The relevance of thought must be obtained (and sometimes wrested) from the circumstance we exist in.
We could then best describe thought as distillation than as searching. To talk about the inner working of thought, we must deal in heuristics (as opposed to, say, universality or validation or platonic enlightenment). Heuristics are the mechanics of circumstances. Heuristics are circumstance-navigation. What does that say to us about knowledge?
Heuristics are hoardable, while at the same time having big discrepancies in power. This means you can keep on acquiring new heuristics and that this is, generally, a good thing, but, at the same time, a new heuristic can often render a big hoard of previous heuristics useless. Like sometimes a cabinet-ethnographer, who has never been to a distant tribe environment, can come up with a new way to understand culture even if he hasn’t ever collected any data on cultures. This is also the relationship between describing butterfly species and describing a general theory of evolution. Hoarding against diversifying, both viable and most likely to be dependent upon each other.
That’s because heuristics can be incompatible, but they can’t be contradictory. In other words, heuristics are non-exclusive. Seemingly opposed heuristics can actually maximize each other’s efficacy when used serially. For example, an heuristic of «search for reasons in the small scale» (similar to reductionism) can be used together with another one of «search for reasons in the large scale» (like holism). Two stances that as ideologies actually fight are good working partners as heuristics. I suspect this is why sometimes people can harmoniously harbour seemingly incompatible ideas.
Heuristics are independent of proof, in fact, proof is one heuristic amongst many. Heuristics are never right or wrong, they just have specific effects in specific circumstances. Plus, the same heuristic can have very different effects in different circumstances, so naive maximization of heuristics (like trying to make heuristics always right or universal or un-leaky) is prone to cause trouble. Heuristics don’t and can’t take the circumstance for granted.
Like any skill, heuristics can and usually must be trained and taught. Taken far enough, the heuristic-hoarding exceeds individual capacity. But this raises the issue of group-dynamics, which is hairy. Add to it the beneficial effects of good heuristic-imparting (that is, teaching) into a group with intense inter-individual competition, and you get complicated, risky dynamics around the heuristics available to the individual and to the group.
A good effect of that is the development of meta-heuristics. Heuristics to apply heuristics. In a certain sense, the individual also needs meta-heuristics, but will usually do it as little as possible, and always subjected to the basal heuristic (“between me and you, me”). As we move to a group effort we can go meta on the meta, an thus complexity exponentiates. The initial inertia of the process is big, but exponentials always beat hoarding in the long run.
There you have all the fixation of Old-Greeks with Logics. Many of the confusions in thought and in the over-valuation of reason come from confusing the study of heuristics with the study of meta-heuristics. Exclusion of the middle is a meta-heuristic, when used as a plain heuristic it leads to silly.
Every sentence, every number, every equation, every ethics discourse is just a temporary abstraction that derives the totality of its value from the circumstance where it exists. Our intelligence then extracts relevance from circumstance, without ever depending on circumstance being “true” or “stable” or “universal” or “unbiased”. Avoiding this dependence is actually getting rid of a vulnerability.
The most important point, though, is that thought is grounded by people. It is people that think, and thus it is people that make thinking relevant and reliable. This of course leads to no idea is completely reliable, but we arrive at this not because we gave up on anything, but because we have better studied the process of thought itself.
Thinking is a very powerful tool. Thinking about thinking is another powerful tool. And through this, we become what we are.