memorable have a power to endure, as cognitive science shows us, and memorizing formulaic labels is easier than remembering entire theories of the good or of good will, which is what actually has to be doing the work.
In any case, as I've also said before, the intelligent person, faced with a relative neologism like 'omnibenevolence', will ask for the underlying account of the term, and only move on the basis of the account, not the similarity of the word to other words.
