One of Friedrich's main (and most compelling) points was that in analogous situations involving human beings, activists pursue both reform and basic change.
- Amnesty International works for better conditions for prisoners on death row (reform), while also working to abolish the death penalty (basic change).
- Feminists work for driving rights for women in Saudia Arabia (reform), while also aiming to change the underlying sexist social structures (basic change).
- There used to be packed orphanages in the UK (and elsewhere) because adoption wasn't legal -- children were consigned to these places with no exit. Children's advocates worked to make orphanages kinder places (reform) while also working to legalize adoption (basic change).
- While waiting for child labor to be outright abolished (basic change), you could of course work for shorter working hours (reform).
- It's perfectly reasonable that anti-slavery abolitionists worked to stop parents and children being separated in auctions (reform), while at the same time working to abolish the whole institution of slavery (basic change).
Francione objects that reforms will satisfy the welfare-sensitive consumer, enticing her to buy more animal products, and so creating even more victims. Friedrich challenges this both on empirical and moral grounds. He says humane reforms have actually caused a decrease in demand for animal products (he cites this Kansas study*). But he wisely also objects to seeing this as a decisive issue.
To avoid speciesism, we have to keep the analogies in mind. Are we really going to reject prison reforms for fear that voters will become more comfortable with the death penalty? Are we going to oppose shorter working hours for children to avoid postponing the day when child labor is abolished? Should we let the Saudi Arabian women continue being prohibited from driving to maintain high levels of outrage about basic gender injustice in that society?
No, these choices would obviously be morally atrocious. We can't use victims of injustice, not even for the noble purpose of ending the injustice. We can't do that to human victims, so why can we do it to animal victims?**
Perhaps the problem is that Francione sees animals (unlike humans) as having just one right--the right not to be treated as a resource. A veal calf in a narrow stall who can't turn around is treated as a resource. But equally, a veal calf in a group stall who can turn around is treated as a resource. As far as rights go -- given the way Francione thinks about rights -- there's no progress there. The calf is a little less miserable, but the rights situation is unchanged.
But this is a very narrow way of thinking about rights. For one thing, the right not to be treated as a resource surely flows from a more basic right-- perhaps the right not be used solely as a means. Suppose Amnesty International decides to turn a blind eye to terrible conditions on death row, thinking this will speed abolition of the death penalty. Prisoners wouldn't be treated as a resource (nobody's going to turn them into dinner), but they would be treated (by AI) as a means. There's a rights violation in either case. It's similarly a rights violation to deliberately turn a blind eye to mistreatment of animals in order to lower consumer demand for animal products.
Even that -- the right not to be treated solely as a means -- isn't all there is to it. We think animals shouldn't be treated solely as a means because they are volitional, sentient beings with desires of their own. What could be more inconsistent with that than tying up or caging animals, making it impossible for them to move? If animals have rights, liberty rights have to be considered basic.
So there's a rights argument for animal welfare regulations after all. There's a rights argument that can hold up even if more liberty for animals meant more demand for animal products from welfare-sensitive consumers and therefore more animals raised and killed for food.
* About the Kansas study. From this summary, it looks as thought the study lumps together all kinds of "animal welfare media coverage". This could be coverage of impending welfare regulations like California Proposition 2. It could be coverage of atrocities in a meat-packing plant. If all this coverage combined lowers animal product consumption, it doesn't follow that coverage of Prop 2 type regulations alone would lower animal product consumption. That's what Francione contests -- he says Prop 2 type regulations will increase demand for animal products. This study very possibly (I need to read the whole thing) neither confirms nor disconfirms his position.
** Would rights have to be respected no matter what -- even if the demand for animal products doubled or tripled after regulations were tightened? It makes sense to think about it in terms of the human parallels. If working for a reform will double or triple the basic injustice, or postpone rectification of the basic injustice for 100 years, or some such ... well, most rights are "prima facie". Other considerations can trump the obligation to respect them and work for them. But in such cases, we've got to have good verification of a large negative effect. I don't think we have that in the case at hand.