Fatalism from Antecedent Truth
The topic at hand is Brian Garrett’s handling of the Argument from Antecedent Truth, a fatalist argument based on notions of truth, namely, the apparent timelessness of truth, that aims to show that free will is an illusion. The goal of this paper will be to retread the argument and examine it – and Garrett’s response – as presented in what is this thing called metaphysics? (WTT). First I will discuss Garrett’s construction and analysis, then I will reconstruct the argument and give a brief critique of it on my own terms.
The goal of the argument is to show that free will is an illusion; we are fated to act in the ways that we do, for whatever reason (mechanical determinism, causal immutability, or something else). The argument tries to show this by way of a priori reasoning from trivially true premises. Garrett puts the argument together in WTT, page 98, in the following form:
“(i) I am currently drinking coffee. So: (ii) In 1800 it was true that I am currently drinking coffee. (iii) I have no choice about what was true in 1800. So: (iv) I have no choice about whether I am currently drinking coffee.”
Garrett rejects this argument in short order by claiming it uses misleading wording to mix together facts about past and present. He asserts that because a person has control over whether they drink coffee, i.e., they control in the present whether (i) is true, they control in the present whether (ii) is true. I believe this means that Garrett is claiming that (ii) simply restates – in an odd syntax – what (i) says. Garrett is reluctant to reject the notion of timeless truth (he says that a Presentist would take this option, but he tables it.) Because of this, when Garrett claims that “it is a fact about the present disguised to look like a fact about the past,” (WTT 98) I interpret this as saying that my decision to drink coffee now determines what the timeless truth of the matter is, and was, and will be, across all times. He may be implying that the current state of affairs was indeterminate in the past; the truth of the matter was decided now, when I made the choice. And if I provide an account of how I made the choice, that clearly precludes (ii) or (iii). (ii) combines a statement about the past with a statement about the present, and when this happens, Garrett implies that we should “defer” to the present clause. Garrett seems to concede that in 1800, it was true that I am currently drinking coffee, but he maintains that this is because I decided that I am going to be drinking coffee currently. There was no determinate fact of the matter at the time, but there is now. A true fact about the past would have to have been a determinate fact of the matter at the time in question. A true fact about the past would be something like “I drank coffee last Monday,” or “In 1800, the library of congress was founded.” The entire sentence stays in the past; there is no awkward meshing of now and then.
However, it seems in this formulation of the argument, Garrett asserts the thing in question. In an argument for fatalism, what is at question is whether we have free choice to determine matters of fact about our actions. Garrett deploys the assertion that we make a free choice to act (e.g., drink coffee) in order to preclude the idea that we were fated to act. But the condition in italics above was that we need an account of how the choice is made, in such a way that would preclude (ii) or (iii). Merely asserting that a choice was made is begging the question against the fatalist. It is far more effective to reject (ii) for many of the reasons A-theorists might – confused wording and tense usage, the contingency of the future upon opaque and contingent causal webs (i.e., there are possible truths realized in the face of agency), or the real coming-into-being of facts at certain times. I shall now move into my recreation of the argument.
Garrett’s formulation, it seems to me, leaves several premises implied that should be made explicit so that they can be properly challenged; in his formulation, Garrett grants or glosses over them. I believe a charitable interpretation of the argument in more clear terms would look like this: (i) For any action x that I undertake at a time tx, there is a corresponding true fact that that I undertake the action x at tx. (ii) Truths, and hence true facts, are timeless and immutable. (iii) It was a true fact at some distant past time t0 that I undertake action x at tx. (iv) I cannot exert any causal effect over what was a true fact at some distant past time. (v) I cannot affect the true fact that I x at tx. (vi) If I cannot affect the true fact that I x at tx, then it was inevitable – fated – that I x at tx. (vii) If it is fated, then any choice I believe I have with regard to a given action x undertaken at time tx must be illusory. (viii) Fatalism is true.
There are two premises that I wish to challenge by highlighting them in the new formulation; several others may be addressed by some formulation of the above objections.
Premise (ii) is a critical and vulnerable premise. In the earlier formulation, it was taken for granted that facts are immutable. However, if we make this premise explicit, we can assess it properly. If we do not wish to adopt an explicitly A-theory of time and reject the timelessness of facts – as Garrett is reluctant to do – we may here consider that consequent facts are contingent upon a web of antecedent facts, but not in a deterministic or fatalistic way; the temporal series of events may be probabilistic relations of antecedents and consequents that are not all equally real in every “throw of the dice,” though they are equally possibly real. That means that facts in the indexically real time-series are mutable, and my xing at tx is not yet determined at t0 in the strong sense that the fatalist needs to prove. Some of the antecedent facts that determine the direction of the mutable or probabilistic time-series may include facts about agency and choice that are not predestined or fatalistic in nature.
(vi) also seems trivially true, but we may consider “being-born” something that I did, had no control over, and yet was not fated – it could have been the result of choices my parents made. This hidden but necessary premise in the original argument has been shown to be weak and in need of refinement.
The argument for fatalism from antecedent truth aims to show that if timeless truths exist, then there are timeless truths about human action – and if it is true at a prior time that I would commit to a certain action at a later time, then fatalism seems to be true. But in the formulation of the argument put forth by both Garrett in the fast version and myself in the more charitable and explicit formulation, there are important objections that need to addressed, whether it be the question of present facts being presented as facts about the past or the possibility of a probabilistic, non-deterministic time-series.