Why is belief necessary for justified true belief?



In justified true belief it is said that for a person to know a fact it must be true, she must believe in it and she must be justified in believing it.

My question is: Is belief necessary? Why is the following not enough:

  1. The fact is true.
  2. She has justifiable reasons to conclude it is true.


Posted 2018-09-23T15:30:14.343

Reputation: 91

2This is an interesting question; but why would anyone not believe something they have reason to believe is true, unless it is something that conflicts with their ideals? I personally find some things extremely difficult to believe for that reason, even with adequate evidence to the contrary. And I don't think that my reaction to, say, unpleasant truths, is all that uncommon. So there are things I rationally believe to be true, which my conscience or subconscious mind rejects. – Bread – 2018-09-23T15:41:53.037

I made an additional edit which you may roll back or continue editing. Good question to ask why all three of these are needed. +1 – Frank Hubeny – 2018-09-23T18:03:15.873

On your proposal we'd have to say that creationists know that the evolution theory is true. That is not a common use of "know". There are theories of knowledge that discard belief as a condition, especially in artificial intelligence contexts, but they put in place more than 1 and 2, see Do machine learning algorithms have knowledge?

– Conifold – 2018-09-23T19:53:53.543

Increase in knowledge can turn a presumed fact false, even if there were justifiable reasons to conclude it had to be true. What would happen with the belief? – Mast – 2018-09-24T06:16:40.577

@Conifold Disagree: evolution just do not seem logical to me. I wonder "How does that pass as good science?" I chalk creationist scientists lack of support up to the fact that they do not get the same level of professional support that their secular counterparts have, it's a spiral: "don't get funding/support because low quality because funding/support because...", kind of like how popularity begets more popularity. Looking past the rhetoric of both, I see creationism fitting history and science better. From my view, I could reverse creationism/evolution in your comment and say the same. – Aaron – 2018-09-24T19:43:06.043

My previous comment should not be taken as an attack on evolution; it would be folly to even attempt such a broad thing in a single comment. Rather, it is merely to point out a counter point to the comment by Conifold, and I am not the only counter point. – Aaron – 2018-09-24T19:44:21.090

What do you mean by "justifiable reasons"? I've run into situations (frequently in historical research) where I had strong reasons to believe two incompatible things. I could have justified either thing with good arguments. I can only believe one of the things, and only one of the things can be true, but there's no necessary connection with my belief and truth. – David Thornley – 2018-09-24T21:43:11.100

@Aaron, you show that you don't understand science when you refer to "creation scientists". Creation is not falsifiable, so it's not possible to hold a scientific belief that it happened. Again, this is far too big for a comment. – David Thornley – 2018-09-24T21:45:15.793

@DavidThornley I didn't come here for a creation/evolution argument. Nobody needs to bother trashing creationism here; it's already done enough, we know all about that point of view. I won't bother to counter you since that would be off topic, but I will merely suggest that such a view is either narrow minded, or it does not know enough about creationism, or it does not admit that a lot of evolutionary science is no more falsifiable than creationism. If you want that debate, this is not the place. And remember to be nice. Telling one versed in science he "doesn't understand science" is rude. – Aaron – 2018-09-24T21:58:32.820

@FrankHubeny Your edit changed the title from "justified true belief" to "Why is belief necessary for justified true belief?". But the new title doesn't really make sense. The question is why belief is necessary for knowledge (or maybe, why is belief necessary on the JTB model of knowledge), not why belief is necessary for justified true belief, which is trivial. – Eliran – 2018-09-26T00:26:47.697

@EliranH Look at the versions. His question was, and still is: "My question is: is belief necessary?" I don't think I changed george's intent. However, either you or george or anyone else can further edit it. – Frank Hubeny – 2018-09-26T02:51:35.187

This question would be more inclusive if gender-neutral personal pronouns were used instead. – Carl Masens – 2018-09-26T06:24:20.067



Your (1) and (2) are not enough. Here is an example: suppose I have excellent reasons to believe that the earth is round (I've seen photos, listened to lectures, etc.), and that it is in fact true that the earth is round, but nevertheless I do not believe it (because I'm irrational). Clearly this is not a case of knowledge.

There is a recent view, however, Knowledge First, that holds that knowledge cannot be analyzed in terms of belief (indeed, that it cannot be analyzed at all). But even on this view belief is a necessary condition for knowledge simply because knowledge entails belief.


Posted 2018-09-23T15:30:14.343

Reputation: 6 164

1For an irrational person (or anyone who doesn't believe the fact), is "She has justifiable reasons to conclude it is true." therefore not applicable? I would say that there's a major difference between "She has justifiable reasons [..]" and "There are justifiable reasons [..]". The former implies that she acknowledges and does not refute those reasons, at which point she has in fact accepted the fact (as the best available explanation). The irrational person would not acknowledge the validity of the justifiable reasons, since they do not believe the fact to be true. – Flater – 2018-09-24T10:28:04.570

@Flater A person may have good reasons but not realize that they are good reasons. Or, a different case: I might have good reasons for X and good reasons for not-X. I have to decide which are better reasons. If I end up believing X (perhaps rationally), that doesn't mean I didn't have good reasons for not-X as well. – Eliran – 2018-09-24T18:09:36.243


According to Eric Schwitzgebel,

Contemporary analytic philosophers of mind generally use the term “belief” to refer to the attitude we have, roughly, whenever we take something to be the case or regard it as true. To believe something, in this sense, needn't involve actively reflecting on it: Of the vast number of things ordinary adults believe, only a few can be at the fore of the mind at any single time. Nor does the term “belief”, in standard philosophical usage, imply any uncertainty or any extended reflection about the matter in question (as it sometimes does in ordinary English usage). Many of the things we believe, in the relevant sense, are quite mundane: that we have heads, that it's the 21st century, that a coffee mug is on the desk. Forming beliefs is thus one of the most basic and important features of the mind, and the concept of belief plays a crucial role in both philosophy of mind and epistemology.

From this perspective the "belief" part of "justified true belief" is what is characterized as "true" and "justified" since we may have beliefs that are neither true nor justified.

It is not something that we, in addition, have to believe in. We already believe the fact but our belief might be incorrect. Now the question is whether that fact, that belief, actually is true and justified.


Schwitzgebel, Eric, "Belief", The Stanford Encyclopedia of Philosophy (Summer 2015 Edition), Edward N. Zalta (ed.), URL = https://plato.stanford.edu/archives/sum2015/entries/belief/.

Frank Hubeny

Posted 2018-09-23T15:30:14.343

Reputation: 18 742


Welcome George.

  1. The fact is true.
  2. She has justifiable reasons to conclude it is true.

These conditions are not sufficient ('not enough') for knowledge because the fact may be true and she, X, may have justifiable reasons to conclude that it is true but those justifiable reasons may lead X only accidentally to conclude that it is true.

Suppose X is walking down a long, dark corridor. At the end there is a hologram of Y, her friend. Suppose that in the given conditions it is perceptually impossible to tell that she is seeing a hologram rather than her friend. In any case, she has visited her friend here often enough and there has never been a hologram before. So X has justifiable reasons to conclude that her friend is at the end of the corridor. How sceptically cautious are we going to require her to be ? By normal standards of evidence she is justified in concluding that her friend is at the end of the corridor. Now, suppose it is also a fact that her friend really is at the end of the corridor, behind and concealed by the hologram.

In this case (1) is true - her friend is at the end of the corridor and (2) is true - she is justified in concluding that her friend is at the end of the corridor. But because the perceptual causal chain is 'deviant' she doesn't know that her friend is standing there.

So (1) and (2) aren't enough for knowledge.

What you might choose to consider is the view, revived and powerfully argued for by Timothy Williamson, that belief is unnecessary for knowledge. The relation between knower and known does not include belief. This view is put forward in Williamson's Knowledge and Its Limits, ISBN 10: 019925656X / ISBN 13: 9780199256563. Published by OUP Oxford, 2002.

Another approach which can in certain versions eliminate belief from the conditions for knowledge is externalism. Here what matters is how a certain state of mind has come about; provided there is the 'right' causal connection between my state of mind and the external world, I can know without believing. If you ask me the French word for 'book', I answer 'livre'. There is an immediate disposition to say, 'livre', and this disposition is (let's assume) causally linked through memory to my learning French rather a long time ago. Of course, if you take a dispositional view of belief then to believe here just is to be disposed to answer 'livre'. But I do not think you are using this sense of 'belief'. And the idea of my having justifiable reasons for my answer simply doesn't apply. I haven't a clue how and when I learned that 'book' translates as 'livre'.

Geoffrey Thomas

Posted 2018-09-23T15:30:14.343

Reputation: 34 276

Are you talking about Gettier Case here? I'm not sure if the OP meant the question to go in this direction. Although it's a very good answer, I think it might miss what the OP asked, which I'll attempt to answer in my own answer – Yechiam Weiss – 2018-09-24T17:36:38.793


Why is belief necessary for justified true belief?

Well, let's take "belief" out. We get

justified true

This is not a grammatically correct sentence.

Let's take out, instead, "justified" and "true", in turn, and we get

justified belief


true belief

These are both perfectly fine phrases.

Belief then is necessary grammatically. Your elaboration of what belief means is just that - a further clarification of what it means to believe in this sentence. Generally it's taken as such in this sentence.

Mozibur Ullah

Posted 2018-09-23T15:30:14.343

Reputation: 1

+1 because your explanation makes perfect sense. And because it is simple enough to easily understand. However, the grammatical construction is a noun phrase, not a sentence. The noun is belief, modified by the adjectives justified and true. – Bread – 2018-09-24T23:16:43.167

I replaced "sentences" with "phrases", but I think your argument still works. +1 – Frank Hubeny – 2018-09-26T00:00:26.707

This answer relies on a fallacious link between universal meaning and the grammatical structure of a single language. For example, taking "belief" out in languages such as French do not produce a grammatically incorrect sentence ("une vraie croyance justifiée" -> "une vraie justifiée"). – Carl Masens – 2018-09-26T06:19:59.647

@CarlMasens: I don't know French well enough to argue with you. The point is that adjectives on their own generally don't make any sense with a noun, something that they modify. They are after all, defined as modifiers. If Chomskys universal grammar is to be believed - and to be honest I would go along with him - there ought to be a similar matching to French, Spanish and so on. Languages have idiosyncrasies and they have to be taken into account. – Mozibur Ullah – 2018-09-27T02:37:40.087


The thing is, when we're talking about knowledge in epistemology we are talking about personal, subjective knowledge. Looking at it that way, although all of the answers here are excellent, I think they miss the core point of the question (except, perhaps, Eliran's) - yes, the fact is true, and yes there is a justified reason to conclude that it is in fact true, but without the proper connection being made by the subject, and without the subject believing that it is in fact true, the subject would simply won't consider it as true. Not because it isn't true, and indeed someone else might actually acknowledge it as true, but without making the connection between the objective truthfulness of the fact and the subjective conviction of said truth, the subject won't consider the fact as acknowledged.

Yechiam Weiss

Posted 2018-09-23T15:30:14.343

Reputation: 3 468


I think there are two distinct nuances in your question and I shall endeavour to address them both. The first is that facts are independent of belief so that by stating a "Justified true belief" we are speaking of a rational hypothesis the person has put forth and this has been borne out by facts and data. For example If I am asked to identify the object in front of me I believe it is an apple. At this point it is a belief since it has not been verified nor tested. It could be a genetically altered banana. After the facts and data show it to be an apple it is then a justified and true belief. It also ceases to be a belief as the data stands on its own without your belief. The second part is that the actual term "belief" tends to be nebulous regarding this concept. It is much easier to use the word "conclusion" for the second part. Going back to the apple analogy, you first say "I believe it is an apple" and after proving so, you then say "I am true and justified in concluding it is an apple". This eliminates your premise that belief is not needed when facts and data have provided a conclusion.


Posted 2018-09-23T15:30:14.343

Reputation: 11


The answer is quite simple. It is the natural consequence of the definition of justified true belief and the requirements that a person must meet to have a justified true belief.

The requirements are:

1 The fact must be true.
2 The person must believe it is true.
3 The person must be justified in believing it is true.

Since your statements (1) & (2) only meet requirements 1 & 3, they are not enough (requirement 2 is missing) to have a justified true belief.


Posted 2018-09-23T15:30:14.343

Reputation: 1 686

Unfortunately, the verification of any one of these requirements requires justified true belief in the requirements themselves. This answer would have been better equipped to answer the question if its reasoning were not circular. – Carl Masens – 2018-09-26T06:22:20.297


Because of logic and common sense. In the same way that it is necessary to have a dog in order to have a big dog, it is necessary to have belief in order to have justified true belief.

Philosopher of science

Posted 2018-09-23T15:30:14.343

Reputation: 168