Sure, I suppose a scientist's threshold for believing he'd proven something could be labeled "beyond a reasonable doubt". But finding that point doesn't seem easy. Earlier I was talking about how consensuses form, and I don't think it's possible to know when enough is enough in terms of evidence, argument and building a conceptual framework of understanding.
I will entirely agree that there is an arbitrary line between proven and unproven (beyond a reasonable doubt). That is part of the art of science, and the human aspect of science. A consensus is a recognition that a particular theory has very compelling evidence behind it, and that the theory should garner respect and attention. The consensus theory sets the bar that challengers must meet.
I've never felt comfortable with this. When I'm only willing to give my "provisional assent" to something, I'm not going to call it a fact.
I find that interesting, given that your position departs from my own. In my experience, there are theories that undergird our collection of facts. A CCD camera that measures the luminosity of a type Ia supernova does so on the basis of many different theories, and your acceptance of the data is based on the provisional acceptance that the camera and methodology will accurately measure luminosity.
Or I could be failing to understand your position and have everything wrong.