I re-read this, in the context of subsequent contributions to this site by others (and my own reactions to these contributions). I think you might be right.
If "science" is presented falsely, we have done no one a service. If there are "non science" ways to gain knowledge, then these should also be recognized and honored.
I guess I also react strongly (see diatribe below)when I sense that science is being defined too narrowly, as in equating with cook-book technology. This trivializes the operation.
Maybe we need to recognize layers of science. Some are quite tidy, others are much less so. These latter layers might blend at the edges with other lines of inquiry (e.g. philosophy). One error may be to think that "science" is a single, unitary THING. Its a word. Maybe we want to make science appear more homogeneous than it can be if "it" is to fulfill its potential. Stumbling "science" could be science in this broadened sense. Probably, when push comes to shove, all science is "stumbling", if for no other reason than we cannot - ever - take all potential factors into account. We can simplify, can isolate, ... and do. The hooker here is that we have then stripped away the contexts within which the items/processes of our focus normally operate. This limits generalization. We can NEVER know all contexts, and their potential influences. So, we have a tradeoff.
If there is a tradeoff, then why not move back and forth between the explorative and tidy. Seems sensible to me. At this point the challenge is to convey to ourselves, and to others, where within this spectrum we think we are. This clarification of our own operational context might be what we, and others, need to make a mature judgement about what is going on. Not easy.
I am not sure this little set of thoughts gets us much further. Damned interesting topic however. More than that, its (thinking about and trying to share what we are doing in our "science{s}") is fundamental. And not a trivial challenge at all. For one thing, it is not a challenge that can be answered "scientifically", at least not in the strict sense of the term.
Weird: can't study (fully study) science with tools limited to the tools of science. [At least not with the "tidy" surface tools of everyday journal science.]