Knowledge, Belief and Truth:
On Knowing and Understanding
How the Pragmatist Theory of Truth Explains Meaning
Stuart W. Mirsky
We all deal with the idea of knowing, of having knowledge, all the time–
from our earliest days though our school years and on to adulthood and our
careers—until we cash it in. So, what is this thing we call “knowledge”? What
is it to know something, or think we do and to be right or wrong about that as
well as about the things we suppose our knowledge is about?
So many of our concepts are like “knowledge,” too, stuff we can talk about,
believe in, but have a hard time explaining. We know things, don’t we? Who
we are, facts about history and about what we see around us. Knowledge
comes easy to us in so many ways. It defines the kinds of creatures we are.
But it’s not always easy to say what counts as something known or as the
condition of knowing it. Knowledge, as in the capacity to have it and use it,
defines us, but we can’t simply say what it is. We can talk to each other about
stuff, describe things we see or have seen, explain ourselves to others and
others to ourselves.
All of this is what we mean by “knowledge” as in having beliefs we take to
be true. But what then counts as something being true? What is it to be true
rather than false, real, not fake? Is a thing we take to be the case, i.e., to be
true, true on our say-so alone—because we believe it? Isn’t knowledge having
something more than just belief?
Knowledge can’t just be what we believe because beliefs can be true or
false and, when they’re false, they no longer count as knowledge, only as the
illusion of it—if we haven’t yet recognized their falseness. So, knowledge, as in
holding a true belief, must also be justified if it’s to be knowledge. There must
be a reason to hold it to be the case.
To be knowledge and not just a belief (something we merely think but do
not know), it must be justified. It’s not enough to think we know something.
What we know must also be arguably true and to be that it must pass some
1
test because just thinking we know something doesn't mean we do. We must
be able to offer a warrant for what we believe, a warrant to others, but not
necessarily to others in every case. We can still believe something if we cannot
supply a warrant for our belief that suffices to convince others. But surely for
a belief to be held true by us it must, in every case, rest on a warrant that we,
ourselves, are prepared to accept as justifying. Only then can we feel confident
that we really know what we think we know.
But warrants (justifying reasons) come in many different forms. I can say I
know how to ride a bike if I can jump on one and ride it (based on what I know
I’ve done in the past and continue to believe I can do in the future). I know
that Tokyo is the capital of modern Japan, though I've never been there,
because there's an abundance of references in reliable sources (sources I've
learned to trust) that tell me this is the case (as Wittgenstein reminded us in
his last work, On Certainty). The same goes for whether the world is round, or
whether a guy named Neil Armstrong once walked on the moon.
But the further away we get from sources we can all agree on, take as
reliable, the more tenuous our claims to knowledge become. Justification,
warrants, depend on other claims we are prepared to accept and they involve
other beliefs in their turn. In the end beliefs are intertwined, entangled in a
web of inferences. Believing one thing depends on believing other things,
whether it has to do with things observed through our senses in the world or
with what others have to say—or on our recollections of what others have
said.
Knowledge is not a stand-alone phenomenon but a complex one that
situates our beliefs within a body of many other beliefs.
And each belief is itself testable, in some fashion or other, if it is to be
believed, though not all tests are the same and sometimes there are delays in
our capacity to test some claim—or perhaps we are stuck with a claim that is
in fact untestable, if not in principle.
In the current political dynamic in America, there are millions of Americans
who deny, to this day, that the 2020 presidential election was legitimate, i.e.,
that it was not altered to produce a fake result by enough voter fraud, thus
disqualifying the result observed. In the absence of evidence of sufficient
widespread fraud, they argue that the outcome is simply not credible to them
so there must have been such fraud and that the fact that no one can prove
2
such fraud did not exist (the voting system in America is complex and votes
are cast anonymously with no way of tracing each vote back to the person
who cast it) means that one is free to believe such fraud must have occurred.
Many of the same Americans will argue that the January 6th protest that
became a riotous, violent assault on the Capitol building in Washington D.C.,
and on Congress meeting within its chambers, didn’t really occur as described;
or that, if it did, it wasn’t what it appeared to be, a riot that turned violent as
its participants broke through police lines and smashed through doors and
windows to enter the building and storm through its halls. Even with the
video evidence, they will tell us the videos were selectively edited or that
some of them were staged. Prove them wrong!
There are still those who deny that Donald Trump, the man whose words
directed the rioters to the Capitol building and then who failed to call a halt to
what he had set in motion, had anything to do with it. In light of what those
watching saw and heard that day, and what the documented video evidence
shows, how are we to understand such claims? Are people just lying about
what everyone with eyes and ears could see and hear on that day? Are they
lying to themselves, in which case do they really believe what they are saying?
Or just think they do? Is thinking we believe something, or telling ourselves
we do, enough?
When does belief fail the tests of truth, of conformance to the evidence?
If the evidence must, itself, be believed, to be credited as evidence, then are
we stuck in a world of made-up facts?
Can there be no actual truth after all? If all knowledge is a function of a web
of beliefs we hold and which are tied into the beliefs of those with whom we
are in community, then where do we hit bottom and uncertainty dissolves into
certainty? Can two or more different groups adhere, with equal justification,
to diametrically opposed beliefs about events and all be right—or no more
wrong than anyone else?
Many of our beliefs lie in the background, untested and maybe even
unexpressed by us but accepted, a function of our upbringing, our education,
our life experiences. All the claims we’re confronted with, to endorse or reject
or just hold in abeyance pending more information, must be fitted into that
background, an activity that occurs via logical inference (the practice of taking
one claim to imply others and others as inferable from it). Such inference is a
3
reflection of the meanings of our terms because it is meanings that link one
idea to another. Indeed, arguably it is that linkage, that inferential
connectivity, that is the meaning. Without it, there are just noises and
markings. But how does this work? Where is this meaning except within the
activity of language users connecting their words, singly or in phrases, to
other words?
And how is this to be applied to a world where we know that it's not
generally a good idea to leap off high buildings or cross heavily trafficked
streets while cars are barreling along them? If we believe we will fly on
leaving the solid ground beneath our feet or that the cars we see will
somehow fail to touch us, is that enough? Can belief be entirely dissociated
from the world which our beliefs are about?
There must be some connection to the world, right? Only some of these
connections are more rarified than others.
To believe that the world is round or that a man named Neil Armstrong
once walked on the moon depends on our buy-in to a broad array of other
beliefs, most in the background, so to speak, all of them themselves testable, at
least in theory if not in actuality.
In the end, all of our claims are themselves only testable by their "fit"
within the still broader tapestry of other background beliefs we subscribe to.
None is unchallengeable in theory but the deeper down they go, the more
difficult it is to test them—and here we can often go off the rails.
If too many of our background beliefs combine to give us a picture of the
world that fails the reality test, we can go off the deep end and yet not realize
it. Because belief is really commitment, we can commit to a picture of things
that seems absurd to others—or that will prompt us to actions with adverse
consequences, where we get smashed on the pavement below when we
choose to leap from that tall building or are crushed by a speeding car when
we fail to grasp that cars are solid objects and can hurt us when they collide
with us. Questions of whether we should jump or hold back, cross or refrain
from crossing that street, are fairly easy to agree on for most of us most of the
time though. The impacts are visible and most of us share enough beliefs in
common to know such things are not worth arguing about. But how do we
prove to folk who argue with us about things that are not so directly
ascertainable that they're wrong if we think they are?
4
Do all beliefs rest on a floating cloud of other beliefs and that’s the end of
it? If so, where's the ground? How do we ever convince anyone of anything
beyond the immediately obvious? In the end perhaps we can't convince others
at this level, not as long as they hold onto enough background beliefs that
cannot, themselves, be easily disproved (by some relatively simple, and so
undeniable observational test). In the end, proof and disproof hinge on the
openness and capacity of each believer to expose his or her beliefs to testing.
And on how deep into his or her personal world of beliefs he or she is
prepared to let the testing go.
So, willingness and willfulness matter.
But this doesn’t deny that there is testing against the reality of our
experienced world to be done. Even on this complex standard, we can test
against the world when it comes to recognizing what’s true. But it's not the
simplified and seemingly straightforward statement-to-world testing that the
correspondence theory of truth envisions. Rather, it’s belief system-to-world
testing and this depends on something else, the holistic nature of meaning, i.e.,
semantic holism which says that all meaning is a function of the entangled
network of inferences, each belief depending on other beliefs for its own
reliability.
When each claim is seen in relation to a broad panoply of other claims, it’s
this broader system that we are putting to the test. William James floated this
approach to explain the mechanics of how truth claims work as his "pragmatic
theory of truth" in the early twentieth century though he was met by severe
criticism from Bertrand Russell and others who subscribed to a form of the
correspondence theory. But, while James was somewhat imprecise in how he
formulated this strategy of explanation, and all too often fuzzy in applying it,
it's worth noting that this strategy of explaining what truth is does not deny
the connection of our words to the world. It just explains the connection
differently than the correspondence theory of truth would have it.
The connection of words to world in this pragmatist account of truth
depends on the entanglement, through the mechanism of inference, of our
claims, our assertions about the world. It rests on the notion of semantic
holism in which meaning is not seen as some parallel phenomenon,
mysteriously attached to the sounds we utter (thus making them words rather
than mere noise). It rests on how we deploy the sounds we make when we
5
utter them, evoking responses from others—and on how they respond to our
utterances.
Understanding others is about acting in expected ways in response to
utterances (or gestures or signs, of course).
But something still seems to be left out in this account and that's the
subjective apprehension we experience on hearing utterances or seeing signs.
After all, we aren't just responding automatons a la John Searle's “Chinese
Room.” Where then is the meaning in our words? Where is the content which,
when we apprehend it, converts the noise or signs into words in a language?
The answer is not that the meaning is in our words but that our words (the
thoughts they express) subsist in a web of other claims expressed in words
(or thought if unvocalized) by other speakers.
Meaning is not in each word as such but in the whole panoply of words that
make up a given language.
If you know the language, you know how to use and react to its words
when they are uttered in appropriate contexts. Our utterances, when they are
words (carrying meaning) occur within a system of interlocked thoughts,
expressible through other utterances, all bound together by an inferential
web—and the practical outcomes they motivate in the behaviors of others.
Meaning is in the whole system of words, not in each word itself, even if,
for convenience’ sake, we assign definitional meaning to words as we do in
dictionaries or in just giving explanations of what we have said. Each
utterance, noise, or sign, must occur within a framework of practices, of
actions. It is its placement within the practices that are each word’s meaning.
Meaning is a "thing" then in the broadest sense of thing. But that thing
we’re talking about is the utterance’s placement, its locus, etc., within a
complex array of other utterances bound up with our thoughts and behaviors
as part of a complex and dynamic web of activity connecting speakers to each
other. Meaning is not a ghostly element connected to a noise uttered nor is it
reducible to this or that particular logical relation of words as in the meaning
of a term is what it refers to (Frege). Meaning is identifiable with how the
system is used in toto and truth lies in its effectiveness, its capacity to guide its
users about in the world more effectively.
Truth, as such, rests on what works. But it's not individual utterances that
we test thus for their effectiveness in depicting the world. It’s combinations of
6
them, articulated or implied, each time we compare some particular statement
about something in the world to that something which we take it to be about.
It is placement in a complex array of other utterances (whether vocalized or
merely potential) that establishes meaning (making noises into words)—and
it is this complex array that we’re testing against the world every time we ask
whether this or that statement is true.
