The Yerushami (Challah 3:5) discusses the case of when bread which didn’t have challah taken (or flour which didn’t have terumah taken from it), fell into a quantity of already separated bread. The gemara says the cases are different whether one takes challah from already separated and permissable dough instead of taking challah from another dough to permit its use, or one is taking this dough’s challah from itself. The gemara asks using the parallel case of terumah (20a-b)

מה בין המוציא ממנו עליו מה בין המוציא ממנו למקום אחר?

בשעה שהוא מוציא ממנו עליו הואיל ואין אותו הטבל ראוי להיעשות תרומה כיוצא בו חולין שבו מבטלין אותו. בשעה שהוא מוציא ממנו למקום אחר הואיל ואותו הטבל יכול להעשות חולין כיוצא בו לא (כ:) אמר רבי יוסי הדא אמרה טבל שנתערב’ בתרומה הואיל ואתו הטבל יכול להעשות בתרומה כיוצא בו לא בטל.

What is the difference whether one takes from it on itself, and what is the difference when one takes from it for the sake of another place[‘s flour]?

When he takes from itself [as terumah] on it, since this un-separated flour is not capable of becoming entirely terumah [as one cannot separate 100% of the flour as a gift to the kohanim, some must be left for the owner], the secular [flour] which is in it annuls it [by bitul, by being in overwhelming proportion].

When he takes from it [as terumah] for [flour] which is somewhere else, since this unseparated flour can become chullin just like it is, no [it does not annul it].

Rabbi Yosi said: this tells [you that] separated flour that was mixed with terumah, since this tevel could become terumah like it — it is not annulled.

This is an intriguing thing.

Tevel, unseparated flour, has terumah within it that is holy, but it must also contain secular flour as well — since one may not declare their terumah to be 100% of the crop. When this tevel falls into separated flour it seems that both come into play into the same speck of flour.

1- The speck is holy, because some percentage of the wheat terumah, and without separation it could well be this speck.

2- Yet, if the flour in question were definitely terumah, it would nullify by falling into a far greater heap of permissable grain. Because it has the potentiality of becoming the secular grain after the separation, it is of the same kind as the greater heap, and we invoke “min bemino eino bateil — something does not get annulled when mixed with its own kind.”

We treat each speck of wheat as though it’s in both states simultaneously. It’s holy because of one possible outcome, and that holiness does not get annulled because of the possibility of the other outcome.

This parallels what we saw about how doubt is treated as a mixture. Or, as the Bavli put it “isah” — literally: dough or mixture — “lashon safeiq hu — is a term of doubt.” When we deal with doubt about the underlying realia of a case, we do not deal with the unknown objective realities, but we instead providing a ruling by treating doubt itself as the phenomenological reality (world as we experience it) about which to rule. And since we think “maybe it’s X, maybe it’s Y” that resulting reality is an admixture of both. As I wrote then:

R YB Soloveitchik (Yarchei Kalah Shiur, August 1982) discussed the topic of bein hashemashos (twilight) as seen in halakhah. The case he brought was that of an esrog that is used only for part of Sukkos. The gemara concludes that since it was sanctified for a mitzvah on a given day, it may not be used for personal enjoyment (e.g. eaten) on that day. And bein hashemashos (Bh”Sh) is included in that day. However, there is a safeiq whether Bh”Sh is part of the previous day or part of the next day. Therefor, since it is holy for Bh”Sh, it is not to be used the entire next day either.

Rav Soloveitchik points out the obvious dilemma. There are only two possibilities:

1. Twighlight is part of the first day. In which case, the esrog is holy and prohibited until the end of the first day, which would be the end of bein hashemashos. Or,
2. It is part of the second day. Then, the esrog should be prohibited until the start of Bh”Sh.

If you want to play safe, then prohibit personal use until the end of Bh”Sh. Prohibiting it the entire next day is declaring bein hashmashos to be actually part of both days. It continues holiness from the previous day, and thereby continues it into the next as well.

According to Rav Soloveitchik, this is not just an oddity about Bh”Sh, but a point about safeiq in general. Safeiq does not mean “either A or not-A, but I don’t know which”. but is itself a third valid state. Similarly, he writes in Ish haHalakhah that halachic logic is multivalent, it isn’t the simple true-vs.-false of Aristotelian logic. There is no law of excluded middle asserting that every claim must be either true or false and no other alternative exists.

Here too, the state of the flour is unknown because it is determined in the future. We do not rule as though it is in one of two distinct states, but treat the terumah and secular states as though they co-exist.