I have no compelling reason to improve on Daniel Bell's … own definition of knowledge: "Knowledge: a set of organized statements of facts or ideas, presenting a reasoned judgment or an experimental result, which is transmitted to others through some communication medium in some systematic form. Thus, I distinguish knowledge from news and entertainment." (17)If I'm reading Castells correctly, then for him, knowledge is based totally on a formal subset of language: a set of organized statements that is grounded in reason or experimental result and that is transmittable or communicable. As I said in my posts about James Berlin's book Rhetoric and Reality, I am uncomfortable with limiting knowledge to language, especially in Bell's case, to a mere subset of language. I think that all sentient beings can learn about their environments, and the things they learn are knowledge and the things they can learn are knowable, even if those knowledges are not expressed, or even not expressible, in language.
I'll give a quick example from my own experience. I have coached youth soccer teams for fifteen years, and I have tried to teach many children to kick a soccer ball. Very little of what I said—or transmitted to these players through a set of organized statements—about kicking a soccer ball did any good. Rather, I was most successful when I allowed them ample opportunity to kick the soccer ball. Through the trial and error of repeated attempts, most of them learned how to kick a soccer ball. I could say of a player that she knew how to kick a ball, or that he did not know how to kick a ball. This was real knowledge, but hardly communicated through a set of organized statements, hardly even expressible through a set of organized, reasoned statements—knowledge nonetheless.