But the analog to being eternal would be omnipresence, or being located everywhere. I saw this coming up a lot in discussions of creating abstract works. People used it in reductio arguments, claiming that creationism entailed that things were created after they existed.

I think there are two ways to take the claim that the work existed before it was created. One is obviously false, the other is, I think, true.

The false way is to suggest that the work exists for some span, and the creation of the work is not an endpoint of that span, but instead exists somewhere along it. The other way of understanding that claim is that it is true, prior to the work's creation, that the work tenselessly exists. However, I think this is similar to the case of Socrates, since it is the case now, that Socrates tenselessly exists, but not the case that Socrates exists in this time period. In the case of Socrates, there is a span of time during which he did exist. In the case of abstract objects, they merely have tenseless existence, and no temporal location, but the case is similar enough.

Newcomb's Problem
There is a man, call him Predicto, who is travelling around offering the following deal to people. An individual will be presented with two boxes, A and B. Box A is opaque, and may or may not contain $1,000,000 dollars. Box B is transparent and contains $50,000. Predicto gives that individual (call him the contestant) the choice of either opening box A, or opening both boxes A and B. The contestent gets whatever money is in the box or boxes chosen. Whether or not their is money in box A is determined by Predicto in the following manner: Predicto has superior predictive capabilities. He is an expert at predicting what people will do, and in fact, his track record is flawless. Predicto predicts whether the contestent will open A or A&B. If he predicts that they will open A only, he puts the million in there. If he predicts they will open both, he leaves A empty.

The question is what the contestent ought to do, when they get to open the box. The puzzle is this. At the time the contestant is making a decision, Predicto has already decided and either put money in the box or not. So, when the contestant steps up to the box, one of two states of affairs obtains. Either there is money in box A or not. If there is money in box A, then, you do better by opening both boxes (you get one million plus fifty thousand). If there is no money in box A, you are better off picking both, because you get 50,000 instead of nothing. So, there are two possible states of affairs, and in either one, opening both boxes leads to the better outcome. Rationally, then, when making the decision, there is no advantage gained (and in fact, advantage lost), by opening just box A.

So, it would seem that opening just A is irrational. However, many people feel that a) opening just A is in their interest, and b) that it isn't irrational at all. If you open both, it's likely that Predicto predicted that, and you get only $50,000. If you open just A, on the other hand, you are immensely likely to get a million. A million is better than fifty thousand, and so, it seems, one is better off to choose just one.

I am going to avoid discussing what is to be said for each view, and instead, explain the analogy to Calvanism. I will however, note, that it seems to be the case that one-boxing, as it is sometimes called, leads in terms of widespread popular/intuitive support.

The aspect of Calvinism that is relevant is God's predetermination of salvation. According to Calvinist doctrine, God determines before souls are incarnated which souls will go to heaven. This means that one's behavior on Earth does not impact whether or not they will go to heaven. However, it is thought that behaving certain ways is an indicator of which decision God made, and thus, there is supposed to be some encouragement to be good, so that you are more likely to have been predestined for salvation.

This notion (which may well be innaccurate as a description of Calvinist thought, but which is widely taken to be a description of such), lacks popular/intuitive support. That is to say, people don't seem to think that this should be good motivation to act well, since it has already been decided.

The claim of analogy is that the calvinist set up is structurally similar to the Newcomb problem. Further, it seems odd that people would diverge as to their opinions on the case, since they both involve actions and behaviors at time t2 impacting a decision that was made at t1. One-boxers, in other words, should see some merit to motivating good behavior in that manner.

Disanalogies in the situations exist. Predicto's existence and predictive capacity is taken as known. Money is the prize in one case, eternal salvation in the other. Etc.

Let me argue for what agreeing to the analogy commits us to:
1) If one-boxing in Newcomb's problem is rational, then the rationality of one-boxing is because it is rational to believe that a decision at time t1 can be impacted by an action performed at some later time t2.
2) If the rationality of one-boxing is because it is rational to believe that a decision at time t1 can be impacted by an action performed at some later time t2, then if someone believes one-boxing is rational and finds the "Calvinist" motivation for behaving well irrational, it cannot be because believing that actions performed after the salvation-decision occurred are unable to impact the outcome.*

Now, I think that we should conclude that they are both irrational, absent evidence for this sort of backwards causation, but the analogy and disparity is interesting.

*(Barring an argument that there is something specifically different about the decision events in the two would make the backwards-causation feasible in one, but not in the other).

Arguments against resolving certain artworks (I think, primarily, performance arts) as discoveries seems to be motivated by the following reasoning:

AD1) If composers (writers, etc.) discover their works, they are not worthy of the praise we give them.
AD2) Composers (etc.) are worthy of the praise we give them.
AD3) So, they did not discover their works (and instead created them).

Another argument could be that this creates an odd division between composing and, say, sculpting, in which the former is an act of discovery and the latter is an act of creation. I am doubtful that such a division is problematic, but I am going to set that aside for a moment and focus on the above argument.

The argument is essentially that discovery is of less merit than creation, but I think that this is an ill-founded disparity. It seems to presume some underlying ease in discovery that is not present in creation. And thus, if composers discover their work, they have not accomplished so interesting a feat after all. The work was sitting around for anyone to discover. I think also the notion that artistic creativity (ingenuity) etymologically requires literal creation plays a role.

Perhaps there is more to be said in favor of why composition must be an act of creation (and if you have something to say on that front let me know), but absent other arguments for it, it seems as though we have relegated discovery (at least, as potentially applied to art) to substantially less regard than it deserves.

Discovery is not easy, and while some acts of discovery can be accidental, others are clearly the result of well organized directed artistic activity, and are not only difficult, but require as much investment from the discoverer as acts of creation require of the creator.