What is Truth?

No word is perhaps so abused in our modern world as is the word truth. Politicians bend it to suit their agendas while they pound their podiums, news columnists promise to reveal it while they trade in gossip, and religious groups praise it as their central concern while they redefine it to be synonymous with charismatic personalities and inspiring texts. Like it is to the fittingly-named Ministry of Truth in George Orwell’s 1984, truth may sometimes appear as little more than a label used to give the illusion of certitude to partisan propaganda. It should come as no surprise, then, that a large majority of Americans – even those identifying as Christians – do not believe that absolute truth exists.1

But what do we make of things that really seem independent of our experience? There certainly are truths that are true for some of us and not for others, like being born in New York City, being six feet tall, having three children, and being of Chinese ancestry. Would it make sense, though, to suggest that it can be true for you that Abraham Lincoln gave his famous Gettysburg Address on November 19th of 1863, yet it’s untrue for me? It rather seems that one of us must be wrong. How we determine what is true, how we understand the nature of truth, and how we see its role in belief and knowledge are questions that date back to the earliest eras of philosophy, and bear upon many subjects of concern and debate still today.

In the preceding three articles of this series, we have looked at the questions of why we are here, what happens after we die, and why we should be good. Although it would have perhaps been more orderly to begin the series with this article, I chose to start with questions on meaning and morality because I believe they tend to be bigger emotional obstructions to an honest and open consideration of atheism. They also help to illustrate the subversive allure of many religious claims that emphasize personal appeal over reason and evidence. Of course, there are plenty of Christians and other theists who are quick to point out that, in their eyes, atheists often make emotional appeals, too.2 But this is just all the more reason to be clear in what we mean by truth and how we arrive at it. Not only will exact areas of disagreement be more apparent this way, but we will arguably be better able to formulate our own views and defend them as well.

A comprehensive discussion of truth, belief, and knowledge is beyond the purview of this article. Rather, the aim here is to provide an introductory overview of sorts, showing the options that are ‘on the table’, so to speak, and, by extension, the significant complexity of the subject that faces theists as well as atheists.

I. What Things Can Be True?

Before we ask what makes something true, it will be useful to consider what kinds of things can be true. To start with, let’s look at a few sample statements that make truth claims.

* Jupiter is the fifth planet from the Sun in our solar system.
* Many diseases are caused by microscopic organisms.
* Jesus is God.
* Michael thinks that Star Wars is better than Star Trek.
* The Grand Canyon is not in China.

Each of the above statements can be classed in different ways. The first can be called a fact, the second can be called a theory (the germ theory of disease), the third can be called a belief or doctrine, the fourth can be called an opinion, and the fifth can be called a proposition. However, as you may already have noticed, these terms often overlap, especially in colloquial usage. It could be said that the Grand Canyon not being located in China is a fact, or that Michael’s preference for Star Wars is a belief. Some of these statements appear to refer to concrete entities, though, or things that exist in space and time, while others appear to refer to abstract entities. For example, opinions and beliefs aren’t things that exist concretely like a table exists, they are abstractions. Jupiter, on the other hand, is something that occupies a certain position in space and time.

How do we make sense out of all these differences? For one thing, it’s not clear how a concrete entity could be held to be true. ‘Jupiter is true’ would not be a sensible statement on its own, but we might infer from it that what one means to say is that Jupiter is real. Yet the statement ‘Jupiter is real’ is a proposition about Jupiter’s existence in space-time, it is not itself a concrete entity. Likewise, the second statement about germ theory is a proposition about the cause of disease. In fact, all of the statements I’ve listed are propositions. They make truth claims that refer to things like opinions, beliefs, facts, and theories. A proposition is a kind of sentence that is constituted of objects and properties.3 In the sentence, ‘The sky is blue’, the sky object has the property of being blue-colored.

This gives us some reason to identify propositions as one sort of thing that can be true, but what about other sorts of things? As noted already, there is difficulty with making sense out of the suggestion that concrete entities can be true. More problematic is that, on such a view, truth is absent where the specified entities are absent. If one takes only concrete information to be true – writing, audio tape, video footage, etc. – then events that have not been recorded in some physical form are not truths. Likewise, if Osiris is identified as the sole truth, then truth would not exist if Osiris did not exist. Of course, Osiris worshipers might not find this a problem, but it nonetheless seems quite a rocky and impractical foundation to rest one’s understanding upon.

Propositions are abstract entities, which means that although they can describe things in space and time, they themselves do not exist in space and time. This means that the five statements above will only cease to be true if they cease to refer to anything, but they will not be non-existent, they’ll just be false propositions. Another reason to favor propositions is that they are restricted to declarative sentences. Some sentences like ‘Shut the door’ and ‘What color is my hat?’ don’t appear to be either true or false. If propositions are what can be true, they address this problem by their nature, whereas sentences themselves don’t. Thus, propositions seem to be the worthiest candidate for having truth-values because they’re sensible, comprehensive in that they often underlie or express other things like facts and opinions, and they’re more ‘permanent’ than some other things. This is, in fact, the most favored view among contemporary philosophers.4

II. What Does Truth Mean?

If propositions are the things that can be true, what does it mean when we say one of them is true?

Perhaps it means that the proposition fits well with the beliefs (propositions we take to be true) that we hold. If you believe, for example, that lakes generally contain fish, that I know how to catch fish, and that people don’t usually lie about small details, you may accept it as true when I tell you, ‘I caught a fish at the lake yesterday.’ This theory of truth is known as coherentism. One obvious strength of coherentism is that when it’s applied carefully, it reminds us to keep a consistent worldview. Sometimes we fail to realize that propositions we have accepted for different reasons actually contradict each other.

But what if someone has beliefs that don’t seem to match well with the real world? It seems perfectly possible for a person to believe many different things for different reasons, all consistent with one another, but which nonetheless fail to describe the universe as we know it. Especially when one considers the vast psychological research showing that our process of belief-formation is not often a rational one,5 it looks quite probable that we hold many beliefs for emotional or neurological reasons rather than based on how they relate to reality or cohere with the rest of our views. Coherentism’s weakness, then, is that it appears confined to working only within specific belief systems, and there is no way of justifiably arguing whether anything in a system is independent of subjective perception.

If the difficulty with coherentism is defining truth in a way that fits the world around us, what about a theory that directly pursues that aim? The correspondence theory of truth dates back to at least the time of the ancient Greeks, and defines true propositions as those which are congruent with actual states of affairs. This means that what’s true about the sentence ‘Jupiter is the fifth planet from the Sun’ is that, among other things, it corresponds to a celestial body that exists in space and time, which is the fifth such body in distance from another celestial body of a different sort, the Sun. The strength of this theory is that it associates our beliefs with reality, and may, at least, give us a shot at ‘fact-checking’ the beliefs we have come to through non-rational means. However, one weakness of it is that it does require a good deal of exposition. While many of its critics dispute the expositions that have been offered, the correspondence theory nevertheless enjoys the endorsement of most philosophers.6

An interesting third theory attempts to tie these two theories together with one additional dimension. The pragmatic theory of truth, as delineated by the philosopher and psychologist William James, holds that a true proposition is one that corresponds with reality, that coheres with our knowledge and beliefs, and also works. By this added last bit, James means that the concepts of correspondence and coherence should not be left simply as theory, but should be applied to actual practice. The reason for this addition he explains by analogy:

The popular notion is that a true idea must copy its reality. Like other popular views, this one follows the analogy of the most usual experience. Our true ideas of sensible things do indeed copy them. Shut your eyes and think of yonder clock on the wall, and you get just such a true picture or copy of its dial. But your idea of its ‘works’ (unless you are a clock-maker) is much less of a copy, yet it passes muster, for it in no way clashes with the reality. Even tho it should shrink to the mere word ‘works,’ that word still serves you truly; and when you speak of the ‘time-keeping function’ of the clock, or of its spring’s ‘elasticity,’ it is hard to see exactly what your ideas can copy.7

Thus, the pragmatic theory of truth not only endeavors to synthesize correspondence and coherentism, it tries to do so in a way that will account for a more complete picture of things than bare sensory experience seems able to account for.

As appealing as this may seem at first, James’ idea of ‘works’ is suited for epistemology rather than ontology. The fact that something serves a useful purpose or can be successfully applied in practice is a sign of truth, but it can’t be part of a definition of truth. You may recognize that it’s morning when your alarm clock goes off, yet it would be mistaken to say that morning is defined as ‘when your alarm goes off’. Something like this can also be said of coherence; it can be a sign of truth that something coheres well with other beliefs and knowledge, but it doesn’t work as a definition of truth. What makes it significant that a proposition coheres with a belief is if that belief is already strongly represented in the real world, and what makes it significant that a proposition works is if there is something strongly represented in the real world that is doing the work (this is perfectly exemplified in the placebo effect, which has deceived many pseudosciences into premature confidence).

For these reasons, as well as others not discussed here, I subscribe to the correspondence theory of truth.

III. Discovering Truth

Let’s briefly recap what we’ve defined so far. Propositions are declarative sentences that have truth-values, they can be either true or false. A true proposition is one that corresponds to the real world. Beliefs involve propositions we take to be true. In this section we’ll consider some further points of discussion, like what the ‘real world’ means, what justifies us to think our beliefs are true, and what constitutes knowledge.

First of all, what should be understood as the real world or reality? This doesn’t just mean what we observe or believe exists, but what actually exists, regardless of whether we observe it or believe in it. Likewise, reality isn’t only what currently is, but is also what has been (and may include what will be, though this is far more difficult to assess). To reductive naturalists, what actually exists is just the material universe; to supernaturalists, what actually exists extends outside the universe to alternate forms of being, like immaterial souls, angels, demons, gods, etc. Correspondence theory is a theory of truth that can be, and has been, embraced by atheists and theists alike. While this definition of reality might seem to make the correspondence theory too broad to be of use, epistemological theories can help to refine its scope.

Epistemology is the study of knowledge, often paying special attention to the question of how we come to know things. Over the centuries, many philosophers have defined knowledge as justified true belief. This means that a belief counts as knowledge when it is true and when we have rational reasons to think it is true. In other words, if you judge the proposition ‘The earth is round’ to be true, you believe that the earth is round, and if you hold that belief for the right reasons – say, you’ve seen pictures from space, or calculated its circumference like the ancient astronomer and geographer Eratosthenes – then your belief is justified and qualifies as knowledge. In 1963, however, Edmund Gettier published a now famous paper providing two counter-examples apparently showing that justified belief can be true by luck and yet not count as knowledge.8 So it seems that this longstanding definition of knowledge is incomplete, although contemporary philosophers have not come to any consensus on what exactly it’s missing.

Has Gettier thrown into doubt even the mere possibility of knowledge? This response is probably an overreaction. Gettier problems, as they are called, involve specific situations that are not identical to many other situations where we do appear able to talk about knowledge as justified true belief in a way that is sensible to us. Perhaps the lesson we take from Gettier, at least in the meantime, is to exercise caution in how we justify our beliefs and how confidently we take some things to be known.

There are two main epistemological theories on how we come to knowledge:

Empiricism is the theory that knowledge comes primarily from our sensory experience. The British philosopher John Locke argued that the mind is like a blank slate on which experience leaves its marks, producing knowledge. David Hume separated the world into ideas and facts, noting that facts are observations of the natural world, and ideas are derived from sensations, both of which are a posteriori (experiential) means to knowledge. In the early 20th century, the logical positivist movement took empiricism to an extreme, claiming that only statements which are logically or empirically verifiable are meaningful. Since the 1960s, logical positivism has been widely regarded as dead by philosophers, due largely to criticisms by W.V.O. Quine, Hilary Putnam, and others.

Rationalism is the theory that knowledge comes primarily from the use of reason. On this view, truth is determined not by sense data, but rather by intellectual and deductive means. Rene Descartes, dubbed the “father of modern philosophy”, believed that some truths can only be known through reason alone, and thus he posited cogito ergo sum (‘I think, therefore I am’) as the a priori ground of all knowledge. The 17th century philosophers Baruch Spinoza and Gottfried Leibniz both constructed elaborate metaphysical systems on the basis of reason.

In addition to these theories, there are the theories of foundationalism and coherentism, which ask a more fundamental question about how our beliefs are justified without piling up into an infinite regress. Coherentism, like its counter-part theory of truth by the same name, is the view that a belief is justified in so far as it is consistent within a set of beliefs. This admits of the same objections raised against the coherence theory of truth, though. Coherentism’s chief rival, foundationalism, is the view that a belief is justified in that it derives from more basic, foundational beliefs. The belief that dead bodies are not alive might be said to rest on the basic belief that something cannot be both itself and its opposite. One weakness with foundationalism is that the choice of a foundational belief may be arbitrary, despite what proponents might claim about appearances of reliability.

Historically, atheists tend to align themselves with the empiricist theory of knowledge, while theists have leaned more towards rationalism (though there are exceptions, to be sure). This difference can be seen quite nicely in two of the oldest arguments over the existence of god. The ontological argument is an attempted a priori proof of god’s existence, whereas the argument from evil is an a posteriori argument against an all-good god. On the opposite end, atheists have made a priori arguments that the concept of god possesses contradictory attributes, while theists have made a posteriori arguments like the argument from design. Similarly, atheists have made foundationalist claims from logical axioms and coherentist claims with other beliefs, while theists likewise claim foundations in faith and coherence of worldview. In short, neither side need be committed to embracing one theory over another, particularly when, as a great deal of philosophers have argued, there can be room for overlapping dimensions in one’s epistemology.9

IV. Putting Truth Claims on Trial

“One unerring mark of the love of truth is not entertaining any proposition with greater assurance than the proofs it is built upon will warrant.”
-John Locke

A central part of the evaluation of truth claims is considering the reasons that support them, whether those reasons come a priori or a posteriori. To believe without reason is, by definition, to hold an unjustified belief. The reasons we hold for our beliefs are often referred to as evidence. A type of theory called evidentialism argues that evidence is the sole justification for belief.

However, the nature of evidence is still a subject of fierce debate in philosophy, and even in many contexts outside of philosophy, particularly when it comes to what is sufficient or insufficient evidence for certain beliefs. According to one survey conducted by Timothy Williamson in 2000, by evidence we appear to mean simply the totality of propositions that we know.10 Another study by Earl Conee and Richard Feldman seems to show that evidence consists of the “occurrent thoughts” one has at a given time.11 Classical empiricists, on the other hand, think that only sensory data counts as evidence. Partly because of such variance, I hesitate to label myself an evidentialist; where evidence is strictly sense data, a problem of justifying the evidentialist principle seems to arise, and yet where evidence is more generally our ‘totality of propositions’, it seems hardly different enough from coherentism.

Nonetheless, we would be wise to ‘proportion our beliefs to the evidence’, as David Hume put it, even if we adopt an epistemology different from evidentialism. In fact, I think there are a few general criteria to look for that increase the reliability of evidence under any of the major epistemologies and conceptions of evidence that we have touched on.

1. Independence – the evidence in question derives from multiple sources. Although something being found in only a single source is not in itself reason to completely discount it, the more sources we have that testify to something, without copying from each other, the more reliable we may consider it to be.

2. Consistency – the evidence does not conflict with other established facts, beliefs, or theories. Things that seem out of place in the overall picture tend to indicate that something, somewhere, is wrong. While it may sometimes turn out that what’s wrong is external to the evidence, other times it can turn out that the evidence itself is unreliable.

3. Usefulness – the evidence is relevant and has predictive power. Naturally, evidence that does not logically serve to establish something cannot be evidence in support of it, but part of how we determine relevance is through prediction. If Gerald is guilty of stabbing his wife, we should expect to find the right kind of evidence showing that he did, such as a bloody knife with his fingerprints and his wife’s blood on it.

Many disputes involving religion, science, and history hinge on these three criteria. Are the four gospels independent accounts of Jesus’ life? Does acupuncture have the predictive power to be considered a legitimate science? Is the story of Atlantis consistent with other known facts about the ancient world? Though these are questions and debates for another time, they illustrate how many arguments over truth claims can come down to arguments over standards of evidence. When we evaluate a proposition’s truth, we essentially put it on trial, weighing the evidence against the counter-evidence. The better and more thorough our standards of reliability are, the better supported our conclusions will be.

V. Whose Side is Truth On?

Often times debates over topics of such intense conviction as the existence of god can devolve into little more than chest-beating displays. During these heated exchanges, it’s very common for both sides to merely assert that they alone possess the absolute truth, that truth is not on their opponent’s side, and any and all critics are simply unjustified in what they believe. It’s almost a fact of human nature, it seems, that it’s far easier to make assertions and criticize the views of others than it is to look fairly and honestly at our own views. Jesus noticed this too, according to the Gospel of Matthew, and offered the sage advice, “first take the plank out of your own eye, and then you will see clearly to remove the speck from your brother’s eye.”

What we need is some humility in how we approach the question. As this article has hopefully shown, truth is a deep, multi-faceted, and complex subject. Even while we think our beliefs are justified, there are enough unanswered questions underlying the entirety of any edifice – theories of truth, epistemological theories, theories of evidence, etc. – to call for some modesty in how we judge our own position and the positions of others. Arguments over objectivity, absolute truth, and so on, are great distractions from the fact that none of us are really that much more privileged to the truth than others that we are actually able to escape the subjective perceptions we all have. We must work with what we’re given, and what we’ve been given only allows us to postulate from where we stand, doing our best to infer beyond ourselves in whatever ways we can. But, at bottom, we are all still inferring.

I cannot, then, in good conscience assume that no one is, or cannot be, rationally justified in the religious beliefs they hold. The most I can do is evaluate individual claims made by individual people, and I think any intellectually honest theist will agree that the situation is the same from where they stand. There is no a priori reason to believe that theists or atheists cannot be justified in the positions they hold, especially when, as Gettier showed, our notion of justified true belief appears incomplete. Of course, though there is nothing preventing the possibility of a theist being justified in his beliefs, the probability of his particular case being justified remains open to debate, so too with any particular atheistic case. Thus, the question of whose side the truth is on seems to have, rather ironically, a relative answer.

But this doesn’t imply that nothing is really true in a robust sense. Perhaps it only means that truth doesn’t actually take sides because it is bigger than the scope of what can be rationally justified to our minds.


Next in the Big Questions: Why Atheism?

For more on questions of truth, belief, and knowledge, see any of the following resources:
Truth and Knowledge at the Internet Encyclopedia of Philosophy.
Truth and Epistemology at the Stanford Encyclopedia of Philosophy.
The Problems of Philosophy by Bertrand Russell.
Sense and Goodness Without God by Richard Carrier.
Theism and Explanation by Gregory Dawes.


1. B.A. Robinson, Does the American public believe in absolute truth, ReligiousTolerance.org (Aug 9, 2005). Retrieved Feb 24, 2014.
2. J. Warner Wallace, Preparing Kids to Encounter Atheism, ColdCaseChristianity.com (Sept 13, 2013). Retrieved Feb 25, 2014.
3. Bertrand Russell, The Problems of Philosophy (Barnes & Noble, 2004), p. 37-38.
4. Bradley Dowden and Norman Swartz, Truth, Internet Encyclopedia of Philosophy. Retrieved Feb 26, 2014.
5. Robert A. Burton, On Being Certain (St. Martin’s, 2008), p. 143-176.
6. The PhilPapers Foundation, Preliminary Survey results, Philpapers.org (2009). Retrieved Feb 27, 2014.
7. William James, Pragmatism’s Conception of Truth, Pragmatism (1907). Retrieved Feb 27, 2014.
8. Edmund Gettier, “Is Justified True Belief Knowledge?”, Analysis 23 (1963): 121–123.
9. For example, Immanuel Kant defended a synthesis of rationalism and empiricism, and Susan Haack has proposed “foundherentism” as a synthesis of coherentist and foundationalist epistemologies.
10. Timothy Williamson, Knowledge and Its Limits (Oxford, 2000).
11. Earl Conee and Richard Feldman, Evidentialism: Essays in Epistemology (Clarendon, 2004).