In a short article "Why vegetarians should be prepared to bend their own rules," Alberto Giubilini claims: "If people perceive vegetarianism as a position that allows for no exception, they are probably less likely to become vegetarian. A flexible moral position is more appealing than a rigid one that allows for no exceptions."
The word "probably" raises questions. Can't we find psycho-social data about whether "flexible" positions are indeed more appealing than "rigid" ones overall and also specifically in regards to choosing a personal diet to support animal welfare? The result could also vary based on the relevant knowledge base, cultural context, and personality of the individual being exposed to the new moral idea. Many people appreciate flexibility to support an abstract goal like "reduce the consumer demand placed on the meat industry" which could allow for eating meat in certain circumstances, but others are more attracted to moral commitments that are simple, concrete, and inflexible like "never eat an animal for any reason." Cultural context is important here because it drives assumptions. Some people are more familiar with vegetarianism as a result of religious commitment, family tradition, or poverty, while others know it mainly as a lifestyle choice prompted by modern concerns about environment and health, and still others know it as a more universal, timeless ethical issue. Dietary flexibility isn't clearly always the approach that will be most readily understandable to others. (Exactly what is a "flexible" religious commitment or a "flexible" moral imperative, after all?) Nor is it obvious that dietary flexibility, even if understood, is the diet that is most likely to be imitated.
Giubilini makes a good point that more animals overall may be helped if a larger number of humans reduce their meat intake rather than a smaller number of humans completely eliminating their meat intake. Indeed, if three chickens have their heads on the chopping block, the chickens care not whether they are to be eaten by a single human or three humans. Therefore, convincing three people to eat one fewer chicken each is surely the same (to the chickens) as convincing one person to eat three fewer chickens. On that point, Giubilini has got it right.
However, as there are different kinds of existing vegetarians and they will convert different kinds of new vegetarians, it isn't necessary to make an "either/or" choice about how everyone should eat. Different behaviors and justification will appeal to different people. Some individuals will reduce their own meat consumption and some will completely eliminate it. A "numbers game" of animals spared suffering is quite valid from a consequentialist point of view, but we still need to see those numbers so we can accurately predict the consequences. If I care about this consequentialist approach, then what I need to know is the precise likelihood that my willingness to eat chicken publicly will inspire a greater increase in other people's future vegetarian meal choices than if I had refused to eat it. Saying "probably" doesn't cut it. Data is needed.
A couple further objections:
If non-vegetarians are most inspired to reduce their meat intake when they see a self-identified "vegetarian" eating meat because they perceive that person as "flexible," isn't the logical consequence that all self-identified "vegetarians" will always eat meat when they are being watched? And wouldn't that ultimately result in a large number of animals being eaten?
The assumption that meatless meals should be consumed in secret so that the effort that goes into making them is kept secret, while meat should be consumed in public so that eating meat-free seems easy, needs to be unpacked. Growing or buying food, cooking it, and serving it always requires effort. As long as meatless options are available, they are easier to prepare because they generally require less cooking and refrigeration for food safety. They are sometimes also cheaper. Eating meat publicly is not a demonstration of simple lifestyle or accessible ethics any more than eating meatless food publicly.
Giubilini may have intended to restrict his argument to the scenario in which a self-identified vegetarian is offered meat in someone else's home, but this restriction is not made clear within the article, as the wide-ranging justifications could easily apply to other scenarios.