What Do You Mean by “God”?: Denotation Switch, Three Forms of Disbelief, and Zealous Agnosticism

Estimated read time (minus contemplative pauses): 58 min.

caute symbol spinoza1. Atheist? Don’t You Mean Agnostic?

I’m sometimes asked if I believe in God by a self-proclaimed agnostic who assumes that I, too, will proclaim myself agnostic. My honest reply, though, is that I don’t believe in God. This makes me an atheist, I presume. But to be sure—to really capture the nature of my disbelief—I must know what’s meant by “God.” Sometimes I’ll request an explanation; but not smugly, I hope, as I’m not a militant or angry atheist. It’s just hard to say what I believe about a thing without knowing what the thing is, more or less.

Some (not all) agnostics respond to my request with something like miffed incredulity, slightly excited frustration, or outright anger, even though it usually turns out that they share my rejection of any standard or traditional account of God. They often go on to point out that I’m failing to appreciate—or worse, am willfully ignoring—the vastness of our wondrously mysterious universe in my (hubristic, egocentric, volitional?) embracement of atheism. (I’d like to think I embrace atheism about as much as I do a-unicorn-ism; if it seems otherwise, it may be because I’m never pressed to defend my a-unicorn-ism. And so I take the accusation of “embracement” here to amount to something like the following: “Even if you do not choose to be an atheist, you accept that you are one without making a sufficiently strong effort to change your beliefs.”)

Those who press me like this are sometimes what I call fanatical or militant or zealous agnostics. I’m happy to tell them that I believe there to be many—insurmountably, infinitely many!—things yet undiscovered or understood by humans. It doesn’t strike me in the least, however, that this means, for example, that the New Testament is true. On the contrary. But, again, to this I’m nearly always told, with certainty, “That book, like all books, gets God wrong.” Sometimes I’m told that no writings, and certainly not the most famous ones, even get close to what, or who, God really is.

So I’ll ask, “What sort of thing, then, are you asking me about? What are those books failing to describe?” And again my questions are greeted suspiciously, as though they amount to a smug evasion or to some sort of rhetorical trap. My aim is not to evade, but I do admit that I often mean such questions as a step towards unmasking a disguised atheist. For that I apologize and I vow to do better. But I don’t apologize for asking what sort of thing I’m expected to have beliefs about.

I write this in the hopes not of persuading anyone to share my beliefs about God, but rather to persuade such agnostics that my question—”What do you mean by ‘God’?”—is a reasonable question from someone being asked to assess their belief about God. To that end, I’d like to touch on two points: the first is a rhetorical maneuver I’ll call the denotation switch (which we might also call the denotation fallacy), the second is a brief outline of three forms of disbelief.

In doing so, I won’t point to real-world instances of zealous agnosticism. Maybe you know what I mean by all this from your own experience with such agnostics. Maybe you are one yourself. Otherwise, you’ll have to take my word for it: I’ve encountered the attitude I describe here (with varying degrees of civility) on numerous occasions, in private and in public.

I should also note that I tend here to talk especially about Christian views of God, as those views are best known to me. I’ll say more about my background beliefs after discussing the three forms of disbelief.

2. The Denotation Switch (Vis-à-Vis God or “God”)

By “denotation,” I broadly mean the activities and facts surrounding the connection a word has with the thing that word refers to in a given instance of usage. Namely, to say that a word denotes X in a given instance of usage is to say that the word refers to or picks out or stands for X.

For simplicity, I use the word “word” here, while recognizing that what I denote with the word “word” may also come down to multi-word phrases; may be hand-signed, pen-scribbled, conveyed with Morse code or semaphore or drumming, and so on; and may itself denote more than one sort of thing, as with a pun or as with my use here of the word “word.”

The denotation switch takes several forms, but essentially comes down to a hidden, subtle, or obvious usage deviation from what a word either usually denotes, or would be reasonably expected to denote in a given context; this deviation may be consciously or unconsciously motivated or performed.

A silly example would be if I told you I’m giving you the fifty bucks I owe you while handing you a penny and saying that “fifty bucks” is what I call pennies. Change the example to one about “God” and imagine my frustration.

You get the idea, but I’ll follow up this post with a Part Two that includes liberal reflections on the scope of this phenomenon; mostly, they’re exploratory notes I made while writing this post. Several examples pop up there—e.g., involving art, pornography, virtual reality, altruism, free will, self, democracy, law, chair, paradigm, nothing, belief, knowledge, child abuse, God (again).

I also briefly speculate on the possible roots of strong agnosticism—e.g., Albert Einstein, who usually (not always) rejected the “atheism” label and at times referred to himself as “agnostic,” though claimed to believe in “Spinoza’s God.” Baruch Spinoza is briefly discussed as well.* (I resist the temptation to mention David Hume, but will recommend, in this footnote, a podcast episode for a discussion that begins with the question of whether Hume was an atheist; the guest essentially answers “yes,” and I’m persuaded that he’s correct.1) Part Two will likely end with a note about a recent live debate frustrated by participants meaning different things by “God.”

[*If you like to define “God” as “energy” and suchlike, and are looking for a role model, I’ll refer you to Spinoza, who described God as a kind of ultimate, universal theory of everything, and did so with rare and awe-inspiring rigor—e.g., see his The Ethics, with its definitions, propositions, proofs, and axioms. The most rigorous answer I know of to the “What do you mean by ‘God’?” question.]

With the basic idea of the denotation switch in mind, let’s return to the central point of this writing—namely, that “What do you mean by ‘God’?” is a reasonable question. Or put another way: “What sort of thing is it that you’re agnostic about?” These questions are reasonable because, if I don’t know what thing the word “God” denotes, I can’t tell you what I believe about that thing.

From those who’ve ventured an answer, I’ve heard quite a range of things that are or may be the target of their agnosticism—that is, that “is God” or at least “could turn out to be God.” But most of those things obviously have nothing to do with what I’m rejecting as an atheist. If you say God is the enjoyment you get from your morning cup of coffee, well, I believe you drink and enjoy coffee in the morning. If you say God is the pleasure you get from Beethoven’s 9th symphony, I can say that I believe your pleasure there exists. Same goes, more or less, for energy and quantum fields and “the fundamental rules of the universe and cosmology” and so on. I figure there must be some basis or another for the birth of a universe. Of course I do. And so does the zealous agnostic. So the existence of those things must not be the target of their agnosticism, so much as, for example, whether those things are God or merit being called “God.”

I find this very confusing, so will try to say it another way. As mysterious as cosmology and love and so on may be, they clearly are not what I’m rejecting when I say I don’t believe in God. Nor are they what more famous atheists than I are rejecting—say, Richard Dawkins in this Ted Talk, “Militant Atheism,” or in this video from Christopher Hitchens’ apartment, featuring four of the current era’s biggest names in Professional Atheism: “The Four Horseman.”

Again, the zealous agnostic, or “z-agnostic” for short (I also considered “m-agnostic,” but maybe “m” for “militant” is too broadly harsh), doesn’t seem to be on the fence about whether the enjoyment of music, quantum fields, etc. exist. Which is to say that z-agnosticism often seems to come down to a difficulty in deciding what known-to-exist phenomenon—or maybe yet-to-be-discovered (natural) phenomenon—the word “God” should or can be reasonably applied to.

What I reject as an atheist is the existence of a certain entity that is presumed to exist whether or not we have a name for it—namely, the one described in holy books. The z-agnostic, too, has already rejected that entity. So it can’t be that the z-agnostic is holding, say, energy up to what’s already understood to be God’s (already rejected) nature in order to compare the two and see if they align. Rather, the z-agnostic is trying to take the word “God” and find a place to put it after having rejected all the other applications of the word they’ve so far encountered, for example, in religious doctrine—again, precisely the same doctrine that I reject as an atheist. A difference here seems to be that I’m happy to also discard the word “God” along with that doctrine, while the agnostic wants to find another thing to denote with the word (and then tell me I’m not an atheist due to their engaging in that project).

Maybe I can put this in an even better way. Suppose you’re z-agnostic (a term I use so I don’t have to continually differentiate with agnostics of a more traditional sort). You might claim to view the word “God” as denoting a certain function—e.g., that of bringing a sense of peace, well-being, and meaning into one’s life; or perhaps, more fundamentally, the phenomena responsible for the genesis of a universe like ours. This comes down to deciding whether or not some thing (or bundle of things) sufficiently performs that function so as to merit the label “God.” This, I suppose you’ll claim, need not require a deity that exists independently of humans and their languages, social interactions, and so on.

Now, if you ask me if there are things that bring a sense of peace and so forth, I’ll say “yes!,” though I’ll reject the notion that I should be agnostic about whether those things are God. In other words, I reject the functional view of God just as a devout believer, in the more traditional religious sense, likely would.

Indeed, many agnostic claims seem to imply that those who do believe in some holy book or another are themselves atheists (at least in my sense of the word), as they are failing to believe in the real God (e.g., quantum fields or energy or whatever), which is something entirely different than the entity described in their holy texts. This is especially true when agnostics emphatically and explicitly reject the supernatural entities described by those holy texts.

Of course, some agnostics (i.e., of the more traditional sort) are agnostic about a deity described in some particular holy book. I’m obviously not addressing such folks here. In fact, they tend to have little problem understanding what I mean when I say I’m an atheist. I’m addressing, rather, the agnostics—whom I, for convenience, am calling “z-agnostics”—who talk in ways that suggest that “something could turn out to be God, though it won’t be the stuff in those holy books.”

I know I’m repeating myself, but I do so both out of confusion and in the hope of not being misunderstood.

And to be clear, the z-agnostic view really does confuse me. My usual understanding is that an agnostic is, to put it simply, someone for whom “Do you believe in God?” is not a yes-or-no question. More elaborately, it is either…

…someone unable to choose between equally compelling sets of opposing evidence about whether God—according to some epistemically viable cultural conception (e.g., it’s never Athena, Inanna, or Odin)—exists;


…someone who thinks the question cannot be addressed by mortal minds or Earthly evidence.

But such conceptions of agnosticism fail when there’s no reference point for the word “God.” An agnostic deliberating over whether some assumed-to-exist, if little-understood, thing, like energy, is the thing the Bible (or some similar book) tries, but fails, to describe seems to keep this definition of “agnosticism” intact. But it falls apart when the z-agnostic rejects those books entirely, particularly when claiming that energy (or what have you) is in fact God (e.g., “Spinoza’s God”). This is less strange in the relatively rare case when the person simply says, “I believe energy just is God, and so I’m a confirmed theist.” But it strikes me as strange indeed when someone (like Einstein) says, “I, like Spinoza, believe energy just is God, and I am agnostic about whether or not God exists.” I take this to be a kind of social maneuver more than a literal statement about belief, but will save this sort of question for Part Two.

Also strange is when claims like “God is energy” come along with the claim that it simply can’t be known whether God exists—particularly strange because the person does in fact seem to believe in energy. So now I’m confused not only about what “Do you believe in God? is asking about, but also about what’s meant by the word “agnostic.” Thus my thought that the z-agnostic enterprise often really comes down to deciding what phenomenon they already believe in merits the label “God,” generally according to some functional role played in their life by that phenomenon.

To be clear, there are some folks who are not agnostic about whether the non-holy-books thing they believe in is God. For example, as I’ll mention in Part Two, some folks (e.g., Deepak Chopra) claim that, by “God,” they mean “consciousness.” They might mean that term in a somewhat specialized way, but in a way that does not involve a deity of any sort, and that is close enough to the usual understanding of “consciousness” that it can be said that no one is an atheist.

Of course I believe in consciousness: I’m conscious and so are you. (If you’re not sure whether you’re conscious, see my recent post, “How to Tell If You’re A Zombie.”) I see no reason, however, to believe that consciousness merits the label “God.” To call me a theist for believing in consciousness would be a paradigmatic example of the denotation switch, which I’ll describe more explicitly in a moment. For now, I’ll offer a basic exhortation: Can we please just agree that we may at times be talking about completely different things in such discussions, even when using some of the same words?

The simple and obvious point that we may very well be talking about very different sorts of things—things that serve entirely different sorts of functions—is a crucial one. I’ll struggle just a few more moments to get across how important this point is; that is, how important it is that we at least try to make sure we’re speaking the same language when discussing things that matter.

Imagine announcing to a Pentecostal congregation that they are in actuality worshipping mindless subatomic quarks. Under a functional view, you might also tell them that the thing they pray to—the actual thing, for example, that produced our universe—is a mindless interplay of energy and physical laws and so on. But this is wrong. It is, rather, a deflationary move implying that the entity they pray to doesn’t exist—the entity, that is, who talks backs when they pray and loves them and sent down a son named Jesus and fills them with the Holy Spirit and heavenly tongues and performs a host of other important functions, rather that just one arbitrary function, no matter how fundamental that function may seem to the z-agnostic.

An analogy. If I found out that the thing sneaking into my backyard to eat my tomatoes—a thing which I’ve assumed is a coyote with all the furry canine features and personality traits therein expected—actually has all the features of a squirrel, I would not begin referring to the little nut-munching creatures scurrying up and down my trees as “coyotes.” The fundamental function sneaks in and eats my tomatoes is not a sufficiently robust feature in itself to change the word “squirrel” to “coyote.”

Though suppose I called the imagined coyote Sam. I very well may start calling the tomato-eating squirrel Sam. But Sam was never necessarily the coyote: it was whatever thing is eating my tomatoes. Of course, if we build into “Sam” all the features of a coyote, than it would not be the same Sam once the name is transferred. The point is that we are applying that name according to a single function, rather than to a bundle of traits (more on this distinction in a moment).

Let’s make these examples more vivid. Suppose you go on to tell the experienced and devout worshipper: “The voice you hear in response to your prayers is actually just you talking to yourself in your mind, and your mind is actually what you mean by ‘soul,’ which really is just identical to your brain. And while the energy contained in your body will disperse into the world after you die, it won’t carry your memories, as those are correlated with physical structures in the brain that will decompose. (Though there may be some hope of a computer mapping your brain so that your ‘soul,’ or really your mind, may persist on hardware.)” But that’s not at all what God or souls are for the worshipper, any more than the squirrel is identical to the coyote. The worshipper means something by the word “God”—some massive bundle of traits irreducible to any arbitrarily highlighted function.

Borrowing some terms from philosophy, we might refer to the functional view as a de dicto interpretation of God, while what we really want is a de re interpretation. Here’s a typical illustration of this distinction: “Marge plans to marry the tallest man in town this Friday” could mean…

de re: Marge plans to wed a particular man—say, Pete—who, at this moment, happens to be the tallest man in town;

…de dicto: Marge plans to wed whoever happens to be the tallest man in town on Friday, which may or may not end up being Pete.

In the de re interpretation, “tallest man” is shorthand for referring to Pete, while under de dicto, “tallest man” refers to whichever man happens to be the tallest on Friday. Should Pete leave town on Thursday night, for example, someone else will be the tallest man in town on Friday (even if that man also happens to be named Pete).

Notice that a careful de dicto interpretation of common uses of the word “God” will converge to a de re output; for instance, a devout monotheist’s use of the word “God” bundles enough traits and functions so that the word can refer only to one particular being.

Contrast with the tomato-eater case, in which the de dicto interpretation of “the thing eating my tomatoes” would lead us to revise “is a coyote” to “is a squirrel” (along with all relevant feature changes; e.g., from canine to rodent), given that the thing eating my tomatoes is the pivotal feature in question. There is no analogous pivotal feature for God.

Again, I admit that the tomato-eater case isn’t perfectly clear-cut. For example, a de re interpretation there might be that Sam was eating the tomatoes all along, even if it turned out to be a squirrel instead of a coyote. But maybe not if it turned out to be, say, a bunch of beetles or your eccentric neighbor’s pet robot or tomato-eating bacteria or really it’s just you doing it while sleep-eating. Or maybe the name would still stick in these strange cases, and would still be considered the “same Sam all along” to some degree. Again, this will all just depend on how many traits are presumed to be inflexibly fundamental to your conceptual picture of Sam when you utter that name in the tomato-eater context. Maybe there turn out to be three Sams. And so on. Now imagine how this silly example might bear on real-world cases where the features of “Sam” are developed over generations and work their way into culture and broader belief systems.

Of course, it’s not difficult to imagine examples in which the observable effects of a mysterious, unseen entity’s activities may earn it a name that rightfully sticks once the nature of the entity itself is discovered. The Wonderful Wizard of Oz is thought-provoking in this context, as are, I presume, certain episodes of Star Trek (maybe you can tell me which). Were some entity revealed to share a substantial number of the right sorts of features with our core concept of God (more about which shortly), then I’d be thrilled to accept that entity as God. There may be some such cases in which the label “God” would stick without significant change in its core meaning, and other cases in which it might stick but with the understanding that the shift involves enough changes so as to imply atheism about the formerly understood entity conceived of as God (just as one may revise “magic” out of the Wizard of Oz concept once the man behind the curtain is revealed; alternatively, one might intuit that the “wizard” label survives the curtain pull).

(All of this works in the reverse direction as well, which essentially amounts to a rose still being a rose by any other name or, “call yourself what you will: if it looks, swims, walks, and quacks like a duck…” I don’t need to say more here for difficult socio-political implications of this discussion to emerge. Indeed, this is precisely what’s going on here: the z-agnostic is telling me I swim and quack like a z-agnostic, and I’m telling them they swim and quack like an atheist!)

Reducing God to one, or even just a few, depersonalized functions strikes me as insufficient for holding onto the name “God.” Calling the blind laws of physics God is, for me, tantamount to saying God doesn’t exist.

Some less dogmatic z-agnostics might respond here that they wouldn’t tell someone that their positive beliefs about God are wrong, but that there is no problem with telling me that my negative beliefs about God are wrong. This falls loosely under a family of stances that includes the there-must-be-something-out-there stance. My general response to that sort of talk is, unsurprisingl, “What do you mean by ‘something out there’?” I don’t understand phrases like “something out there,” “ultimate truth,” or “ultimate reality.”

Interestingly, what comes to mind here is an instance in which a self-proclaimed believer in the traditional Christian God once told me, “God doesn’t care what you believe in, so long as you believe in something.” I said I believe in the nonexistence of God. She said this didn’t count. I take it she meant the belief  should have some transcendence about it, particularly of an at least vaguely religious or “spiritual” sort, though I didn’t ask what she thought God would say about fully rejecting the Bible while believing mindless particles or energy or information or some such to be God. But she did give every indication of being a believer who had specific beliefs about God that included the feature “He who would approve or disapprove of the ‘something’ I (choose to) believe in.” This fundamental specificity is what the z-agnostic rejects, a move I don’t know how else to interpret but as entailing that their most important positive beliefs about God are wrong, thus implying that that prayer-hearing, all-loving, heaven-ruling, etc., entity does not exist.

To reconcile this implication with the those specific religious beliefs, some z-agnostics (and self-proclaimed believers; the line at this stage of discourse eludes me) have told me that holy books such as the Bible were intended as—or at least these days should only be viewed as—metaphors, and what those metaphors stand for is up to a given reader. One can, then, validly interpret the Bible as being about one’s enjoyment of Beethoven’s 9th. That’s fine, but such a view allows for infinitely many valid interpretations, with no reason to favor one over another as the singularly true interpretation; in other words, I have no reason to believe any one of them. Nor do I feel remotely moved to concoct any such interpretation.

And so, I’ll again point out that if you see the Bible as “just a metaphor,” particularly when it comes to descriptions of a sentient universe-creator, I don’t see how this makes you agnostic (much less a believer); that strikes me as atheism. Or, if you claim that your personally chosen metaphor is literally correct—literally accounts for “the real God”—then this strikes me as robust theism, far from agnosticism.

Allow me to again emphasize that I’m not arguing for the truth or falsity of any given holy writing. I’m simply arguing that if you claim to interpret all those writings in such a way that none of them is literally true—particularly when it comes to the existence of sentient deities—then I’d consider you an atheist. If you claim they are all true in some deeper, pluralistic sense, then I’d say you’re a theist (one obviously need not be sure about God’s nature to be a theist; more on this in about 15 seconds).

To this line of thinking, I’ve encountered the response: “But isn’t this a nice way to view scripture so that people don’t become textual literalists and start stoning one another for silly reasons?” Sure it is! Though I believe such books were meant literally, I favor the idea of it being acceptable, or even compulsory, to ignore or blur out the most socially damaging aspects of those books. This is also tantamount, however, to telling devout believers that their view of the Bible as the best—if not perfect—description we have of the one and only true God, is wrong. (And many of the folks I grew up with swore every word of the Bible was literally true: “Show me one wrong word in this book and I’ll throw it out the window,” I was once told.) I take it the death of the author, as that ethos is commonly understood today, begins here.

Such difficulties have led some to posit that the more we think we know God, the less we really do (consider Søren Kierkegaard’s theistic, or Christian, existentialism), and that the only way we can talk about God is by saying what God is not (see Moses Maimonides‘s apophatic, or negative, theology). Such views are far too complicated to get into here, much less how they might bear on z-agnosticism (in which phenomena about which we know quite a lot might be nominated as “might be God” candidates). I will say, though, that those who adopt such views, however, whether believers or not, do seem to have some core concept of God in mind, with varying degrees of basic or essential details in place. (The notion of a core concept is important to keep in mind when discussing the denotation switch).

What might this core “God” concept be? I suppose I can only speak for myself. By “God,” I roughly mean a sentient supreme being who knows you and I exist, created the physical world from nothing, is all-loving, all-knowing, all-powerful, all-present, and so on. Some variation thereof. A being to pray to and who may or may not send me to Hell for eternity and who, according to some (but not all), sent down His or Her son to be crucified and resurrected. And so on.

This covers the core concept, at least, with some flexible cultural additions and filtering. It’s familiar enough that I don’t really see a need to explain it any further. Nor need I explain how my description relates to or differs from, say, the deities of  Norse mythology. Until, that is, a z-agnostic tells me everyone is getting even the core concept wrong (though scrutiny of the z-agnostic’s attempts to describe what God might be usually reveals an adherence to some subset of a core concept much like the one I just described).

To address this, some z-agnostics (or maybe this is more common among believers) embrace a pluralism that claims all sincere uses of “God” refer to the same thing, whatever the user’s core concept may be—i.e., provided, that is, they use the word with the appropriate sense of awe and submissiveness and so on. That view, however, still insults the monotheistic traditions in which, for example, Jesus cannot simultaneously be a divine manifestation of God (i.e., one third of the Trinity) and a mortal prophet. (I won’t—though I could—call out specific instances in which globally revered practitioners of an inherently pluralistic religion have told monotheists, “Hey, no worries. We’re all ultimately practicing the same religion anyway.” This is, of course, not something a devout monotheist can get on board with. Particularly when followed with, “One day you’ll reincarnate into a being who practices a more correct expression of that broader religion.”)

At any rate, the pluralistic route leads to the same problem for the z-agnostic as does that route which claims all interpretations of the Bible to be correct. And it clearly poses a problem for any approach in which God “might turn out to be” something in particular—i.e., the actual thing our world’s holy books are failing to describe.

I’ve obviously only scratched the surface, by the way, of what, for some z-agnostics, God “might turn out to be,” thus inadequately representing that view. For example, I’ve not yet mentioned that idea that we likely live in a computer simulation, and, if so, our programmer(s) is (are) God. I’ll briefly address this view in order to give a sense of how such a discussion might go.

I find the view incomprehensible, even if we presume those programmers are still physically alive and can actively influence the program, and even if we take the functional (or de dicto) view that God is, strictly speaking, whatever entity created and has massive influence over the reality we live in. It seems to me, rather, that God must be, at the very least, the entity that created the conditions that make physical existence possible—and thus that made possible the existence of programmers as well as the laws of physics (and logic and metaphysics?) that limit what those programmers can create in the world external to our simulation-running hardware, and so on. God, the “unmoved mover,” is whatever created the first programmers (who may be the programmers of the programmers of the programmers of our programmers, who may themselves be simulations).

I’m more open, however, to the idea that some feature of the program itself might merit the label “God,” particularly if there is involved an indefinitely sustained afterlife assigned by the program according to how you’ve lived your life. Such a feature might meet enough of the traditional functions of God to qualify and, in fact, could turn out to be the true source of many things written about in a given holy book, as that book was written within the simulation. I’m more open to this meriting the label “God” than the programmer whom I imagine to be a pimply-faced space-teenager sitting in front of a fancy X-Box hooked into a planet-sized battery.

Though there still remains here the critical problem of the program not existing eternally—not being an unmoved mover—and the impossibility of miracles (i.e., the program is still, in significant ways, 100% subject to the laws of physics that exist “out there” in the “real” universe, even if those laws are very different than the ones we observe in our simulation). This may not be a problem, however, for the sorts of deities whose origins are thought to arise after the existence of the material world, not to mention that, for all we know, there have always been universes in or outside of the simulation (but we’re especially ignorant on this count if we are in the simulation: the Big Bang might just be when the pimply teenager hit clicked “run”).

Time to wrap up this section.

When given a description of God, I can generally give some account of my belief. For example, I can say that I don’t believe in the deities described in any holy texts I’m familiar with. Nor do I believe our living in a computer simulation would open up a space for such an entity, though, as I’ve noted, there are things I must grapple with there. I can’t grapple with them, though, unless you tell me what you mean by God! When I’m told, however, that God must posses some bundle of attributes that throughly defies any and all attempts at description, then I can confidently say I don’t believe in such a thing, or at least that I don’t believe in its being God, if only trivially so (though I do believe that there may exist bundles of attributes that will forever frustrate human attempts at description). I explain trivial disbelief in the next section, where I outline three ways of not believing.

I’ll close this section with a silly illustration of the frustration I feel when people pull the denotation switch in the “God” context. It’s certainly a caricature. Belief in God is obviously a far more complicated—e.g., socio-culturally and historically loaded—matter than the below conveys. What it does convey is my frustration.

Earl: Do you believe in vampires?

Joan: What, in their literal existence?

Earl: Yup.

Joan: I mean… no, right? Is this a trick question?

Earl: You don’t believe vampires exist?!?

Joan: Of course I don’t. Do you?

Earl: What if vampires just turn out to be vampire bats or even just mosquitos or some other blood-sucking animal? Who knows?

Joan: Uh… yeah, I believe those things exist. But I thought you were asking about the monsters referred to in titles like Vampire Diaries, Buffy the Vampire Slayer, Interview with a Vampire. Stuff like that.

Earl: Oh, no, I don’t believe in those kinds of vampires. But I saw online that there are some people who like to drink blood and sleep in coffins, and others who are allergic to sunlight.

Joan: Those are humans.

Earl: They call themselves “vampires.”

Joan: Ok, I believe those humans exist and do that stuff and call themselves “vampires.” Why not?

Earl: So you believe in vampires!

Joan: I don’t believe in immortal, undead humanoids that catch fire in sunlight and literally must ingest blood to survive and have super-strength. But if you tell me you’re now calling tomatoes “vampires,” then I’d have to say I believe in the existence of vampires insomuch as they are the fruit we English speakers usually call “tomatoes.” Anyway, it sounds like you’re talking about the sorts of phenomena that inspired the creation of the fictional beings we usually mean by the word “vampire.” I believe in those phenomena, but not in the fictional beings themselves—not literally. See what I mean? And I know you also believe in those real phenomena (such as vampire bats), while you apparently don’t believe in the fictional beings we usually call “vampires.” Right? Right??

You also seem to be talking about people who are, in turn, inspired by and name themselves after, those fictional beings (e.g., the humans who sleep in coffins). We agree that the humans are real and that the fictional beings are not real.

So I’m not sure what it is we’re trying to figure out. Is it whether mosquitos and vampire bats and such are literally the mythical, nonexistent monsters we call “vampires,” or is it whether they inspired the imaginative creation of the mythical, nonexistent monsters we call “vampires”? Or is it simply whether mosquitos and vampire bats and such share enough salient attributes—e.g., feeding on blood—to merit being called literal “vampires”? Seems to me they don’t. For example, vampire bats aren’t immortal monsters who can only cross a privately owned threshold if invited to. Just feeding on blood and sleeping during the day doesn’t seem sufficient for meriting the label “vampire,” certainly not in any literal sense. Plus they’re cute little critters who practice reciprocal altruism. (Well, maybe that’s true of some fictional vampires as well.)

Earl: I don’t know. You seem too sure of yourself. How do you know vampires don’t exist? What if they turn out to be my shoes?

Joan: Ugh. This is making me appreciate a story my mom used to tell when she was frustrated with men who would contort language to save face: A man visiting a forest preserve urgently phones the park rangers’ office: “I’m up a tree with a hungry grizzly bear headed this way. Help!” The office prides itself in their forest’s zero-bear-attack track record. What to do? “Don’t worry,” says the ranger, “we’re now calling grizzly bears ‘squirrels.’ Everyone knows squirrels are harmless. Enjoy your stay.”

3. Three Forms of Disbelief (and a Note on Belief’s Passivity)

Recall that the point of discussing the denotation switch is to show that, if I don’t know what thing you mean to refer to by the word “God,” I can’t tell you what I believe about that thing. The point of discussing ways of disbelieving is to show that agnosticism about God is not a viable option for me, even if you can’t tell me much about what the word “God” denotes.

There are at least three ways to not believe something. I’ll call them Trivial Disbelief, Agnosticism, and Explicit Disbelief. I’ll give a brief account of these, followed by a note on the passivity of belief.

In a more rigorous discussion, I might formalize these with expressions like: ~[Bsp] (i.e., “It is not the case that the subject believes the proposition true”) and [Bs~p] (“The subject believes the proposition false”). That might be particularly useful for zeroing in on the difficulties in nailing down Agnosticism. I think informal descriptions are enough here.

3.1 Trivial Disbelief comes in at least two forms. I’ll call these Disbelief from Ignorance and Disbelief from Fact-of-the-Matter-less-ness. Awkward but descriptive.

3.1.2 Disbelief from Ignorance is when you don’t believe something simply because you have no beliefs at all about that thing. Before reading this, you only trivially (or “vacuously”) disbelieved in the god Flifketslÿpz Blurry Worm. I know this, because I just invented her. Now you explicitly disbelieve in her. And now you explicitly, rather than trivially, disbelieve that there are sixty million tiny green monkeys building a robot butler on your right shoulder. In short, you can’t explicitly disbelieve or reject an idea that never crosses your mind.

We might think of this form of disbelief as overlapping with dispositional belief, which I’ll define simply as that which you don’t yet explicitly believe, but are disposed to believe. This is often taken to mean something like: those beliefs entailed or implied by your existing system of beliefs. In which case, perhaps trivial disbelief would be best seen as a kind of subset of dispositional belief. I favor the overlap conception, however, for the following reasons.

I can ascribe trivial disbelief to someone who is not disposed to have any belief at all about a proposition. Confucius didn’t believe me to be an insincere person, as he had no beliefs at all about me; indeed, he was not—or at least certainly is currently not—disposed to have any beliefs about me. Perhaps similarly, it would be strange to say that a man on his death bed who never learned math beyond basic arithmetic is disposed to disbelieve that the first derivative of sinθ is –cosθ, though he would disbelieve that false claim trivially (notably, he would also trivially disbelieve the true claim that the actual derivative there is cosθ).

On the other hand, it may be instructive to consider a broader view of dispositional belief. For example, the dying man would have disbelieved the false claim about –cosθ with a bit more math training, or maybe merely even if his math-inclined niece told him it’s false, even if he didn’t understand her explanation. And I might be disposed to believe all sorts of strange things (true or false) with a bit more “training,” i.e., with the right switches flipped in my brain.

It’s not clear to me, then, whether dispositional belief relies primarily on what’s implied by one’s existing system of beliefs. The dying man’s present beliefs about basic math—e.g., he thinks 20 equals 0—may not entail proper calculus, but they would entail proper calculus with some small corrections and a little more training, even should he never be subject to those corrections and training. He’s also disposed to continue believing those wrong things with a likewise misinformed teacher who misadjusts his current beliefs.

Given such observations, while perhaps trivial belief is best thought of as falling within a theory of dispositional belief, even if as a sub-genre, I’ll for now consider them to merely partially overlap. In other words, I don’t think trivial belief should be thought of as primarily a kind of dormant or potential explicit belief (a little more on this below, when I discuss explicit belief). I also make this assumption given the second form of trivial disbelief I’ll consider, some aspects of which clearly involve explicit (and not just implicit or dispositional) disbelief.

Whether that assumption is correct or arbitrary is unimportant. Again, I don’t intend this to be a rigorous taxonomy of clearcut concepts, so much as an attempt to demonstrate that there are various ways to disbelieve and that this fact has implications for the “What do you mean by ‘God’?” discussion. My speculating about dispositional belief is likewise intended to illuminate my discussion of trivial disbelieve, rather than lay down rigid lines of distinction.

3.1.3 Disbelief from Fact-of-the-Matter-less-ness is a sort of trivial disbelief that arises when there’s simply no fact of the matter about a proposition. For example, you might think there’s no fact of the matter about whether vanilla or chocolate ice cream is objectively better for 21st-Century Humans (even though you far prefer vanilla). To be clear, you’re not agnostic about which flavor is better—you believe there’s no answer to the question, and therefore you trivially disbelieve that one is better than the other.

If you don’t like that example, how about something like: Do believe a typical adult giraffe to be taller than the dreams of a stone? Do you believe 100 U.S. pennies weigh more than the sourness of an unripe strawberry? Do you believe that 3 ° 2 = 7  (where the ° symbol is undefined)? And so on. You’d be technically correct to answer “no” to these questions, though “no” obviously doesn’t quite capture the nature of your disbelief.

That nature may be even more complicated than it at first seems. I don’t believe that the giraffe is taller than the dreams of a stone, and I explicitly believe that neither is taller than the other, as it is a category error to compare them; and I explicitly believe it false that the dreams of a stone (which have no height) are taller than the giraffe. And so on. Triviality allows for all sorts of beliefs and disbeliefs, including explicit ones and ones that might turn out to look more like agnosticism, given the difficulty of pinning down that concept, which I’ll now consider.

3.2 Agnosticism is the messiest of the three, particularly (I think) when it comes to God. For starters, recall the difficulty of discerning whether an agnostic is trying to determine whether God exists, versus trying to determine which known-to-exist thing merits bearing the label “God.” With that and related complications well in mind, I’ll try to sketch out an overview.

Here’s how I would generally characterize agnosticism:

A subject is agnostic is when aware of a proposition (e.g., “God exists”) while being undecided about that proposition’s truth, for whatever reason (with some qualification, more about which below); such reasons are usually either the presence of equally strong evidence for the proposition’s truth and falsity, or a lack of (persuasive) evidence entirely.

Here’s a succinctly stated implication of this characterization, in the “God” context: An agnostic is someone for whom the denotation of the word “God” has been at least roughly established, and yet for whom “Do you believe in God? is not a yes-or-no question.

The above-noted “awareness of the proposition” criterion differentiates agnosticism from trivial disbelief. Otherwise, you could simultaneously trivially disbelieve a proposition and be trivially agnostic (i.e., be undecided) about it, simply because you’ve never heard the proposition, even if there is some easily discernible fact of the matter about the proposition’s truth. In general, unawareness of the proposition is the most important distinction between trivial disbelief and the other two forms of disbelief. Note also that this accounts for the aforementioned “qualification” about being undecided for “whatever reason”: the reason cannot be mere ignorance of the proposition.

A subtler, but also important, observation is that, if compelled to give an answer about an undecided proposition, an agnostic would express what looks like disbelief. For example, forced to answer “yes” or “no” to, “Do you believe God exists?,” an honest agnostic would say “no.” The problem with this is that the answer is also “no” to, “Do you believe God does not exist?” Again, “Do you believe in God?” is not a yes-or-no question for the agnostic.

Suppose we reformulate the question into a set of true-or-false statements: “God exists” and “God doesn’t exist.” The agnostic cannot be forced to truthfully respond in this case. What strikes me as especially interesting here is that the sincere answer to “Do you believe God exists?” is “no,” while the sincere response to “God exists” cannot be “false.” This gets at a defining feature of agnosticism that, I think, shows the answer “no” in the “Do you believe God exists?” case to be independently uninformative**—the “no” there being a kind of trivial disbelief that is ruled out not only by awareness of the proposition, but by the fact that the agnostic does not believe the assertion “God exists” is false. Any good account of agnosticism should make sense of such tensions.

[**I have heard folks argue, and have myself in the past argued, that agnosticism must really amount to atheism on the grounds that an honest agnostic must answer “no” to “Do you believe in God?” This argument is at least logically misguided. How God feels about agnosticism, should She exist, is another question.]

That said, with a reminder that my thoughts here constitute a sketch and not a sculpture, I submit three joint criteria for being considered agnostic about some proposition p (e.g., “God exists”):

[1] it’s not the case that you believe p true;
[2] it’s not the case that you believe p false;
[3] you are aware of [1] and [2] (which entails being aware of p).

The agnostic, then, is aware of not believing p to be true or false. Here again, this distinguishes agnosticism from trivial disbelief. (I think it also distinguishes explicit disbelief [see the next sub-entry] from trivial disbelief, but I take this to be a more controversial claim, as there are interesting arguments to make about belief being at least sometimes principally behavioral, so that examples of explicit belief could emerge without a person—or some nonhuman animals and even some non-conscious entities—being aware of having the belief.)

I think these three criteria jointly necessary, but I don’t consider them sufficient. For example, it seems to me that agnosticism can’t emerge from mere intractability, though I struggle to think of a good example. Indeed, my attempts to do so make me wonder if intractability isn’t exactly what agnosticism is usually about.***

[***By “about,” I don’t mean that I’m trying to define the word once and for all, nor to come up with jointly necessary and sufficient, universally applicable conditions for agnosticism. Rather, I’m trying to get a handle on some of the most basic criteria that most of us can agree on as users of the word or, at least, of the epistemic concept or semantic category we take, in standard contemporary usage, the word to refer to or establish.

Part of the difficulty for me here is that the word “agnostic” raises a dilemma similar to that raised by the word “knowledge”: In defining the word, are we attempting to establish a semantic category that draws a border around some arbitrarily designated bundle of behaviors and cognitive states (i.e., certain flavors of belief and disbelief)? Or, rather, are we attempting to un-arbitrarily pick out some independently existing, singular phenomenon? To be clear, the aforementioned behaviors and cognitive states will exist whether or not we draw the border around them. An important question here is whether the border is arbitrarily drawn, so that “agnosticism” is whatever we want it to be, or whether the bundle is naturally bordered off that way, so that there is some singular, independently existing phenomenon picked out by the word “agnosticism.” The difficulty here is in separating the word “agnosticism” from the in-the-world phenomena the word does, should, or is thought to denote.

I’ll say more about these sorts of difficulties in Part Two, but I belabor the point here to show that there is yet another word I need to know the meaning of before I can say how it relates to me: “agnosticism.” For the z-agnostic, at least. Again, I do think that, for most folks, the word is fairly well understood.

Now to return to the question of what role intractability plays in agnosticism.]

How many human-sprung sneezes where there in Vermont on March 4th, 1891? There is some fact of the matter (given some reasonable operational definition of “human” and “sneeze” and “in Vermont” and “March 4th, 1891”), but good luck finding it. Still, there are more or less believable candidate answers. Suppose someone proposes answers of precisely zero or 20 quintillion. I’ll confidently reject those candidates. But given even just a relatively narrow range of plausible estimates (along with facts about the world population at the time, etc.), I would assume the estimate to lie in that range, but wouldn’t be able to form the explicit belief that any given number in that range is true or false. This doesn’t strike me as agnosticism, however, because I do believe that one of the numbers is correct.

Replace “sneezes” here with something that can be counted, like “coffee beans in the bag.” Is this an analogous example, given that one need only count the beans to find the answer? Or does the question need to be intractable to be analogous,: “coffee beans in the bag on the ship that sank in 1752” or “coffee beans in the bag in the safe whose combination is too complex to crack”?

It seems to me that none of this counts as agnosticism.

Similarly, were I to look at a word problem on a multiple choice test in a language I don’t know, I’ll believe that one of the multiple choice options is true while the rest are false, though I can’t decide which. Am I agnostic about each answer choice? Or do I simply not know the answer to a question that it may turn out any 8-year-old student who speaks that language could answer? (I don’t have to understand the problem, by the way, to form a conclusion about the answer—e.g., I could look at the answer key and form a belief, even if the belief is wrong because I’m looking at the wrong answer key.) This isn’t an intractable problem in the way that, say, counting the number of relations in the universe is, but it gets at the possibility of intractability being a condition of one’s relative perspective.

I find these examples disanalogous to the question of God’s existence, because what we are trying to determine in the God case is the broader fact of the matter of whether God exists. Above, I assume some positive number of people sneezed, some quantity of coffee beens is in the bag, and one answer choice is true and the rest false. These are easily answered questions in principle, though circumstances may make them impossible to answer in practice.

On the other hand, maybe this is no different than the God question after all: “How many deities are there in the world?,” where “deity” may be defined in such a way that options are zero or one or are zero to some arbitrarily large number. Whatever the case, I need a definition—some criteria—in order to determine the number of elements in the set, “deities in the world.”

I an’t settle the question here of what agnosticism is or should be—more specifically, of the role played by intractability, as I’m using that term here—but I think it’s useful to think a little more about the question, particularly in the God context.

The agnostic considers a broad question: Does God exist? I presume the agnostic would not say, “I believe that some known conception of God describes, more or less accurately (insomuch as is humanly possible), a being that exists, and I believe that existing being to be God; but I cannot say whether I believe God exists.” More simply put, this is tantamount to saying, “I believe God exists, I just don’t know which of the many plausible descriptions of God is correct.”

For one thing, note that this might force us to count as agnostic those believers-in-God who think humans can’t know God’s nature (see again Kierkegaard). The agnostic, rather, believes that one of these statements is true: “God exists,” “God doesn’t exist”; but isn’t sure which is true (for whatever reason).

On the other hand, this seems to be just what many z-agnostics claim when they say things like, “What if God is just energy?” Again, they seem to be trying to figure out which existing phenomenon merits application of the word God, rather than trying to decide whether they believe God exists. This is, again, where I find myself most confused by the z-agnostic, who seems to be saying things like, “I believe in God, I’m just not sure which thing in the world is God” or “I believe energy exists, I’m just not sure if energy is God.”

Or most confusing of all: “I believe energy exists. I believe that energy to be God. I am an agnostic about the existence of God.”

All of this goes against what I’d expect an agnostic to say, which, again, is something more like, “I believe some plausible conception of God might be true, though I can’t decide (for whatever reason) if it is true,” where there may be a family of plausible conceptions that are significantly similar—e.g., it could include all Abrahamic conceptions of God while excluding, say, Norse-historical and Spinozistic conceptions. Notice that this is not the same as saying, “I believe some conception of God is true, but I’m not sure which,” where the competing conceptions could be, again, in some family (e.g., of Abrahamic accounts); I would, again, consider this theism, not agnosticism, as it really is in line with something like, “I believe God exists, but am not entirely sure of God’s nature.” This seems to be precisely what many z-agnostics are saying when they say things like, “I believe God exists, but am not sure of whether to characterize it as quantum phenomena, energy, love, or all of the above.”

The z-agnostic just seems to be (in this case) a theist with a certain range of viable conceptions of God—a range that excludes, for example, the deities of Norse and Abrahamic histories, but that includes something like the Spinozistic conception (see again the famous Einstein quote: “I believe in Spinoza’s God…,” more about which in Part Two).

Of course, viable conceptions are often narrow, as determined by the agnostic’s position in a given culture at a particular point in history. I assume most North Americans don’t give much thought to the many mystical creatures believed to exist in other parts of the world (e.g., elves in Iceland). This same principle extends, of course, to those who are agnostic within a given subculture (e.g., depending on the religious beliefs of their childhood household community). I’m not agnostic about entities I’ve never heard of, not even trivially so (that is clearly ruled out by the most uncontroversial conditions for agnosticism).

Taking in the discussion of viability and intractability and so on, further criteria for agnosticism suggest themselves, such as: [4] You are aware of viable alternatives and supporting reasons thereof, relevant to whether p (itself a viable belief for you) is true or false.

Perhaps, then, I’m misconstruing the examples of intractability I’ve chosen. Maybe “agnostic” is the right word in the sneezers and multiple choice cases, and I’ve just chosen bad examples for bearing out my intuition. For example, I’m not agnostic about 20 quintillion as the number of sneezes, as this is not a viable belief for me, but I might be agnostic about whether the figure is 60,112 (given a 1900 population of 343,641 and that allergy season doesn’t start ramping up in Vermont until April and so on). With more information, this becomes a not-so-bad Fermi problem.

We can’t produce an exact number, but this doesn’t in itself preclude agnosticism—recall that one common reason for not explicitly believing p true or false could be that the answer simply can’t be known. Perhaps this should be that it can’t be known in principle. But that might be going to far. (Also, it’s interesting to note that we can know God’s existence in principle, for example by going to Heaven after death; but cannot, after death, learn of God’s absence.) The other common reason for agnosticism is having equally compelling evidence for the truth and falsity of the proposition.

This suggests another sort of example of intractability, regarding my belief about this fair quarter I’m about to flip. I believe it will land Heads or Tails, but I don’t have an explicit belief which of those it will be, nor am I agnostic about which it will be. At least, I don’t think I am. With enough of the right kind of information, I might be able to form a belief, but as it stands, my evidence is equally persuasive for those two outcomes. So the most I can believe is that it’s 50-50. That in itself is a proposition that one can believe or reserve judgment about (most coins are probably not exactly fair).

This now strikes me as a bad example, because it’s all too easy to find the answer: just flip the coin. Perhaps an even better example would be to invoke a subjunctive conditional: Had I flipped this fair coin (that I in fact decided not to flip) one minute ago, how would it have landed? I don’t believe “Heads” here, nor do I believe “Tails”; but I do believe “Heads or Tails.” Is this agnosticism? I don’t think so, but maybe it is if we believe there is some fact-of-the-matter answer to that question. I take it that there is only a fact-of-the-matter answer if it is possible, at least theoretically, to answer the question. I don’t know if this is possible.

Using a 50-50 example might be too easy, because it’s easy to say “As a good Bayesian, I believe Heads and Tails to equal degrees.” When playing a lottery with a one-in-a-million shot, I don’t believe I’ll win; more to the point, I disbelieve that I’ll win. Even more to the point, suppose it’s a pick-five with 510 (i.e., 9,765,625) possible permutations: I do not—and absolutely cannot—explicitly believe all at once that each permutation will will to a degree of 5-10, but I can certainly claim that this is my belief if you ask me, “Do you believe 73648 will win?” I can confidently say I believe it won’t. But I can also confidently say that one of the numbers on this list will win. What I cannot truthfully say is that I believe to a degree of 5-10 (that is: .0000001024) that the number 73648 will win.

Compare this to the comparatively simpler example of rolling a fair die. I don’t believe the die will land 6, but I don’t disbelieve this. But, again, am I agnostic, or just waiting to see what happens, as there is some very easily discoverable fact of the matter, which is discovered by simply rolling the die?

Here are two more examples of intractability.

I dump two apparently identical bottles of the same brand of aspirin into a large bowl, so that there are now n pills in the bowl. I am then asked to redistribute each pill back to its original bottle. I have no way to figure this out. Am I agnostic about each pill’s origin? Or is it 50-50 for each, and so n2 for any given final redistribution of pills into the bottles?

What if I only had mixed together one pill from Bottle A and 1000 pills from Bottle B? My bet for each pill would be that it came from Bottle B, but I couldn’t be sure. Am I agnostic about each pill’s origin?

Suppose I come across a puddle of water that I know to be from a melted ice sculpture of some famous human (past or present). I have no idea which human, but could certainly guess. Am I agnostic about the shape, or just clueless? I think clueless.

I could go all day with these, though I’m not sure I could easily come up with one that is clearly not an instance of agnosticism. Nor am I sure any of these examples are illuminating or if this is even a useful exercise. (Maybe if I had an editor, I’d be told, “delete 90% of this section.”) I think it is useful, however, at least for demonstrating the sincere difficulty I have in applying this category of belief to myself in relation to an unspecified application of the word “God.”

I’d like to again make clear that I’m most confused about the z-agnostic stance; it is this stance that requires a deeper look at what is meant by the belief category “agnostic.” In general, though, I, like most people, do understand what most people mean, more or less, when they say “I’m an agnostic about God.” Similarly, I understand how someone could be agnostic about things like the many-worlds interpretation of quantum mechanics or views on mereology/composition (e.g., a position recommended by Rosen & Dorr in their 2000 draft “Composition as a Fiction,” which I touch on here) or scientific realism or even whether we live in a computer simulation. (Though I think when scientists and philosophers and the like say “I believe in theory X,” I think what they most often really mean is, “I accept theory X as the most plausible working interpretation of the evidence and arguments”—though what “believe” means in such statements may depend on whom the speaker is talking to [e.g., colleagues, physicist groupies] and how many beers have been consumed.)

I also take it that I can be agnostic about some complicated idea because I simply haven’t thought enough about it and wish to not pretend like I have a well-founded opinion, such as: whether a given description of black holes is correct; which competing view on whether nonhuman primates have theory of mind is correct; whether memes—or genes for that matter—are best thought of as metaphors, broad semantic categories, or literal and independently existing things in themselves; which historian’s account of why an 11th century Japanese village had this or that code of etiquette is correct (well, this might simply be too hard, as it asks us to read the mind of long-dead people); and on and on. With many such things, I simply haven’t been moved to do the hard work of forming an opinion, and I’d rather not pretend to have an opinion, and I’d rather not skimp on the hard work needed to formulate, despite being asked to weigh in on them in on them in the first 15 minutes of learning about them in an undergraduate college lecture.

Perhaps this, though, is what the term “suspending judgment” is for, while “agnosticism” is best reserved for after one has thought hard or at least for a good while about a topic and has weighed some of the viable options. Criterion [4] that I added above is looking more and more appealing.

In fact, with many such things, agnosticism, rather than increased certainty, might be the best response to learning more about a proposition—see again the Rosen & Dorr paper on mereology; they don’t recommend agnosticism to someone who hasn’t thought about mereology, but rather to those who’ve thought about it a lot.

I actually am starting to see some coherent picture of agnosticism come into focus here. But for today’s purposes, I think I’ve given an adequately—though by no means thoroughly—messy account of agnosticism, while allowing room for instances of more or less clear usage. That in mind, I’ll wrap up this sub-section by pointing out that what’s thought of as agnosticism might sometimes be better thought of as an explicit belief of the following sort: The proposition could turn out to be true or false. Compare this, and my four conditions, with the Stanford Encyclopedia of Philosophy‘s observation:

“Nowadays, the term ‘agnostic’ is often used (when the issue is God’s existence) to refer to those who follow the recommendation expressed in the conclusion of [T.H.] Huxley’s argument: an agnostic is a person who has entertained the proposition that there is a God but believes neither that it is true nor that it is false.2

3.3 Explicit Disbelief is the easiest category here. Explicit disbelief about a proposition means you are aware of the proposition and you believe the proposition to be false. This is where atheism falls, or at least where I think it should be generally understood to fall in our current era.

Notice that in order to have such a belief about God you need to know what “God” denotes, even if as a baseline or core concept. In other words, you don’t need to know a whole lot about God’s “true” nature—I can tell you whether or not I believe George Washington existed without knowing much about his “true” nature.

Notice also that the word “explicit” distinguishes this category from the phenomenon of believing something without knowing you believe it. For example, you might, without noticing, consistently perform behaviors that are in line with your felt attitudes and that affirm the truth of a proposition, but without noticing; if brought to your attention, however, you might then become aware of your belief so that it can be expressed or felt or conceived of etc. explicitly rather than implicitly.

There may also be an interesting discussion to be had here about “dispositional belief,” namely involving a contrast “dispositional” with “occurrent,” where “occurrent” means something like “actively present in one’s conscious awareness.” In other words, if we view “occurrent” and “explicit” as essentially synonymous. I don’t think these should be viewed as synonymous, but I’m not going to get into this here, nor into other potential ways of parsing out or conceiving of explicit disbelief. I think it’s clear enough what is meant by the idea, though there may be room for discussion about how it relates to other notions about disbelief mentioned here.

3.4 Believe Is Often Passive — and Some Background on My Beliefs

Belief is often something that happens to you rather than something you choose, despite common ideas about what one “chooses to believe” (though I do think there is something to the notion of willful—i.e., self-imposed—ignorance, more about which in a moment). This, by the way, is not a claim about free will. Even if you believe in free will, there are plenty of behaviors and cognitions you know you can’t control. You can’t force yourself, by sheer exertion of will, to believe you were born in 1700.

While God’s existence is obviously more complicated than that, I simply can’t help but not believe in God, even if by “God” we mean “energy” (I believe energy exists, but I’m not yet persuaded that it merits the label “God,” even in 2019).

An objection to this might be that I’m willfully ignorant. Which is to imply that were I to thoughtfully consider certain evidence, I would come to believe in God (much as when Marx, after holing himself away with a stack of writings to affirm his disapproval of Hegel, emerged a convert; see also the below passage from van Inwagen). Moreover, goes the accusation, I don’t want to come to believe in God, so I willfully avoid such evidence. I think this is wrong, as it presumes I believe the evidence might be convincing enough to persuade me. I believe no such thing. There’s much more to explore about willful ignorance, which I’ll do another day, but that question is moot in my case. I was raised to be a believer and I wanted to be a believer.

What follows is a personal note that I hope supports the claim that my way of engaging with discussions about God is reasonable, while also revealing some of my biases. Which is to say that I cannot talk about others’ motivation for asking “What do you mean by ‘God’?” I can only speak for myself.

Growing up, I was expected to believe in various accounts of the Christian God. I was expected to read the Old and New Testaments, sometimes with parallel translations that explained the philology and such of the original text, with the aim of losing less in translation. I at times color coded with highlighters and I went to churches both big and living-room-based (and witnessed exorcisms at both) and I was moved to speak in tongues (just as the adults around me were) and I distinctly recall being an 11 year old in Meridian, Mississippi declaring my intention of being a preacher when I grew up (though I also seem to recall this goal really being about showing off my color-coded scholarship).

As I got older, my experience of those texts and the world stirred doubt, which I dealt with by seeking out writings claiming to reconcile, for example, the Book of Genesis with the Big Bang and accounts of the Christian God with the language of other religions. I tried to take on board such reconciliations—tried, for instance, to take on board a pluralistic view of God, in which all the world’s religions are, deep down, talking about the same fundamental phenomenon that monotheists call “God.” Such efforts only made things worse.

And so I read and read and listened to the adults debating and pontificating and all of this only stoked my doubts so the more I learned the less sense it made and the more I doubted—much in the way erecting roads and bridges breads, rather than relieves, traffic.

Eventually, belief was off the table for me. Though I could lie and say I believed. I lied often, for a while. When I stopped lying and reported my disbelief, I was accused by some of lying about that. Trust me: if I truly thought there was even the slightest chance I might end up going to Hell—of being subjected to the eternal torture that was the relentlessly cited consequence of disbelieving in the version of God I was raised to fear, with vivid warnings like, “it’ll be as when putting your hand on a stove’s hot coil but the nerves in your hand will never die”—I’d do as Pascal suggests and tend best I could to that little bit of credence in the desperate hope of making it grow.

It turned out that the soil of my mind was unfriendly to such beliefs. Still, I for years prayed each night and I pretended to believe just in case the terrifying, Hell-creating, eternally torturing, all-loving Almighty was in fact watching and might be moved to show pity on one of His creations who was trying hard to kill his doubt. But belief was eventually off the table for me, and the fear went with it.

(A fear that has always made me suspicious of self-proclaimed Christian agnostics, especially those who think the question of God’s existence literally unanswerable. How can this be, given that embedded in the idea of that God is that one believes by faith rather than with rational or empirical certainty? If such faith is literally unattainable—even the tiny bit of faith it would take to instill real fear—it must be that such a God does not exist. Right? It seems to me that Christian agnosticism must involve enough faith—must involve enough belief in the possibility of God’s existence—to sustain such fear. No? Couple this with the fact that I can understand, in a society like ours, pretending to be an agnostic rather than admitting to atheism. At least, though, I have a rough idea of what such agnostics mean by “God,” though, again, they themselves might be unsure of what the precise nature of that entity is—which may in itself be a source of the person’s agnosticism. I explore such questions in Part Two. For now, I assume most agnostics, of whatever sort, are sincere.)

In high school, I found myself in a less religious environment that allowed me to look in new places for ways to reconstruct my model of the world into something I could deal with. I hung out at the local New Age bookshop and read about transcendental meditation (which I practiced), astral projection (tried and failed to make it happen), lucid dreaming (managed to have precisely one), and the occasional tie-in to Christianity (e.g., about atheistic scientists converted to Christianity by their failed attempts to prove the Shroud of Turin unmiraculous).

I so wanted this stuff to be about something more than material reality. I didn’t know anything about academic philosophy or the history of ideas, but perusing libraries and bookstores felt like it was getting me closer to the sorts of things I should be contemplating. At about 16, I read Alan Watts’ The Way of Zen (originally published in 1957) and was greatly moved. I remember sitting in math class reading that book and getting goosebumps (instead of doing what I usually did in class—composing chamber music or trying to memorize the words to the opera “Carmen” or some old Zarzuela tune, which I’d also discovered at the library; but these are other stories).

In the years after high school, I turned more and more to music and film and literature and so on, and eventually gave only sporadic thought to God (the last push I can remember was at 18, during my roughest patch in the military, when I turned to the only book I had on hand, a small Gideon Bible; though this, too, is quite another story).

I continued to be as contemplative as ever, to overthink things and to look for and try to understand the world’s hidden connections and so on, but it would be nearly 20 more years before I began to delve into what we call “Western” philosophy and into psychology and mathematics and such, and before, in fact, I would finally attend college, where I majored in philosophy. I took courses on the world’s religions and philosophy of religion and I learned a tremendous lot in those classes from engagement with writings and in discussions with classmates from a variety of viewpoints (i.e., with practitioners of a wide range of religions, with agnostics, and, less commonly, with fellow atheists), and I continue to learn and to reflect, but my disbelief in God has not been challenged.

Now, with quite probably more years of consciousness behind me than ahead of me, I do not believe in any conception of God or gods of which I am aware. But I’m open to suggestions, but this requires telling me roughly what is meant by the word “God.” Who knows, maybe I’ll look back on these words 20 years from now and wonder at my spiritual naiveté. (Hi future me!)

3.4 Conclusion to Section 3: The Point of Discussing Disbelief

How does this discussion of disbelief bear on the question “What do you mean by ‘God’?”

Agnosticism is currently not a viable option for me, though I have to know what you mean by “God” to be sure. For example, I might be agnostic about whether the computer simulation we’re living in contains some features so functionally similar to traditional conceptions of God that the collection of those features merits the label “God.” (This, again, is still not a straightforward case, as there would exist the critical problems of this “God” not pre-dating the simulation, thereby not being an unmoved mover, and of the impossibility of miracles [i.e., violating the laws of physics, whatever they may be, that limit what the program can do].)

So far, then, my beliefs about things denoted by “God” either come down to trivial or explicit disbelief, depending on what I’m being asked. That is, I explicitly disbelieve in anything like the God described in monotheistic texts. And when I’m asked if the word “God”—with all its historical and cultural attachments—is most appropriately used as a reference to quantum physics, love, life, mathematics, pure existence, consciousness, energy, imagination, dancing, the experience of music, or whatever a zealous agnostic happens to find interesting or moving or meaningful at a given moment—I may explicitly believe in, or at least be willing to take an expert’s word about, the existence of those things; but I also explicitly disbelieve that they are God.

And I trivially disbelieve descriptions of God that have never been introduced to my attention, or whose content is so vague that I can form no concept at all—e.g., “There must be something out there that is God, but we can’t say what.” We all trivially disbelieve the concepts that never have and never will occur to anyone. That’s a lot of concepts.

Finally, my disbelief in God isn’t a choice. At least, it certainly doesn’t feel like one. Though I suppose that, in some robust sense, I do choose to accept the label “atheist” and to behave like one (e.g., by writing this post). I consider this acceptance to be a basic and honest reflection of the fact that the word does indeed apply to me, but I admit that it is a choice to accept the word as a label. (Don’t worry, my free will skepticism is compatible with this observation. One way to see this is to notice that if the consequence of such an admission were the execution-by-fire of me and my loved ones, I seriously doubt I’d have the courage to make the admission. In fact, I need not go that far: if I were running for political office, I might not feel free to make the admission.)

Also a choice is the performance of the denotation switch, particularly when the switch is explicitly called out as such (as I’m doing now).

4. Closing Thoughts

I hope I have convinced you that my question,“What do you mean by ‘God’?” is a reasonable one. If not, I hope I’ve at least convinced you that I believe it to be reasonable, and am not being disingenuous, smug, or evasive in posing the question.

If you are a zealous agnostic and can grant me this, I’ll grant in return that, even though you and I might might share nearly identical beliefs about the physical world, it is reasonable for you, for whatever reason, to identify as “agnostic” while I nonchalantly accept the label “atheist.” In other words, maybe it is possible to be a z-agnostic without performing the denotation switch, particularly in such a way aimed at forcing me into a z-agnostic corner.

In other words, I hope that, if you’re a z-agnostic, you’ll view my question “What do you mean by ‘God’?” as an invitation to converse—and to work on better understanding one another—rather than as fighting words.

That in mind, my point here has not been to convert anyone to any particular belief about God. My concern, rather, is with how agnostics, especially z-agnostics, and atheists discuss the question of God’s existence. How believers might enter into the conversation is another matter, and one I’ve touched on, but inadequately.

To make up for this a little, I’d like stress that I do not hold it against those who would aim to convert me to theism. I can see why some believers would make a huge effort to guide their neighbors away from the dire consequences of disbelief. If I sincerely believed you were on a path to eternal Hell fire, I’d be in a constant panic to save your soul. I also understand that such efforts are often obligatory, in fact. But how to convert with words?

I’ve often noted that my favorite argument for God goes something like this: Words don’t convince me; rather, I look at the world around me and I look inside myself and, mysteriously, I am filled with faith. I recently encountered a wonderful expression of something like this from Peter van Inwagen, a philosopher I greatly admire. The following passage shows up in at least two collections, including this one, in which he follows the passage with additional commentary and speculation: God and the Philosophers: The Reconciliation of Faith and Reason (1994), Chapter 3, titled “Quam Dialecta” (page 41):

I am a convert. For the first forty years of my life I was outside the Church. For much of my life, what I believed about the Church was a mixture of fact and hostile invention, some of it asinine and some of it quite clever. Eventually, I entered the Church, an act that involved assenting to certain propositions. I believe that I had, and still have, good reasons for assenting to those propositions, although I am not sure what those reasons are. Does that sound odd? It should not. I mean this. I am inclined to think that my reasons for assenting to those propositions could be written down in a few pages—that I could actually do this. But I know that if I did, there would be many non-Christians, people just as intelligent as I am, who would be willing to accept without reservation everything I had written down, and who would yet remain what they had been: untroubled agnostics, aggressive atheists, pious Muslims, or whatever. And there are many who would say that this shows that what I had written down could not really constitute good reasons for assenting to those propositions. If it did (so the objection would run), reading what I had written on those pages would convert intelligent agnostics, atheists, and Muslims to Christianity—or would at least force them into a state of doublethink or intellectual crisis or cognitive dissonance. Perhaps that’s right. If it is, then among my reasons there must be some that can’t be communicated—or I lack the skill to communicate them—, like my reasons for believing that Jane is angry: something about the corners of her mouth and the pitch of her voice, which I can’t put into words.

Maybe there is a similar experience that could find its expression with certain agnostics who do not believe in traditional conceptions of God, and so are not sure where to put the wonder—the faith—that fills them when they look at the world and inside themselves. They tend to be rational beings, students of science and philosophy and so on (not to mention art), and so think they can give some rational, persuasive argument for their z-agnosticism. It strikes me that some version of what van Inwagen expresses above would be more appropriate. I’ll be sure to raise this question the next time a z-agnostic presses me.

That does it for Part One of this entry. In Part Two I’ll share further notes on the denotation switch.

In that spirit, I close here with the first denotation switch I remember experiencing. One day, while readying the floor at a fine Catalan restaurant at which I was an ayudante de camarero, I put into the house stereo a tape from the popular Le Mystère des Voix Bulgares series (this would have been about 1989, the year the second volume of the series won a grammy). A chef bolted over to turn it off. He thought it was terrible. I said, “But you told me you love all music.” He said, “I do love all music! But this… is not music.”

Enjoy or find this post useful? Please consider pitching in a dollar or three to help me do a better job of populating this website with worthwhile words and music. Let me know what you'd like to see more of while you're at it. Transaction handled by PayPal.
Or click the banner to shop at Amazon (at no extra cost: it just gives me some of what would have gone to Amazon).

Further Reading


  1. This is one installment of a four-part The Philosopher Zone series from 2011, commemorating David Hume’s 300th birthday: “Hume and God.”
  2. From SEP’s “Atheism and Agnosticism”; Section 2: Definitions of “Agnosticism”—where you’ll also find a brief discussion of the word.

Share your thoughts:

Deprecated: Directive 'allow_url_include' is deprecated in Unknown on line 0