Knowledge, Belief, and Their Relation to ‘Truth’ Nicole Scott Free Inquiry

In an article titled “What Is Faith?” (FI, October/November 2021), Richard Packham suggested a tripartite breakdown of faith or belief: (a) necessary, unavoidable faith; (b) harmless faith; and (c) dangerous or stupid faith. As these labels indicate, his typology incorporates judgments or evaluations of the consequences, the utility, or, we might say, the “worthiness” of each type.

The following framework offers an alternative perspective by drawing a sharp distinction between belief (or faith) and knowledge. This distinction, which first appeared in 2013 as part of the philosophy I call pragmatic rationalism,1 utilizes impartial definitions that eschew value judgments.

This approach to defining belief and knowledge also differs from mainstream philosophy in that virtually all epistemological theories assert that knowledge is some sort of “true belief” or “justified true belief”—that is, a proposition about the world or reality that reflects what the world or reality actually is.

More precisely, all the major theories of knowledge in conventional philosophy (a) consider knowledge a form of belief, (b) depend on identifying a world of facts that are true independently of anyone’s opinion, and (c) identify knowledge as a proposition that accords with the facts of this autonomous reality. Hence, knowledge is described as a “true belief,” and each theory proposes a way to justify a belief as “true.”

Over the years, a variety of problems have been identified in this approach, mainly having to do with what counts as a valid and reliable justification; and several refinements or alternatives have been suggested—again, all concerned with devising a better way to justify a “belief” to support calling it “knowledge”—that is, all are concerned with confirming whether a belief is “true.” In addition, the glaring, insoluble problem with such a perspective is that none of us has any way of determining what “reality” is, as opposed to, and independently from, our opinions about what is.

Pragmatic rationalism’s approach avoids these problems by defining knowledge and belief not as variations of each other but as separate and distinct types of opinion, without reference to an objective reality or truth (which, in any event, is beyond human ken): belief (or faith) is an opinion held in the absence of evidence; knowledge is an opinion dependent on evidence and/or cogent argument derived from evidence. To elaborate:

Belief, as an opinion in the absence of supporting empirical evidence, is a matter of faith, not reason; hence, it can’t be proved or disproved. Knowledge is an opinion also—a working hypothesis—but it’s one for which substantial, public (or potentially public), cogent, empirical evidence exists to support it; that is, knowledge is a conclusion warranted by a preponderance of evidence and experience.

Accordingly, knowledge in the sense of infallible apprehension or grasp of reality—what is ontologically “true”—is probably not possible. Instead, knowledge seems always to be contingent: new evidence as it becomes available may further support it or may disprove it. To use philosophical jargon, knowledge is an opinion that is “epistemically justified”—it reflects, is derived from, and is in accord with known evidence.

The advantage of this approach is that it doesn’t rely on an opinion’s relationship to something called “truth” to determine whether it is knowledge. All other attempts to define what knowledge is depend on being able to take a cosmic or god’s-eye-view of a situation to adjudicate whether the “belief” that something is so really is so—e.g., is the time “truly” 1:17 p.m., or does it only appear so (e.g., the clock is broken and the hands are stuck at the position indicating the time of 1:17).

To declare something to be knowledge requires that one knows when one is right.

Because we cannot summon an omniscient judge to tell us whether a belief is really true or not, the correspondence theory of “truth” is of no help in determining a definition of knowledge. The best we can do in any situation, therefore, is fall back on the principles of probability and induction to determine what appears to be most pragmatic and rational to serve as bases for decision making and guides for behavior. Where empiricism supports an opinion, that opinion is accepted as “knowledge”; where empiricism does not support an opinion, that opinion is categorized as “belief.”

Because knowledge is not static but contingent and based on probability, and additionally can change as the evidence changes, certainty—an essential criterion for some philosophers—plays no part because certainty is beyond human reach. In short, pragmatic rationalism accepts the form of skepticism that says that we apparently can never be sure that what we know corresponds to what is really true. Pragmatic rationalism does not deny that truth or reality exists; it only doubts our ability to know it with complete confidence. What is, is; that’s reality or truth. But I cannot know what is, only what seems to be, based on empirical evidence and reason acting in concert with evidence.

Implications: The Nature of Reality and Skepticism

What is real? The short answer is I don’t know. Reality is that which is; it’s an observer-independent existent. It’s beyond my apprehension, however, because I cannot transcend my observer status. I cannot attain an objective view of reality; I’m stuck with a subjective view: I’m aware only of how things appear to me, not how things are independent of me. This perspective on the impenetrability of the nature of reality has implications for the entire scope of living but is especially relevant to epistemology (what we know and how we know it), as follows.

If knowledge is defined as “true belief,” as a belief (opinion, proposition) that corresponds to the truth of what actually is, I can never know anything because I have no way of establishing what actually is. And neither does anyone else. Accordingly, because I wish to speak of “knowing,” I define knowledge as an opinion (a proposition or set of propositions) supported by a preponderance of empirical evidence and constructed on the bases of induction and probability as those principles are discussed below.

Nevertheless, though I can’t know it precisely, I act as though something called the “real world” exists independently of me and as though I apprehend something about the probable nature of this external reality through the exercise of my five senses. Though I cannot prove that such an external world actually exists, I act as if it does because doing so is both practical and rational. I act as if something is true to get on with the day-to-day business of living.

In developing my opinions, I confine myself to those opinions I’ve labeled “knowledge.” I try never to transcend (a) empiricism—that which is based on, supported by, and demonstrable through data accessed by means of the five senses; (b) that which can be plausibly and cogently argued from these sense data by means of induction, by observing previous specific cases, and generalizing these observations to formulate regularities; and (c) probability—that which can be convincingly generalized by extrapolating from specific known instances of empirical phenomena to like cases not yet observed or otherwise experienced.

Going beyond this threefold set of parameters for knowing requires one to enter realms of belief, unprovable propositions, a priori reasoning not connected to concrete or tangible experience, metaphysical assumptions, conjecture, fabrications, mysticism, and fantasy—all considered in pragmatic rationalism to be epistemologically unreliable and misleading and therefore invalid.

Accordingly, pragmatic rationalism holds that if we can’t arrive at a satisfactory, functional, usable resolution to a matter by restricting ourselves to (a) induction, (b) probability, and (c) empiricism, then our only recourse (the recourse that is pragmatic and rational) is skepticism—that is, withholding judgment on the issue. The alternative to skepticism in such a case is faith, which is the adoption of a belief, something that’s not supported by a preponderance of evidence and experience.

Proximate and Absolute Truth

A closing note about “truth” is in order because both belief and knowledge involve opinions about what truth is or is likely to be. Pragmatic rationalism treats two kinds of truth: proximate or immediate truth and absolute truth.

We can determine who’s right and who’s wrong about certain concrete issues—for example, the number of seats in Yankee Stadium or, to use an example from the philosopher Simon Blackburn, the time of high tide—by restricting ourselves to the delimited realm in which these matters occur and not concerning ourselves with the nature of truth in a generic or more abstract sense. That is, I can believe a guardian spirit hovers at the intersection of Main and Elm, over the structure my senses tell me is a grocery store at that location. But I can’t know that a spirit is there. On the other hand, I don’t believe a grocery store is located on the corner because sufficient evidence exists for me to conclude (i.e., to know) a grocery store is there. The existence of the store in that place at this time is an immediate truth, subject to verification through sensory tests; the belief that a spirit guards it is not.

I can ask determinate questions to establish if there’s a grocery store on the corner of Main and Elm Streets. Does the city map I just purchased accurately depict the intersection of these two streets so that I can navigate to and from that location with confidence? Is the downtown bus scheduled to stop at that corner at 11:15 a.m. in front of what I perceive to be a grocery store? Can I enter the store and actually purchase groceries? Immediate truth can be confirmed by comparatively simple sensory experience: I can follow a map, buy groceries, and catch a bus or not depending on whether the information is or is not true. In other words, the test of immediate truth is its functionality, its utility, its verifiability for practical purposes. If the information I have works, if the sense-data allow me consistently to do what they promise I can do with them, then the information is “true.” It confirms for me in an immediate, proximate sense that there is a world “out there,” a reality, with which I can predictably and reliably interact.

What I can’t do with this sort of “immediate truth” is convert it to “absolute truth”—prove incontrovertibly that the world “out there” actually exists as I conceive it to exist. I can only behave as if it does as long as the predictability and reliability continue. And that’s enough for me as a pragmatist and rationalist. I’m content to live without absolute truth. I’m also happy to live without belief or faith of any sort.

Living without Faith—and with Induction and Probability

When I assert that I have nothing to do with belief of any sort in anything, some accuse me of deluding myself. But I can justify my assertion as follows.

We gain knowledge in only one way. We apprehend stimuli through our five senses, and we apply reason to the resulting sensations. The result of this process, the sense we make of what we have seen, touched, smelled, heard, and tasted, is knowledge: it is all we know and, apparently, all we can know.

Faith is belief in something we do not and perhaps cannot know. Belief is an unverified and unverifiable opinion. I can believe in the existence of Santa Claus or a god, even though I don’t know either exists, but I can’t believe in shoes, a triangle, or the color blue. The former imagined objects are not susceptible to proof, whereas the latter can be known through our senses. Because I say I reject belief of any kind and rely instead on knowledge, I plausibly claim to live without faith.

Objections are often raised at this point, claiming that no one can live without faith and that we all must believe in something even if it’s as simple as believing that when you get on a bus it will take you to a specific destination. This objection is easily refuted; that the midtown bus will take me to Times Square is not a matter of belief but of probability. This is where the bus usually goes, so I expect that today it will go there too. But I’m fully aware that it may go somewhere else today. However unlikely that is, given past performance, it’s still quite possible. Just as it’s possible that the Sun may not appear tomorrow, though it has appeared, or so I’ve been told, every day for about five billion years; there’s nothing that says it must appear tomorrow. But I’m betting that it will, and I’ll behave accordingly. Hence, I maintain that one can, and many do, live their lives without faith or belief systems of any kind.

Living without recourse to faith doesn’t mean living in a world without regularity or predictability—a chaotic world in which everything occurs randomly and without meaningful relationships or patterns. Though I don’t rely on a belief system to order the world or on any necessary connection between one fact or event and another fact or event, it nevertheless allows for regularity and predictability based on induction and probability.

Inductive reasoning is the process of forming generalizations based on observations of specific instances. To borrow from David Hume, every time I knock one cue ball into another, the second ball moves. Every time I let go of a pencil from my hand, it falls down. Every time I put my hand in a flame, my hand hurts. From observing these same things happening over and over again, I draw conclusions (general inferences): the movement of the first cue ball coming into contact with the second causes the second to move; things fall down; fire burns my hand.

I have inferred a connection or association between certain facts or events, such that I expect, based on past observations, that when similar facts or events occur in the future, the same things that occurred every other time I witnessed them will occur in concert with them. That is, the generalization I have formulated is likely to hold true; and I use my confidence in the veracity of that generalization to predict future events. Based on this probability, then, I’m able to adjust my behavior in accordance with it. In other words, I can be reasonably assured that when I knock the white cue ball into the numbered ball, the numbered ball will move. I don’t believe it will move; I calculate what it is most likely to do based on what I’ve seen it do in the past.

Note that inductive reasoning (moving from the specific to the general) is the modus operandi of science. Religious thinking, or faith, is based on deduction (moving from the general to the specific): a general statement (a rule, a law, etc.) is accepted as true on its own authority—without being based on documented evidence or independent empirical observation; in other words, it’s true because I say it’s true (or someone or something else asserts it’s true)—and then this general rule, law, or belief is applied to specific instances to explain or understand them.


1. Frank Robert Vivelo, Pragmatic Rationalism: An Introduction. Bradenton, FL: Verlaine Publishing, 2013.

In an article titled “What Is Faith?” (FI, October/November 2021), Richard Packham suggested a tripartite breakdown of faith or belief: (a) necessary, unavoidable faith; (b) harmless faith; and (c) dangerous or stupid faith. As these labels indicate, his typology incorporates judgments or evaluations of the consequences, the utility, or, we might say, the “worthiness” of …