Reliabilism is defined by several sources as follows:

Reliabilism is an approach to the nature of knowledge and of justified belief. Reliabilism about justification, in its simplest form, says that a belief is justified if and only if it is produced by a reliable psychological process, meaning a process that produces a high proportion of true beliefs. A justified belief may itself be false, but its mode of acquisition (or the way it is subsequently sustained) must be of a kind that typically yields truths. Since random guessing, for example, does not systematically yield truths, beliefs acquired by guesswork are not justified. By contrast, identifying middle-sized physical objects by visual observation is presumably pretty reliable, so beliefs produced in this manner are justified. Reliabilism does not require that the possessor of a justified belief should know that it was reliably produced. Knowledge of reliability is necessary for knowing that a belief is justified, but the belief can be justified without the agent knowing that it is.

Source: https://www.rep.routledge.com/articles/thematic/reliabilism/v-1

One of the main goals of epistemologists is to provide a substantive and explanatory account of the conditions under which a belief has some desirable epistemic status (typically, justification or knowledge). According to the reliabilist approach to epistemology, any adequate account will need to mention the reliability of the process responsible for the belief, or truth-conducive considerations more generally. Historically, one major motivation for reliabilism—and one source of its enduring interest—is its naturalistic potential. According to reliabilists, epistemic properties can be explained in terms of reliability, which in turn can be understood without reference to any unreduced epistemic notions, such as evidence or knowledge.

Source: https://plato.stanford.edu/entries/reliabilism/

A broadly reliabilist theory of knowledge is roughly as follows:

One knows that p (p stands for any proposition—e.g., that the sky is blue) if and only if p is true, one believes that p is true, and one has arrived at the belief that p is true through some reliable process.

A broadly reliabilist theory of justified belief can be stated as follows:

One has a justified belief that p if, and only if, the belief is the result of a reliable process.

Source: https://en.wikipedia.org/wiki/Reliabilism

Can a reliabilist have a reliably justified belief in God? Is there any reliable process that can justify such a belief?

Context: this question has been inspired by this answer.

  • 3
    No. (Based on evidence and arguments I've seen, presuming reality to be the way I perceive it to be, assuming the observer has a reasonably solid grasp of our shared reality, and excluding some sort of pantheistic view where one redefines "God" to be the same as the universe.) But God is a conclusion, and reliability describes a process - we can only evaluate the God claim under known processes; it would be up to the theist to present a reliable process if they wish to convince a reliabilist.
    – NotThatGuy
    Commented May 24 at 13:04
  • 1
    @NotThatGuy Are you sure the answer is "No" instead of "I don't know"?
    – user66156
    Commented May 24 at 15:36
  • 2
    @Mark It's "No" in the same sense that I would answer "No" if someone asks if Bigfoot is real or if the government is run by lizard people. We just need to establish from the start what can or cannot reasonably (if at all) be definitively proven and where the burden of proof should be, and then "No" could be considered a reasonable answer to such questions. If you want, you can consider "No" to roughly translate to "I have no good reason to think that's true, and every reason to think it's false". I did also add many caveats, and explicitly mentioned the possible existence of such a process.
    – NotThatGuy
    Commented May 24 at 16:07
  • 3
    With all these definitions, I'm still confused how a reliabilist can justify anything. They all seem to require a "reliable process" which is only "reliable" if it arrives at the truth, but there still doesnt appear to be any way to actually ascertain the truth in the first place. How do they find the reliability of statements? To me there seems to be an implication that they already know the "truth" and then compare the results of reliable processes to that truth... but how do they establish the truth in the first place?
    – JMac
    Commented May 24 at 16:34
  • 2
    @NotThatGuy Its the part about referencing "truth" that I'm not really getting the point of. I have no problem with the scientific method. But to me a big value of it is that it doesnt really rely on "truth" in the way reliabilism seems to (and maybe I just misunderstand the point). Science seems to decouple the reliability from truth, in the sense that if its reliable, its reliable whether it is true or not, and that's a useful property in a lot of ways on it's own
    – JMac
    Commented May 24 at 18:38

4 Answers 4


Reliabilism seems to me to be an entirely vacuous concept. Of course it is possible to have a belief in god which is 'reliable' in the intended sense. We hold very many of our beliefs through the process of accepting what we are told by people in positions of authority over us when we are young. On that basis, I believe a very large number of things to be true, such as Queen Victoria having been married to Albert, the Moon orbiting the Earth, the source of the Congo not being in Yorkshire, black widow spiders being poisonous, and so on. An extremely high proportion of the beliefs I have acquired as a consequence of my schooling are true, so on that basis the psychological process of accepting what we are told by our schoolteachers seems to be a 'reliable' one. If, therefore, you are unfortunate enough to attend a school in which a given religion is drummed into you along with your times tables and the rest, then yes, you can have a reliable belief in God.


Yes, of course, reliabilism is a relativistic approach, and pretty much any belief can be justified that way, given the right conditions.

As an example, a member of a cult will typically be restricted by the cult to only read media permittes by that cult and only entertain relationships with members of the same cult, reducing the chances of learning of a better source of truths than the cult doctrine.

I believe that in the modern age to a person with access and exposure to a broad range of media sources, reliabilism would justify agnosticism or atheism much more easily than faith.

However beliefs in gods are typically not justified with reliabilism anyway. Instead, the epistemology of religion is a topic that is complex and has a lot of history: https://plato.stanford.edu/entries/religion-epistemology/

(More elaboration as requested) A modern person is typically exposed to various sources of information, old an new. There is scripture from various religions, scientific writings, news channels, opinion pieces by influencers, TV shows, movies, and so on. This is a stark contrast to human life 200 years ago, were beyond newspapers people would mostly be exposed to gossip from the village they stayed their whole life in.

Since there are opposing views and beliefs in all the channels of information a person is exposed to, all people would need to keep track of how many claims from each channel turned out to be truths or lies, as a means to find the best process for reliabilism. Science would typically be easily the most reliable channel about all observable reality, all nin-scientific channels erred much more than science on that subject. But science says nothing about the spiritual world (it does not claim there are no gods). For all other channels of information, skepticism would be an obvious way to defend oneself against falsehoods (also called fake news). For claims about God's, this just leaves some spiritual sources of information. The problem with those is that they either only make claims that cannot be verified (like the afterlife), or they make supernatural claims about observable reality that can be verified, but have turned out to be falsehoods too often to be taken seriously (else they would be part of science). For reliabilism, any spiritual channel that makes only claims that are not verifiable cannot be counted as a process that typically generates truths.

From a skeptical point of view, all beliefs in gods even in prehistorie were not based on reliabilism, stone age people in their caves had no good processes that yielded a lot of truth all the time. Instead they had a lot of open questions, and gods (good and bad ones) were fill-gaps for the lack of explanation, mostly. Disease spreading, some evil god did it. Earthquake, evil god. Person recovering from. Disease, a good god did it. And so on. This is not reliabilism. Reliabilism became relevant once "wise men" or "wise women" had some true knowledge about this world (cures, agricultural methods, historic knowledge) to pass on, so that they could count as most reliable sources of truth at the time (often without local competition). Reliabilism would then mean believing in those people would follow the best available process at the time that was known to lead to at least some truths. But with secular education, the people telling identifiable truths are no longer (necessarily) also preaching about gods, because that's not their job in modern pluralistic society.

Of course religious people will say that churches are the last bastion or truth in morality. But since Kant philosophy has a robust framework of morality with gods and fear of hell, and all old scriptures are full of misogynistic, homophobic, racist, fascist, discriminatory crap and don't even cover the basics of human rights. Where as New Age spirituality has no defining authority and boils down to the individual picking whatever suits them.

So that's where we are as modern society in most places of the western world.

  • I believe that in the modern age to a person with access and exposure to a broad range of media sources, reliabilism would justify agnosticism or atheism much more easily than faith. - Can you please elaborate more on this? Can you explain, for example, why you believe this?
    – user66156
    Commented May 25 at 0:03
  • I have elaborated more in the post.
    – tkruse
    Commented May 25 at 12:53

Yes, remarkably so, for certain beliefs about God.

What really is a reliable process? That question is at the heart of the matter. It seems to be simple, but it is not. The problem is that it tries to start by presuming that what is and is not true. If we do not know what is true, then how do we know what process most often arrives at it? If we do know what is true, then why do we care about the process? So, really, Reliabilism is more about consistency than about reaching any specific target.

As an oversimplification, though one I believe to be generally applicable, the alternative to believing in God would be something similar to a Naturalistic or a Materialistic view of the world. These views would also often start with presuming the effectiveness of our empirical senses. The problem here is that the conclusions reached by these methods seem to contradict these very senses. Light seems at first to move in a straight line, but further explanation shows it to move as a wave or a particle. Seemingly solid things are revealed to be masses of moving particles with space in between them. So, "truth" in this view is more malleable, varying with perspective. That doesn't mean that the approach is wrong, but that its reliability is questionable because the results are not consistent for all applications even using the same method, and there is no way to know how the knowledge would morph with further observation. This seems to me to be the case with most inductive processes.

For these alternative approaches, I also want to point out Alvin Plantinga's Evolutionary Argument about Naturalism. Essentially, these methods tend to lead to a contradiction in their own reliability. In other words, to make them work, you have to break the model and use a presumption rather than trust the method consistently.

In that light, consider beliefs about God. Not all beliefs in god are reliable. Many will succumb to the same faults. However, some do not. Consider the Transcendental Argument for God, which will often cite the existence of God as being the factor which makes our ability to understand the world at all reliable. Consider the belief in a creator being which created both the universe to be consistent and follow order, but also which created our minds with the ability to reason which relates to that order, and that the use of such reason reveals the existence of God. In such a case, using our reason and observation of the universe to arrive at God would be consistent. Even though observation using that method might reveal more detail upon greater application, it arrives at a similar picture such that previous ones were also true.

Furthermore, the process remains the same for all beliefs, even in our theory of knowledge and reason itself. If our minds were created by a force biased toward creating minds which can accurately reason, which reliably arrives consistently at such minds, then we can even use the same method of Reliabilism to affirm our belief in our ability to reason, in away that alternate methods do not allow.

Of course, this all depends upon the method used to arrive at the belief in God. It is certainly possible to believe in God for emotional reasons, cultural reasons, or many other reasons. One can also believe that the Earth is round for the same reasons, and as such, it would not be a reliable method. The successfully reliable method itself might be in question, but it does seem as if Theistic arguments, such as perhaps TAG, might be able to make this claim as well if not better than arguments against God, at least without allowing unreliable ad hoc requirements.

I am personally skeptical about Reliabilism, but it seems to me that if that is the case, a belief in God is the most preferred belief.


Yes; below are three assumptions which are reliable, by the definition you quote, and which reasonably lead to belief in God:

  1. Complex functional machines have a designer. This is the famous watchmaker analogy. Spaceships, watches, computers, and cars have designers; and if we found one such machine in a junkyard, we would know it wasn't made by a tornado! It is thus reasonable from experience to assume that biological machines have a Designer, too. Biology is full of micro-machinery which surpasses the complexity of amazing man-made machines. Even many evolutionists will admit that nature has a surprising level of apparent design. Thus they indirectly (and maybe sometimes, directly) admit that belief in a Designer is reasonable, even if they think it is wrong. But this logic is "reliable" by the definition given, because it results in an extremely high proportion of true beliefs in other areas.
  2. Language implies an intelligent author. Everywhere we find genuine language, we infer authorship. If we read a large book or find a complex and coherent message scribbled in the sand, we know an intelligent mind was behind it. Thus, it is reasonable to infer that the language of DNA has an intelligent Author; and considering the variety, unity, creativity, and mind-bending complexity of DNA--in everything from trees, to hummingbirds, to lions, to humans--it is reasonable to infer that this Author is God. Because this thought process works so well in areas other than DNA, it is "reliable" by the definition given.
  3. All natural things and events have a cause. This also works so well in everyday life, and it is thus reliable. If we see a turtle on a fencepost, we know that there is a cause for it, and that the cause isn't the turtle's climbing ability! If we find the glass window broken in and valuables missing, we know that the cause was probably a burglar. Certain things may happen by complex and seemingly "random" precursor events, but we assume that every natural thing and event was caused by something adequate to the result. Thus, it is reasonable to assume that the existence of the universe itself has a Cause outside of itself; and considering the immensity of the universe, what a Cause that must be!

All three of these assumptions result in "a high proportion of true beliefs" in the uncontested situations where we know the details by firsthand experience; that is, where we can prove whether the conclusions are true or false. Thus, they are "reliable" by the definition given.

You must log in to answer this question.