That's a bit hyperbolic so I'm sure I'm wrong, but I have an ace: if you point me at very smart people who argued against types I'm gonna say that they weren't serious. I think it's not possible, if you have the relevant experience of working on both typed and untyped codebases of at least moderate complexity with at least one collaborator, to come away seriously believing that the untyped way is superior (unless you were forced to use a really bad typed language, I guess). And arguing that untyped languages are better without that experience is also not serious, in the sense that anyone can unseriously say anything if they don't care about being well-informed enough to be right.
That's a bit hyperbolic so I'm sure I'm wrong, but I have an ace: if you point me at very smart people who argued against types I'm gonna say that they weren't serious. I think it's not possible, if you have the relevant experience of working on both typed and untyped codebases of at least moderate complexity with at least one collaborator, to come away seriously believing that the untyped way is superior (unless you were forced to use a really bad typed language, I guess). And arguing that untyped languages are better without that experience is also not serious, in the sense that anyone can unseriously say anything if they don't care about being well-informed enough to be right.