The Meaning & Justification of Truth
A Domain Perspective
The notion of truth is vexed in many ways — ways that have been extensively debated over the centuries. This article aims at some reorientation and clarification.
One notable area of confusion arises from not clearly distinguishing between what we mean when we say a proposition is true (meaning), vs how we determine whether a proposition is true (justification). Saying that a proposition is true means that it correctly or adequately describes whatever exists. Saying that a proposition is true is justified if it is arrived at by rational methods.
All human knowledge is usefully subdivided into coherent domains, each with its own conjectures about what purportedly exists and its own methods for justification. For example, physics studies physical objects and properties by analyzing experimental evidence, biology studies organisms using controlled trials, and mathematics studies mathematical structures using axiomatic systems. Within each domain, meaning (existence) and justification (method) go hand-in-hand in circumscribing what is accepted as true in that domain.
Claims as to what exists and the accepted methods of analysis typically differ between domains. The truth that there is a smallest prime, involving the existence of a number, and the truth that the Earth is round, involving the existence of a material object, are justified in different ways in the two domains. We accept both truths, but intuitively understand that the justifications for those truths are arrived at in very different ways.
Domains of Knowledge
It seems worthwhile, if somewhat tricky, to extend the boundaries of human knowledge to other more nebulous domains as well. In fictional domains the basis for truth is typically the widely accepted narrative describing the goings-on in a story, plus common sense inferences. In our shared quotidian common-sense domain we usually rely on consensus within our perceptual and conceptual intuitions about the world. In a religious domain, accepted truths are normally derived from traditional historical accounts. In a specific political domain (e.g., liberal vs conservative view) truth is generally considered to be what is consistent with some stipulated platform of ideas.
What seems common to all such domains is that each can be thought of, at least theoretically, as based on a set of grounding assumptions (both as to what exists, and the accepted methods for analysis). The grounding assumptions are often implicit and difficult to ferret out, but their presence and acceptance is the grist for grinding out new truths in a domain.
Many problems concerning conflicting truth claims can be explained in terms of an overlap between domains. Since a domain’s coverage is ultimately governed by its grounding assumptions and their elaboration, the source of the conflict traces back either to existence claims or the methodologies. Unfortunately, both can be difficult to dissect. But at least identifying the source offers some guidance. If the grounding assumptions of each domain can be made more explicit, it can help clarify whether a conflict is based on form or substance. For example, the existence of God is usually a grounding assumption in a religion, whereas it is a claim to be established by evidence in science. Without some serious focus and agreement on grounding assumptions, it seems futile to argue the point.
As an example of merely apparent conflict, consider the domain of Newtonian mechanics. Within this domain it is true that velocities add. But within the domain of relativity, that statement is false. The conflict is resolved by recognizing that relativity supersedes Newtonian mechanics as a theory of motion. So the statement is simply deemed false in the wider contemporary context.
If all human knowledge is domain knowledge, then there is no such thing as absolute truth independent of the specific domains we talk about. (Note that this still leaves open the possibility of larger and larger domains, such as contemplated by the misnamed Theory of Everything.) The intuitive appeal of a simple, global, univocal grasp of “truth”, independent of domain perspective, is therefore illusory.
Comparing Theories of Truth
This conclusion appears to be consistent with, but also diverges somewhat, from most of the other leading theories of truth, which are not overtly domain specific. It will be convenient to refer to the above characterization of truth (by domain) as the Domain Theory for comparison with other more traditional theories of truth:
• In the Correspondence Theory, a proposition is true if it corresponds to a fact or state-of-affairs in the world. While this theory seems reasonable for physical scientific and common-sense domains, it is at least questionable for fictional and mathematical domains, where truth is derived from accepted stipulations and chosen axioms. Whether the abstract objects in mathematical domains, such as numbers and infinities, actually exist in the world, and therefore enter into facts about the world, is controversial. The Domain Theory recognizes the significance of correspondence, but it is between true propositions and postulated domain elements, rather than between true propositions and the world.
• In the Coherence Theory, a proposition is true if it is part of a whole system of interrelated true propositions. This theory comes close to the above proposed Domain Theory, if an interrelated system of propositions adequately describes the content of a domain of interest, and is based on a set of assumptions. It is the grounding assumptions that justify the truth of the whole system, instead of simple coherence. Coherence alone is insufficient to warrant truth, since a set of coherent statements can misrepresent a domain (e.g., as often used to advantage in war strategy). But the Domain Theory implies coherence.
• In the Pragmatic Theory, a proposition is true if it is part of the final explanation of reality at the end of inquiry. But it may also acknowledge the practical value of truth in the here and now. This view is closely aligned with a scientifically based correspondence view, which claims to present progressively more accurate accounts of reality over time (aiming ultimately at a Grand Unified Theory), as well as provide practical technological advantages along the way. It is somewhat more difficult to apply it to abstract domains like mathematics, which are open-ended in the sense that, according to Godel’s incompleteness theorems, some truths are not derivable from the axioms. Such truths, in principle, are beyond any inquiry.
• And finally in Tarski’s Theory, a sentence is true if the referents of its terms satisfy its predicates. For example, “snow is white” is true if the term “snow” refers to snow, and snow satisfies the predicate “is white”. This amounts to a domain independent characterization of truth, but many feel that it does not get at the underlying nature of truth itself, because the concepts of reference and satisfaction are not well understood. The same could be said of the so-called Deflationary Theory of truth, which claims that truth has no underlying nature over and above how it is used in language. Thus, a proposition ‘p’ is true, iff p. For example, ‘snow is white’ is true iff snow is white. There is then nothing else of significance to say about truth. However, Tarski’s Theory is clearly compatible with the Domain Theory, since it could apply to any domain whatsoever.
