I also find it quite strange. I remember my teachers telling that we're making a little bit of progress in software engineering research, but comparing it to the state of research in fields like Math or Physics would be like comparing a fully grown man and a toddler.
We may hardly know anything, but, comparing any field that involves humans to math or physics is an apples to oranges comparison. All it takes to do math research is time, knowledge, writing material, and, preferably, the support of an institution which allows one to do research full time. There are no pesky experiments to run, except possibly inside a computer. Physics can be the same way, but, even when it involves "pesky experiments," those experiments only depend on the physical laws of the universe, and the labor of a fleet of graduate students who have great incentives to see these experiments through. SWE research has either of these properties.
In other words, we hardly know anything