Why is predictive capacity so forceful for Hard Determinists?

People's everyday intuitions about free will are not really consistent.

On the one hand, we make predictions all the time, and accept that some people are better than others at often making guesses that turn out to come true. We also accept that having more knowledge about any kind of process will enable us to make guesses (about that process) that come true more often. But on the other hand, many people find it hard to apply that kind of perspective to their own choice actions: they cannot quite accept that other people would be able to guess their "free" (i.e. voluntary and deliberate) actions correctly.

This inconsistent view on "free will" seems to be the basis both of hard determinism and of a libertarian view. It seems to me that the problem is mainly just psychological, a belief like: "While we are deliberating over something, we do not know what we're going to do. If we ourselves don't know, how could anyone else know?" The basic mistake made by both hard determinists and libertarians is that epistemic unpredictability does not entail indeterminism and that being in control of one's actions (in the everyday sense of "being in control", the kind of sense that would count in an ethical or legal dispute) does not preclude determinism (some form of local determinism or causality). In fact, being in control, autonomous agency, doing something "out of my own free will", presupposes a form of local determininism, causal links that allow me to more or less reliably predict what will happen if ...

We see "as in a mirror darkly" and only have partial self-knowledge. But in so far as we have some self-knowledge, we can predict correctly how we ourselves will react to some (not all) future events. And nobody seems to think that that ability contradicts "having free will" ^B.

Apart from this, there may also be a strategic (game-theoretic) reason for keeping the inconsistency alive. In an adversarial social situation in which two or more actors know they have several options, it's definitely in one's advantage if the other can not predict one's choices. More than that, the mere assumption that the other can (somewhat reliably) predict my choices may be demoralizing^A. So, a bias towards viewing one's own choices as unpredictable, may be adaptive. On the other hand, most social situations are not adversarial; many have the character of coordination games where we want to be predictable to others and want others to be predictable (simplest example: driving on an agreed side of the road).

(A) See Ted Chiang's story What's Expected of Us.

(B) Is my ability to predict my own future reactions (or actions) fundamentally different from my ability to predict the reactions (or actions) of others? I have a lot more (and more reliable) information about myself, but is there any other difference apart from that?

Predictive capacity is an essential part of competition (among other things), and the concept of competition only makes sense in a non-deterministic worldview. In a deterministic world, every interaction between Barry Bonds and Greg Maddux was preordained: the outcome was given since the dawn of the universe, and any thoughts that Bonds had about the interactions were also preordained (and thus entirely empty and meaningless).

Far from taking away another's free will, the effort to predict another's behavior demands that they have free will, and predictions become a contest of skill in which one tests the limits of the other person's ability to predict.

Note that there is a subtle but vital distinction here. Science often talks about predictions in the sense that a good theory is supposed to be able to predict the (mechanical) behavior of (material) objects. There's no imputation of free will to these objects, merely the idea that these objects are difficult to understand properly, and better science does a better job of it. But predicting the behavior of subjects adds different wrinkles. If we try to treat subject as though they are objects — effectively reducing them to animalism — we will inevitably discover that subjects just don't like that, and the effort will blow back on us. It works to a certain extent, in certain ways, and then it just doesn't work anymore.

As I understand it:

• you've encountered an argument to the effect that if a scientific theory with some empirical data fed in predicts an outcome it was determined and there was no free will to do otherwise,
• you've compared it to another argument to the effect that if a different predictive approach (e.g. rooted in understanding a person's motives, reading their movements/body language etc.) is as reliable we again lack free will,

and your question is why anyone would take the first argument seriously when they don't take the second seriously.

I don't think the first bullet point properly captures the strongest version of the deterministic argument against free will. I'm not saying you didn't see it put that way, but it can be strengthened to address your question.

An early model of scientific explanation modelled it on prediction, but the problem is that, when a theory implies two statements have equal truth values, often one is the cause and the other is the effect, and deduction doesn't capture that asymmetry. We might get around that with a fix such as "an explanation has to deduce what is explained from a theory plus prior events", but whatever we do explanation is not equivalent to prediction.

If a batter predicts the pitcher's behaviour, the thinking isn't that the predicted behaviour is caused by the observable reality that informed the batter's prediction, but that it has the same cause as that reality. What's more, we definitely don't have a causal model saying the batter's inference caused whatever was inferred to happen, though we may have an adequate model to predict the batter won't be mistaken.

By contrast, determinism implies there is an explanation running from cause to effect, which we might not even understand well enough to make the prediction. But in principle, our known physical laws imply events determine subsequent events. So there is a metaphysical guarantee (i.e. the ball couldn't move differently because of the laws of motion, including their effects in the pitcher's body) as opposed to an epistemic guarantee (i.e. the batter couldn't be mistaken, as they're just too good at reading opponents).

Determinism (Causal Determinism)

https://plato.stanford.edu/entries/determinism-causal/

In order to get started we can begin with a loose and (nearly) all-encompassing definition as follows:

Determinism: Determinism is true of the world if and only if, given a specified way things are at a time t, the way things go thereafter is fixed as a matter of natural law.

The italicized phrases are elements that require further explanation and investigation, in order for us to gain a clear understanding of the concept of determinism.

Laplace probably had God in mind as the powerful intelligence to whose gaze the whole future is open. If not, he should have: 19th and 20th century mathematical studies showed convincingly that neither a finite, nor an infinite but embedded-in-the-world intelligence can have the computing power necessary to predict the actual future, in any world remotely like ours. But even if our aim is only to predict a well-defined subsystem of the world, for a limited period of time, this may be impossible for any reasonable finite agent embedded in the world, as many studies of chaos (sensitive dependence on initial conditions) show. Conversely, certain parts of the world could be highly predictable, in some senses, without the world being deterministic. When it comes to predictability of future events by humans or other finite agents in the world, then, predictability and determinism are simply not logically connected at all.

The equation of “determinism” with “predictability” is therefore a façon de parler that at best makes vivid what is at stake in determinism: our fears about our own status as free agents in the world.

Personally, I don't think it is true that predictability and determinism are not logically connected at all. I think a hard determinist holds a belief or conviction that unpredictable events are products of a deterministic natural process despite the human experience of unpredictability or other forms of uncertainty.

I hold with folk psychology, such as expressed in law, where we look for moral sources of cause (adult moral agents) to blame for adverse outcomes, and we praise or blame moral agents, but we cannot hold animals, young children, or natural events accountable for moral benefit or harm. If we had no capacity to recognize moral agents, perhaps via the mechanism of Unconscious Inference (see below), then we would not make distinctions between adult humans and animals, young children, or natural events in folk psychology incorporated into ethical judgment and social institutions of law.

Original Answer

Hermann von Helmholtz argued persuasively that our real-time perceptions of speed, size, and location of objects are generated as the product of an unconscious process called Unconscious Inference. To make this inference conscious I often hold up my thumb so it obscures the image of the full moon. In my mind the moon is very large and far away but in the visual 2D image my thumb is larger than the moon! Some philosophers argue that the unconscious physical, chemical, and biological processes are ultimately governed by physical determinism and that our conscious mind is an epiphenomenon. Then the ultimate unconscious physical mechanism is driving the apparent experience of predictable versus unpredictable human behavior patterns. These philosophers compare the experience of free will to an optical illusion. They discount the experience of moral agents with conscious uncertainty in favor of their belief in unconscious physical determinism as the sole type of cause. Modern science does seem to hold that consciousness is the product of an unconscious physical process. But science does not yet compel faith or belief that consciousness has no causal effect or that free will is an illusion.

The difference is that hard determinists believe that nothing they do, or that happens to them, could have happened any other way.

If the player drops the ball, that was because they were destined to drop the ball; that may have been caused by a fumble which may have been caused by an injury which may have been caused by being distracted which may have been caused by worrying about their great aunt Sophie's parakeet which... But all of that was predestined.

If you chose to play the three of clubs, the same kind of chain of causality applies; you chose that because you had to, because you are who you are and are in the situation that you are in. Whether the dealer is cheating is part of that situation, but even if they aren't the cards are in a specific order because they had to be in that order.

This probably doesn't change how you should live your life, since your ideas about that and your opportunities to apply those ideas are also predestined. Including your belief, or disbelief, in hard determinism.

Assuming you accept hard determinism in the first place. That's the definition of the belief.

"I am a rock, and I must deeble." (Lyric based on Roger Zelazny's SF story about a collector who picks up more than he bargained for.)

To focus on the prediction issue: irrelevant, as far as I can tell. Hard determinism does not mean we can make reliable predictions. It means that if we had the near infinite amount of data representing the relevant parts of reality, and the near infinite computational ability to run that model, we could make perfect predictions. But we don't and we can't. There are essentially no practical implications; it's entirely a matter of how you define free will, whether you think God plays dice with the universe (She does, but the universe cheats), and how well you cope with not knowing.

Suppose that human action is perfectly predictable. If so, why? One answer is: determinism. It's not the only possible answer, but it's a plausible one.

In other words, there's an abductive argument from perfect predictability to determinism. What's the best explanation why human action is perfectly predictable? One plausible answer: Because such action is determined, and determined action is perfectly predictable.

A similar argument for determinism can be made even if human action is highly, though not perfectly, predictable. But that argument isn't as strong as the argument from perfect predictability.

Hard determinism presupposes that the total state of the universe determines all future events with absolute necessity. But this assumption becomes unstable once we consider the temporal and epistemic structure of observation itself.

When I close my eyes, I experience myself in the present. When I open them, I see the past. Light takes time to travel; perception always arrives late. What we call “the present” is thus a reconstruction of what has already happened. Phenomenologically, consciousness never coincides with the event — it observes from a displaced frame, slightly beyond it. From this standpoint, our experience of time is not fully contained in the causal chain; it interprets that chain from a temporal offset.

The apparent determinism of reality arises from this causal interrelation mediated by light. It creates a coherent illusion of necessity: events appear linked because perception stitches them together through memory and inference. Yet this coherence belongs to consciousness, not necessarily to the underlying world.

Quantum mechanics deepens this issue. At the fundamental level, entities are not merely particles or waves, but probabilistic fields of potentiality. Their behavior is only defined in relation to measurement, not in advance of it. This means the world is not entirely fixed before being observed. Indeterminacy is not just a failure of knowledge but an ontological openness — a margin where prediction cannot reach.

Predictive success, therefore, does not demonstrate determinism. It only reflects how often our internal models align with probabilistic regularities. What seems like “causal necessity” may simply be correlation within uncertainty. In this sense, the capacity to foresee another person’s action is not proof against free will but a coincidence between patterns — a momentary synchronization within an undetermined field of possibilities.

In short: determinism describes the grammar of the past; freedom lives in the syntax of the present.

Prediction is not proof of fate, only a reflection of coincidence.