login about faq

What is the proper definition of "justification" in an epistemological context? In other words, "justification" in the following sense: "I am justified in holding this belief." It seems very difficult for me to find a definition that isn't arbitrary or non-essential.

asked Oct 25 '10 at 13:56

ttime's gravatar image


edited Oct 25 '10 at 14:01

Greg%20Perkins's gravatar image

Greg Perkins ♦♦

The question asks about justification "in an epistemological context." But in an epistemological context, as in "justified belief," a more accurate term would be validation, which means identifying an idea's relation to reality.

When I look up "justify" in my dictionary, I find that it follows directly from the term "just" (as an adjective), which basically means fair or ethically right or righteous. Using that term in an epistemological sense may actually be an expression of an implicit premise that one must gratify or placate some conscious authority (possibly social) greater than oneself before one can hold true to a belief. Primitive man might feel obligated to "make an offering to the gods" before implementing a judgment about a course of practical action for living -- implying that the gods are the final authority on truth or falsity, not individual human reason and focus on reality.

The same question that Objectivism asks about values can also be applied to justification: "Justified" to whom and for what?


After reading the answer posted by Valerypublius, it occurred to me to check a dictionary of philosophy (in addition to the ordinary dictionary that I had originally checked). In The Harper Collins Dictionary of Philosophy, 2nd Ed., by Peter A. Angeles, I found the following:

justification 1. defense; that which is offered as sufficient grounds for an assertion (claim, statement, conclusion) or for one's conduct. 2. logical proof; in logic, the procedures applied to the premises of an argument that show the proof for the conclusion. Compare with EXPLANATION; PROOF.

In other words, "justificatin" in this sense refers mainly to whatever one offers as a basis for one's claims, or the act of offering it (if offered as a serious attempt to tie one's claims to some basis in reality). Variations like "justify" and "justified" simply mean to offer a basis, or having offered it, respectively.

Such a definition allows one to recognize when evidence or argument has been offered, but tells us almost nothing about whether or not the evidence and/or argument objectively validates the claim. In other words, "justification" can be valid or invalid, while still qualifying as a justification in either case. Having a justification doesn't necessarily mean that the justification is valid, and one cannot determine the validity of the justification merely by checking the definition to see if what was was offered is a justification or not. Validation is an additional process, and it is precisely on differing theories of validation that academicians disagree. This is similar to defining rationality as "adherence to reason," and "reason" as "the faculty that identifies and integrates the material provided by man's senses." These definitions allows us to know clearly what we denote by the terms "rational" and "reason," but do not tell us what constitutes valid or invalid reasoning.

I mention this because the original question vaguely suggested that the validation of a justification might be sought in the definition of justification, i.e., to determine the validity of a justification, look at the definition of justification and apply the definition in any specific case.

Another example is Ayn Rand's definition of "value" as "that which one acts to gain and/or keep." After one has identified a valid standard of value, it may be tempting to redefine value as "that which one acts to gain and/or keep, in order to preserve and strengthen one's life." The problem with the latter definition is that it removes the concept of "value" from applicability to differing theories of value, requiring the use of some other concept (or expression) to denote alternative theories of value. Furthermore, identifying an objective standard of value depends crucially on an analysis of values in general, without a priori restriction to life-serving values.

answered Oct 26 '10 at 01:18

Ideas%20for%20Life's gravatar image

Ideas for Life ♦

edited Nov 03 '10 at 01:46

The term "justification" in this context seems very similar to the term "validation", which has this helpful entry in *The Ayn Rand Lexicon:*

“Validation” in the broad sense includes any process of relating mental contents to the facts of reality. Direct perception, the method of validating axioms, is one such process. “Proof” designates another type of validation. Proof is the process of deriving a conclusion logically from antecedent knowledge.

Leonard Peikoff, “The Philosophy of Objectivism” lecture series (1976), Lecture 3.

answered Oct 26 '10 at 02:34

jasoncrawford's gravatar image

jasoncrawford ♦

In my view, "justification" is a term to be avoided in epistemology. I don't think it adds anything but a difference of connotation to the existing term "validation," and it introduces potential confusion.

As others have pointed out, the etymological root of the term "justified" is "jus," which in Latin means "law," "right," or "justice." In ethics, "justification" literally means a vindication or explanation for actions that are, or might appear to be, immoral. "I'm sorry I refused to answer your questions, but I'm not allowed to discuss company business." In Christian theology, the first "justification" is baptism, which clears away original sin (so you can start loading up on all the non-original ones). In politics, "justification" means establishing that you have a right to do something. "I'm justified in forcing you to move, because you're standing on my property."

In academic epistemology, a "justified" belief is a belief that you have an epistemological "right" to hold. It is thus a very broad term, including false beliefs that one nonetheless has legitimate reason to hold. I don't think this, as such, is a problem.

The problem is that this opens the door to subjectivism. As a term directly imported from ethics, "justification" focuses on the status of the believer, not the truth or falsity of the belief. In epistemology, the primary question should always be "is X true?" not "is it okay for someone to believe X?"

Of course, "validation" and "proof" also have analogs in ethics. However, both of those are primarily terms from epistemology, whereas "justification" is primarily a term from ethics/politics.

(Among academic philosophers in the past century or two, the move toward discussing "justification theory" was an attempt to avoid (or "bracket") the question of whether one's ideas actually correspond to reality, focusing instead on whether they are "justified." I think this is actually an offshoot of John Locke's rejection of intrinsicism and "innate ideas." Locke held that an idea could not be knowledge, nor even strictly speaking a belief, until it had stood trial in your own mind like a sinner on trial before God. It's a wonderful image, and one of many intriguing ways in which Protestant theology directly influenced the secular Enlightenment. Locke's argument is mostly correct on this topic, but once the battle against intrinsicism had been won, philosophers slid right over into subjectivism.)

Leaving aside the context of academic philosophy, as long as we have the perfectly useful concept of "validation" I think the term "justification" should be avoided. Use the latter term in your classes, if you must, but not in your own thinking about philosophy.

answered Oct 26 '10 at 20:49

Robert%20Garmong's gravatar image

Robert Garmong ♦

"Justification" in academic epistemology does mean the epistemological right to hold a belief. I do not think there is anything non-objective about the term or the concept. It is, I think, a useful concept.

Others have noted that the term is introduced by way of an analogy to ethics, but that as such, it focuses on the status of the believer, not the truth or falsity of the idea. This is true as far as it goes. The primary question in epistemology should always be: is something true or false, not: is it justified or not?

But: how do we determine if a belief is true or not? Unless the belief is derived directly from observation, it requires the use of a logical method. When the answer to a question is not self-evident, but depends on a method, to answer it, we have to think about what our evidence justifies in order to determine the answer.

Note, also, that the concept "proof" is not always adequate here. Our evidence may only establish a claim with probability, not certainty. In that case, our evidence may not prove anything. But it may still support the conclusion to some degree.

When we decide that our evidence only establishes a claim with probability, we know we are not justified in holding that belief with certainty. We are justified in holding it with a degree of probability. There is such a thing as right or wrong in the way of belief. When we consider whether our beliefs are justified on the available evidence, we are considering whether we are using the policies which, in the long run, are likely to maximize our acquisition of important truths.

"Justification" is an epistemological concept that refers primarily to a fact about thinking agents, not about the truth or falsity of a belief. But the fact about agents it refers to is the extent to which an agent is using truth-conducive methods. Thinkers may have false but justified beliefs, and it's important to know this. When a belief is false but justified, we know that the thinker is on the premise of using the right methods, and will likely soon replace his false belief with a true one.

In this respect, "justification" in academic epistemology serves much the same purpose as "objective" does in Objectivist epistemology. The main difference is that it has a verb that comes with it, "justify." It's easier to say that evidence justifies a given belief, but harder to say that evidences makes objective a given belief. There is also the verb "validate," but this doesn't come with a noun form that applies to a believer. We can say that a person is justified in believing something, but not that a person is valid in believing something. "Justification" united the evaluation of an act with the evaluation of an actor, and this is something we need to be able to do. The kinds of cognitive actions people engage in reflect their cognitive policies and character, and this matters for how they will reason in the future.

There are many perverse theories of what justification is among academic epistemologists, but that is a problem with their theories of justification, not with the concept of justification as such. For the better theories, look for philosophers who describe themselves as "foundationalists," as "internalists," or as "evidentialists." (Unfortunately epistemologists don't always hold all of these views at the same time--they should.)

answered Oct 27 '10 at 11:53

valerypublius's gravatar image

valerypublius ♦


I agree with your grammatical point. Note, however, that one can use the term "justified" while rejecting "justification."

(Nov 02 '10 at 09:19) Robert Garmong ♦ Robert%20Garmong's gravatar image

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here



Answers and Comments

Share This Page:



Asked: Oct 25 '10 at 13:56

Seen: 4,781 times

Last updated: Nov 03 '10 at 01:46