Polonius: ... What do you read, my lord?
Hamlet: Words, words, words.
I suspect that the strings of words we use to explain and justify our behavior may be insignificant froth compared to the deep neural sources of our convictions and our actions. As split-brain and similar experiments have shown, we have a natural tendency to confabulate, to rationalize behavior after the event - and the stories we tell (and believe!) are often utter fictions.
Arguably, most of our decision-making occurs below the level of conscious awareness. As a schoolboy, I was taught to make important decisions by creating lists of fors and againsts, and weighing them up rationally. This never worked, and I thought the fault was with me. But the recommended technique was based on an utterly false view of the human brain, one that saw consciousness as all.
Intuition is an important element in certain types of complex, real-world decision-making, especially in relation to pattern-recognition and also in relation to judgements about other people. The trouble is that, though intuitions can usefully draw on sub-conscious processing of complex data, they also incorporate input from primitive, value-related systems of the brain which were appropriate to the sorts of quick, rough-and-ready judgements required by our ancestors, but are inappropriate to people living in technologically-advanced societies.
Of course, language affects cognitive processing (conscious and unconscious) in complex and subtle ways. But it is a mistake to think that the structure of language - its concepts and categories and so on - reflects the underlying nature of reality. The most one could claim in this regard is that it enhanced our ancestors' ability to deal with their environment in a practical way. And language itself is a key component in the social environment which makes human intelligence possible.
Natural language is good for many things - for sharing knowledge and enabling complex group operations, for inspiring and encouraging, for misleading and discouraging, for cementing social bonds, for fomenting rebellion, etc. But one thing it is not good for (except as an adjunct to scientific methods) is discovering new truths.
Metaphysics was the ultimate armchair discipline. Throughout the 19th and early 20th centuries metaphysicians in Europe and America created vast linguistic and conceptual structures which purported to explain the underlying nature of reality and much else besides. But they were blinded by the structure of language and their own rhetorical flights, and their prolix works now gather dust in library basements and warehouses.
Of course, there have always been those, of an empirical cast of mind, who saw little value in metaphysics, but perhaps the most formidable attack on this style of thinking was mounted in the 1920s and 1930s by the so-called logical positivists. They sought to discredit metaphysics entirely and had a huge impact in the middle decades of the 20th century.
In the latter part of the century, however, certain philosophers tried - with some success - not only to attack logical positivism but also to rehabilitate metaphysics. They claimed, for example, that the logical positivists' anti-metaphysics campaign was flawed and indeed ultimately metaphysical itself.
In general terms, though, I think the logical positivists were right - science (broadly understood so as to incorporate the historical disciplines) is the only way to push back the frontiers of knowledge. Not only do metaphysics and philosophy lack the empirical dimension of science, they are generally pursued via the medium of natural language, whereas the underlying structure of the world is amenable only to approaches which are heavily reliant on quantitative and mathematical methods.
But science has little or nothing to say about right and wrong, or about political and social ideals. By contrast, natural language might be seen to be ideally suited to deal with such issues. Perhaps, but caution should be exercised in this area. The trouble (as I see it) is that value-based political and ethical theories proliferate, not unlike the metaphysical theories of the 19th century, without any way to test them and weed out the rubbish.
I am inclined to think (like Wittgenstein) that value-related issues are real, and they can of course be discussed, but they are not the sorts of things one needs to have theories about.
Value systems are real and inescapable aspects of social relations, built into the manners and customs and expectations of any society. And, of course, individuals and groups may and do challenge certain conventions, but always within the context of a complex network incorporating many other conventions. (Such a view seems to me compatible with virtue-based approaches to ethics, by the way.)
This process of review and revision happens. Conservatives may prefer that it happen more slowly, liberals may prefer that it happen more quickly, but it is an inevitable part of the life of society.
And though language constitutes a key element in the social environment and so profoundly affects cognition, and though words may inspire, instruct, clarify, warn and all the rest, the ultimate driving forces of human existence will always be wider and deeper than language, in the still obscure dynamics of social and somatic processes.