The Gemara (2a-4a) discusses the impact and the scope of doing one of the 4 avodot ha’dam with the intent of another sacrifice (or another owner). Some issues that emerge include:
- The sacrifice still has its full prior status, so if one slaughtered an olah with the intent of shelamim, the korban is still fully an olah (although the owners do not fulfill their obligation). Given that it is a full-fledged olah, it is forbidden for the Kohen to do another avoda with the intent of a shelamim, even if the first avoda was done that way. (2a-2b)
- According to R. Shimon (the pshat of his statement, and the way Rabbah understands it), if the Kohen’s intent is obviously wrong – i.e., can in no way be reffering to the sacrifice in front of him, then the korban is kosher (2b-3a).
- According to Rava and Rav Ashi, however, R. Shimon’s statement is to be understood differently, and the opposite is true – the more obviously wrong the statement is, the more invalidating it should be (3b).
- The Gemara in the opening sugya of Zevachim (2b) states that if a Kohen did the avodot of a korban with no intent the korban would actually be valid, and R. Yossi states that it is prefferable, so that a Kohen should not accidently think the wrong thing. The reason having no intent works here, and not in other cases – such as the need to write a get specifically for woman who is being divorced – is because of the principle staman lishman omdin, they are “naturally” designated for their purpose. Tosafot (ad. loc., s.v., Zevachim and Stam) assumes that this means that when working with a korban, the circumstances and context can normally be assumed to shape what the Kohen is thinking, unless we know otherwise, but that this assumption cannot be made with a get, because it is always possible that the husband or sofer will use another get.
I believe that all these issues revolve around the question of why shelo li’shma creates invalidity, or detachment from the owner. There are two ways of framing the problem:
(A) Since the Torah requires that the Kohen have the proper intent, when this requirement is violated, the violation impacts – to a greater or lesser extent – the status of the korban or
(B) When the Kohen does an avoda having a different korban in mind – when he slaughters an olah with the intent of a shelamim, then he has defined his act as a shelamim ritual, and thus – by extension – has to some degree redefined his olah to no longer be fully an olah.
This is actually a phenomenon that we find by korbanot, that under certain circumstances, an avodah done with a certain intent can redefine the korban. So, for example, if an asham that could no longer be used was slaughtered with the intent of an olah, it becomes an olah, and if a korban Pesach is slaughtered during the rest of the year as a shelamim, it becomes a shelamim. This concept is known as “akirat shem,” uprooting and redefining the “name,” that is, the prior status.
When the Gemara insists (point 1) that the animal is still fully an olah and it is not acceptable to do another avodah with the intent of a shelamim, the Gemara is embracing approach A. No matter what the Kohen did, the identity of the Korban does not change. [The halakha could still be true according to B, because the korban is still primarily an olah, but the Gemara’s language is more in line with A.]
The debate about whether a patently false intent is better or worse (points 2-3), is best understood as a debate exactly about approach A or approach B. If the problem is the violation of the Torah’s demands – approach A – then a more blatant violation is worse. If, however, the problem is that the thought redefines the act, and also – to some degree – redefines the korban – then if the thought is totally absurd, it cannot redefine the act. The physical reality – that a Kohen is taking the kometz from a deep-pan mincha and not a dry mincha – defines the act as a kometz of a deep-pan mincha, much more powerfully than the thoughts or words of the Kohen to the contrary.
Finally, when it comes to staman lishma (point 4), Tosafot’s approach emphasizes our concern with what the Kohen is thinking – approach A. However, a simpler explanation is that as long as one does not have contrary intent, the act is defined by the very reality itself. The slaughtering of an animal sanctified as an olah is the slaughtering of an olah, even if the Kohen’s mind is blank. Only if the Kohen redefines it will the act be considered something else. This seems to be Rashi’s understanding (Zevachim 2b, s.v. biStaman lishman omdim). In the case of a get, in contrast, since the paper has not been “sanctified” and has no prior identity, the act of writing cannot define itself to be for the sake of a particular woman (since another woman might have the same name). Only the intent of the sofer can define the nature of the act.