1) DISQUALIFYING "EDIM ZOMEMIM"
QUESTION: Rava and Abaye disagree about whether a witness who is found to be an Ed Zomem becomes disqualified retroactively (Nifsal l'Mafre'a) or whether he becomes disqualified only from now on (Nifsal mi'Kan ul'ha'Ba), from the time that he is found to be an Ed Zomem and not from the time that he testified. The Gemara explains that Rava's logic is that in truth there is no reason to believe the second set of witnesses more than the first set; the only reason why Beis Din accepts the second set over the first is the Gezeiras ha'Kasuv. Once Beis Din does not believe the first set of witnesses, Beis Din no longer can consider them valid witnesses. However, the reason why Beis Din does not believe them is because of the Gezeiras ha'Kasuv and not because Beis Din knows that they are lying. Therefore, it is enough to invalidate them from the time at which they became Edim Zomemim, and not from the time at which they testified.
Why does Abaye argue with Rava's logic and disqualify the witnesses even retroactively, from the time that they testified?
ANSWER: The most straightforward explanation is that Abaye reasons that even though the law of Edim Zomemim is a Chidush, part of the Chidush is that the testimony of the first set of witnesses is not believed. Consequently, the Chidush itself dictates that Beis Din should disqualify the witnesses from the time of their testimony.
However, the LECHEM MISHNEH (Hilchos Edus 18:2) suggests that according to Abaye, the law of Edim Zomemim is not a Chidush of the Torah, but it is based on logical grounds. There are logical grounds to believe the second set of witnesses over the first. This logic is described by the RAMBAN (in Devarim 19:18). The Ramban writes that if two sets of witnesses contradict each other about the details of an event, Beis Din has no reason to rely on one set more than on the other. If, however, the second set of witnesses gives testimony about the first set of witnesses themselves (for example, that they are thieves or slaves), Beis Din believes the second set since the first witnesses cannot testify with regard to their own status (they are Ba'alei Davar). The same applies when the second set of witnesses testifies about the whereabouts of the first set at the time the event occurred, without saying anything about the details of the event. The first witnesses are not believed to say where they were at the time of the event because they are Ba'alei Davar with regard to testimony about themselves. The TUR (CM 67) also writes this reason.
The Lechem Mishneh suggests a practical difference between these two approaches in understanding the reasoning of Abaye (that is, whether the law of Edim Zomemim is a Chidush or it is not a Chidush). What will be the Halachah when the second set not only testifies about the whereabouts of the first witnesses, but they also contradict the testimony of the first witnesses with regard to the event itself? For example, a second set of witnesses says about the first set of witnesses (who testified that Reuven gave Shimon a certain amount of money at a certain time) that "you were with us at that time, and, furthermore, Reuven and Shimon were with us at that time as well and no money was loaned." What will be the Halachah in such a case? The Lechem Mishneh suggests that according to the second approach, the second set of witnesses should be believed since the first witnesses cannot defend themselves with regard to their own whereabouts. The RAMBAM (Hilchos Edus 18:2), however, rules that in such a case the second set of witnesses is not believed. Perhaps the Rambam maintains that Beis Din trusts the second pair only because of the Chidush of the Torah, and that Chidush was said only when the testimony of the second pair does not relate to the testimony of the first pair, but to the whereabouts of the first pair.
However, the Lechem Mishneh's suggestion is not infallible, for two basic reasons. The first flaw with the Lechem Mishneh's suggestion is that even according to the Ramban who says that there is logic behind the law of Edim Zomemim, Beis Din should not believe the second set of witnesses if they contradict the testimony of the first set and challenge their whereabouts. The reason for this is that once the two sets of witnesses argue about the facts of the case, any other testimony that they say together with those facts should be suspect, because it is known that one set must be lying (as is the case whenever two sets of witnesses contradict each other). (See also KOVETZ SHI'URIM to Bava Kama #43, and NESIVOS HA'MISHPAT 38:2.)
The second flaw with the Lechem Mishneh's suggestion is that the Ramban and the Tur might not mean to say that the law of Edim Zomemim is not a Chidush. In fact, in his commentary on the next verse (Devarim 19:19), the Ramban himself refers to the law of Edim Zomemim as a Gezeiras ha'Kasuv! When he explains the logic behind Edim Zomemim, he is simply explaining what logic caused the Chachamim to differentiate between Hakchashah (contradictory testimony of two sets of witnesses) and Hazamah, such that the verse of "v'Hirshi'u Es ha'Rasha" (Devarim 25:1) should be assumed to refer to Hazamah. The Ramban, therefore, gives a distinction by saying that Hazamah bears a similarity to testimony about a person's integrity. However, the Ramban himself admits that without the verse of "v'Hirshi'u Es ha'Rasha" the second set of witnesses would not have been given more credibility than the first set, since it is not entirely comparable to a case in which witnesses testify about another person's integrity. (It is not completely comparable because in this case, the second pair of witnesses are relating directly to the specific testimony that they are trying to disqualify, and not just to the character of the witnesses.)

27b----------------------------------------27b

2) RELATIVES NOT MENTIONED IN THE MISHNAH
QUESTION: The Mishnah gives a list of a person's relatives who may not testify for him. RASHI explains that although the Mishnah mentions that the husband of one's mother ("Ba'al Imo") may not testify for him, it still mentions that the son of one's wife ("Chorgo") may not testify for him, even though "Chorgo" is the same relationship as "Ba'al Imo" (it is just the reciprocal relationship). This is because the Mishnah expresses the reciprocal of every case, as Rashi explains at length. Rashi says that the only exceptions are the case of a man's wife's nephew -- the son of his wife's brother ("Ben Achi Ishto") -- testifying for him, and the case of a man's son-in-law ("Chasno") testifying for him, which are derived from other cases mentioned in the Mishnah and which are not mentioned themselves.
Rashi's words need explanation. Why does Rashi not mention that the Mishnah omits the reciprocal relationship for the son of one's mother's husband ("Ben Ba'al Imo" in the Mishnah; see Chart #2, 7B), which would be the son of one's father's wife ("Ben Eshes Aviv")? The Mishnah says that he ("Ben Ba'al Imo") may not testify for the son of his father's wife, but it does not say the reciprocal case, that the son of his father's wife may not testify for him!
It is clear why Rashi does not mention that the Mishnah omits the reciprocal relationships of the sons-in-law that it mentions (column C of the chart). Rashi is not discussing the sons-in-law mentioned in the Mishnah at all, but only the ten listed relatives themselves and their sons (as the MAHARSHA writes). However, Rashi should mention that the Mishnah omits the reciprocal relationship for "Ben Ba'al Imo" (the son of one's mother's husband).