Despite the fact that it is not clear why there is a need for a position at all, determinists think of determinism as the default position. Honderich says that it is ‘remarkable’ that apart from the concern with ‘whether choices and actions are free … what we find is determinism and hardly anything else.’Footnote ¹ What he means is that we do not find much support for indeterminism (or a third alternative). However, it is unclear what we are looking for when we find hardly anything but determinism.

Why are so many philosophers determinists, when determinism, as a metaphysical thesis, seems hard to understand, let alone defend? After explaining what is problematic about determinism, I conclude with some remarks about its real appeal.

1. Determinism as a Metaphysical Thesis

William James uses the metaphor of pregnancy to express the basic idea:

The ‘womb’ is where the future is developing; and the ambiguity that determinism rules out is that there is more than one future. Future what? Future of what? We may wonder about the future of certain demographics, institutions, or a person's career. But, James seems to have something much more comprehensive in mind, even though I think that it is hard to understand just what that could be.

Van Inwagen doesn't seem to help matters with his definition of determinism.

His definition only applies to what is ‘physically possible’, which suggests that he is not thinking about the future as far as inflation, population growth, or a friend's career, if they are not ‘physical’. What ‘physical’ future does that leave, and how is it to be separated into instants? And why would we be interested in a determinism that applies only to it?

When Van Inwagen gives determinism a more ‘precise’ definition, what seems to make it more precise is that he is clarifying what is meant by the ‘future’.

What he refers to as the ‘state of the world at some instant’ is what James refers to as the ‘totality’ that is the ‘present’ or ‘future’, or what Van Inwagen's own less precise definition refers to sas the ‘(physically possible) future at any instant’. Earlier, Van Inwagen explained that a ‘proposition’ is ‘a general term for the things people assent to, reject, find doubtful, accept for the sake of argument, attempt to verify, deduce things from, and so on.’Footnote ⁵ Here the reference in the less precise definition to ‘exactly one possible future’ is clarified as an entailment between propositions expressing the state of the (physical) world at earlier and later instants.

Ironically, the ‘precise’ definition seems to be more problematic than the others, because it presupposes that there is something to talk about that has the state-of-the-world-at-an-instant as its subject matter. Each of the definitions assumes something that I want to call into question in this paper, namely, that there is such a thing as the state-of-the-world-at-an-instant. However, when it comes to his more ‘precise’ definition, what is even more questionable is that we have something to assent to, find doubtful, etc., in connection with something as mysteriously comprehensive as the state-of-the-world-at-an-instant.

These different definitions all may be understood to say that given the state of the (physical) world at the present moment there is only one way it could be at the next (or any future) moment. And the argument for each is of the it-stands-to-reason variety: the only possible explanation for the transition of the state of the world from moment to moment is that each state is the only one possible.

Explanations of transitions in the state of the world from moment to moment are so remote from our experience that determinists refer to more familiar kinds of explanation when they argue for determinism. However, these kinds of explanations do not seem to be deterministic, a point we will develop in the next section.

2. The Argument from Explanation

According to Honderich all events have an explanation in the ‘fundamental sense.’ In this sense, an event is a ‘standard effect’, in that it

So, a fundamental explanation is distinctive in that it is supposed to make clear how what happened had to happen.

As support for determinism Honderich tells us that he has ‘never known of a single event to lack an explanation in the fundamental sense.’Footnote ⁷ Of course, he is saying that what he knows we also know; presumably, anyone with a scientific bent who is not drawn to the occult, can say the same thing. The fact is that events just do have explanations. Even when it comes to the most mysterious or inexplicable, we still believe that an explanation does exist, and with what seems to be good reason.

However, there are problems when Honderich is understood to be talking about fundamental explanations that make clear how what happened was made necessary by the circumstances. Consider, for example, the case, cited by Weatherford, of a ‘spoon (that) has mysteriously levitated at breakfast.’Footnote ⁸ No doubt spoons don't levitate unless there is some explanation, i.e., some trickery is involved. However, it is unclear how Weatherford or Honderich could know that such a phenomenon has a fundamental explanation that makes clear that the spoon had to levitate, at least in the sense that the trick could not fail. That there is an explanation, whether or not it has yet to be discovered, is one thing, that the explanation is deterministic quite another.

Moreover, the idea that every event has an explanation is problematic. Honderich defines an ‘event’ as ‘something in space and time, just some of it’, and adds that ‘it is rightly said to be something that occurs or happens.’Footnote ⁹ His use of ‘event’ is different from our actual use of it to refer, for example, to a celebrity wedding or convention lecture, if only because determinists never confine their use of the term in this way. Rather, he would refer to my brushing my teeth, or throwing the ball back when playing catch, as an ‘event’ (or as something that ‘happens’). If I did not brush regularly, or if I had been holding on to the ball, then my brushing or throwing might be referred to, perhaps sarcastically, as an ‘event’. However, the concept, as Honderich uses it, applies even to what is unremarkable, and he needs this application of ‘event’ because determinism is supposed to apply universally, even though there does not seem to be anything to explain in connection with the unremarkable.

And if there is nothing to explain, then there is no explanation. How can we explain, for example, why I threw the ball back? As I pointed out, the question is misconceived when I was not doing anything else that might make it call for an explanation. So, Honderich or Weatherford is wrong in thinking that because we will eventually be able to find explanations for what seems mysterious or unexpected we also will find explanations where there is nothing to explain.

What confuses matters is that kinesthesiology or neurology does find something to explain about my throwing the ball, namely, what happens to my shoulder, arm, muscles, nerves, brain, etc., when I throw it. The explanation involves a change of subject, from explaining why it happened when where there was nothing to explain, to explaining how organs and bodily systems operate when it happened, which does call for an explanation.

And the same is true of ‘events’ where, by contrast with my throwing the ball, nothing seems to happen or occur. A few moments pass as I read the newspaper or my cup remains where I put it. As I imagine it, there is nothing to suggest that I should not have continued to read it or the cup should not have remained there. And yet, it would be considered by Weatherford or Honderich to be an event, i.e., as something in some of space and time, and so in need of explanation. I say this despite the fact that if asked about whether anything happened, we would have to say that nothing did. We can find something to explain if we think of what biology or physics tells us about why a heart continues to pump or a tabletop interferes with the pull of gravity. But neither of these is the ‘event’ we were discussing, which is supposed to have to do with why I continue to read the paper or the cup remains where I put it. That there is something to explain in connection with each ‘event’ does not argue for the fact that the ‘event’ itself is in need of explanation, let alone that it has one.

So, the argument that every event must be given a fundamental explanation is not supported by the fact that explanations for puzzling phenomena have been or will be discovered. The explanations that have been discovered are not deterministic; and, with some ‘events’ there is nothing to explain, and so no explanation.

3. Why Determinism is not Defined in Terms of Causation

The definitions of determinism that were cited earlier introduce refinements in what is already believed, namely, that the rule of causation is universal. I quoted Honderich as saying that to explain an event it is necessary to show that it was a ‘standard effect’, i.e., that it was made necessary or determined by its cause. And even though reference to causation is dropped in Van Inwagen's definition, it is best understood as evolving from the determinism that, as McFee, puts it, ‘begins with the idea of universal causation (i.e., that every event has a cause)’,Footnote ¹⁰ and is reformulated because of problems with treating causes as determinants.

Why is determinism defined in terms of the way the world is from one instant to the next, and not in terms of causation? The answer, I want to suggest, has to do with the fact that something's being an effect of a certain cause does not mean that it was made necessary by the circumstances, that it was the only outcome possible.

To see why, consider, for example, the divorce that was caused by the husband's affairs. Was the breakup inevitable, given Moshe's unfaithfulness? The answer may be ‘no’, because of underutilized resources, such as marriage counseling. However, even if we don't know of any such resources, we can imagine a different outcome given certain unforeseeable developments. How we think about it, or what we say about it, depends on the kind of thing a breakup is, what kind of history it has–we are implicitly comparing and contrasting this troubled marriage with others. If we say beforehand that the breakup is the only possible outcome, we may be expressing a lack of hope; if we say afterwards that no other outcome was possible, we may be offering a kind of reassurance to the couple that they did all they could. In any event, if they stay together because of something we could not have anticipated, such as a head injury to Moshe that completely changed his personality, that is not grounds for questioning what the person said. ‘They did stay together after all. You were wrong to say it was hopeless.’ Someone who says this is joking, if he knows the extraordinary circumstances.

Even if nothing happens to affect the outcome that does not mean that the infidelities determined that the breakup would take place just because they were the cause of it. ‘What he did determined what would happen.’ ‘What he did made a divorce necessary.’ In either case Mary is saying that the divorce was her only option, given her husband's behavior. Even so, she did the determining of what her husband's behavior necessitated; his infidelities did not do that for her.

Determinists will want to say that the husband's repeated infidelities were not by themselves the determinant; and they will go on to remind us of other marriages that did not dissolve when the wife found out about her husband's betrayals. Other factors also may have contributed to the breakup of the marriage, and the total or complete cause, should make reference to all of the conditions that were factors in the divorce. As Honderich says,

He goes on to refer to (1), the set of things, which included the cause, as a ‘causal circumstance’ or ‘total cause’.Footnote ¹¹ Richard Taylor's definition of (total) cause specifies just what the relation is between Honderich's (1) and (2): If A was the cause of B, then, ‘A was the set of conditions from among all those that occurred, each of which was necessary, and the totality of which was sufficient for the occurrence of B.’Footnote ¹²

However, it seems impossible to specify all of the relevant antecedent conditions. The husband may have mixed feelings about his wife and marriage, which surely contributed to the breakup. But was his ambivalence a necessary condition? And the same question can be asked about other features of this particular marriage. The wife could not forgive the husband, and because of what happened, she felt very differently about their relationship. Is the fact that she couldn't forgive him or came to feel differently, a necessary condition?

The problem with these questions is that they treat each condition as separable from the story of this particular relationship. Did the couple communicate very well? To answer we have to know what needed to be communicated. Were they really close? Again, the answer depends on what kind of people they were, and what form that closeness could take. For each possible reason for the breakup, we need to consider what it could amount to as far as the story of this particular couple is concerned.

So, suppose we try to tell the story to make clear that the breakup is inevitable or the only possible outcome. And suppose we isolate from the story certain factors or conditions, treat them, collectively, as constituting the ‘complete cause’ of the breakup, and refer to this complete cause as making it the only possible outcome. Isn't this cause a determinant?

That the answer is ‘yes’ is not obvious. As we pointed out, what makes them conditions or factors is dependent how the story is told. And, there is no reason for thinking that there might not be another marriage where this ‘cause’, as isolated from the story, would not result in a breakup. So, it does not seem possible for there to be a set of conditions, individually necessary, and collectively sufficient, for the breakup to have taken place.

What reason is there for thinking that there must be such a total cause? The answer that otherwise there would be no explanation for the ‘event’ is question begging because it assumes that the only adequate explanation is one that makes clear why nothing else could possibly have happened, no matter how unexpected developments might be. We already cited a perfectly good explanation, namely, the husband's infidelities. If that explanation is rejected it is because of the question-begging assumption that an adequate explanation has to take the form of identifying the conditions which are such that if all of them were present in any marriage then it must have failed.

The same point applies to a natural disaster like an earthquake. To identify its ‘total’ or ‘complete’ cause we need to be able to rule out any possibility that the same conditions might have obtained without the earthquake taking place. The question begging argument for determinism is that there must be a complete cause for the divorce or earthquake because otherwise there will not really be an explanation for it.

To avoid begging the question the determinist talks about the way the world is just before the breakup or earthquake. By referring to everything existing just before the breakup, it becomes unnecessary to assume that the necessary conditions can be individuated in the way that we have argued is questionable, or that the total cause must exist. Moreover, references to causal (or other explanatory) concepts are inapplicable when it comes to the way the world is from one moment to the next. If I push hard enough I may be able to move the refrigerator. But nothing prepares us to understand talk of the way the world is at a given moment, for example, causing, making, or producing the next state; all it could mean for the way the world is at a given moment to be a determinant is that there is only one way the world could be at the next instant—the world or its state at a moment does not seem to be the kind of thing to which a causal concept such as ‘make’ or ‘produce’ applies. So, given that there must be an explanation, the only possible one is that given the way the world is at the previous moment, there is only one way it could be the next one.

Rather than question whether such an unfamiliar kind of explanation is an explanation at all, or why it is better than no explanation at all, I want to treat as problematic any reasoning about the state of the world from moment to moment. There are several reasons for doing so. For one thing, a breakup or forest fire doesn't seem to be the kind of thing that exists at a certain moment and not the moment before. For another, there is nothing in our experience to prepare us to understand, let alone think in the relevant way, about the way the world is at an instant. These criticisms will be developed later. All I want to point out now is that we do not know that there is anything to explain without knowing what a ‘state of the world’ is.

4. How the World Works

Determinists almost never talk about the cause(s) of something like the breakup of a marriage, if only because they want to illustrate the concept of causation with a case where there would be no question as to what ‘caused’ it. This kind of case is suited to explain the sense that things are determined, even though, as we will see, it is the explanation of how things work that is responsible for it. Whatever we decide is the cause of the divorce, we do not expect that explanation to be based on some model of the workings of marriage (or divorce); whereas, that is just we do expect with the kinds of causal explanations that seem to support determinism.

What there is to explain is not made clear when a determinist like Honderich uses an illustration to raise questions about causation.

He obviously thinks of ‘strike’ in terms of cause-and-effect: the cause is the striking and the effect is the igniting, where each of these must be an ‘event’. However, his use of the passive voice obscures the fact that someone struck the match, and by doing so may make it seem that ‘striking’ is logically independent of ‘igniting’, even though ‘striking’ refers to a technique for lighting a match. The requirement that they be independent explains why Taylor refers to the cause as ‘the rubbing of a certain match against a rough surface in a certain way’,Footnote ¹⁴ which is a misleading way of referring to the striking of the match. Honderich's use of the passive voice, and Taylor's use of ‘rubbing’ for ‘striking’, distract us from asking what there is to explain in connection with the igniting of the match.

There is a problem not only with what there is to explain but also with why striking the match was a determinant, as far as the match's igniting was concerned. Judy ignited the match by lighting it, so, she made it happen or brought it about. Did she also determine that the match would light? This question seems to echo what we asked earlier when we wondered whether the spoon had to levitate because a magnet was used to pull it upwards. Then we pointed out that the trick with the magnet was not fool proof, which also applies to lighting a match–something might go wrong, and the match not ignite.

The fact is that the explanation of why the match lit makes no use of the word ‘cause’. A key part of the explanation tells us what a match is: a slender piece of wood or cardboard with a tip that is made of a chemical mixture that ignites easily. When the tip is rubbed against a rough or specially treated surface, heat is generated to a temperature where the chemicals (including a phosphorus compound) in the tip react and produce more heat than can escape into the atmosphere, and so they burn with a flame. After the tip ignites, the paraffin in which the match has been dipped carries the flame from the head to the rest of the match.

Note that I did not use ‘cause’ in the telling of this story, although I did refer to heat being produced, chemicals reacting, and flame being carried. However, the fact that the story may be considered a causal one by philosophers does not depend on whether any aspect of it would be described in explicitly causal terms.

What we may think of as causally deterministic is to be found in the story of how the match (or some other device such as a toilet) is supposed to work when it is struck (or flushed). In the story things happen as they are supposed to happen, which generates the illusion of determinism (a point which also applies to the seeming deterministic nature of a toilet or vending machine). The subject of the story is any match to which the schematic of how it works applies. And a similar point applies to explanations of something like a volcanic eruption, something that is not manufactured to work in a certain way: there is a schematic of the geological fault that depicts what happens when an earthquake takes place.

Even if things go wrong, there must be a story of how that works as well. The match didn't light when struck. Perhaps its tip was wet. If so, we can give a causal account of how that works, by explaining what moisture does to the phosphorus compound. As we keep on asking how things can go wrong, we are going to find it necessary to incorporate references to chemical reactions or physical laws, which will apply to the terms in which the references are couched. We are going to find it necessary because we are no longer giving an account of how a particular device or system operates, but of how the ‘world’ functions.

However, no contrast is possible between how the world is supposed to work and what actually happens. If the match does not ignite when it is struck, then we can explain what the effect of dampness is on the eye of the match, by talking about the reaction of phosphorus with H₂O. Once again, we are talking about what is supposed to happen, this time on the atomic or sub-atomic level. And if things don't happen with the match and its dampness as this story says that they are supposed to happen, then still another explanation needs to be provided. This is because we are talking only about part or an aspect of the (state of the) world. When we consider how the world itself works, we are alluding to everything going on, in which case there is no room for a contrast between what is supposed to be happening and what actually happens: no factors or circumstances have been excluded which would account for why things did not happen as they are supposed to happen.

How are we to understand talk of the world (and how it works)? We have been suggesting that the feeling of determinism is derived from the fact that we are really talking about how things are supposed to work. Since there is no distinction between how a state of the world is supposed to work and how it really works, a thesis about transition from one state of the world to the next seems to have a deterministic flavor. In what follows we have trouble understanding the concept of the state-of-the-world-at-a-moment, and so whatever momentum determinism derives from the use of the concept seems to be stalled.

5. The State of the World at an Instant or Moment

We can think of familiar references to the ‘world’ or its ‘state’, but they do not help us to understand the concept of the ‘state-of-the-world’ as it used in the definition of determinism. As it happens, I have not seen much of the world, nor would I be said to be a man of the world. However, as determinists refer to it, the ‘world’ is not a geographical term or a reference to social sophistication. Nor, is it be understood as a reference, for example, to the insect world or any other subject of a body of knowledge. Furthermore, it is not to be understood as a reference by an economist (to world-wide inflation), political scientist (to increased instability of governments), environmental scientist (to global warming) or religious observer (to the rise of fundamentalism). Each of these familiar references to the world or its state at a certain time is much too narrowly conceived for it to serve the purposes of the determinist.

Rather than think in terms of actual references to the ‘world’ or its ‘state’ (or ‘way’ it is), we should try to think of it as being constituted by what really exists. This is how I imagine determinists to respond to the foregoing. According to this response, the state of the world at a moment consists of everything that really exists at that time.

However, there are problems with this approach. One has to do with determining what exists; is a theory needed to make that determination, and, if so, how can that theory be justified? Another has to do with how are we to determine whether we have everything; what are we to include or exclude? Still another has to do with a difficulty alluded to above, namely, that some things that seem to exist do not exist at a moment. Although the problems seem serious enough, what I think is most significant is that the determinist has not really thought about them. This suggests that his position comes equipped with a conception of what does and what does not really exist, and it is in terms of this conception that we are supposed to understand the state of the world at a moment.

Consider, for example, the lack of hope expressed beforehand or the reassurance expressed afterwards about the marital breakup being inevitable; is that inevitability something that exists? The question is hard to understand, let alone to answer, because it is hard to imagine someone actually asking about it. And if that inevitability does exist, in connection with the expression of a lack of hope or of reassurance, does the inevitability also exist in connection with what a marriage counselor, religious teacher, or even the woman's husband says about it? The determinist may want to introduce his own concept of inevitability, and rule that only in terms of the use of this concept does the inevitability really exist. If so, then the problem has been shifted from determining what exists in the world to understanding and justifying the application of this special concept. So, in the absence of a theory about what sorts of things really exist it does not seem possible to answer a question about whether it really was inevitable; nor is it clear why we need to do develop such a theory, except, of course, for the sake of determinism.

Even if we could somehow determine what exists we still are confronted by the problem of whether it could be the kind of thing that exists at a moment. The problem arises most obviously in connection with things that depend on a pattern of events or actions, some of which happened a while ago, some of which have yet to happen; it also arises because even what may be true or not true of something at a certain moment is not necessarily the kind of thing that can be true or not true of it at other moments. Consider, for example, Ali's character. He is a narcissist. This does not mean that he never acts selflessly, that he is always being narcissistic. Moreover, there are times, such as when he is asleep when there can be no question of whether or not Ali is being self-absorbed, even though he is self-absorbed. So, the determinist must be excluding the latter as something that really exists in the world because it need not exist (or manifest itself) at a particular moment or instant.

Or consider Leilani's belief in her son, in the fact that he will make good decisions. That faith is revealed in what she has said and does over a period of time; it is likely to be tested by what she has yet to say and do. That it exists or does not exist at a moment is an idea that is not easy to comprehend. Presumably, the question of whether she believes in him at a certain moment is asking about something different from whether she believes in him. Here, too, it seems to be the former that exists in the world and not the latter, although it is not clear why.

Or consider my anger at a colleague's implication that my earlier lectures went badly, when he tries to compliment me by saying that students have told him how much better my classes have been recently. That I am annoyed reflects the fact that I take what he said as a suitable object of anger, i.e., as a grievance. Am I annoyed only when I show that I take it that way? The question is misconceived because being annoyed with someone is evidenced in a pattern of thought and behavior, not by what happens to be going on at a particular moment. However, once again, it is what I reveal on a particular occasion that somehow exists in the world, not what is revealed over time, a conclusion that is puzzling, if not obviously wrong.

In the next section we look critically at what one of the more thoughtful determinists says when he explains that it is easy to think about the concepts that have given us so much difficulty in this section.

6. Van Inwagen on the State of the Entire Physical World at an Instant

Van Inwagen offers an explanation of ‘the state of the entire physical world at an instant’ that seems to avoid the problems discussed in the previous section. He relies on a logical point about ‘state’ and on the concept of ‘change’: ‘given that the world is in a certain state at a certain instant, nothing follows about its state at any other instant’ and, ‘If there is some observable change in the way things are—if a white cloth becomes blue, a warm liquid cold, or if a man raises his hand—this change must entail some change in the world.’Footnote ¹⁵ That the cloth becomes blue or the liquid becomes cold is a change in the cloth's color or the temperature of the liquid. It also is a change in the way things are. Van Inwagen's approach seems appealing because it relies only upon the concept of ‘change’–presumably, ‘change’ is an easy concept to understand and use.

However, ‘change’, as he employs it, has its problems. Consider, for example, my raising my hand (to ask a question of the speaker); the raising of it is not a change in me; is it a change in the (physical) world? The answer seems to be ‘no’: the world did not change because of what I did. For me to think otherwise would mean that I greatly exaggerate the impact of what I do. Van Inwagen must be thinking that the action did not change anything; rather, it was (at least part of) the change. The world now includes my raising my arm, whereas before it did not. I was assuming that the world was distinct from my action, and so I wondered what if anything my action did to the world. Now I realize that my assumption is mistaken: the ‘state of the world’ includes, along with everything else, my raising my arm.

However, when we consider what I am doing then it would seem that Van Inwagen's requirement for the state of the world at a moment is violated. What am I doing? I am seeking to be recognized to ask a question, and if that is to be a change in the world, then it is something I can do only if there is a speaker and I am at the lecture where she is speaking (and a host of other conditions). That my action is what it is because of where I am and what has been going on seems to violate Van Inwagen's requirement that from the fact that I am raising my hand nothing follows about the state of the world before or after I am doing so.

An even more serious problem has to do with the idea that because there is a change in the state of the world, there must be an explanation for that change. If my action is supposed to be the change then, as we saw earlier, the mere fact that I raised my hand to ask a question does not call for explanation, as would be the case if, for example, I had vowed not to ask a question or it turned out that I did not have one to ask. If the action is the change, then there may not be anything to explain, and so no explanation for it.

Van Inwagen would insist that the change he has in mind consists of the various (‘physical’) movements of my arm, hand, muscles, nerves, etc. when I raise it. The state of the world at the given moment includes this change as one of its constituents.

The problem is that now there seems to be a distinction between two ways the world is supposed to be, one involving my raising my hand, the other my bodily movements. The distinction may be made clearer by contrasting my raising my hand with its moving to the position it attains when I raise it, as may be pointed out, for example, by a physical therapist who is explaining how the exercises she has given me should help with the soreness in my shoulder. The question we should ask is, which is in the world, my raising of my arm, or my arm moving, as the latter figures in the explanations, for example of my physical therapist (tennis coach, or anatomy and physiology professor)?

Perhaps we do not have to choose between these answers because both the raising of my hand and its moving in certain ways are in the world. The idea seems plausible when my raising my hand is the subject of the physical therapist's explanations.

However, if my action is considered to be in the world, then the fact that the lesson is indifferent to whether I raised it or it moved does not argue for there also being a change in the world consisting of my hand moving in a certain way. If anything, all we are entitled to conclude is that I raised my hand, and that, for certain purposes, my raising it may be treated as movements of certain parts of my body.

Or, perhaps we do not have to choose between them because my raising my hand is reducible to bodily movements. The difficulty with this idea is that it treats the way the world is in terms that are in conflict with the world as we find it: my raising of my arm is misdescribed by the second answer as my arm moving or my moving it in certain ways. The latter is something quite different: it is something that happens, for example, because of a neurological disorder; when I am pushed or I do something else such as a shoulder exercise; that I do without realizing it when, for example, I am sitting for my portrait and have been asked to keep my arm still; or that I do when I comply with a request to move my arm by someone sitting behind me at the movies. So, to describe my arm as moving, or me as moving my arm when I raised it to ask a question, is to misdescribe what I did.

It might simplify matters if we qualify determinism, as Dennett and Van Inwagen do, by confining it to a thesis about how the world can be ‘physically’. However, it is unclear what a ‘physically possible future’ is. We have had trouble applying the thesis of determinism to such things as a marital breakup or my raising my hand to ask a question. If neither of them is ‘physical’, then presumably many other things also are excluded, including an economic recession, a religious revival, social injustice and a lack of hope. These exclusions leave us with a determinism that seems to have much too limited an application.

Rather than defend himself, the determinist would take the offensive by asking, what else is there to us other than what is described by our anatomy and physiology? This is difficult to answer, because of the assumption behind the question that our ‘physical’ selves can be conceived of independently of our selves. After all, my organs or extremities are identified not as they are to be found on shelves or drawers in a hospital or morgue, but as mine. So, this resolution of how actions can be reduced to bodily movements has the paradoxical consequence that the ontology upon which it depends is undermined by the reduction: without me there is nothing left into which I can be analyzed in biological terms.

The objections I have been making against determinism are based on the assumption that determinism is supposed to apply to the world as we find it. However, these objections do not seem to arise in connection with a world or universe defined in the terms of scientific explanation, and in the next section we consider the significance of this fact.

7. Determinism and the Particle-Universe

Dennett explains determinism by relying on Laplace's idea of a demon who knows all the ‘forces’ that ‘animate’ the particle-universe.

The laws of physics obviously are uniquely qualified to explain the collisions, rebounds and near misses of particles, and they enable the demon to do the deterministic plotting, at least in theory, that Dennett (or Laplace) has in mind. And, presumably, the determinism of the particle universe implies the determinism of everything else.

Dennett thinks that there are ‘fudge factors’ in that Laplacian version of determinism, including whether every sub-atomic particle is to be plotted, and which properties of the particles need to be included in a ‘state description of the universe’. This explains why Dennett confines his determinism to Quinean ‘Democritean universes’ where ‘atoms’ that are unsplittable and unsliceable are conceptualized as moving about in digitized space.Footnote ¹⁷

However, the questions I want to raise in this section do not depend on how particle-universes are defined, provided that the problems with determinism that we have been uncovering do not arise when we are talking only about such universes. Consider, for example, the objection that sometimes there is nothing to explain, such as when I do nothing remarkable in throwing the ball back. The sciences explain what happens when I throw it back with my arm, shoulders, legs, back, nerves, muscles, and so on, as well as what happens to the ball. And, presumably, when those things happen, things also happen in a Democritean or particle-universe. However, what happens is not the same ‘event’ as my throwing back the ball. So, because there is an explanation on a (sub)microscopic level, it does not follow that there also is one on the macroscopic level: what is or may be true of the former is not necessarily true of the latter.

Let me further illustrate the problem I am raising by referring to the distinction between raising my hand and its being moved. If that distinction cannot be made in connection with events in the ‘physical’ universe, then that argues against conceiving of that universe as all there is. It would be more prudent to insist that the distinction does not disappear when the particle ontology is adopted. So, either the ontology itself fails because of what it does to the distinction, or it fails as a strategy to avoid this obstacle in the path of determinism.

This dilemma exposes a tension in the relationship between the (sub)microscopic and the macroscopic. The snapshot the demon examines is of the location (etc.) of particles that are referred to in connection with an explanation, for example, of why certain substances are soluble in water, an explanation that, ultimately, will refer to certain atomic properties. Note that the direction of the explanation: we start with the familiar world, and then arrive, eventually, at molecules and atoms (and other particles) to explain how things are supposed to work and how things can go wrong.

However, a determinist like Dennett must be thinking that once we arrive at the sub-microscopic world then the other world somehow drops out and ceases to really exist, at least as far as our understanding of things is concerned. And so, the dilemma is avoided by denying the assumption behind its existence, namely, that there really is a distinction, for example, between raising my hand and its being moved. By adopting this particle-ontology the obstacles that we encountered macroscopically turn out not to exist at all because the only reality is to be found in this (sub)microscopic universe.

However, if the reality of what is being explained is questioned, then the explanations themselves become questionable. What this means is that a particle universe ontology seems to undermine itself if it requires revising our understanding of the very things that the sciences are supposed to explain in terms of particles, their location, velocity and mass. That is to say, a revisionary approach towards the supposed obstacles to determinism seems to sabotage the very basis for that approach.

In response, a scientific philosopher or determinist would cite the fact that science is responsible for changes in our understanding of macroscopic phenomena. Consider, for example, sunrise and sunset. When we look to astronomy to explain what is really happening, then we have to alter or revise our understanding of what is to be explained. On this revised understanding supposedly provided by science, the sun does not rise or set; so, what is explained no longer exists under the explanation.

A related point is that what is explained embodies a folk theory that mistakenly conceives of the sun as an astronomical body that moves in relation to a stationary earth. The explanation alters our understanding because it is made in terms of an astronomy that displaces that folk theory.

However, astronomy does not encourage correcting anyone who describes or directs attention to a sunset or sunrise. As Ebersole explains, when we suppose that anyone needs correcting,

That we have to imagine ourselves in outer space becomes apparent when you consider that the distinction between what is moving and what is stationary, when it involves ourselves, is made in terms of what is on the earth, not the earth itself. ‘You moved.’ When an artist complains in this way to her model, she could not be talking about the fact that her model was standing on earth, which was moving around the sun. We might sometimes be mistaken and experience what we later discover to be a false sunset or sunrise. However, that is highly unusual, and not what we have in mind when trying to understand how any mistake is being made when we refer to a sunset or sunrise: science should not be understood as correcting us on this point.

The fact is that a reference to sunrise or sunset does not presuppose any theory about what the sun is or whether or not it revolves around the earth. To think otherwise is to suppose that when say, ‘look at how beautiful that sunset is’, we really are saying that the sun is orbiting the earth and has reached a point in that orbit where it is about to disappear from view. So, as evidence that the sciences may require us to drop distinctions that we would otherwise make, or explain what otherwise is not in need of one, this example fails. And, I submit, the same point applies when philosophers use discoveries of the sciences to rule, for example, that there is no difference between raising my hand and its moving up, that everything needs to be explained, that events are the kinds of things that exist at a moment, or that an explanation fails when it is not deterministic.

Let me summarize this section by returning to the reasoning that if what happens to particles when I throw the ball is determined, then surely the same is true of my throwing it. The reasoning is mistaken and the mistake is in saying that the particle-universe is determined. That universe is conceptualized in the course of offering explanations of how things are supposed to work, and of how things go wrong. And, there is a powerful pull towards determinism that is exercised by the idea that the objective is to explain how the world works. This is because, as we saw earlier, when we explain how the world works no distinction is possible between what is supposed to happen and what does happen. However, the concept of ‘how the world works’ is without any real application: the ‘world’ is what we end up talking about when we want to refer to a limit of inquiry and explanation, even though we have no reason for supposing that such a limit exists. So, the mistake with the idea that the particle universe is deterministic is that that universe is conceived of as existing independently of the program of explaining how the world works.

8. The Belief in Determinism

What explains the appeal of determinism? It is hard to believe that so many philosophers are determinists because they are convinced of a thesis about states of the world from moment to moment. As I explained, this way of defining determinism is introduced by philosophers because they think it offers a more philosophically respectable version of determinism, a determinism that they already embrace. So, what is the real basis of the appeal of determinism?

It is easier to explain why the opposition to determinism has not been effective. The explanation has to do with the rhetorical strategy of determinists. An important part of that strategy is to challenge the opposition to defend indeterminism. However, we have the same reasons for questioning indeterminism as we do for questioning determinism, when it, too, is defined in terms of states of the world from moment to moment.

Also effective is the strategy of treating the threat to determinism as coming from its supposed incompatibility with free will. Does the fact that there is only one way the world could be at the next moment imply anything about the behavior of someone who acted at that time? Without a better idea of how to think in terms of the way the world is from moment to moment, I do not see how I can be clear about whatever it is about our agency or autonomy that determinism is supposed to threaten. So, the strategy of challenging the opposition to show that determinism is incompatible with free will, or that free will really exists, also is effective.

Why be determinists in the first place? I suggest that the answer has to do with the appeal of the idea that there are laws of nature that govern everything that happens, that make everything comprehensible or explicable, at least in principle. This idea seems to account for the comprehensiveness of the deterministic principle, for the fact that everything falls under its jurisdiction. If this is what science tells us, then an allegiance to science seems to be in back of the idea that nothing is fortuitous, that the laws of nature govern everything.

The stimulus for the further idea that these laws are deterministic seems to be derived from what determinism seems to displace, namely, the idea of the sovereignty of God. When it is God who rules, His rule may be described, in psalms or prayers or elsewhere, as unbounded and absolutely general: He made or is responsible for everything. However, the fact that it is God who makes something happen means that it had to happen. I am not saying this because I think that it is true, I am saying it because that is what many religious philosophers believed: when God does it, He determines that it take place. Spinoza even refers to the laws of nature as ‘divine’, by which he seems to mean not that the laws are due to God but that they are God. If Spinoza may be understood to be replacing the God of transcendence with what might be thought of as the laws of immanence, then it would seem that he retained something from what he replaced, namely, the idea that whatever God wills must take place, and that His will has no bounds (even if these ideas are beyond our understanding).

If my suggestion that there is a religious basis for determinism seems far-fetched, if not a bit perverse, my excuse is that I am trying to account for how powerful its appeal is even to a non-believer like myself who wonders whether he is missing something. I hope that determinists are not put off by my suggestion and so fail to appreciate what I see as my achievement, namely, that of showing that there are problems with thinking clearly about the key metaphysical concepts used to define determinism.Footnote ¹⁹