Oct 23, 2013 | By

Does Knowing = Having Justified True Beliefs?

Justified True Belief is the Holy Grail (to borrow a religious allegory) of epistemology – “knowing” is to believe something, on the basis of rational evidence and thought, that is in fact true.  While this concept has taken a beating over the past forty years, starting with some paradoxical thought experiments by Edmund Gettier, it generally describes the model for knowledge that philosophers have applied for more than 2,000 years.  But is it helpful?  Is there a better way to define “knowing”?

In determining whether an idea or statement meets the test of being a justified true belief (“JTB”), identifying a “belief” is the easy part of the problem.  A belief is something that an individual holds to be true.   “Justification” is more subtle – in the broadest sense justification could include sensory (empirical) evidence, logical analysis, subjective (personal) experience or potentially indirect sources such as  testimony or revelation, although most philosophers put little or no value on subjective or indirect justification.   But by far the most difficult task is to determine when something is, in fact, true.  For many important questions, this task may be impossible.

Consider what happens if you examine a statement such as “the sky is blue.”  It is likely that if my eyes are open and I look at the sky on a sunny day I will be able to say “I believe the sky is blue and I have a good justification for that belief.”  But in order to know this is a true statement, I would have to add some additional information, perhaps by validating the fact that I am not color-blind, and confirming that I am not asleep and dreaming or merely looking at an overhead projection.  Backing up our findings on these facts then requires an additional level of analysis – “I am not dreaming because…” and “I know it is not an overhead projection because…”   Indeed, each fact statement generates the need for another “justification” – in order to back up the reasoning that led to my first conclusion that the sky is blue.  Each layer of justification, from the standpoint of determining the absolute truth of the statement, yields an additional layer, ultimately resulting in an infinite regress of statements attempting to prove the truth of the first.

Philosophers have responded to this problem in a variety of ways and the theoretical convolutions are quite intricate.   The simplest and perhaps best response is the practical one – to acknowledge that perfect knowledge is impossible and the best we can do is say that a true statement is one that is supported by a convergence of “justifications” that by necessity must be finite.

Consider, however, what happens when you examine a statement such as “miracles are real” or “human beings have free will.”  These statements are topics of recent ISAS Forum articles (see Miracles and Free Will).  The difficulty here is that these, and many other important questions, may not yield any convergence of justifications and, in fact, may be subject to a diverging chasm that will never yield to rational inquiry.  As C.S. Lewis postulates in his book On Miracles, there is no proof for a belief in either “Nature” or “Supernature” – but to answer the question of whether miracles are real one has to adopt one or the other belief system.

The conclusion is that there is a class of problems (most of the important ones in philosophy and religion) that fall into this category of divergent justification.  These problems are not accessible through the JTB model of epistemology and can never be “known” in that sense.  In what sense, then, can they be “known”?

Four possible avenues (and there may be more) for “knowing” the truth of statements in these categories are possible:

(1) The preponderance of considerations suggests that the statement is more likely to be true than not.  The statement cannot be proved “beyond doubt” to everyone, but considering the available evidence suggests to me strongly enough that it is true, so I will believe it.

(2) I have personal experience that may not be available to everyone that supports the truth of the statement.  This “private evidence” is sufficient justification for me to believe in the truth of the statement but this justification may not be useful to others.

(3) There are consequences to me in concluding that a statement is true or false, and in weighing those consequences I conclude that I am better off believing the statement is true and I therefore choose to do so.  This somewhat back-door justification was featured in the recent Free Will discussion.

(4) The rational evidence and analysis from (1), (2) and (3) may not be entirely compelling, but the statement is something that I believe is true based on intuition, inspiration, revelation, community norms, feelings or other inexpressible or intangible factors.  These may or may not be relevant to others.

These approaches to knowing the truth may seem unsatisfactory, but they are critical in giving us tools to address problems with divergent justifications under the JTB model.  The difficult challenge to us as human beings with the gift of free will is to explore and to balance all of these factors in order to achieve a coherent, consistent and non-contradictory set of beliefs as close to “what is, in fact, true” as possible.  We will never know with certainty “what is, in fact, true” – at least not during our lives in this world – but we are obliged to make an effort to learn what we can of the truth in order to make the most of what we have and who we are.

Significantly, we need to recognize that all human beings are in the same situation.  The answers that we so desperately yearn for are not directly accessible – we are all faced with having to “muddle through”.  Some of us may have access to more information and education, or to teachings (including religious teachings) that better accommodate or address these questions, but many do not.  That suggests that sympathy may be a more appropriate response than antipathy.  After all, we do share in the amazing (indeed, to many, miraculous) journey of life.

4 Responses to “Does Knowing = Having Justified True Beliefs?”

  1. […] have confidence in the validity of our rationality. However, as pointed out in an earlier post on  knowledge, this standard is not feasible. Facts are not as easily determined as we might wish. Proving that […]

  2. George Gantz says:

    Hi Brian –

    You ask – “Are you claiming that superstition (known false belief) is knowledge?” No. Anything that someone believes that is demonstrably false (beyond a reasonable doubt) is not knowledge, but ignorance. This applies to the stay off the mountain or avoid sex examples you offer, and to any lies that someone might want to concoct.

    I would agree that on most practical questions, public evidence can approach the standard of “convergent justification”. For example, we have very good evidence and a strong framework for explaining how and why the sun rises and sets every day, so believing that it will do so is strongly supported. However, the evidence is subject to the infinite regress problem of justification – proof beyond doubt could only be achieved as the number of days observed approached the number of all days. Someone may believe, and could be correct, that the earth will be destroyed by an asteroid tonight, or that the laws of physics will inexplicably cease.

    Divergence of justification is something else. Divergence occurs on questions like the ones I cited – the existence of free will or the existence of miracles. There may be little relevant public evidence in support of a particular belief, but there is also little relevant public evidence that can be called upon to contradict the belief. We have to move our justifications from consideration of the public evidence to considerations of private, consequentialist or intuitive/revelatory evidence.

    Most questions of ultimate origins, purpose and meaning are in this category – the question of the existence of God being the big one. I have no doubt that there is one answer to the question of whether God exists, but it is not an answer that can be reached through consideration of public evidence.

  3. Brian Macker says:


    “(3) There are consequences to me in concluding that a statement is true or false, and in weighing those consequences I conclude that I am better off believing the statement is true and I therefore choose to do so.”

    Are you claiming that supersition (known false belief) is knowledge just so long as it results in someone being “better off” by believing it? Examples might be: a) staying off a dangerous mountain because you believe there are yeti wandering on it. b) Not having sex with prostitutes because you think they like to hide razor blades in their vaginas (actual example of a kid I knew). Both these examples will reduce various dangers which therefore can make one “better off”. I doubt that kid paid to catch any STDs if he kept believing that.

    In fact, there are lots of lies out there tailored specifically to benefit people. The very best scam artist is the one who believes his own lies (for they no longer are lies). He can keep a straight face as he benefits from those beliefs. Maybe you need to tailor your criteria in 3) so that the false belief must benefit everyone, and not just the person who believes it.

    I also don’t see why you are not classifying 1) as “convergent knowledge” based on the first sentence, “The preponderance of considerations suggests that the statement is more likely to be true than not.” If that is the case then hasn’t it “converged”.

    I’m having problems with that second sentence, “The statement cannot be proved ‘beyond doubt’ to everyone, but considering the available evidence suggests to me strongly enough that it is true, so I will believe it.” That sentence does not seem to describe what you described in the first. Also it seems overly broad. Much of our knowledge can’t be proved “beyond doubt” for everyone. There always seems to be someone who can’t understand, or can’t believe what is obvious to everyone else, or doesn’t believe someone speaking the truth.

    With 2) that’s just a case of you know something someone else doesn’t know. Like if I peeked at my poker hand and did not let anyone else see. I could bluff (or not) and if I win the pot on a fold I am not obliged to show it. Everyone understands I knew what my hand was even if I can’t prove what it was later. I’m not sure why you see this as a problem. For something to be knowledge it doesn’t have to be known by everyone.

    I’m just not understanding what you are getting at with “divergent justification”.

  4. […] is no obvious answer to these questions.  But to paraphrase what I said in my last post on Knowing –  “We will never know with certainty what is, in fact, THE RIGHT ANSWER – but we are obliged […]

Join the Discussion

Why ask?