There are several halakhic domains that require the evidence of two witnesses, including monetary claims and cases relating to marital status. However, the general rule is that one witness is sufficient in order to verify the status of a normal prohibition (issur) - i.e., eid echad ne'eman be-issurim. As such, one witness may attest to the status of grain: whether the necessary tithes and terumot have been selected thereby rendering the remainder of the grain permissible to eat. Yet, the mishna in Demai (4:7) cites an example of the testimony of one witness which is rejected. The case involves two peddlers who enter a town to sell their grain. Each attests that his own grain is prohibited but that the other's grain is permissible. As per our discussion above, each should be believed as an eid echad for the other's grain; but, the mishna informs us that this situation is too suspicious and, therefore, neither is believed! The mishna describes that this is so because they are "gomlin" - they are reciprocating favors, "partners in crime." Let's explore the essence of this fascinating halakha.
Our initial urge is to explain this halakha in a very straightforward manner: though an eid echad is generally believed there are factors which cast doubt upon his reliability and subvert his testimony. For example, if a person has been caught lying in the past we might be less inclined to believe his future testimonies. In the case of "gomlin" the peddlers' 'sincerity' seems very fishy. That they each speak negatively about their own grain while approving the ware of their friend, gives us the sense that the event has been staged. We suspect that this charade was choreographed well in advance; we have been duped, setup and, therefore, we reject their testimonies.
But, if indeed this is the rationale we would expect the qualification of 'gomlin' to appear in several locations. Interestingly enough, it doesn't. The mishna in Demai is elaborated by a gemara in Ketubot: this is the lone mention of our halakha. Thus, we might be dealing with a more case-specific halakha; a concept with a more limited, specialized scope.
Rashi in Ketubot posits a reason for rejecting the peddlers' testimonies. He writes that since they are amei ha-aretz (individuals ignorant of halakha) they are not trusted. At first glance, this comment is perplexing. After all, above we stated that they are not believed because we suspect a setup, not because they are ignorant!?! Perhaps, one might suggest that our suspicion is magnified in the case of ignorant people; but clearly this is not always the case. In fact, many uneducated people are more trustworthy and less likely to pull off a scam than their sophisticated counterparts, precisely because they are uneducated and, thus, more innocent!!! Why, then, does Rashi explain the disqualification of the testimony based upon their ignorance?
The Rambam, in his explanation to the mishna in Demai, elaborates: the reason the peddlers' testimony is not accepted is because they are ignorant of the laws of teruma and ma'aser and are thus suspect to have not properly selected these tithes. In fact, this is the reason amei ha-aretz are never believed in teruma cases. However, in our case we might be inclined to believe them despite their ignorance since they are attesting to another person's grain. In general, we assume that 'a person will not lie solely for another's benefit.' If a person is testifying on behalf of someone else, without any benefit or interest whatsoever, he is less suspect and more reliable. Thus, the only reason to believe an am ha-aretz in a case such as this is because he is afforded a special trust when he testifies about SOMEONE ELSE'S teruma. Hence, we might disregard their ignorance and accept their testimony based simply on their other-ness.
But the concern of gomlin does not allow us to afford them this SPECIAL reliability. Even though they are technically speaking of the other's grain, the adjacency of their testimony is not merely coincidental. Ultimately, each person HAS an interest in approving the other's grain since he has pre-arranged with his partner to reciprocate. The SPECIAL reliability one is afforded when testifying solely about another person clearly can not apply. Ultimately, then, according to the Rambam these people are not believed because they are (1) amei ha-aretz and (2) can not be supplied with special trust. We cannot afford these amei ha-aretz any extra trust and must reject their testimony as we do that of any am ha-aretz who testifies about HIS OWN grain. This might have been Rashi's understanding as well. Ultimately, the gomlin concern does not torpedo their testimony: it simply redefines the case as the type of testimony for which amei ha-aretz are not supplied special believability.
To this point we have suggested two distinct models for appreciating the halakha of gomlin. We might view the suspicion as that which independently disqualifies any and every testimony. Of course, we might then wonder why this halakha is so seldom referenced in the Shas. Alternatively, we might claim that taken alone this suspicion is not sufficient to thwart 'regular' testimony. Instead, it disallows boosting a witness's reliability based upon his testifying about someone else.
The most significant nafka mina to these two possibilities is the scope of our concept: gomlin. On the one hand, it might subvert testimony any time we notice this reciprocal partnership: any eid who is normally believed would be rejected if he testified together with another, when each is speaking about the other. Alternatively, according to the Rambam, a regular eid would be unaffected by the rule of gomlin. Only an am ha-aretz who requires a booster of reliability (for otherwise he is not trusted) does not receive this reinforcement in the case of gomlin and REMAINS disqualified.
The Yerushalmi in Demai asserts that any eid echad should be hampered by the gomlin factor. However, the Rash (in his commentary to the mishna in Demai) agrees with Rashi in Ketubot and qualifies the Yerushalmi - the concern only applies to an eid echad who is also an am ha-aretz. The Gra, in his Sefer Shnot Eliyahu (a commentary to the Yerushalmi), actually changes the text of our Yerushalmi. He posits that the Yerushalmi never applied gomlin to eid echad (abbreviated with the letters [ayin - alef]) but only to an am ha-aretz (abbreviated [ayin - heh]). By changing one letter (alef - hei), the Gra alters the Yerushalmi and interprets it in a manner similar to Rashi and the Rambam.
Our discussion is whether to apply gomlin universally or merely in cases of an am ha-aretz who is distrusted to testify about grain. This issue fuels a Tosafot in Ketubot about the content of the testimony of the two peddlers in the mishna in Demai. According to the simple reading, they were testifying whether their grain was last year's crop (yashan - which is permissible) or the new crop (chadash - which is forbidden until the korban omer is sacrificed on 16 Nissan). Other commentators assert that they were testifying that their grain was not harvested during a shemitta year. Rabbenu Tam (cited in Tosafot Ketubot 24a s.v. Sheli) opposes these possibilities since amei ha-aretz are NOT suspect regarding these rather simple laws of chadash and shemitta. And since they are not suspect, the law of gomlin should not apply. According to Rabbenu Tam, this halakha only applies to testimony regarding teruma and ma'aser - a complicated process with which amei ha-aretz are not familiar. Apparently, Rabbenu Tam agrees that the disqualification of gomlin is limited in scope: it only applies to an eid who requires a 'reliability boost' through his testifying about another. [The Pnei Yehoshua in Ketubot adopts a stance similar to that of the Rabbenu Tam.]
An interesting offshoot of this question might be the degree of suspicion necessary to invoke the gomlin clause. If we disqualify an otherwise acceptable testimony we might require a case in which the gomlin issue is highly probable. However, if gomlin merely prevents the augmenting of an eid's level of reliability we might apply the halacha more freely -anytime there is even a remote chance of gomlin. Thus, in order to completely scuttle testimony we might require a clear signal, whereas to prevent an artificial increase in reliability we might require less suspicious circumstances. The gemara in Ketubot limits gomlin specifically to a case in which they come to town with the tools of a peddler (weights and markers, etc.). Otherwise, two regular people are trusted about each other's grain. [Tosafot argue (and interpret the gemara in a different light).] Hence, according to Rashi at least, only a sharp gomlin suspicion, highlighted by the peddlers' arrival in town with their sales pitch 'ready to go,' can be classified as gomlin. This might indicate that gomlin scuttles eidut entirely on its own and, therefore, only applies if the suspicion is very real. (If so, this would run slightly contrary to our previous discussion of Rashi's position!!)
There are other factors which might affect the degree of suspicion. Clearly, simultaneous testimony is more suspicious than had they made successive statements. Additionally, we might focus on whether they freely offer this testimony or simply respond to our questions when asked. If they initiate the discussion we are more likely to suspect that they are staging a performance in order to sell their grain. The mishna in Demai 4 - i.e., immediately prior to the gomlin mishna - describes a case which should also qualify as gomlin, yet the testimony is accepted. Perhaps, the difference between these two adjacent mishnayiot lies specifically within one (or both) of these factors: gomlin might only apply if the testimonies are simultaneous and/or are offered rather than solicited. Gomlin, then, reminds us to be on the lookout for a choreographed 'set' of testimonies.
1. A halakha (e.g. gomlin, which impedes a person's reliability) can either be the actual CAUSE or instead might COMBINE with a different halakha to achieve a certain result (e.g. prevent boosting one's ne'emanut despite the principle that a person would not lie purely for another's benefit).
2. Be sensitive to the 'scope' of a halakha. Very often, a halakha which intuitively should apply in many cases is cited by the gemara in only limited contexts. Might this limited scope signal a different understanding of a halakha (e.g. gomlin only applies to an am ha-aretz)?
The gemara in Ketubot (23b-24a) discusses gomlin in the case of two women who testify that the other was not compromised during captivity. Given our discussion regarding gomlin, what are the various ways to understand the gemara's debate?