Alan Kay uses an interesting analogy to explain the distinction between "Computer Science" and "Software Engineering":
UCLA has one computer science department, but 25 full departments of biology (not counting medical school stuff). Why? Biologists are smarter then we are. When things are bogging down, the best thing to do is to go create a new department. (http://www.windley.com/cgi-bin/printthis.pl?url=http://www.w...)
I think the pharmacy:"software engineering"::biology:"computer science" metaphor is pretty insightful. So to answer your question - people discount a formal education in CS because it isn't necessary to write good code, just like pharmacists don't need the formal education that research biologists go through.
A lot of disciplines are organized into strata, where once you are on a particular stratum in terms of what you know, you can get a lot done without learning new things. The truly new ideas tend to occur in corner cases or logical extremes, and generally people discount such things as being too focused on insignificant details. However, these corner cases often lead to new strata. I suspect corner cases define the history of ideas.
So, when working on practical problems, since they are the general case, it is hard to see the value of special, stratifying issues.
