It is not necessarily clear to me that monkey patching (edit: or bit twiddling) is appropriate either. You can insert an expression cassette for a transgene int

It is not necessarily clear to me that monkey patching (edit: or bit twiddling) is appropriate either. You can insert an expression cassette for a transgene into a plant, predicting that the gene will be expressed. You can then test that prediction, detecting the transgene mRNA. Then you can predict that the mRNA will be translated into protein. You can test that prediction by performing a Western blot. You can then predict a phenotype, for example roundup resistance. You can then perform all sorts of other tests to see if the nutritional content of the plant products are detectably different from the parent strain, and if feeding the plant products to rats results in unhealthy rats. To me, that is very different from making random changes. I’m not trying to make a value judgement there about whether we should do all that. I can imagine someone approaching this from the precautionary principle holding the position that there is some currently undetectable, unwanted, and important change occurring in the plant. If you make decisions solely based on that, you would effectively never get any potential benefits from GMOs. Again, to some people the potential political/social/economic risks outweigh the benefits. I just think people should be clear about the rationale for their position. (Edit: grammar and clarity)