The notion of truth comes up in many contexts, not just philosophical, but very often a discussion can come to a grinding halt when it becomes apparent that differing understandings of the term are being employed and the dreaded question rears its genuinely ugly head: what is truth? Pilate found that no-one could provide him with an answer and, as we shall see, the answers we may give have consequences for both how we may tackle problems of wiseacring and what we are ultimately aiming at when wondering in the first place.
What is truth?
Before we can move on to consider the various theories of truth offered, we need first to look at the basic aspects any version relies on and some of the initial difficulties that turn up. To begin with, we have truth values: the truth or falsity (or otherwise) of something. Here we mean "value" in the same sense as in a sum of how many times we've given up on a debate as soon as truth was mentioned. A truth value could thus be "true" or "false", but other options exist in different logics (see the fourth article in this series) or the value could be "indeterminate"—like saying "don't know" in response to "true or false?", or even "can't know".
What kind of things can be true?
There are many options we could consider when asking what exactly has a truth value; for example, beliefs, statements, sentences, propositions and theories. These are all candidates but some have proven problematic; take sentences or beliefs, for instance. When we ask if a sentence or belief is true, do we mean a specific instance or a general one?
In the first case, we could have a remark like "Hugo is dull", which is most certainly a candidate for being true, but what if there is no-one around to make up such sentences? The same would go for a comment like "philosophers cannot fly"; it seems like the kind of thing we might want to say is always true, but if there are no philosophers or skeptics around to compose the sentence or utter the belief (or if no-one chooses to do so) then it cannot bear a truth value. Another problem potentially could be that there are (presumably) far more true sentences or beliefs than could ever be stated or written, leaving us falling well short.
In the general case, the concern lies in the very generality itself. If we want to say "I think Hugo is duller than a winter graveyard" then it may be true for some people when they take the place of the "I", as in "I, Count Duckworth-Smedley of Ditchwater, agree with the aforementioned sentiment", but not others—Hugo's mother, hopefully.
In order to get around these issues, some people have suggested instead that propositions are what we assign truth values to. One benefit of this is that several sentences in different languages all end up describing the same proposition, instead of being distinct statements, sentences or beliefs. The hope for propositions is that they express something outside of time and not dependent on the existence of human (or other) observers; thus, "the earth orbits the sun in just over 365 days" would be true (or later false, in the event of some cosmic occurrence) whether or not people exist to think of or write down this particular notion, whereas "honestly, Hugo really takes the cake" depends on the existence of several persons and their valuations.
Problems and paradoxes
Propositions are subject to some of the same criticisms as we saw above and some thinkers find them deeply problematic, but they are typically employed in discussions of truth and so we shall do likewise to get any further in our analysis. In order to do so, though, we need to give a little more thought to the distinction between what is or is not a proposition.
Usually we separate declarations, instructions and questions, saying that the former (declarative statements like "it was raining when Wilkinson dropped the winning goal") express propositions while the latter (imperative comments like "please land that drop goal" or interrogative remarks like "what was the weather like when Wilkinson kicked it?") do not. However, some famous examples of difficult declarative statements that may or may not do so have led to some disputes and paradoxes.
Suppose a proposition employs a term that does not refer to any extant circumstances. The oft-used example is Russell's "the present king of France is bald"; there is no king of France today, so can we still say that the proposition is true or false, or is it meaningless? We could find other similar instances, like the properties of (supposedly) mythical creatures: can the propositions "griffins have bad tempers" and "pixies are helpful" have truth values? Some thinkers argued that these do not give us propositions, while Russell thought that they did—only false ones.
Another issue studied by Russell (and others) concerned the liar paradox and similar "liar sentences" in general. If we consider the proposition "I am lying to you", we run into difficulties when we try to assign a truth value: if true, it would imply that the speaker is lying about lying and therefore telling the truth, rendering the statement false; if false, it means that he or she is in fact telling the truth, which would imply that he or she is lying to us, rendering the statement true. This is a paradox that seems to trap us every way we look at it.
Although much ink has been spilled on this issue, one way around it that we have already touched on is simply to say that the original statement was not a proposition after all. Another could be to note when paradoxes come up but still use the approach where it works for the vast majority of cases.
A third area to look at concerns ethical or aesthetic statements, such as "it is wrong to use animals in medical research" or "Beethoven's symphonies are more valuable than Elvis' work". Do these express propositions? Can we assign truth values? Some thinkers have argued in the negative, insofar as such remarks merely tell us the opinions of the person saying so. On the other hand, those who consider that ethical or aesthetic values can be determined in some way (for example, moral realists—as we saw in the sixth article and will touch on again soon) suggest to the contrary; after all, if we can decide what is right and wrong (to also be discussed in the next piece) then it will be a relatively simple matter to compose ethical propositions that are true or false.
Lastly, what of propositions concerning those events that may or may not have happened in the past or might happen in the future? We glanced at this matter in the fifth article of our series, but for now we could consider the proposition "Caesar sneezed when crossing the Rubicon", or something similar: what can we say about its truth value? It's trivially apparent that if he did sneeze then it's true, or false if he didn't; what, though, if we don't have access to the information to help us decide? Short of a time machine, it's hard to see how we could come by anything to help us.
Much the same occurs when talking about the future: take the proposition "Hugo will be more interesting tomorrow than he is today"; how can we know whether this will be the case or not? It certainly seems unlikely, but that doesn't help us assign a truth value—unless we are happy with "indeterminate" or something along those lines.
One way around this is to note the way we actually reason about future events. We might say: if it rains tomorrow, we had better do the washing today; moreover, we have reason to believe that it will rain—a weather forecast, for instance. If someone then asks "why are you doing the washing today?" we could reply with the proposition "[because] it will rain tomorrow". This seems like a valid argument (and one that we use often enough in a similar form) but the proposition may not have a determinate truth value; it seems folly, though, to discard the argument on these grounds.
Forms of truth
So far we have discussed truth values and what may be true (or false), but what do we mean by truth itself? There have been many versions of truth put forward, each subject to critique in general terms or in favour of another that purports to address these shortcomings. We cannot cover them all, but in this section we'll take some possibilities and consider their strengths and weaknesses, trying to understand why we might choose one or more of them.
The correspondence theory
Many people use a form of the correspondence theory when speaking of truth: a proposition is true if it corresponds to the facts, or reality, or how things actually are. There is a subtle difference here, though: are "facts" and "how things actually are" the same thing? If we look at how we would state that a proposition is false, we can see the distinction: in the first case, the proposition would be false if it does not correspond to any fact; in the second, it would be false if it corresponds to how things are not. In the latter, then, we have a kind of comparison to something that doesn't exist—the way things are not.
Although there is some dispute on this issue, we won't consider it in any further depth here. What, though, are facts? Looking back to our third article, we could ask what the ontological status of facts is supposed to be: take a proposition, say, like "England beat Australia in the 2003 World Cup Final"; is this a fact? If so, is it a fact consisting in a fact? If so, does it correspond to the same fact as the proposition "Australia were beaten by England..."? If so, does it correspond to the facts that "[team x] were not beaten by England", for any other team x? And so on. When we consider instead "the ways things are", we have other difficulties. If we say that a proposition corresponds to "the ways actually are", surely there is only one way, to which all propositions must correspond. This doesn't appear to say much beyond a triviality.
Another objection could be to question the nature of the correspondence relation itself, which seems mysterious. It hardly seems likely that the words in a proposition correspond to the facts, but rather the entire proposition does. In and of itself, the correspondence theory doesn't appear to say much until we expand on it and note what it means: in our example, we want to say both that the propositions means what it says and that England actually did beat Australia. Lastly, is the correspondence theory true itself and, if so, what does it correspond to?
The semantic theory
In 1944, Alfred Tarski proposed his semantic theory as a successor to the correspondence theory, expanding on it somewhat but dropping the problematic concepts of facts and correspondence. He suggested that a proposition is true if and only if a claim about the world holds. Thus, the proposition "Hugo is dull" is true if, in fact, Hugo really is dull; conversely, if Hugo is dull then the proposition "Hugo is dull" is true. More generally, we have "p is true if and only if p", where p represents some proposition. A similar rendering would apply to falsity.
This is an improvement on the correspondence theory because we can write the condition as follows:
- The proposition ("Hugo is dull") is true (1)
- if and only if (2)
- Hugo is dull. (3)
In this layout, only (1) is talking about truth, while (3) is a claim (which may or may not be accurate) about the world; any reference to facts or correspondence is gone. Note that we are not
saying "Hugo is dull if and only if Hugo is dull", which is trivially so (a tautology) but says nothing about truth. Tarski was concerned to separate what he called the object language
(the part in quotations, describing the object of discussion) and the metalanguage
(the rest of the sentence, containing the object).
The semantic theory is a good deal more complex and involved than this sketch explains, but one of the other issues considered is that of contingent
truths: the former are those that may or may not be true, while the latter are necessarily
true. Examples could be the proposition we looked at before, "Australia were beaten by England..." which may or may not have been so, contrasted with "twice two is four" which must
be (leaving aside certain special number systems). Can the semantic theory account for non-contingent truths, which appear to be true by definition?
It seems that a distinction between these two may be drawn insofar as we can imagine a world in which Australia beat England (for instance, one in which Wilkinson was born in Perth), but twice two seems the kind of thing that must be four in any
world—or all possible worlds
, as the terminology often goes. How we go about finding out whether a proposition is true in each case differs slightly (i.e. we need not appeal to experience to justify mathematics, or so some thinkers say) but it does not follow that there is a differing form of truth at work. Moreover, the semantic theory is not telling us anything about how
to go about such things, but only that the success of non-contingent truths like those of mathematics or logic may be due to their accurately describing our world.
The coherence theory
The main contender to the semantic theory of Tarski is the coherence theory
. In general terms, the theory says that a proposition is true if it coheres
(or agrees) with other propositions we already hold to be true.
The easiest way to appreciate what this means is to consider an example: suppose a person drops an expensive vase when browsing in an antique shop and is asked to pay for it. Instead, the person offers the explanation or proposition "I dropped it because an African elephant knocked it from my grasp, since we were arguing over who should buy it". Why might we not accept this story?
- African elephants are not known to talk.
- African elephants are not known to be patrons of antique shops.
- African elephants are not found in this part of the world.
- No elephant was known to be within a certain number of kilometres of the shop.
- No-one else in the shop saw the elephant.
And so on. Each item in the list is some other proposition we already
hold to be true, or approximately so. Given, then, that the person's claim conflicts
(or fails to cohere) with the set of propositions we have previously accepted, we reject it and call it false.
Note that this is much the same way as we usually come by knowledge, especially on a day-to-day basis. Moreover, this has nothing at all to do with Ockham's Razor
or the likelihood of different explanations; indeed, it seems that when people appeal to Ockham they are usually employing a coherence theory instead.
To make the coherence theory general we say that "a proposition is true if and only if it coheres with x
"; the problematic aspect of the theory—and the resulting critiques—come from what to use in place of x
and what we mean by "cohere". The first—and obvious—objection, though, is that a proposition may still be true (in another sense) even if it fails to cohere; perhaps the propositions we already accept are mistaken and need to be rejected in favour of the new one? This has happened very many times in the history of ideas, as we saw earlier in the series, and often we need to see common facts in a new light in order to reinterpret or discard them altogether in favour of a different approach. Insisting on a coherence theory in this way, then, would be poor methodological advice and would have halted some of the changes in our knowledge that we now tend to regard as progress.
Now we come to the question "cohere with what
?" If we answer "those things we already know" then who decides what we already know? After all, there is hardly agreement about lots of things, least of all what truth and knowledge mean in the first place. What if individual people have conflicting sets of beliefs or ideas that they hold to be true and against which they compare new propositions? For instance, the claim "Australia lost because they had a bad game" might cohere with the prior proposition "Australia are too good to lose to England unless they have a bad game" which is accepted by one person (Campese, perhaps) but not another; in that case, we would have a proposition which is true for one and false for the other, which hardly makes sense if we want to maintain the notion that a proposition is either true or false.
We could say instead that we mean cohere with the majority of people's ideas, or the judgement of experts, but why should we expect either to be a good choice? Moreover, many people believe contradictory things—like some of those we saw in the aesthetics introduction
—and the idea of coherence with a group of propositions that are themselves contradictory scarcely makes any more sense. We could try suggesting that we use those propositions that are
consistent and believed in by the largest number of people, but those people may still be wrong. Alternatively, we could say the same thing but with the caveat that the propositions accepted are those that would be arrived at when the limit of inquiry has been reached, a position put forward by Hilary Putnam; of course, we then have to decide how we know when this limit has been reached. Another approach could be to appeal to the set of propositions that we would accept if we were omnipotent, or which would be used by an omnipotent being, but we have the same difficulty. If we do not agree that this set—even if it is unattainable—is the one we must be aiming at, then we have to reject the idea that a proposition is, in the final analysis, either true or false.
Lastly, what do we mean when we use the term "cohere" in this way? We could respond that it means "agree" or "consistent with", but what then do these
mean in the context of truth? We want to avoid having to say that two propositions cohere because they may both be true together, since then we already assume the concept of truth in trying to define or explain it. In general, does the coherence theory help us to answer the question "what is truth?" or does it just give us a way to test for it?
The pragmatic theory
The principal proponents of the pragmatic theory
were William James and C.S. Pierce; according to this version of truth, a proposition is true if it is useful
to believe it. Another way of putting this is to say that those propositions that best justify what we do and help us to achieve what we are aiming at are true.
For example, many people wonder if there is a God—however they understand the idea—and are want to insult one another on the subject rather than discuss it. Instead of throwing toys and running around the playground calling each other names, though, we could say that it is useful
for some (indeed, many) people to believe in God; perhaps they want to make sense of their lives, or justify a moral code, or understand why a loved one has been lost. In this case, the proposition "God exists" would be true. Alternatively, they might choose to explain the way they live their lives or the goals they hope to attain on the basis of God's existence, much as an artist might; then, also, it could be true that God exists.
One difficulty with this conception is that not everyone finds it useful to believe in God, for whatever reason. This would mean that some people find the notion useful while others do not, rendering the proposition both true and false—a disagreeable prospect just as before. Another criticism is to note that a belief we know to be false could still be useful: for instance, we could tell a dying patient that she is going to get better, or anything at all that might help ease her passing. This renders a false belief true, which is absurd. Lastly, and as with coherence, the pragmatic theory gets us no closer to understanding what truth is.
The deflationary theory
Another version of truth having been advanced or defended by many recent thinkers is the deflationary theory
. It has many forms that differ slightly from one another, such as the disappearance, redundancy, minimalist and disquotational theories, but the basic idea is that we can deflate
the notion of truth: to say that a proposition like "Australia were beaten by England" is true is just to say "Australia were beaten by England", and no more.
Some deflationists consider that our attempts to puzzle out the nature of truth are never going to get anywhere because they are based on the assumption that such a nature exists; in fact, truth is just another piece of conceptual baggage that adds nothing to our understanding. Others say that the theory is to be favoured because it shows that a great philosophical problem can have the air taken out of it, so to speak, showing that there was no puzzle after all.
It is easy to see the appeal of the deflationary theory. When we say that the proposition "twice two is four" is true, we just mean that twice two is four—there is no need to talk about truth at all, it seems. It is also useful
, though, insofar as we can use it to make general a whole series of specific propositions. For example, suppose we want to say that the current England side will beat any opponent; to do this in propositional form, we would have to say something like "if England played France, England would win; and if England played Australia, they would win; [etc...]", which is much the same as "the proposition ‘England would beat France' is true, and
the proposition ‘England would beat Australia' is true, and
[etc...]". For some such propositions, we would be at the task for a very long time, especially if the intention was to involve an infinite number of teams (for instance, any team past or in the future).
On the contrary, the deflationary theory allows us to reduce this to the common sense (and as we would actually say it) proposition "the current England side would beat any opponent". Moreover, it tells us the total content of the proposition without having to write it all out and without needing to involve any notion of the nature of truth.
One way of formulating the deflationary theory is via a schema
- x is true if and only if y
is a name for y
; that would reduce to "y
is true if and only if y
"—for instance, if x
was "two plus two is four" and y
was "twice two is four". This would imply that we could describe falsity in a similar fashion:
- x is false if and only if x is not true; or
- x if false if and only if not x is true.
A strong objection to the deflationary theory arises from these and concerns those problematic areas we looked at earlier, particularly the possibility of propositions that lack a truth value. Take an ethical proposition, say, that does not have a truth value; that is, it is neither true nor false. In that case, following the schema above, the proposition is neither true nor not true, which is a contradiction. To avoid this we could dispense with the deflationary version of falsity, but it hardly makes sense to accept the deflationary account of truth while so doing. There are other difficulties for the deflationary theory that are still being investigated.
Some of the many additional theories under study include the revision theory
, the identity theory
and various versions of the deflationary theory. We have considered the main ones and some of the arguments for and against them, but work in this field continues to advance our understanding of what we mean by truth and what we can do with the concept, if anything.
Later in this series we'll look at postmodernism, so-called, but for now we may consider one of the points made on the subject of truth by thinkers associated with this area that has since become known as postmodern.
It has bothered some people that the notion of truth tends to have a kind of power
to it, insofar as it could be interpreted as saying "this
is true; your ideas are not". That is to say, truth is power just as much as knowledge is. This would be important politically and socially if one or more groups intended to supplant the ideas of another, or force their own on them; the sanction of truth being applied to them may make this easier, or at the least give the groups a justification for their actions that may convince them to be less than scrupulous. Alternatively, the ideas of certain experts or people of influence may have a prestige attached to them that may not be due to their merits. Thus the acceptance of what is true depends on many social and other factors; moreover, what we accept and hence becomes the consensus is what is true: if everyone believes that we are and always have been at war with Eurasia then who cares if an omnipotent being would know otherwise?
The main criticism aimed at this thinking is that it is a good deal more plausible in certain areas than others. It seems easy to agree, for instance, that a thinker with many vocal supporters who shout down or ridicule their opponents may come to enjoy a greater standing in the intellectual community than a fair appraisal of their ideas might otherwise permit, or even that areas of research were chosen at the expense of others because of factors like envy, dislike, friendship with those controlling funding, and so on, leading to a true theory being neglected for study of a false one; however, this does not imply that we cannot fly simply because jealous academics have prevented the study of superhero properties in individuals. The standing of an idea in the social sciences, say, is a lot more likely to be due to factors other than its truth than one in physics or biology.
Truth as a goal
What is the goal of our efforts to learn about our world, in whatever way we choose to do so? Are we aiming for the truth after all? Perhaps instead we could try for useful ideas, or just those that help us get by according to the notions we happen to hold at a particular time?
As we saw in our discussions of epistemology and the philosophy of science, there is by no means an agreement on this issue amongst philosophers, scientists or most other investigators. Our theories may be only approximately true, if they are not actually false anyway and shown to be so somewhere down the line. If we can in fact be content with usefulness, or theories that are adequate
for the purposes we have, then should we worry about truth or finding true
theories? This question is not easily answered and appears to depend on the valuations of the answerer.
Even if truth is problematic to define or explain, or even not really required, we still have the vague idea that some theories are better than others—closer
to the truth, whatever it is, or less wrong
. This is what we mean by truthlike
, or stating the degree
of truth rather than truth or falsity of a theory; in Popper's terminology, as we saw previously, it is called verisimilitude
Consider the problem of discovering the temperature at which water boils at sea level, along with two estimates: 105 and 150 degrees. The propositions "the boiling point is 105 degrees" and "the boiling point is 150 degrees" are both false, but it seems that this doesn't say enough; in fact, 105 is a better guess, and so to be preferred (we would think). Can we find a way to analyse this in a way that makes it a meaningful tool to use in our studies?
The approaches taken towards this issue by various thinkers have generally been too complex to go into here, but we can understand how to go about attempting to answer this question. An immediate problem would be that in order to say how far away from the truth a given suggestion is, would we not first need to know the truth itself? If so, we would have no requirement for truthlikeness anymore. On the other hand, if we do not
(and perhaps cannot
) know the truth, then how are we to measure
the difference between it and the suggestion?
Several possibilities have been considered that use mathematics to model the situation and to try to describe the notion of getting closer to the truth without necessarily knowing what it is. It was mooted in the past that a discussion of the content
of two theories could help decide which was the better, but this is now recognised to be insufficient. Much work today is ongoing in the study of truthlikeness and it appears to be demonstrating the manner in which various disciplines such as mathematics, science and philosophy are interdependent.
As we said at the beginning, truth is a concept that comes up more often than we may suppose and upon which many a philosophical ship has foundered. Although, like most important ideas, it is subject to dispute and its nature is far from clear (if, indeed, it has a nature at all), it is one of those terms that we use all the time even in everyday speech and hence is sure to be the focus of much analysis for the foreseeable future. That proposition, of course, is quite true.
Dialogue the Seventh
The Scene: Now outside The Drunken Bishop, Steven has offered to walk Jennifer home. Trystyn and Anna head off in the opposite direction.
Whereabouts do you live?
Over by the Ferris wheel. It's a long walk.
I don't mind. (He is looking at her awkwardly on occasion, trying to avoid her noticing.
(A short silence.
How do you think those two will get on?
Pretty good, I guess; she was curious about him before they even met.
Oh? How's that? Would it have anything to do with you?
I don't know what you mean...
He's my cousin, remember.
Well, he doesn't do much of anything except reading those books.
... so you decided to give him a helping hand?
I guess so, but it's up to them. She's nice girl; maybe they'll hit it off.
... and fall in love, do you think?
I don't know about that—I doubt he even believes in it.
Love? I don't follow you.
All that bunkum about "true love", I mean.
don't believe in it, then?
) I can't see any reason to, but all the same I hold out hope.
The real deal; the genuine article. I'm sure you know.
Perhaps I think I do, but that usually means I don't. How do you know when it's true
love, and when it's just love? What makes it the real deal, as you say?
I guess you just feel it; sometimes it's accompanied by the swelling of the score, if you're in a movie.
) Surely "feeling it" isn't much help, though; suppose the other person doesn't
feel it—then you'd have one person saying "this is true love" and the other saying "I assure you it isn't", or something. Those statements are contradictory, so they can't both be true.
I guess it depends on what you mean by "true", then. (He shrugs.
Probably. What do
Hmm. I'm learning not to take you philosophical types on when it comes to questions like that. What options do I have?
Well, you could say it means something that accords well with what you already know, or think you know. That way, when you say "it's true that I love you" you're saying that the love is consistent with what you already have—like respect, admiration, devotion; that kind of thing.
Okay. What else? (He is still stealing furtive glances.
You could say that truth is an agreement with the facts, whatever they might be. This time, then, you say "it's true that I love you" because the fact is that you really do love the person; that means the statement is true by virtue of agreeing with this fact.
It seems that I'd rather just say "I love you"; what's the point of the additional worry about the truth of it if I mean it and say so?
That's a possibility too, and a pretty plausible one. In that case, then, truth would have nothing to do with it.
It'd just be a rhetorical flourish.
Still, it might make you feel better, or even both of you...
Sure: if a philosophical analysis fails, it doesn't mean some mean-spirited academics are likely to turn up at your door every time you use a word they say is meaningless or flawed in whatever way.
I guess I'd be a mute if that ever happened.
) All of us, probably.
You could say that truth is determined by the circumstances, or the use you want to put something to. That's often what people have in mind, I think, especially when it comes to love.
So I'd define true love for myself?
Right. Otherwise we have the difficulty of distinguishing between love, true love and somewhere-in-between-but-not-quite-there love. What measure or test are we going to use? Take true love, which someone presumably knows the definition of, and compare yours to it to find out if you have the real deal, a close approximation or just a poor imitation—common or garden love.
It defeats the object of it and kills the notion, I think.
You're right, I'd say; people aren't talking in such terms when they say "I love you" or "it's true that I love you". Probably you have an idea in mind of what you mean when you say "this
is true love" and it becomes the truth by being in accordance with the use you have for it, or the circumstances in which you're going to employ it.
What if the significant other has a different idea in mind?
That's the problem, isn't it? Do you say something and risk the other person having a completely different idea of what you see in your relationship, or do you take a chance on it? What are you aiming at anyway? Do you have to have exactly
the same idea, or is there a compromise to be made? Perhaps your version of the truth is close enough to theirs to be compatible, and that will suffice?
It's quite a step to take, though: what if you both have completely different understandings but you talk about true love as though you're speaking the same language—when, of course, you aren't?
That's probably part of the attraction of the very idea of truth in the first place: no grey areas, or dispute, or uncertainty—this
is the truth, and none other. In the context of relationships, it's quite comforting to think that there's a match out there somewhere, that one other to complete you. It's far less romantic to suppose there are plenty of people who'll do the job.
Here we are.
Oh. Well, what do you think of all this?
We never really talk about it.
) I should've said something.
Oh. (A pause.
) Nevermind. Thanks for talking to me tonight; maybe I'll see you again sometime.
Sure; I hope so.
Goodnight. (He turns and walks away quickly.
By Paul Newall (2005)