Ohalos, Chapter 16, Mishnah 4. The preceding mishnah taught us that when a person digs and comes upon three meisim lying together, this is treated as a burial ground, and he must examine the surrounding area for twenty amos in every direction to see whether there are further kevarim there. Our mishnah now fills in the practical details: what exactly does this bedika look like, and how deep must the digger go? Along the way the mishnah adds a halachah about a kohen who finds himself doing the digging.
Not a solid twenty amos of digging
The subject of the mishnah is "habodeik," the person carrying out this survey of twenty amos. We might have assumed that he is obligated to overturn the ground continuously, working through every last tefach of those twenty amos. Instead the mishnah reveals an enormous leniency. His method is to excavate a square measuring "amah al amah," and then "umeiniach amah," to pass over the next amah entirely, and then to open another square of an amah, alternating in this fashion until the full twenty amos have been covered. The Bartenura reads the wording of the mishnah in exactly this way: one amah is dug out, the following amah is left as it is, and the cycle repeats.
Why is such a spotty search sufficient? Because of the way people buried their dead in those days. It was their minhag to leave a full amah of space between one kever and the next. Since no two kevarim would ever be crowded closer than that, a digger who tests every second amah cannot possibly pass over a kever without striking it. So he begins an amah away from the three meisim that were found, and continues in this rhythm outward.
A second leniency is folded into this halachah. Since the digger advances along a single line and probes only squares of one amah, we might worry that a kever sits slightly off to the side and will escape him completely. Should he therefore be obligated to open a trench the length of a kever, four amos, with a width of one amah? The mishnah rules that no such demand is made of him. Testing one square amah and then skipping over the next is sufficient, once again because of that dependable empty amah that was invariably left between one meis and the next.
How deep?
As for depth: within each square that he does open, the mishnah says he keeps going "ad shemagia," until he arrives, "lesela o lisula." One possibility is that he strikes solid rock, and then it is clear beyond any doubt that nobody was ever laid to rest there. The other is that he hits "besulah," untouched earth, soil that he can recognize as never having been broken open by human hands. Whichever of the two he meets, that square is finished and he moves on to the next.
The kohen who did the digging
Tumas meis is something a kohen is commanded to keep his distance from. Should a kohen come forward volunteering for this task out of a sense of duty toward his fellow Jews, we would have to tell him outright that the job is not his to take. But what if his kehunah went unnoticed, or nobody informed him of the halachah, and this kohen carried out the examination himself? Where does that leave him?
The mishnah says: "hamotzi es he'afar mimakom tumah, ochel bidim'o." One who removes earth from a place where there is only a suspicion of tumah, and in the end found no further meisim there, may still eat his terumah. He should not have been the one to check, but the fact that he did check does not forbid him from terumah, because there is no chazakah that tumah was present in that ground at all.
Clearing a fallen pile
The mishnah then sets against this a different case: "hamefakei'ach bagal, eino ochel bidim'o." The word "gal" refers to a heap of rubble, and the scene is one of urgency: a wall has come down on top of somebody, and people are frantically pulling away the debris to save a life. Should a kohen take part in that effort, terumah becomes forbidden to him. In this instance it is established, nebech, that a meis lies beneath those stones. We cannot pinpoint the moment of death, and it is possible that the victim was still breathing while the kohen was working; nevertheless a chazakah of tumah attaches to that heap, and the kohen loses his terumah.
The word "dim'o"
And why should terumah be labeled "dim'o"? The term is borrowed from the Torah's own language. In Mishpatim we read, "mele'ascha vedim'acha lo se'acher," a pasuk that teaches the sequence in which each of these portions must be removed, and there terumah is called "vedim'acha."
There is a shitah of the Raavad that "ochel bidim'o" does not mean that he may eat actual terumah, but only chullin into which a bit of terumah became mixed. The simple reading, however, and this is how the Rambam learns it, is that he may eat terumah itself.
Reviewing the mishnah
To summarize what we have gained: the requirement to examine twenty amos does not mean twenty amos of continuous digging. One digs an amah by an amah, leaves an amah, and repeats, relying on the fixed amah of space that was always left between kevarim, and he digs down until he reaches rock or untouched soil. If a kohen improperly did the digging in a place that merely might contain tumah, and nothing was found, he continues eating terumah. But a kohen who helped clear a heap of stones under which we know a person is buried may not eat terumah, since he may well have passed over the meis. Only where there is no certainty that meisim are present does he remain permitted, and so rules the Rambam.