Rewilding would increase total suffering through r-selected species
A serious argument from within negative utilitarianism, and one that bears on land management decisions rather than on diet. It does not favour factory farming, which is larger in scale and better documented.
If people stop eating animals, pasture and grazing land will be abandoned and left to rewild, and the large herbivores that once used those resources will be replaced by insects, rodents and other small creatures that breed in vast numbers. Most of those offspring die within days or weeks from starvation, predation, parasites or exposure, so a single hectare of wild land can hold far more short, painful lives than the farmed animals it replaced. Anyone who cares about suffering itself, rather than about how natural it looks, has to count that cost, and it may outweigh what veganism spares.
The reasoning is that removing large herbivores frees resources exploited by small, r-selected species (insects, rodents, small prey) which reproduce in enormous numbers with most offspring dying young and unpleasantly. Wild animal welfare researchers take this seriously, and it is not a frivolous objection.
It does not reach the conclusion it is offered for. The comparison is between a speculative and unquantified effect and a documented one involving tens of billions of animals annually under conditions of severe confinement. It is an argument about ecosystem management, not about whether to buy chicken. And if the reasoning were accepted in full, it would license conclusions its proponents rarely endorse (against habitat conservation and against wilderness generally) which suggests it is being applied selectively.
Introducing a point that is not relevant to the question under discussion, drawing attention away from it rather than answering it.
How it applies here: A question about wild animal population dynamics is raised in answer to one about farming practices. The two are connected only at several removes.
Treating two things as comparable when the differences that matter to the argument are large.
How it applies here: Sets a speculative, unquantified effect against a documented one of very large scale as though the two were comparable quantities.