The Gemara challenges: How can one say that the very decision to partake of the fruit establishes it as fixed with regard to tithes? But isn’t it true that even if one declared his intention to eat it the following day, it can nevertheless be assumed that the remaining fruit is restored to the pile? And we have heard that Rabbi Eliezer explicitly said: Anywhere that its leftovers are restored, it is not established with regard to liability for tithes at all.

This is as we learned in a mishna in tractate Ma’asrot: One who removes olives from a vat [ma’atan] where they are temporarily stored before being pressed may dip them one by one in salt and eat without tithing them first, since he is eating them one at a time. Although he is eating them with salt, it is not considered a fixed meal. And if he dipped and placed several olives before him, such as ten, they are liable in tithes. However, Rabbi Eliezer says: One who eats from a ritually pure vat is liable to separate tithes; one who eats from a ritually impure vat is exempt because he returns the surplus to the vat.

And we discussed it: What is different in the first clause of the mishna and what is different in the latter clause; why is the issue of purity relevant to this case? And Rabbi Abbahu said: The first case is referring to a ritually pure vat and a ritually impure person, who transfers his impurity to the olives he touches. He may not return the olives to the vat because he would thereby render all the remaining olives ritually impure. Therefore, from the outset he takes only the amount he wishes to eat. This is enough to consider it a fixed meal, and he must tithe them.

However, the latter clause is referring to a ritually impure vat and a person who is ritually impure, who may return the olives to the vat, as the olives it contains are already ritually impure. He is not particular to take the exact amount he wants to eat, since he knows he may return the remaining olives, and they are therefore not considered fixed for tithes. For the purposes of this discussion, it can be seen from here that Rabbi Eliezer maintains that whenever one may restore the food, it is not considered fixed until its work is complete.

The Gemara answers: The mishna also deals with the case of a ritually pure storage area, containing pure food, and a ritually impure person, who may not return them to the vat. The Gemara challenges this answer: But aren’t they already returned? This is not a case where a person takes all the fruit and replaces what remains after his meal; rather, he takes the amount he explicitly designated the day before, while the rest remains in place.

Rather, Rav Shimi bar Ashi said: The previous explanation is to be rejected, and it should be understood as follows: Rabbi Eliezer, you said? There is no difficulty according to his approach. Rabbi Eliezer conforms to his standard line of reasoning, as he said that the separation of teruma itself establishes the work of fruit as completed, so that one may not eat it even in a casual manner without first separating the other tithes. And, all the more so, Shabbat itself establishes food as fixed with regard to tithes, as we learned in a mishna: Fruits from which teruma has been separated before their work was completed, Rabbi Eliezer prohibits eating from them in a casual manner without separating the rest of the tithes, as the teruma establishes the food as fixed; but the Rabbis permit it.

The Gemara suggests a different answer to Rava’s question as to whether Shabbat establishes an obligation to tithe food whose labor is incomplete: Come and hear a resolution from the latter clause of the mishna, which states: And the Rabbis say: Even in the Sabbatical Year, when teruma and tithes are not separated from fruit, a declaration from the day before is not enough to render the food prepared for Shabbat, unless one marks the fruit he is preparing and says explicitly: From here to there. The Gemara infers from this: The reason is that the eve of Shabbat during the Sabbatical Year is not fit for tithes; but during the other years of the Sabbatical cycle, which are fit for and obligated in tithes, the fruits are prohibited. What is the reason? Is it not because Shabbat itself establishes them as fixed? If this is the view of the Rabbis, one may not reject it in favor of a minority opinion such as Rabbi Eliezer’s.

The Gemara refutes this: No, this is no proof; there it is different. Since he said: From here to there I will eat tomorrow, he has thereby established his meal, and the reason is not due to Shabbat. The Gemara asks: If so, why discuss particularly Shabbat; even on a weekday as well the same law applies? The Gemara answers: This comes to teach us, as stated above, that untithed produce is not fundamentally muktze because it is prohibited to remove the dues and tithes on Shabbat; rather, it is considered prepared with regard to Shabbat, in that if one transgressed the words of the Sages and tithed it, it is tithed.

And we raise a contradiction from a different source, in which it is taught: If one was eating from a cluster of grapes, whose work is not completed, as grapes are designated for juice extraction, and came in from a garden, where one may eat fruit in a casual manner without separating tithes, to a courtyard, Rabbi Eliezer says: He may finish eating the cluster, as the courtyard itself does not establish the fruit with regard to tithes, if their work was not completed beforehand. Rabbi Yehoshua says: He may not finish. He maintains that a courtyard does establish the fruit as fixed for tithes, even if their work has not been completed.

Similarly, if it grew dark on Friday evening, the night of Shabbat, while one was eating the cluster, and eating on Shabbat is considered a fixed meal, Rabbi Eliezer says: He may finish, as not even Shabbat establishes fruit as fixed if its work was not completed. And Rabbi Yehoshua says: He may not finish. He holds that Shabbat does indeed establish fruit as fixed for tithes even if its work has not been completed. This indicates that Rabbi Eliezer maintains that Shabbat does not establish food with regard to tithes, whereas the mishna here indicates that he agrees that the beginning of Shabbat does establish them as fixed.

The Gemara answers: There, the reason is as he taught explicitly that Rabbi Natan says: It is not that when Rabbi Eliezer said: He may finish, he meant that he may finish in the courtyard itself; but rather he meant: He may exit the courtyard and finish. And similarly, it is not that when Rabbi Eliezer said: He may finish, he meant that he may finish on Shabbat itself; but rather, he meant that he may wait until the conclusion of Shabbat and finish. If so, this source does not contradict the mishna here.

With regard to the halakhic ruling in this case, when Ravin came from Eretz Yisrael to Babylonia, he said that Rabbi Yoḥanan said: Whether with regard to Shabbat; or with regard to the separation of teruma from fruit; or with regard to a courtyard into which the fruit is brought; or with regard to a transaction; all of these cases establish a requirement for tithes only for items whose labor is completed.

The Gemara notes that each of these details teaches a novel halakha. How so? Shabbat comes to exclude the opinion of Hillel, as it is taught in a baraita: One who gathers fruits from one place to another in order to reap them, and the day sanctified upon them, i.e., Shabbat commenced, Rabbi Yehuda said: Hillel prohibits the food from the gatherer himself. In other words, Hillel alone prohibits eating the fruit in that case until its tithes have been separated, for he believes that the beginning of Shabbat itself causes the fruit to be considered completed. Rabbi Yoḥanan teaches that the halakha in this case is not in accordance with the opinion of Hillel.