This is an interesting point. I've taken to cooking beans from dry in a pressure cooker and usually in a "meal" that gets cooked with it at the same time. So I can't discard the water. The toxins in uncooked beans are real, but does the heat of cooking destroy them? It looks like the answer is yes [1], which says that "No significant lectin changes were observed after soaking the beans" but that the lectin levels are undetectable in cooked beans. (I.e. cooking, not soaking is the important part.) That does leave your question in somewhat of a limbo: maybe the water hasn't absorbed any of the lectin and so is safe?

Interestingly, [1] says:

> It appears that cooking beans to the point where they might be considered edible is more than sufficient to destroy virtually all of the hemagglutinating activity of lectins.

But then the authors seem somewhat puzzled as to why anyone is reported symptoms of lectin poisoning - who's eating beans that are inedible in texture and hardness? They hypothesize raw food junkie's in salads.

EDIT: I'm more skeptical now that any conclusions can be drawn from this paper. They mention leaching as an important factor at one point, so maybe heat just makes leaching happen at an appreciable rate. And your uncooked water might still have lectin at levels they couldn't observe (since we're looking at it indirectly by only measuring the bean's content) and still be toxic. You only need 5 raw kidney beans to have symptoms.

[1] https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2621....