Peer Disagreement, or: Why Isn't the Other Person Right Like Me? (Column 247)

With God's help

In column 244 I discussed judging a person who acts from a different starting point than mine. I argued there that when we come to judge such a person (on the deontological plane, not the teleological one), we should take into account mainly his motives and the way he formed his position, and less our own opinion. Our own opinion is relevant mainly for teleological judgment (whether his act was harmful and what its consequences were), but in deontological judgment, which concerns the person himself and his motives, it carries little weight. At most, if my position seems self-evident to me, that has some weight in shaping my attitude toward him, because if he is making a clear mistake there is concern that he did not weigh his position properly, though even here caution is required (see, for example, here and my reply).

In this column I want to conduct a similar discussion, but from a different angle. Here I will not ask how I judge the other person, but how the fact that there is such another person should affect me and my own position. That is, should the existence of a person or group who sincerely believe in a position different from mine have implications for me? Does this not mean that I myself should doubt my position, since they are no less intelligent than I am? This is the question called in the philosophical literature peer disagreement (= disagreement among peers), and in the aforementioned column I brought three references to articles that deal with it (Example 1, Example 2, Example 3. I mainly mean articles 2 and 3.[1] Article 1 deals with the question of the effect of disagreement on ethical realism, that is, whether the existence of disagreement teaches—or at least confirms—the thesis that there is no single ethical-evaluative truth. Hint: absolutely not)[2].

This question troubles many people, and the answer to it is far from simple. Therefore, in my opinion this is an important column, and it is worth thinking about what I will write here. Let me say in advance that my conclusion will be that in quite a few cases there is room to regard my own position with skepticism merely because there exists a significant dissenter. The question I want to discuss is whether this is always the case, and if not—why not, and what the criteria are.

An evenly matched disagreement: Is a person a truthometer?

This question can concern disagreements in various philosophical fields, but also in the moral context. Even in the discussion about Yigal Amir, which is ostensibly an ethical discussion, we saw that considerations of assessing reality are involved (what the consequences of the Oslo Accords will be and whether the murder can prevent them) together with evaluative considerations (whether preventing those consequences justifies murder and harm to democratic values and the public's right to determine its own fate). In the moral context, it is clear that such a question presupposes ethical objectivity, that is, that there is right and wrong with respect to moral values (if there is no ethical truth, the question loses most of its significance). In addition, such a question also presupposes that this is a substantial disagreement, that is, that the other side is no less intelligent and capable than I am, and no less equipped with the relevant data than I am (which is why in the philosophical literature this is called peer disagreement. Peers, for this purpose, are people with roughly equal skills and knowledge).

I remarked there that I do not think you will find in those articles any interesting or novel arguments. It seems to me that among all of them there is only one novel argument that purports to give some reason to prefer my own position over that of the other person. I mean the argument in David Enoch's article about thermometers. He argues that this question is based from the outset on viewing human beings as thermometers. When we see two thermometers, one of which reads 35 degrees and the other 38 degrees, it is obvious to us that one of them is wrong. But if they are of the same quality (peer thermometers), we have no way to decide which of them is mistaken, and we are left in doubt. In such a situation it is not reasonable to adopt the result of one of them. The dilemma of peer disagreement basically assumes that human beings are a kind of truthometer, that is, instruments for discovering truth. Given that I have two instruments of similar quality for discovering truth (people of equal skill and knowledge—peers), it is not reasonable to give preferred status to one of them (= me). If I were looking at a disagreement between two of my peers, both of whom have similar levels of thought and information, I would not prefer the opinion of one of them, exactly as in the thermometer example. So why, when I myself am one of the disputants, is there room to give me preference over the opinion of the other? Seemingly this is only a subjective bias. The obvious conclusion is that in every case of disagreement between two peers, each of them ought to give up his position and remain in doubt about the issue under discussion.

Enoch's argument

I mentioned in that column that Enoch argues that the solution to this dilemma lies in the fact that a person is not a truthometer, and therefore it is incorrect to treat such a disagreement like a contradiction between two thermometers. The argument on which he relies is that I, as one of the disputants, cannot (and therefore perhaps ought not) ignore my own uniqueness in the discussion. Unlike my hesitation in the face of a disagreement between two other people, where both have a completely symmetrical status (like thermometers), in a situation where I am one of the disputants that symmetry is broken.

To sharpen this point (in the aforementioned column, questions came up in the comments that wondered about the issue), I will add another clarification here. Mathematicians and physicists define symmetry mathematically as follows: suppose there is a function of two variables, f(x,y). It will be defined as symmetric with respect to these two variables if, whenever we exchange them, we always get the same result; that is, f(y,x)=f(x,y) (for all values of x and y). In the case of a dispute between Reuven (x) and Shimon (y), where Levi (z) is the person who must make the decision, this can be denoted as follows: f_z(x,y). The function describes Levi's decision when there is a dispute between Reuven and Shimon and they are of equal ability. The question is what Levi should decide given that there is a dispute. It is now clear that if we exchange the roles of Reuven and Shimon, the function should not change, since given that they are peers, we have: f_z(y,x)=f_z(x,y). In such a situation it is not reasonable to assign the function a value that is the position of one of the two disputants (that is, that the result depends on x and not on y, or vice versa). But if one of the disputants is Levi himself, the resulting function is: f_z(x,z). Now it is clear that if we exchange the variables we get a different function. That is, z has a different role from x in this game, and therefore symmetry between the two peer disputants cannot be assumed.

From here Enoch argues that the status of the person who is supposed to form a position in light of the dilemma is different from that of the other person, who merely expresses a position. From my point of view, my own position has a unique status, and therefore it is incorrect to treat such a dispute as a contradiction between thermometer readings, unlike a case in which the dispute is being conducted between two people other than me.

Notice that Enoch does not really answer the problem here; rather, he undermines one of its assumptions. He argues that the main assumption in the problem (the thermometer assumption) is the symmetry of the function, and points out that at least on the formal level it does not necessarily exist when I myself am one of the disputants. But by this he only shows that there is no formal symmetry here. He does not show why my own position has priority, and therefore he also does not show, by this argument, that my position should not be unsettled by the existence of a dissenting peer.

A critique of the argument

As I wrote in that column, in my opinion this is merely a clever dodge. It is true that I have a special status and that it is hard to neutralize it in the mathematical-formal sense, but as I noted at the end of the previous section, that does not mean that in practice it is not correct to neutralize it. After all, from an objective perspective on my disagreement (= Levi) with Shimon, any other person (such as Reuven) would have to remain in doubt as to who is right. So why, if that 'other' person is myself, should the conclusion be different? Even if Enoch were right at the psychological level—that I really cannot technically detach myself from my special status (and I am not sure of this)—that still does not imply that this is also what ought to be done in a straight search for truth. Seemingly, from the standpoint of seeking truth, a person is indeed a truthometer, and that is how he should be treated. For this purpose, the human being, both I and my disputant, is a thinking machine whose goal is to reach the truth. Nothing more. In the formulation I presented above, I am basically arguing that the z-index of the function is irrelevant and should not appear there. The functions representing the point of view of all the peers in the group are identical: f_z(x,y) = f_w(x,y) = f_x(x,y) = f_y(x,y).

Of course one can talk about forming a personal position irrespective of objective truth, and one can also argue that I have no way of implementing this (because of psychological biases), but I do not see how from either of these there emerges a claim that this is also how one ought to proceed; that is, that it is truly appropriate to give myself preferred status when what I am seeking is the truth. At most this is an unfortunate necessity (indeed, a blameworthy one), or a psychological justification, but not a substantive justification.

The significance of the way I formed my position

At the end of that column I raised the possibility that the claim I made against Yigal Amir (was he not rash in forming his position?) might perhaps provide a justification for the distinction Enoch makes. I explained that in that discussion the question is not only who is right, in regard to which we are indeed both truthometers (and therefore of equal and symmetrical status), but that there is also another meta-question in the background: did that person do everything required when he came to form a position and make a decision of this kind? After all, Yigal Amir saw that many people disagreed with him, and it is not clear whether he took that into account with the seriousness it deserved. Notice that implicitly, this is exactly the peer-disagreement claim presented here. It in effect assumes that he ought to have wondered whether he might be wrong. In other words, the mere existence of a dissenting opinion should undermine his own position. But, paradoxically, I argued there that this argument is relevant only with respect to his actual position. On the meta-discussion plane (which concerns me and not him), that is, on the question whether he did enough to examine his position, there is precisely an asymmetry.

I explained it as follows: I am not required to do everything to examine my position, because I am not only saying that Yigal Amir is wrong. What I am claiming is that beyond my view that he is wrong, he himself also cannot be certain that he is right, because he did not examine the matter sufficiently. And if he asks: perhaps you too did not examine it sufficiently? I would say that indeed, perhaps I did not. But from my point of view it is enough for me to remain in doubt, and in order to remain in doubt the duty of inquiry is weaker. It is not comparable to the duty of inquiry that applies to him when he reaches a sharp positive conclusion, especially when he decides on that basis to take such an extreme action. I explained that here the symmetry between us is broken, because he decided and acted, whereas from my point of view it is enough merely to declare a doubt. I sit in the judge's armchair, and therefore for me a tentative position is enough.

There I stopped the discussion, but here I want to continue it from that point.

Peer disagreement again

The great question that arises here, and which we have not yet answered, is: when there is a disagreement between me and someone else, do I really have to suspend judgment or give up my position (or at least weaken it)? After all, if Enoch's argument—which, as noted, is the only one I saw there—did not answer the difficulty, then the question remains in place. Notice that in the analysis up to this point I assumed that this is indeed the case (for the thermometer model is the correct model when what we seek is truth and not psychological justifications for error). In other words, in the context of seeking truth we ought, in my opinion, to regard a human being as a truthometer, and therefore the mere existence of a conflicting opinion held by a peer should weaken my own position, and in fact put me into a state of complete doubt (either he is right or I am. If we are peers, that is, people with similar abilities and information, the objective chances on the two sides are equal). Is that really the case? Is my own position truly not preferable in a state of disagreement? Are we all condemned to yield to every dissenting opinion (at least every opinion of a peer) and remain in doubt?

One can take this question a step further, as Oren presented it in a comment on that column:

This reminds me of the story you often tell about Rabbi Yonatan Eybeschutz, about whom it is said that a non-Jew asked why we Jews do not abandon our faith and follow the non-Jews, who are the majority, since in our Torah it is written follow the majority ('follow the majority'). Rabbi Yonatan answered him that following the majority is a rule of conduct in a state of doubt, whereas we are not at all in a state of doubt regarding our faith. On a matter about which we are certain, there is no place whatsoever to follow the majority.

And seemingly this is difficult, for if human beings are truth-meters, and no person can be sure that his intellectual tools are superior to his fellow's (that is, there is doubt here), why should we not follow the majority even where we ourselves have no doubt (since even the understanding that we have no doubt is a product of our own intellectual tools)? Perhaps one could answer that indeed one should follow the majority, only not the majority of legs but the majority of wisdom. Except that now the same question returns recursively as to how one weights each person's wisdom, and since this has no end, a person must begin from an initial anchor point in which he himself weights every person's wisdom, and from that point onward can make decisions according to the majority of wisdom.

He too, in the end, arrives at the same lack of choice (the inability to detach myself, and the necessity of giving myself preferred status). To be sure, with respect to this broader version of the question (why not follow the majority?), there are a few more points that have to be taken into account (for example, does each member of that majority arrive at his decision independently?[3] How many wise people are in that majority group? And so on). But I will not go into all of that here, because if I answer the difficulty, or the basic argument, I consider myself exempt from addressing the broader question.

One may wonder why I view this argument as a 'difficulty' at all. After all, human beings are truthometers, and when there is a disagreement I must enter a state of doubt and not adopt a position of my own (certainly not forcefully). The problem with this picture is that it leads, at least de facto, to skepticism. It implies that a person cannot believe the position he himself has formed simply because there are others (peers, that is, people with a level of thought and knowledge similar to his own) who disagree with him. This almost completely neutralizes our ability to form positions and act on them. That seems unreasonable to me. For me to be convinced of such a claim, it must be free of problems, and as I will now explain, in my opinion it is not really free of them.

Does the disagreement itself refute the assumption of peerhood?

Enoch raises in his article yet another claim against the argument of peer disagreement. Suppose I did indeed reach conclusion X (say, that abortion is permissible under certain circumstances), and now I encounter a peer of mine who holds the opposite position. Can I not infer from this that he is not a peer, that is, that he does not possess skills and abilities similar to mine? The very fact that X seems right to me and my peer reaches a different conclusion may indicate that he is not truly a peer.

One can ask more than that: how can I infer about someone in the first place that his abilities are similar to mine? How do I arrive at such a conclusion? Seemingly this happens only because I have seen in the past that he was right. If so, that determination itself is also the result of my own judgments (again the symmetry between us is broken). If so, our present disagreement on the issue at hand can undermine that conclusion, for here my judgment says that he is mistaken. In that way, seemingly, I rid myself of the challenge that a dissenting position poses to my own position.

This argument looks, at first glance, like begging the question. I assume that I am right, and because the dissenter's position undermines that on the assumption that he is indeed a peer (that is, of skill equal to mine), I reject the assumption that he is a peer. Seemingly I am changing my basic assumption only in order to preserve the conclusion I value. But it should be noted that begging the question is not necessarily problematic in such a context. After all, I do have some position on the issue. Now someone comes and challenges it by invoking the existence of a peer who disagrees with me. So I continue to assume that my position is correct and infer that the dissenter is not a peer. In this way I have defended my position, and that is enough for me. This does not prove that I am right, of course, but it does repel the refutation of my conclusion. If this is indeed my conclusion, then at least from my point of view the burden of proof lies on the person who comes to challenge it. So rejecting a refutation can be done by way of begging the question.

And yet there is something problematic here. After all, until now I thought that the dissenter really did have skills similar to mine. That was the result of my experience with him in previous cases. It seems strange that the very moment a disagreement arises with him, I abandon that. It may depend on the question how convinced I was that he really is a peer (if that happened on the basis of just one case, then perhaps one could say that there is one case in favor of his peerhood and another against it. But if this is the result of long years and many discussions, it is harder to say that), but it still seems that there is no convincing and general solution to the problem here.

An explanation of the error

There is a logical move that appears, at first glance, fairly similar, but in an important way different from the previous one. I want to accept the picture of the two disputants as truthometers, and therefore if the person who disagrees with me is equal to me in his abilities, that should indeed threaten my own position. But there may perhaps be a way to continue holding my position with the same strength despite the existence of a dissenting peer, if I can offer a specific explanation of why he is mistaken on this issue despite his abilities. If I have such an explanation, then it is reasonable to continue holding my position despite the disagreement.

Let us take as an example a dispute between someone who believes in God and someone who denies Him. If I, as a believer, have reached the conclusion that God exists, the existence of an intelligent and capable atheist can certainly undermine my confidence that I am right (and of course my position should likewise undermine his confidence)[4]. If I want to continue holding my position with the same strength, I must offer an explanation of how my peer, who is no less philosophically skilled than I am, went wrong on this point. Notice: I am not rejecting the very assumption that he is a peer. On the contrary, I continue to hold it, but I offer an explanation of why he is nonetheless mistaken in this specific dispute.

In my opinion this is a stronger move than the previous one (presented in the previous section), and it is certainly relevant to quite a few disputes. In the religious-secular context, for example, each side can claim that the other side did not seriously examine the opposing position because of social influence, lack of exposure to the relevant arguments, lack of willingness and openness to consider them seriously (perhaps because of the implications for one's education and beliefs), and the like. These are serious arguments, and I think that on both sides they have a real basis. This is certainly not a straw-man rejection.[5]

The conclusion is that if I have a convincing explanation for how my learned peer, despite his skill, came to make a mistake on this specific issue, I have justification to continue holding my position. Where I do not have such an explanation, then we have two disagreeing truthometers, and there I must reconsider my position on the issue.

Seemingly, even with respect to this consideration itself there is symmetry between the disputants. After all, my peer can also offer an explanation of why I am mistaken, and here once again we have returned to a symmetrical situation and to the question of peer disagreement. But I think this is not correct. To understand why, we must remember two things:

1. It is not always the case that when I have a good explanation of why the other is mistaken, he also has a good explanation of why I am mistaken. This is not necessarily symmetrical. Remember that I am speaking here about a consideration that takes place inside one of the sides (= me), and not one that arises in the argument itself. This is not a proposal that Reuven should tell Shimon that he has an explanation of why Shimon is mistaken, and thereby win the argument. The explanation of Shimon's error is an explanation that Reuven finds when he reflects on the matter within himself. He must examine whether he really has a convincing reason that Shimon is mistaken on this issue or not. It is not enough for him merely to declare it in order to win the argument. This is a consideration that each of the two sides makes with himself, and there is no necessity whatever that both will have a good explanation of why the other is mistaken.
2. But even if both sides do have such an explanation, from my point of view I still have justification to hold my position (and of course he, from his point of view, does too). This is because on the plane of reasons for error there is no assumption of 'peerhood' (that is, the assumption that the two sides are evenly matched). Even if both sides have the same philosophical skill and the same knowledge, there is still no reason to assume that they have an equal likelihood of erring for extraneous reasons (and here we are dealing with errors stemming not from philosophical reasons but from various biases). Therefore, the mere fact that the other person is arguing with me, and even if he has an explanation of why I am mistaken here, I am nevertheless entitled to see my own arguments for the position itself (which seem strong to me) as confirmation that he has made a mistake on this issue. Because the assumption of peerhood does not exist here, this does not seem to be begging the question, as we saw with respect to rejecting the very assumption of peerhood itself (which was discussed in the previous section).

How did the other person arrive at his position?

The argument that comes up here is a particular case of the previous section, and it returns us to column 244. When I hear that a peer with abilities similar to mine, or greater than mine, holds a different position, I certainly ought to reconsider my own position. Anyone who does not do so suffers from arrogance, and that is certainly not a recommended path for forming a reliable position on important issues. But after I have reconsidered my position and my peer's arguments, and I have reached the conclusion that I am still right and he is mistaken, I am entitled to hold my position if, in my assessment, my peer did not do what I did. That is, in my estimation he did not seriously consider my position and my arguments after hearing them (he was 'locked in'). In such a situation I am entitled to maintain my position, and the existence of a dissenting peer need not necessarily undermine it.

This is a particular case of the previous section, for there is an extraneous reason for error here ('lock-in,' or insufficient judgment). Therefore the two explanations I gave in the previous section for why there is no symmetry here, and no begging of the question, apply here as well: 1. It is not always the case that when I reconsidered my position, he did so too. 2. There is no reason to assume that the likelihood that I will seriously reconsider my position and not be 'locked in' is equal to the likelihood that he will do so (the assumption of 'peerhood' does not exist here, because the likelihood of being 'locked in' is not necessarily equal among peers. Therefore in such a case the very existence of a disagreement can serve as a basis pointing to my peer's 'lock-in'). One must remember that the most natural thing in the world is for a person to be locked into his position and not attentive to opposing arguments. There is nothing unreasonable about this assumption. Therefore, if I examine myself and become convinced that I am not 'locked in,' and that I weighed the other side's arguments properly and nevertheless remained with my position, there is nothing wrong with concluding that the other person, who continues to disagree with me, is probably 'locked in.'

This returns us to the discussion in column 244 about Yigal Amir and abortions. I ended my remarks there by saying that my claim against Yigal Amir is not about his acting in accordance with his positions. That is what a person ought to do, assuming those are indeed his positions. My claim was against the certainty with which he held them, and against the way in which he formed these positions (especially when it is such an extreme and grave issue). In the terminology I have presented here, what I am claiming is that Yigal Amir did not seriously consider the peer disagreement when he formed his position. How do I know? First of all, from the very fact that I am sure he was wrong. Beyond that, that is indeed my actual assessment of what happened (in light of my impression of the way he operated, the environment in which he acted, and the people with whom he discussed the matter. I very much doubt how far he listened to and seriously considered the positions of people who disagreed with him without being 'locked in'). Notice that there I raised this consideration as a basis for a deontological judgment condemning Yigal Amir, whereas here I raise it as a justification for my holding a position that disagrees with him (even if we are peers in terms of skill and knowledge). It is the same logic, but it serves different purposes.

The surprising conclusion is that the argument about the way Yigal Amir formed his position, which was presented at the end of column 244 as a special situation illustrating a case in which the truthometer model is not correct, now turns into a general claim that is relevant to many disputes. The question of how the other person formed his position is relevant in all types of disputes, and therefore it is relevant when I come to decide whether to remain with my own position. It is now no longer a special situation, an exceptional example in which the existence of a dissenting peer has no effect; it is a consideration that can arise in any disagreement. Of course, it is not correct to apply this in every disagreement. It can be relevant in any disagreement, but only if I really reached the reasonable conclusion that the person with whom I am speaking in this dispute did not form his position out of a serious examination of the full range of considerations. As I explained above, this is a relevant consideration only if I really think that this is the case, and not when I raise this suspicion merely as an ad hoc justification for my stubborn insistence on holding my position. In such a case I myself suffer from the same defect of which I accuse him (since I myself am not seriously considering the position of the dissenting peer).

One of the indications that the other person is not seriously considering my own position is when I raise a strong logical argument and he does not accept it without showing where it fails, or when I point out a logical refutation of his arguments and he nevertheless does not retreat. This is a case where it is clear to me that if he were seriously considering my arguments, he would certainly have to accept them (as opposed to a situation of disagreement due to different starting points or different premises). In such a case it is highly reasonable, from my point of view, to assume precisely because I assume he is a peer (that is, someone whose intellectual skill is equal to mine), that he is 'locked in'—that is, he is not seriously considering my arguments and therefore is not forming a position with the required seriousness. In the next column I will bring an example of such a philosophical disagreement (not in the field of ethics).

I would note that in my article on tolerance I raised a similar claim, according to which a person who does not seriously consider his position is not entitled to a tolerant attitude toward it. And again, there it arose in the context of my relation to a peer's position that differs from mine, and here it arises with respect to the question whether my own position is undermined by the existence of a peer who disagrees with me. But the logic in the two cases is similar.

An example: the dispute between Beit Shammai and Beit Hillel

It seems to me that I have already mentioned this here in the past, but it is hard to ignore the relevance of the passage in Eruvin to the discussion I have been conducting here, especially since it also brings us to the festival of Sukkot.

The Talmud in Eruvin 13b states:

Rabbi Abba said in the name of Shmuel: For three years Beit Shammai and Beit Hillel disputed. These said, "The Jewish law is in accordance with our view," and those said, "The Jewish law is in accordance with our view." A heavenly voice came forth and said: "These and those are both the words of the living God, but the Jewish law is in accordance with Beit Hillel." And since these and those are both the words of the living God, why did Beit Hillel merit to have the Jewish law established in accordance with them? Because they were gentle and humble, and they would teach both their own words and the words of Beit Shammai; moreover, they would mention the words of Beit Shammai before their own. As we learned: If a person's head and most of his body were in the sukkah, but his table was inside the house, Beit Shammai disqualify it and Beit Hillel validate it. Beit Hillel said to Beit Shammai: Did it not once happen that the elders of Beit Shammai and the elders of Beit Hillel went to visit Rabbi Yohanan ben Ha-Horanit and found him sitting with his head and most of his body in the sukkah, while his table was inside the house? Beit Shammai said to them: Is that proof? They too said to him: If that is how you acted, you never fulfilled the commandment of sukkah in all your days. This teaches you that whoever humbles himself, the Holy One, blessed be He, exalts him; and whoever exalts himself, the Holy One, blessed be He, humbles him. Whoever runs after greatness, greatness flees from him; and whoever flees from greatness, greatness runs after him. Whoever forces the hour, the hour forces him; and whoever yields before the hour, the hour stands by him.

A long-running dispute raged between Beit Shammai and Beit Hillel over many different issues, and they did not succeed in reaching a decision. In the end, a bat kol (heavenly voice) goes forth and instructs that Jewish law follows Beit Hillel. I have explained in the past (see, for example, here, here, here and here) the meaning of the dispute and the resort to a bat kol in this case, and I will not return to that here.

What is important for our purposes are the words of Rabbi Yosef Karo ('the Mechaber'), who explains the ruling in favor of Beit Hillel on a monistic assumption (that there is a single truth in Jewish law). He explains that the fact that they put Beit Shammai's words before their own is not merely a matter of politeness. It is the better way to arrive at the correct decision. The fact that you weigh your opponent's words before you formulate your own position makes it more likely that you will come closer to the truth (even if you are less acute, as in the case of Beit Hillel vis-à-vis Beit Shammai). Therefore the bat kol ruled precisely in accordance with them. This is exactly what I am arguing here. In order to reach the truth, you must be attentive to the opinions of those who oppose you and to their arguments, and weigh them seriously; only then should you formulate your own position. If you have done that, you have an advantage over a peer who did not, even if he is your equal in wisdom.

A concluding note

Notice that there is no claim here that is relevant to a third party. If Levi sees a dispute between Shimon and Reuven, and in that dispute each side raises arguments and also claims that the other does not seriously consider his arguments and did not form his position seriously, Levi has no way to decide. The arguments I have raised here are relevant only to situations in which I myself am involved, and that is because with respect to myself I have the ability to examine whether I was sufficiently attentive—that is, whether I was not 'locked in.' With respect to others I have only indirect indications of that, and I can never be sure that I was right. All that remains for me is to conjecture that if I am not 'locked in,' then if I raised good arguments and the other was not convinced, he was probably 'locked in.' Under such circumstances, it is right from my point of view to decide to remain with my position. But none of this can be said with respect to a third party who observes a disagreement between two peers. In such a case he is supposed to assess what is in the heart of each of them, and there is no side that is exposed to him (as there is when he himself is involved in the argument). Of course, if he has convincing indications regarding one of the disputants that he is 'locked in,' he may adopt the position of the other (as in the case of thermometers, when there is an indication that one of them is broken).

[1] Warning: Van Inwagen's article (3) is truly childish. He asks the question exactly as I have heard it dozens and perhaps hundreds of times from various teenagers (and somewhat from adults too), and adds nothing significant beyond that.

[2] As an aside, I would note that Enoch's article is devoted to rejecting all the arguments advanced to undermine ethical realism on the basis of the phenomenon of disagreement. But what was missing for me in the article was the most basic argument for the opposite thesis: the very existence of disagreement shows that, in the view of both sides, there is ethical truth—for otherwise, what are they arguing about?! Suppose I think that abortion is permitted in a situation of the mother's economic distress and someone else thinks it is not. If there is no moral truth, then there is no disagreement between us. He feels one way and I feel another, just as I like chocolate or a certain woman and he does not like it or her. That is not a disagreement. If we are arguing, both of us implicitly assume that there is something to argue about, that is, that there is ethical truth, except that we disagree about what it is. The same is true in a case where we have a dispute over the question who is a Jew. Here too, both of us assume that there is such a thing as a 'Jew' (that is, that this is not merely an arbitrary subjective definition); otherwise there is no disagreement here and nothing to argue about. Therefore, in my view, the existence of disagreement not only does not weaken the assumption of ethical realism, it even strengthens it (the fact that both sides think there is a truth does not, of course, mean that there really is such a truth. Perhaps they are both mistaken about that. But if the very existence of disagreement has any effect at all, it is probably not in the direction opposite to ethical realism). One might perhaps think that although both sides think there is ethical truth, and in that sense there is some strengthening of ethical realism here, the very existence of the disagreement nevertheless undermines it. On that matter one can consult Enoch's aforementioned article.

[3] Also in Jewish law, when votes are counted in a religious court, a master and his student are not counted as two separate fingers, because these are not two independent opinions. The same applies in deciding about an error or corruption in a Torah scroll (or even a book of Jewish law), where several halakhic decisors wrote that one should follow the majority of textual versions. But there too it is clear to everyone that one must not take into account versions that copied one another as versions counted separately. This is also accepted in philology, which seeks to uncover more original and authentic textual versions. The 'son' (= the copied version) has no status that adds anything to the 'father' (= the original version).

[4] By the way, in this case it is very common on both sides (but especially on the atheistic side) to infer from this very dispute that the other side is not a peer (that is, not as intelligent as I am). On the religious side, the argument I am about to describe is more common.

[5] I have more than once heard people attribute religious belief to religious education, and see in that an argument against it—as though this were mere conditioning from an early age. This also comes up among people who grew up with a religious education. All of them, of course, ignore the parallel secular conditioning, and it is not entirely clear to me why. Moreover, in my opinion in this dispute the religious side has an advantage (even if not an absolute one), because the secular person is indeed less willing to seriously consider religious belief. To him it seems utterly bizarre (roughly as if I were seriously to consider some Indian or pagan belief). In order to believe, one usually needs religious education and early habituation to the idea of transcendent entities, and a person who grew up secular rejects this out of hand. In my view, precisely the religious person can consider the secular position with greater seriousness, since he knows it very well. He lacks neither information nor the mental preparation for a serious examination of it. Of course, all of us also have other natural biases, that is, the question of openness and costs, and these exist on both sides.