The Evidence for Hinduism

defending the sanatana dharma

Me in this blog post

---

Last week, I stumbled across Jonardon Ganeri—one of the foremost philosophers of the Indian tradition—politely discussing Hinduism with a Christian street apologist. I googled Ganeri just to make sure it was him, and then sidled over to ask if I could listen in. Ganeri explained that the apologist had just asked him to define Hinduism, and asked me how I’d define it.

I started trying to list necessary conditions on Hindu belief (ridiculous of me) when the apologist saved me at “reincarnation” and proceeded to explain why reincarnation was “illogical”, on the grounds that it just keeps going, and “you never get off the bus”. I pointed out that most Hindus believe in moksha, which says precisely that you do get off the bus. He then asked what the evidence for karma was, to which I suggested that if you believe in retributive desert (a point I’ll discuss in section IV), then karma is straightforwardly predicted by theism, since you’d predict a just God to give people their just deserts, whether in this life or the next.

Ganeri had to go soon after that, but I stayed with the man for almost two hours, going back on forth on a range of issues. (Gay sex, it turns out, is wrong because it causes anal cancer, violates the law of non-contradiction, and involves “putting something clean in something dirty”, which is “an expression of hatred towards the other man”). Since Ganeri wasn’t there, I spent most of the time steelmanning Hinduism in his absence.

The discussion got me thinking, and I decided to write an updated Bayesian case for Hinduism, pun intended. Here is why I genuinely think Hinduism might be the most probable religion, and a live religious option for me going forward.

---

To get the ball rolling, here are two facts about Hinduism that weakly count in its favour (stronger stuff is coming, I promise):

1. It’s big. Suppose the true religion is Wollenism, a religious creed according to which God wants you to like my article and subscribe to Going Awol. Conditional on Wollenism being true and its God being loving, would you expect Wollenism to be popular or unpopular? Popular, right? If Wollenism is the fullness of the truth (I submit that it’s at least part of the truth), and God wants us to live in the fullness of the truth—which seems natural if he loves us—then from the armchair, you’d expect Wollenism to be pretty popular. In contrast, if all you knew was that Wollenism were false (I assure you: this is a counterfactual we’re talking about), then given the infinite range of possible religious beliefs people could have, you’d expect no one believe it—let alone that it would be popular. The general lesson is: the more widespread a religion is, the more evidence you have for its truth. Not decisive evidence, granted, but evidence. So the fact that Hinduism is the third largest religion—with 15% of the world’s population, and 1.2 billion members—sets it alongside Christianity and Islam as one of the big contenders.

2. It’s old: A similar logic applies here, but this time, Hinduism gets the gold not the bronze. Hinduism is the oldest living religion. This is likewise a point in its favour. On his website, Christian astrophysicist Hugh Ross quips: “That Hinduism is presently the oldest major religion does not make it true any more than prostitution being the oldest profession makes it honorable.” But the argument here isn’t that Hinduism being the oldest living religion makes Hinduism true, or proves it to be true; the claim is that its age is evidence for its truth, since a priori, you’d expect God’s truth to be available for more of history than for less of history, all else equal. The same thought doesn’t map on to prostitution, since there’s no reason to expect ethical trades to be practiced for more of history than for less of it.

But these are small potatoes. What gets me about Hinduism is that it’s right—or plausibly right—about so many philosophical questions. I’ll start with the most important.

I. From Theism to Henotheism

To grease the wheels for me, suppose that God exists—where ‘God’ is defined a perfect person, maximal in power, knowledge, goodness, and so forth. (Note: the argument I’m about to present [which was invented by Matthew Adelstein, and which we've co-authored a currently unpublished paper about], is intended to work as an argument for the existence of God. If you’re not convinced of theism already, check out Matthew’s blog post on it, where he sums up the argument nicely. To save you some time, I’m presenting a trimmed-down version of the argument, where I’ll assume theism, and then try to derive from it a form of ‘henotheism’: the view on which a capital-G God exists who creates [or emanates, or incarnates as, or causes] a vast plurality of lower case-g gods—a standard claim in the Hindu tradition.)

Here it is, warts and all¹: If God has created all possible people, that makes the probability of your existence 1., since you are a possible person. In contrast, if God has created fewer people than that, the probability that you would exist is less than 1. Since you do exist, you should update in favour of the view on which God has created all possible people. But if God has created all possible people, then not only has he created infinitely many people like you and I; he’s also created infinitely many persons who are sufficiently powerful, good, and wise, that they satisfy the lower-case ‘g’ god-concept.

To see the intuitive motivation, consider three rival theories about how many people God would create:

Theory #1: God creates no people.

Theory #2: God creates 12 billion people.

Theory #3: God creates all possible people.

Conditional on the fact that you exist, which theory do you have the most evidence for, all else equal?

Well, given that you exist, Theory #1 is off the table. If theism is true and you exist, then, given some widely held assumptions about theism vis a vis God being a creator, God must have created you.

Theory #2 does slightly better. On Theory #1, you had no chance of existing. On Theory #2, you had some. Given that you exist, then, you should update in favour of Theory #2 over Theory #1.

How about Theory #3? Well, as long as the class of all possible people is greater than a measly 12 billion (and note: I mean possible in the logical/metaphysical sense, not the physical/nomological sense), it seems intuitively that Theory #3 should be preferred, given that you exist. Why, you ask? Because if all possible people are created—and if you’re a possible person, which you are—then the probability of you in particular existing would be 1., which is higher than the probability of your existence on Theory #2. Hence, you should update in favour of Theory #3 over Theory #2.

From the fact that you exist, the theory you have most evidence for on anthropic grounds is Theory #3: that God creates all possible people. On the most plausible reading of this claim, say I, we should think that God has created a vast pantheon of minor gods, or devas.

Here are two ways of stating the claim that God creates all possible people:

1. God creates all the possible people who you yourself could be, but not every other possible creature.

2. God creates all the possible people who you yourself could be, plus every other possible person.

Both 1 and 2 are equally well supported by the anthropic reasoning I just employed. As such, both theories fit the anthropic data sublimely. Are there any grounds on which to prefer one over the other?

There are. While both theories fit the anthropic data equally well, (2) has a higher prior probability than (1), because it’s a more plausible prediction of theism. Think about it: if all you knew was that (a) there’s a God, and (b) he’ll create people, what reason would one have to predict that, in all likelihood, God would create all (and only) the kind of possible people who you, in particular, might be? Nearly none. The prior probability of such a theory is titchy.

On the other hand, the prior probability of the theory that God would create all possible people (or perhaps I should say: all creatures great and small, since many creatures—squirrels, for instance—don’t satisfy the concept of a ‘person’), is decently high. After all: it’s good to create a person if you can give them a good life overall. Since God is unsurpassably morally great (I take that to be a core divine attribute), it seems reasonably likely that he’d take an unsurpassably good action and great all possible people (of which, plausibly, there are unsurpassably many), and give each of them a good life overall. I’m not saying this is definitely an entailment of theism; I think it is an entailment, but ‘definitely’ is way too strong. All I’m saying is that the prior probability of God doing this isn’t vanishingly low.

If you grant this, you should grant that the theory: “God creates all creatures great and small, including you” has a higher prior probably than: “God creates all the possible people who you yourself could be, but not all the others”, even though both fit the anthropic data we’re trying to explain: that you, in particular, exist.

One objection you might have is that some possible creatures are irredeemably corrupt, such that even God couldn’t give them a good life overall. In reply, however, insofar as one thinks that such creatures are possible (which I doubt, but regardless), you should just modify the theory from: “God creates all possible creatures” to “God creates all possible creatures he could give a good life to”. This wordier theory still has a higher prior probability than the theory that God creates all the possible people who you yourself might be, but not all the rest.

Now here’s the exciting bit: if we think—on anthropic grounds—that God has created all possible conscious creatures that he could give a good life to, it’s plausibly the case that among those possible creatures are creatures who are extremely powerful, extremely knowledgeable, and extremely good, in virtue of their intrinsic features. And intuitively, so long as these creatures are knowledgeable, powerful, and good enough, they will satisfy the concept of lower-case-‘g’ gods.

In other words: given that you exist, you should reason that—probably—all other people who you yourself could be exist, since that’s the theory that makes your existence most likely. But, assuming theism, the most intrinsically probable theory that makes that happen is one where God has created all possible creatures, including a pantheon of gods. This all looks suspiciously Indian.

You might wonder why God wouldn’t also have created a vast plurality of capital-‘G’ Gods, on top of the lower-case ones, an idea which doesn’t look especially Indian. My response is threefold: (1) Even if God did create other beings who were just as powerful, knowledgeable, and good as himself, these wouldn’t count us upper-case-‘G’ Gods since they were created, and a God must be uncreated; (2) Perhaps God couldn’t create other Gods for logical reasons, because doing so would create an unsolvable ‘paradox of the stone’ (“Can God #1 create a stone so heavy that God #2 can’t lift it? If yes, then God #2 isn’t all-powerful, and if no, then God #1 isn’t all powerful.”); and (3) perhaps God didn’t create a pantheon of God’s because doing so would compromise his aseity, or supremacy, or something like that, and would therefore be bad.

With that, on to some arguments for reincarnation.

II. Reincarnation

(a) Huemer’s Proof

This one falls into the “plausibly right” bucket. Conditional on time being infinite in both directions, you, dear reader, have near certain anthropic evidence that you’ve been reincarnated an infinite number of times in the past, and will be reincarnated an infinite number of times in the future. Such was proved by Michael Huemer in his recent Nous article, Existence is Evidence of Immortality.

For the proof to work, you need three assumptions:

1. You exist now.

2. Time is infinite in both directions.

3. There is a non-zero epistemic probability that some theory of personal identity is true on which you could be infinitely reincarnated.

From these, you get the fun result that—with probability 1.—you’ve been reincarnated an infinite number of times, and will be reincarnated an infinite number of times in the future.

Why? Here’s the quick an dirty version: if time is infinite in both directions, you exist now, but your existence isn’t repeated across infinite lifetimes, the probability you’d exist now would be 0, since—given infinite time—you’d either have already occurred in the past, with probability 1 (after which point you couldn’t occur again, given the assumption that you are non-repeatable), or you’d never occur at all. Either way, the probability you’d exist now would be 0. But you do exist now. Hence, you must’ve existed across infinite lifetimes.

Why think you exist now? Well, because you obviously exist now. You might worry that affirming “I exist now” commits one to this or that controversial theory of personal identity. But that’s not so at all: pick any theory of personal identity you like, and you’ll still get the result that you exist now. (Even reductionist, no-self Buddhist views on which there is no permanent, unchanging core of selfhood will have to say something about how you can affirm indexical like “I exist” in evidence-gathering, since sometimes our evidence is indexical. Whatever they say generally, they can say here.)

Why think time is infinite? Huemer’s argument is simple: (i) it’s intuitive that time is infinite, and (ii) there are no good reasons for thinking it’s finite. I’m not sure on (ii), but I share Huemer’s intuition about (i). Since intuitive seemings—like all seemings—give us prima facie justification for beliefs in the absence of counterevidence, then should there turn out to be no good argument for the finitude of the past, we are (or at least: I am) justified in thinking time is infinite. Why is it intuitive that time is infinite? Well, as Huemer writes:

[T]he idea of a beginning of time seems to me metaphysically impossible in a manner similar to that of an edge of space. For any location in space, it makes sense to ask what is, for example, one meter to the left of it (even if the answer is that nothing is there). If there were an edge of space, what would happen if you approached it and tried to gaze past the edge? When we try to imagine this, we find ourselves trying to imagine a place where there is no space, which is of course impossible. The idea of a bounded space is perfectly mathematically consistent; it simply does not seem that Space, that is, all of space, could have this structure.

Similarly, for any moment in time, it makes sense to ask what happened, say, one minute before (even if the answer is that nothing happened). If there were a beginning of time, it would have to somehow not make sense to ask what happened before that time. When we try to conceive of the beginning, we find ourselves trying to imagine time coming into existence. But time could not have come into existence, because a thing’s coming into existence implies a time when the thing does not exist, followed by a time when the thing exists. Of course there could not have been a time when time did not exist. Sometimes it is suggested that God, existing outside time, created time. But even God could not do this, because any act of creation – or any other action or event – presupposes a time at which the act or event may take place. One can identify, of course, a consistent mathematical structure containing a first time; it just does not seem that Time, that is, all of time, could have this structure.

Finally: why think there’s a non-zero epistemic probability that some theory of personal identity? Because there are a number of theories of personal identity—believed by smart philosophers, who give arguments to back them up—on which reincarnation makes sense. Examples include: the theory that we’re immaterial souls, the theory psychological continuity grounds personal identity over time, and the theory that persons are gappy spacetime worms. As long as you’re not 100% sure that all of these theories are false, which no mere mortal should ever be, you have to allow a non-zero epistemic probability that you could be reincarnated.

Putting the pieces together, there’s a decent argument for the view that you’ve been reincarnated infinitely many times, and will be reincarnated infinitely many times in the future.

Along one dimension, Huemerian reincarnation doesn’t mesh well with Hindu theology: according to Hinduism, the utmost aim of our lives is moksha, or liberation—liberation, that is, from the cycle of rebirth. If moksha is taken to be final, then given Huemer’s proof, we can deduce with probability 1 that we’ll never attain it, since, with probability 1, we’ll keep reincarnating forever and ever and ever and ever.

The best move here is to revise moksha a bit: Maybe we’ll never escape the cycle of rebirth permanently, but maybe we will escape it for ridiculously long periods of time—times so long they might justly be thought of as ‘eternities’. I have no idea if any Hindu thinkers have thought this, but it doesn’t seem to do do violence to the concept too much, such that the revision puts moksha beyond recognition.

You might think it clearly does do violence to the traditional idea of moksha. I admit: the argument does take the wind out of its sails a bit. But, I don’t know, if release for the cycle of rebirth is temporary but very, very extended, it still seems a goal worth centring one’s life around, and treating—metaphorically—as “an eternity”. So I’m not too worried at this stage.

(b) Adelstein’s Anthropic Argument

Here is a quick and dirty argument for reincarnation (possibly too quick and dirty…) which was suggested to me by Matthew Adelstein. Unlike Huemer’s Proof, it doesn’t assume time is infinite in both directions.

Here is the data point:

DATA POINT: You exist now, with your evidence.

By “you” I mean you, by “now” I mean the present, and by “your evidence” I mean whatever information you currently have about reality that gives you reasons to believe things. This includes your memories, sense perceptions, intuitions, and so on.

Now consider two hypotheses, and ask yourself which of them makes DATA POINT more likely:

1. You only exist at one time with the evidence you have now.

2. You exist an infinite number of times with the evidence you have now.

Plausibly, hypothesis (2) makes DATA POINT infinitely more likely than hypothesis (1), since if you exist at an infinite number of times with your evidence, the odds that now would be one of those times is higher.

But (2) entails reincarnation. Thus, from the fact that you exist now with your current evidence, you should update in favour of reincarnation.

(c) Gupta’s Gambit

In conversation, Akshay Gupta suggested the following type of argument for why theists should accept reincarnation.

If you want to solve the problem of evil—which you should if you’re a theist, since it’s the best argument for atheism—you’re probably going to have to think that there’s some benefit God wants to give you (of the value soul building, intimacy with God, or whatever) from the suffering you endure in your life.

You’re also probably going to want to affirm the following principle, on the grounds that it’s obviously true:

OBVIOUS PRINCIPLE: If God can benefit you by giving you the the life of a joy and flourishing just as much as he could by giving you the life of great suffering, he should do the former, all else equal.

But if you affirm OBVIOUS PRINCIPLE, you’re just a hop, skip, and a jump away from affirming a plausible theological argument for reincarnation. Think about it: if you live a happy life, God could’ve benefited you more by giving you a life of misery. If that’s the case, then if you’re not living a very miserable life, that’s prima facie surprising. What would make it not surprising is if you’ve lived a life of extreme suffering before this life, or will live one in the next life. That way, God isn’t depriving you of any suffering-induced benefits by giving you a happy life.

If you happen to be living a life of extreme suffering, first of all, I’m sorry. Second of all, just apply this logic to those who are suffering less than you. If you’re attracted to the kind of theodicy where God benefits people in the long run by letting them suffer, why are other people getting less of a benefit, if this is the only life they have?

(d) Pre-existence

Part of the problem of evil is explaining why, even if the evils we suffer are worth it because they spawn greater goods, it was permissible for God to use as as means to those ends. This is the deontological problem of evil, and it’s very troubling. By my lights, the theist’s best bet is to say that each of us pre-existed our earthly lives, and, before we were born, consented to God allowing evils to befall us (after which he wipes our memories—again, with our permission—and sent us into the world). I’ve written more on this theodicy here. In any case, if you’re a theist who finds this plausible, you have evidence for reincarnation over the boring, milquetoast view that we started our lives in the womb, since reincarnation entails pre-existence, while the boring view rules it out.

III. Karma

Conditional on theism, and we have good reasons to be theists, prior probability of God laying down a karmic law is fairly high. (The law might be identified with God’s actions, or be something God puts in place and lets run—either way, it doesn’t matter much.)

Why? Well, if you believe in desert—a word which here means: the ethical doctrine that its intrinsically good for wrongdoers to be harmed in proportion to their wrongdoing or vice, and intrinsically good for do-gooders to be benefited in proportion to their virtue or good deeds—then since, given God’s other goals (soul-building, the permission of morally significant free-will, etc.), he might not be able to satisfy the demands of desert in this life, we’d expect him to satisfy demands of desert in the next life. Hence, karma.

And if Huemer’s argument for reincarnation is right, we have infinitely many lifetimes ahead of us, meaning if it’s not convenient for God to give us our due in the next life, he could hold off and give us what we’re due in the life after that.

I take it most Hindus think we’re going to get our karma sooner rather than later, such that we have more reason to expect bad karma for the evil we did in this life in the next life rather than, say, a billion lives on. Is there any reason to expect this on theism? I think so: after all, God—omnirational as he is—is not a procrastinator. Hence, if there were ever a lifetime where God could be satisfying the demands of desert, had no countervailing reasons not to, but didn’t, he’d be slacking; failing to do the very best he can. But God is the very best. Hence, on theism, you’re more likely to get your just deserts sooner rather than later, since the more lifetimes God waited, the more probable it would be that he would’ve had an opening to give us what we deserve, without needing to sacrifice anything of equal or greater importance.

III. Religious Experience/Common Consent

Here are two much-discussed lines of evidence for theism:

1. Religious experiences. An enormous number of sincere and sane people report having experiences where it seems to them they’re experiencing a divine being. Since we’re prima facie justified is believing what seems to us to be the case, many of these people are prima facie justified in believing in a divine being. But even if you’ve never had a religious experience yourself, the fact that many sincere and sane people report having them is evidence for a divine being or beings. sa

2. Common consent. As discussed above, if lots of people believe something, that’s evidence that it’s true. Since most of the world believes in a divine being, that’s trivially some evidence for a divine being.

The trouble begins when people trying to use these arguments in support of Anselmian Monotheism, or on behalf of an exclusivist religious tradition. Many religious experiences are reported as being as of not just an Anselmian God, but as of limited, minor deities, or nature spirits. The most straightforward interpretation of this evidence is that it best supports a fusion of Anselmian Monotheism and polytheism about lesser deities. This is exactly what Hinduism gives you, unlike most other religions, which hold either that there’s (i) one God, (ii) many gods, or (iii) no God/gods at all.

The same goes for common consent: most people in the world either believe in God, gods, or—in the case of Hinduism—God and gods. Hinduism hoovers up this common consent data more naturally than any other world religion: it predicts that there is a God, a plurality of gods, and isn’t an exclusivist religion (traditionally understood), meaning it predicts that Muslims, Christians, Jews, and Mormons would genuinely come into contact with the divine during their devotional practices.

It also hoovers up the common consent data in its strongest form. For almost any proposition, common consent evidence is at its strongest where the belief commonly consented to was arrived at independently, by groups who had no contact with one another. This makes the common consent argument, say, for the Abrahamic God, and that God alone, especially weak, since widespread belief in the Abrahamic God is traceable back to the same historical root: uncle Abe. In contrast, Hinduism is supported by widespread theistic, polytheistic, and animistic belief, much of which did arise independently in separate communities.

V. Final points: no meat and a small potato

Not all Hindus are vegetarian, but about 4 in 10 Hindus in India are, and 8 in 10 place some restriction on meat in their diets. Hindu scriptures praise vegetarianism, and it’s a plausible practical entailment of ahimsa, the ethical doctrine of non-violence and compassion towards all beings. Since ethical vegetarianism is correct, this is a mark in favour of Hinduism, since the more plausible ethical views a religion endorses, the more likely it is that God revealed it, since God—being omniscient—has all-true moral beliefs, and you’d expect him, if he did communicate with us at any point, to communicate the most important of those to us.

Under the last incarnation of this post, Dustin Crummett wrote:

Unfortunately, this is probably net negative (as vegetarianism promotes egg consumption, maybe the worst thing from an animal welfare perspective, and the dietary restrictions discourage beef consumption, probably the least bad thing from an animal welfare perspective). Hard to know exactly how to think about this, as the underlying principles are good and the bad consequences are maybe because of things unique to modern conditions.

I’m not sure on the empirical question of how much higher vegetarianism drives egg consumption (my hunch would be that it does quite a bit, just based on how many backyard eggs I consume as a vegan when I get the chance.) But the data point I’m conditionalizing on isn’t that [Hindu’s do best from an animal welfare perspective]. It’s that [Hinduism explicitly includes an opposition to animal cruelty in its tradition, which, if implied consistently, would obviously rule out the factory farming of eggs]. This fact, I think, is evidence for Hinduism.

In “Confucianism and the Liturgy: An Analectical Argument for the High Church Traditions”, Tyler McNabb and Joseph Blado argue that—given the centrality of ritual to human nature, its contribution to social cohesiveness, and its positive effects on human virtue—you’d expect to some degree, from the armchair, that whichever religion God revealed would be pretty ritualistic. Since Hinduism is highly, highly ritualistic—more so than some other religious traditions I could name—that’s an evidential chip in favour of Hinduism. I find this point decently plausible, but it doesn’t favour Hinduism that uniquely.

In a recent blog post, “Steelmanning Christianity”, Matthew Adelstein argues that there’s a reasonably high prior probability that God would become incarnate. Matthew takes this to support Christianity:

Christianity is unique in claiming that God becomes incarnate. But it’s decently likely that God would become incarnate. A God who stays aloof in the clouds, not experiencing the suffering on Earth seems in some way deficient. This is particularly because God’s experience as an incarnate being would help strengthen our relationship with him, just as two people who have undergone the same experience might bond over that. The prior probability that God would become incarnate can’t be too low—and so the prior probability of Christianity, the only reasonable view on which God becomes incarnate, is also not too low.

I don’t find this argument convincing, since I think that God is both omnipresent and passable—meaning he’s everywhere and he suffers. If God is omnipresent and suffers, then God is neither emotionally aloof nor up up in the clouds. He suffers with us in our darkest moments, and is present while he does so.

Nevertheless, if you find Matthew’s claim plausible, it should raise your credence in certain forms of Hinduism, not just Christianity. Many Hindus—especially Vaisnavas—think God has become incarnate and interacted with us in the form of a range of avatars. (On this view, the avatars aren’t distinct divine creations; they’re forms God himself takes on.)

On the one hand, incarnation is more weakly predicted on Hinduism than on Christianity, since one can be a non-heretical Hindu—insofar as “heresy” even makes sense in an Indian context—and deny that God himself takes on the form of avatars, while the same can’t be said of Christianity.

On the other hand, the logical constraints imposed by the various Christological Councils—and the ongoing disagreement on whether there’s a logically consistent, theologically acceptable model of the incarnation—should also decrease the prior, for those who aren’t experts on Christology, of the Christian incarnation making sense to begin with. Hindus, in contrast, have fewer logical constraints hemming them in.

(For the record, I suspect Christianity wins the predictive tug-of-war for this data point—if it is a data point at all, which I deny. Still, it might help raise the prior of Hinduism over a religion like Islam, where incarnation—excluding, perhaps, Allah’s Word in the text of the Qur’an—is off the agenda.)

---

I’m not a Hindu—at least, I’m not a Hindu yet. But from the standpoint of now, I find it pretty plausible. It does better evidentially than many other major religions, including Christianity, I think, which is another plausible contender for The Religion of Greatest Promise.

Hinduism might actually be true.

1. Note: this argument has no warts; it’s perfect.