In the early 1990s, a fellow named Ian Barbour, author of the book Myths, Models and Paradigms, delivered a series of Gifford lectures addressing the topic of 'Religion in an Age of Science'. Barbour's main concern was to find a context in which two distinct disciplines could achieve a forum for dialogue. To that end, he proposed that it would be helpful to reflect on how we arrive at conclusions about what is 'true' or 'false'; he had long contended that all assertions about reality were in fact best viewed as cognitive models or relative constructs.
To understand this, reflect for a moment on your own existence from the standpoint of perception psychology. Everything that is 'real' to you - trees, flowers, the ground, the sky, your toast - is in fact a representation, a model, of what is actually real. Your senses gather data on light emissions or air vibration or chemical constituency and transduce that information into neural impulses which are then sent to your brain for processing. Your brain takes all this data and assembles a 'picture' from it which it then uses as a 'map' to find its way around reality. Most of the time we assume that this map is pretty accurate. If you can see a baseball bat coming for your head, and feel it connecting with your head, a reasonable deduction is that you were hit in the head with a baseball bat.
Nothing's True, Nothing's False
It gets more confusing to try and figure out what's going on when we try to reach beyond the limits of our tools. This happens in quantum physics, for example, because the only way to gather information about a photon is to bounce another photon off of it, thus disrupting the photon being observed. The important thing to keep in mind here is that everything you deem to be 'real' is actually an approximation of reality suspended in the synaptic pathways of your brain. In other words, subjectivity is pervasive and objectivity is a pipe-dream. To this end, Barbour proposes we do away with quaint concepts like 'truth' and 'falsehood' and discuss instead what is meaningful. If all of reality is a perceptual 'model' then it follows that all of our ideas and internal precepts are also 'models' with more or less meaning.
Barbour described three significant modes of building and interpreting models that are common to discussion of such matters in western thinking. He strongly felt that a study of these various modes of gathering and interpreting information would aid those seeking to find a forum for dialogue between different and distinct disciplines such as 'religion' and 'science', because proponents of any opinion would recognise the value of critical rationalism, wherein their opinion is not deemed 'truthful' or 'factual' but more or less meaningful.
Logical positivists contend that only that which can be empirically verified can be said to be true, or meaningful. The empirical method involves using experimental data derived under controlled, reproducible conditions, to construct theories.
At first glance this approach appears to offer a powerful method of building models which are a very close approximation of reality. However, there are several problems with this seemingly simple and elegant approach. First and foremost it denies that subjective interpretations can be considered in any sense meaningful; a logical positivist cannot, for example, say truthfully that he loves his children, or that he likes toast. Secondly, it defines empirical methodology as an absolute, and by so doing makes the assumption that the senses by which we gather data (product of a seemingly random, or at least chaotic evolutionary process) are suitably equipped to perceive absolutes. Furthermore an assumption is made that the cognitive processes involved in developing the data into theories are able to do so in an entirely objective manner. In a sense, the process of empirical verification cannot itself be empirically verified; therefore by its own definition, it would seem that logical positivism is a meaningless stance.
Fideists contend that an absolute can be defined by their personal beliefs. Fideists commit themselves to a model of reality and then interpret data in terms of how it affirms and supports that model. A strong fideist, for example, might espouse the theory that the Bible is an infallible, divinely inspired text, and offer as evidence the fact that a verse printed in the Bible makes the claim that scripture is divinely inspired and infallible; therefore, the text is self-authenticating.
Fideism is relatively common, because it neatly sidesteps reason and logic as having any relevance to the process of determining what is true. In a sense it is thus the easiest way to assemble cognitive models of reality. The strength of fideism is that it allows one to make meaningful statements (relative to the framework of their model) with little requirement for semantic wordplay or exhaustive supportive argumentation. A fideist is quite free to love another person simply by believing that they do.
Fideism's major weakness is that its models are closed systems, and fideist models quite often tend to be more than a little arbitrary in their depiction of reality; for example, the notion (or rather, one particular notion of many) that a wrathful, jealous, loving, alpha male war-god is the supreme manifestation of the universe does not offend the sensibilities of a strong fideist. Therefore it is not terribly useful as a way of making assertions about reality that will be in any way meaningful to a person who does not share in the fideist's particular belief system.
Critical rationalists (such as Ian Barbour, who used critical rationalism to define these concepts) contend there is no absolute that can be meaningfully ascertained. Therefore, they tend to look for relative meaning in any cognitive model of reality, accepting them as models but not as truths or falsehoods. Note that a critical rationalist would find very little relative meaning in a model of reality that proposed, for example, that a pink elephant just flew by the window, unless he/she was assessing someone psychiatrically. A critical rationalist doesn't necessarily assert that there is no 'truth' or 'falsehood'; he/she merely asserts that 'truth' and 'falsehood' are relative values useful in establishing a meaningful context.
A critical rationalist sees all models of reality as having some relative value or merit as well as limitations and weaknesses. Critical rationalists tend to compare and contrast various models on a fairly constant basis. A strong critical rationalist alternates freely between logical and fideist modes of thinking, recognizing the strengths and weaknesses of each; a critical rationalist also recognises that critical rationalism is itself a model of limited value, although he/she would assess it as the model with the most flexibility of the three.
You don't know everything. You don't know anything, in fact. The best you can do is make guesses about things. If those guesses keep you from being run over in traffic, then they can be considered extremely meaningful, relevant, and otherwise accurate, but 'truth' and 'falsehood' went out the window with Heisenberg's uncertainty principle.
It is worth considering that many of the issues Barbour has mentioned stem from the fact that Western logic is founded on bivalent Greek principles. The Greeks had strong views on absolutes, eternals, and all-encompassing theorems being the foundation of ultimate reality. Philosophy was a form of geometry for them. This is the form of 'logic' with which we are all familiar. However, other civilizations have in the past developed other forms of logic, such as the trivalent model. In trivalent modes of logic there are 'maybes' as well as yesses and nos, so assertions can be made about ideas that are logical but not confined to the arbitrary imprecision of having to be considered either true or false. Trivalent logic is a useful tool for considering such issues as the wave/particle duality of light, for example, because two conflicting things can be defined as 'truths' at the same time without a breakdown of the logic involved.
The Aymaran Indian tribe had nine different words in their syllabus for 'yes' and 'no'. Their complex language offered a valency field of some 27,000 different variations of truth and falsehood as opposed to the 16 offered by the Greek logical model.