So, if any definition of truth is necessarily circular (my definition is true because it meets my definition of truth) we have a bit of a problem in defining the term. I'm taking a roundabout route to grappling with it. First, I said that the medium of truth is language. Now I'm going to talk about how humans actually use and interact with potential truth. I'm not talking about different "kinds" of truth, just different uses for potentially true statements. Then I'll talk about some properties truth must have in order to actually be useful. And as a sidenote, all through this post I am talking about potentially true statements, even if I don't say it. I use "belief" as a synonym for "an accepted, potentially true statement."
I think there are two things humans use truth for. One is interaction with the physical world; both manipulating it through science and technology and explaining it. We not only want to know how to generate electricity, we want to know what electricity is.
Systematic observation and reason (ie, science, informal logic, etc) are the best tools we have to develop and apply truthful statements. We strive for simplicity and explanatory power in order to connect the dots of our world (As a sidenote and an addendum to the "Webs" post, "simplicity" doesn't necessarily mean "easy to understand").
When interacting with the physical world, it can be very easy to discover that statements are untrue. False statements that attempt to interact with the world fall apart and become useless when actually applied. Potentially true statements that are false are are eventually exposed.
Concerning the second use of truth, each of us holds certain propositions that we believe are really, truly important. These beliefs are the ones that are the most useful in our pursuit of our imago (the "me" I am trying to be). Our imago, along with these beliefs, form our personal centre. Others might insist on using the term "religion" here; indeed, I'm borrowing the term from the theologian Paul Tillich. I don't use the word religion, however, because I think that word is better off being used for a specific human activity.
We all need to use potentially true statements to aid us in our pursuit of our imago, and we also use them to justify the value of our imago. There is a difference here from the first use of truth; unlike interactions with the physical world, the statements of our personal centre can hold only a superficial resemblance to truth and still be useful. All that is required is that the person believe these statements.
So we use potential truth for interacting with the physical world, and in pursuit of particular personal goals. What qualities must statements have in order to fulfill either of these goals? There is one common quality, and one quality that is slightly different for each goal.
One, a statement must be universal. Statements about the physical world cannot be true in one time and place and false in another. In order for a belief to be useful in one's personal centre, one must believe that the belief is unconditionally superior to all others. Exactly how one comes to believe that a certain belief is superior can be different depending on what the truth is being used for. One can either encounter something with their senses, or the belief can be "warranted." Warrant is heavily subjective and relative; it is a question of what kind of justification one requires to believe something.
Everyone has beliefs they consider superior to all others, including the most ardent "postmodernist." Discovering exactly what these beliefs are is partly a matter of introspection, partly a matter of watching your own behavior.
For example, I consider the idea that meaning does not reside solely within language to be unconditionally superior to the opposite statement. I consider my beliefs about language to be true for all people in all times.
Two caveats: belief that a statement is unconditionally superior does not mean that one insists others share the beliefs. Insisting that others share your beliefs is a matter of your imago, and sometimes a matter of a lack of power (understood as a kind of metaphysical capital). It also does not mean that the belief must be complete; in other words, one does not have to have complete knowledge to find a particular statement valuable.
The second quality is slightly different for both goals. When interacting with the physical world, statements must be as concise as possible. You don't want meaning to slide around when an engineer is working on your airliner. Statements in historical, scientific and even psychological works must be rigidly controlled by a particular context, or they will begin to lose their value in interacting with the world. For the purposes of interacting with the world, the ideal is a scientific detachment; ie, reducing things as much as possible to numbers.
On the other hand, statements in one's personal centre cannot be a simple matter of numbers and detachment. Such statements are far too cold and grey; they are useless in forming the upper layers of our value hierarchy. Even the most ardent metaphysical naturalist who insists on accepting nothing but scientific findings will repeatedly speak of being in awe of the natural world.
Now, that is not to say that precision and clarity are not always goals in one's personal centre. Everyone, even the most philosophically apathetic person, needs to believe that at least they understand themselves. Being confused by our beliefs is an uncomfortable experience, and we always seek to remedy it. It is only to point out that the beliefs in our personal centre always carry a certain passion with them that a mathematical equation does not. One can find excitement and fulfillment in their study of math, but it is their beliefs about the value of math that are useful to them, not the math itself.
It is that desire for precision and clarity that raises so much ire against so-called postmodernism, because it seems to suggest that precision and clarity are always doomed to failure. The basically academic practice of deconstruction really freaks a lot of people out, because it suggests that something seemingly vital to them is unattainable.
I would insist that an unmediated expression of truth is not necessary for one's personal centre. It is the very richness of potential meaning that language slides on that offers us the ability to have a personal centre to begin with.
I'm writing down ideas that are important to me; the fact that there is an irreducible split between my words, my intent and my audience does not reduce the value of the ideas, or even of the words. To say otherwise is a matter of one's psychology.
So: truth must be universal. Striving for clarity is necessary, but the ambiguity is necessary also; the tension is perhaps something like the id/superego relationship.
The medium of potential truth is language. The uses of potential truth are environmental and personal. The characteristics of potential truth are universality and the clarity/ambiguity tension.
Not quite a definition of truth, but I'm not sure where to go from here, so I'll just bite the bullet and accept the most common definition - correspondence with reality. The correspondence theory is compatible with my description of truth thus far.