OPINION RE WHETHER ADR LOCAL RULES PERMIT A PARTY TO DISCLOSE TO A SETTLEMENT JUDGE AN ASSESSMENT BY AN ENE EVALUATOR

I.

Introduction

A party to a civil case has delivered to the undersigned a letter of complaint alleging that opposing counsel violated the ADR Local Rules by disclosing to a settlement conference magistrate judge the assessment of the case that had been formed earlier in the pretrial period by an evaluator during an Early Neutral Evaluation conducted under this Court’s ADR program. The undersigned has issued a separate ORDER (disclosed only to the parties) that sets forth the disposition of the letter of complaint. The purpose of this OPINION is to apprise the public of how the undersigned interprets the pertinent provisions of the ADR Local Rules.

It is important to emphasize, at the outset, that the lawyer who disclosed the evaluator’s assessment of the case did not first seek permission to do so from opposing counsel and from the evaluator. Had such permission been sought and given, no violation of the ADR Local Rules would have occurred. Thus, the pertinent Rules create an opportunity for the participants in an ENE session to share the product of the evaluator’s efforts with their settlement judge if all the parties jointly conclude that it would be appropriate and helpful to do so. In other words, the Rules provide an avenue for capitalizing on contributions made by the evaluator, or for making subsequent use of any communications made during an ENE, if all participants agree on such a course.

It is against this backdrop that the issue we address in this opinion takes clear shape: in the absence of a stipulation by all parties and the evaluator, do the ADR Local Rules prohibit a party from disclosing to a settlement judge any communication made in connection with an ENE session and any view expressed by an evaluator? For the reasons set forth in the pages that follow, the answer is “yes.”

The letter of complaint that sought intervention by the ADR Magistrate Judge in this matter alleged that defense counsel committed two violations of the ADR Local Rules. The first consisted of describing the evaluator’s assessment of the merits of the case in the body of the “Settlement Conference Statement” that was submitted to the magistrate judge who had been assigned to host the settlement conference. The second alleged violation consisted of attaching the evaluator’s written evaluation as an exhibit to that Statement — -thus making the evaluation itself available for the settlement judge to read.

The body of the defendant’s Settlement Conference Statement includes these two sentences:

Defense counsel then attached as an exhibit to his Settlement Conference Statement a copy of the evaluator’s written evaluation — a closely reasoned discussion of relevant law and evidence that filled seven single-spaced typed pages.

Before beginning the substantive discussion of the issues before the court in these ADR proceedings, I feel constrained to take a moment to praise and thank the evaluator who so obviously devoted such conscientious effort to this assignment. The evaluation that he presented in writing to the parties is an extremely impres sive document — reflecting a systematic, detailed consideration of what the evaluator understood the relevant law and evidence to be. Without implying anything about the ultimate accuracy of the views the evaluator developed, the Court takes this opportunity to publicly acknowledge the significant amount of work and the considerable care that are reflected in this lengthy written evaluation. It is powerful evidence about how much our evaluators give of themselves in this program and how valuable to litigants their efforts can be.

As I hope subsequent sections of this opinion make clear, there is, ironically, a direct correlation between the character and quality of an evaluator’s written evaluation and the level of risk that would be entailed if that opinion were disclosed to a settlement judge.

II.

The Meaning of the Two Relevant Rules

The parties point to two separate provisions in the ADR Local Rules that are relevant to the pending dispute. The first and most obviously relevant provision is ADR Local Rule 5-12, which addresses squarely the subject of “Confidentiality” in ENE proceedings. This Local Rule begins by setting forth, in its first major subsection, a generally applicable prohibition; then, in the second major subsection, the Rule identifies a few specific exceptions to that general prohibition.

The general prohibition states, in pertinent part, that “this court” and every person involved in the ENE process must “treat as ‘confidential information’ the contents of the written ENE Statements, anything that happened or was said, any position taken, and any view of the merits of the case formed by any participant in connection with any ENE session.” The Rule proceeds to prohibit disclosure of “confidential information” to “anyone not involved in the litigation” or to “the assigned judge.” (emphasis added). In addition, the Rule declares that “confidential information” may not be “used for any purpose, including impeachment, in any pending or future proceeding in this court. ” (emphasis added).

The second of the two major subsections of ADR Local Rule 5-12 sets forth five specific exceptions to the general prohibition that is announced in the first subsection. None of these exceptions purports to permit a party to disclose any “confidential information” from an ENE session to a settlement judge.

In framing these provisions of ADR Local Rule 5-12 it was the Court’s intention to impose a broad prohibition on disclosing ENE communications and then to identify, in the second subsection, the only exceptions to that broad prohibition. This intention is made clear in the first few words of subsection (a), which declare that “Except as provided in subdivision (b) of this local rule [everyone must treat ENE communications as confidential information].” Thus, if the general prohibition applies, disclosure of an ENE communication is permitted only in the five circumstances specified in the “exceptions subsection”— and in no others.

ADR Local Rule 7-5 is the second provision in the ADR Local Rules that arguably has some bearing on the matters in issue here. That Local Rule addresses “Settlement Conference Confidentiality.” Its structure parallels the structure of the local rule that addresses the confidentiality of ENE communications: its first major subsection sets forth a general prohibition, then the second subsection identifies a few specific exceptions. The general prohibition expressly applies not only to the parties and their agents, but also to “the settlement judge.” That prohibition requires anyone who has access to any settlement conference communication to treat it as “confidential information.” And like the largely parallel ENE rule, this general provision announces that “confidential information” may not be “disclosed to the assigned judge” except in a circumstance that is specifically identified in the second subdivision of the Rule.

Because its confidentiality provisions expressly apply to “the settlement judge,” Local Rule 7-5 appears to offer settlement conference communications the same level of protection against disclosure to “the assigned judge” as Local Rule 5-12 offers to ENE communications. On closer inspection, however, we find one potentially significant difference in the scope of protection offered by these two rules. As pointed out above, the general prohibition in the first subdivision of the ENE rule begins with the phrase “Except as provided in subdivision (b) of this local rule,” [everyone shall treat as confidential information ....]. Thus, as framed, the only qualification of the broad promise of confidentiality that is made by the ENE Rule consists of the exceptions that are expressly identified in the second subsection of the Rule itself.

In sharp contrast, the general prohibition in the first subdivision of the settlement conference Rule begins with the phrase “Except as provided by a case-specific order or in subdivision (b) of this local rule,” [everyone shall treat as confidential information ....]. (emphasis added). There is no commentary to ADR Local Rule 7-5 that explains or limits the reach of the qualifying phrase “[ejxcept as provided by a case-specific order.” Nor is there any other provision in the Court’s Rules that explains or limits this exception. Thus, the Rules leave parties, their counsel, and settlement judges with considerable uncertainty about what the circumstances might be in which a “case-specific order” might be issued that would compel disclosure to “the assigned judge” of otherwise confidential settlement conference communications. As we explain in a subsequent section of this opinion, this uncertainty undermines one of the principal arguments that defense counsel has advanced in support of his view that it should be permissible under our ADR Local Rules for a party to disclose an evaluator’s views to a settlement judge.

The Rule that is directly in issue here is ADR Local Rule 5-12. As noted earlier, this Local Rule commands all persons with access to ENE communications to treat them as “confidential information” that may not be “disclosed to anyone not involved in the litigation” or “used for any purpose, including impeachment, in any pending or future proceeding in this court.” Neither of these provisions was intended to permit a party, without securing the agreement of his or her opponent and the evaluator, to disclose ENE communications to a settlement judge.

When crafting the provision of this Rule that prohibits disclosure of ENE communications to “anyone not involved in the litigation” the framers’ intent was to limit the field of persons to whom disclosures were permitted to the parties and the persons who were acting as agents of the parties for purposes of the particular case (e.g., additional attorneys for the parties who did not attend the ENE session, staff in the attorneys’ offices, retained experts, and retained investigators). Thus, the phrase “anyone not involved in the litigation” was intended to exclude everyone who was not a party or a party’s agent. So the thrust of this provision is to inform litigants that, unless one of the exceptions that is specified in paragraph (b) applies, it is only the parties themselves and their agents in the litigation who may have access to ENE communications. It never occurred to the drafters that someone might suggest that a settlement judge (or anyone else working on the Court’s staff) was “involved in the litigation” as that phrase is used here.

The same disposition of this dispute is dictated by the language of ADR L.R. 5-12(a)(3) — the separate provision that prohibits use of ENE communications “for any purpose, including impeachment, in any pending or future proceeding in this court.” (emphasis added) The phrase “any purpose” contains no limitations. Read literally (as we must read words in legal rules), this prohibition forecloses use of ENE communications for such purposes as trying to influence a settlement judge’s view of the merits of the case, or trying to explain to a settlement judge why a party has embraced a given line of reasoning or believes that a case, or some disputed issue in the case, likely will be resolved in a given direction.

Nor does use of the word “proceeding” in this prohibition support a different view of its reach. In a different setting, one might suggest that the word “proceeding” should be construed to embrace only the formal, on-the-record components of the litigation process. But a different objective informed the drafters of this provision. Their use of the words “any pending or future” to modify the word “proceeding” reflects their purpose, which was to make it clear that the prohibition against use of ENE communications “for any purpose” applies not only to the case in which the ENE was held, but also to any other case over which this Court might have authority in the future. The phrase “in this court” was added simply to signal parties that the judges of this Court cannot control how judges in other jurisdictions will resolve issues related to the confidentiality of communications that are made in the ENE program here.

III.

Policy Considerations that Inform the Rules

Despite the intended reach of its pertinent provisions, should the court interpret ADR Local Rule 5-12 to permit parties to disclose ENE communications, including the views of the evaluator, to a settlement judge? For several reasons, the answer is no.

A rule that permitted parties to disclose ENE communications to a settlement judge could compromise significantly the quality and utility both of ENE proceedings and of settlement conferences — even though a settlement judge has no formal power over the parties’ rights. Even when they have no formal power, judges can have considerable influence on litigants — and parties and their lawyers might well fear that “influence.”

Why do the rules that govern ENE proceedings go to such lengths to protect ENE communications, including those emanating from the evaluator, from disclosure to any judge who may exercise power during the litigation? The answer has several elements. One is that a primary purpose of the pertinent Local Rules is to encourage litigants and their lawyers to be as forthcoming and frank as possible in ENE sessions — forthcoming and frank about the most sensitive topics: what the relevant evidence and law is and how the material issues in the case might play out. The productivity of an ENE session — how much benefit it can deliver to the parties— depends in substantial measure on the willingness of litigants and lawyers to disclose to one another and to the evaluator what they have learned about the evidence and how they could argue the law. In addition, the parties must be willing to distinguish between the claims and defenses that are most viable and significant and those that they can afford, at least off the record and for the time-being, to relegate to a back-burner.

In considering these issues, we need to bear in mind that an ENE session is intended to occur relatively early in the pretrial period — so litigants can use it to reduce the disproportion between litigation transaction costs and case value and to craft the most critically focused and efficient case development plan possible. In the case at bar, the ENE session occurred many months before the deadline for completing non-expert discovery — and even longer before the staggered deadlines for completing expert discovery and for filing potentially dispositive motions. Often, as happened in the case at bar, the ENE session will be scheduled before the parties have undertaken substantial discovery and before the court has addressed any significant motions. In such situations there will have been relatively little formal development of the pertinent evidence and no occasion for the court to have established “the law of the case.” So even if the parties in an ENE proceeding think they know what most of the relevant facts are, they often will not have gone through the pretrial processes that test the accuracy of their views — pretrial processes that include sources of discipline that can sometimes result in surprises about what witnesses say under oath and about what kinds of documents surface.

Stepping back, we see that we have identified two significant factors that our effort to construe the court’s ENE rules must take fully into account. The first is the importance of maximizing a “sense of safety” (protection against adverse consequences in the future) that can free litigants and lawyers to be relatively forthcoming about the merits of and support for their positions. The second important factor is that, by rule-design, in many cases the ENE session will occur before the parties have formally developed and tested the evidence and law through discovery and motions. There is significant play between these two factors. We can expect parties to feel a greater need for a “sense of safety” when their ENE session is held before they have been able to “secure” the evidence and to fix the court’s view of the law. The earlier in the pretrial life of the case, the more guarded and cautious we can expect the instincts of counsel and clients to be — about the relative viability and importance of their claims or defenses, about possible sources of information, about ambiguities in the relevant law, and about what each party has privately learned, through its own investigations, about what the testimonial and documentary evidence might be. The more guarded counsel and clients are, the less likely they are to benefit from an ENE session.

There is another significant implication for the “sense of safety” that the ENE rule is'designed to foster that is rooted in the fact that an ENE session is intended to occur relatively early in the pretrial period. It is the Court’s practice, generally, not to order litigants to participate in a judicially hosted settlement conference until after they have participated in a mediation or an ENE proceeding. As a result of this practice, often the ENE session will be held many months before the settlement conference. In most ENE sessions, the informational/evidentiary bases on which an assessment of the merits can be developed will be infirm — sometimes quite infirm. That fact intensifies the risk, sometimes considerably, that an evaluator’s assessment of the merits will be precariously superficial or will turn out to be wrong. In this context, “wrong” means not an accurate predictor of outcome and/or flawed analytically as a result of key evidence not being available (or later changing), the law of the case not being established, or parties later adding new claims or defenses. In the case at bar, for example, plaintiff amended the complaint well after the ENE process had been completed — a turn of events that might not have been foreseen by anyone during the ENE session.

In deciding whether to participate in an ENE (or how fully to engage in the ENE process), a well-advised litigant must assess both (1) the likelihood that an evaluation from the court-appointed neutral will be poorly premised or wrong and (2) how much harm an evaluation of that kind might cause. The weight of this second factor is likely to vary to a significant extent with the likelihood that “anyone not involved in the litigation” will learn the substance of the evaluator’s views. While risk of disclosure to the judge with case-dispositive power likely would inspire the greatest fear, it would be naive to suggest that risk of exposure to another judge of the same court — the settlement judge— could not inspire considerable concern. In the case at bar, it was counsel for plaintiff who asked the court to refer this case to ENE. During the proceedings before me, that lawyer declared emphatically that he would never have asked the court to refer this case to the ENE program if he had known that the views developed by the evaluator could later be shared with the settlement judge.

Several considerations may inform concerns about disclosing an evaluator’s opinions to a settlement judge. One set of factors is rooted in assumptions, fears, and perceptions about the behavior of settlement judges and the dynamics of settlement conferences. Even though a settlement judge formally has no power over the parties’ litigation rights or their substantive rights, many lawyers and clients believe that judges in any role, even a role that is supposed to be primarily “facilitative,” can and do exercise, informally, considerable influence over litigants. Many lawyers and clients also assume that settlement judges can and do exercise considerable control over the character of the negotiations that occur in a settlement conference. For example, a settlement judge who uses private caucusing may determine, to a significant extent, what information flows between parties during settlement negotiations and how that information is prefaced, packaged, or qualified.

Moreover, some lawyers fear that a settlement judge will pressure both counsel and client to make concessions or change their positions as the judge pursues, sometimes over-zealously, the goal of getting a settlement. Some lawyers and litigants will fear that a settlement judge will search for any available basis for encouraging parties to change their settlement positions — and, in that effort, would treat an evaluator’s opinions as weapons to turn on parties whose reluctance to move created a barrier to reaching an agreement. Similarly, some litigants will fear that a settlement judge would later use as weapons to push for movement any honest ex pressions of uncertainty by a party during an ENE session, or any conditional or temporary concessions made for the purpose of advancing the goals of the ENE process.

There is another pertinent set of factors at play here. Lawyers and litigants often expect settlement judges, unlike purely facilitative mediators, to form opinions about the merits of the case (or some significant issue in the case) and to share those opinions with the parties. Those opinions, while not binding, can have a significant impact on the direction and chances for success of settlement negotiations. Opinions expressed by settlement judges can increase the pressure on one litigant to make concessions while reinforcing his opponent’s resolve not to move. Anticipating that a settlement judge might well form and express opinions on the merits of the case, a lawyer who knew that ENE communications (including the evaluator’s assessment) could be shared with a settlement judge is likely to be especially careful to do nothing in an ENE session that might be used to influence a settlement judge’s views on the merits — or that might indicate uncertainty or vulnerability that a settlement judge might try to exploit. Thus, a lawyer who knew ENE communications could be disclosed to a settlement judge would be more likely to play her evidentiary cards close to her vest and to engage in the self-protective posturing that often attends proceedings on the record. She also would be more reluctant to indicate that some of her client’s claims or defenses might not be as significant as others. All of these kinds of inclinations would jeopardize the benefits that parties can secure through ENE.

Counsel for plaintiff has expressed precisely these kinds of concerns in the case at bar. Referring to some of the evaluator’s opinions as “findings,” plaintiffs counsel informed the court that he was quite worried that these conclusions by the evaluator would carry considerable (but, in his view, unjustified) weight with the settlement judge. Counsel’s use of the word “findings” cannot be dismissed as the product of a paranoid adversarial anxiety. In his written evaluation, the neutral used the phrase “[t]his evaluator finds” at least three times. He also prefaced one conclusion with the phrase “the evidence shows.” While in other places the evaluator used more qualified or guarded phrases to express his opinions, it is understandable that a party would fear that matters articulated as “findings” by an expert in the field would enter the mind of a settlement judge accompanied by at least a modest presumption of accuracy. Because they can affect litigants’ behavior, we cannot ignore foreseeable fears when we are fashioning a court rule — even if we believe the fears in question are groundless. So we would need to attend in some measure to parties’ fears about the role our neutrals’ findings might have in the mind of our settlement judges even if we were confident that every settlement judge was sufficiently independent of intellect to ascribe no significance to an evaluator’s opinions and conclusions.

Moreover, it is unrealistic to suggest that learning an expert evaluator’s assessment of the merits of the claims and defenses in a given case could have no impact on a settlement judge’s views. Lawyers and litigants won’t believe this — and shouldn’t. A settlement judge is likely to be quite interested in learning an evaluator’s assessment of the case. Why? The source of such interest is not idle curiosity; nor is it merely a desire to better understand why a party views the case as he does or is taking a particular settlement position. A settlement judge often will be interested in an evaluator’s views primarily because they might reflect insights or perspectives that might elude the judge— or might be based on information or knowledge that was unavailable to the judge. Most judges are well aware of their own limitations. They also understand that the task of analyzing a case and trying to predict the outcome of litigation is fraught with analytical and informational perils.

It is awareness of these fundamental circumstances that primarily drive a settlement judge’s interest in learning the anal-yses and conclusions of an independent expert. In short, a settlement judge is likely to be keenly interested in the evaluator’s views precisely because the settlement judge believes that knowing those views will increase the likelihood that the judge’s assessments of the case will be “right.” As settlement judges, we want to be able to put this important information in the data bank from which we. will try to build our opinions. In other words, we are likely to treat the evaluator’s views as a species of “evidence” about the merits of the parties’ positions — influential evidence.

Lawyers and most clients will understand that the information base that underlies the views formed by both evaluators and settlement judges is never complete. Litigants will know that neither evaluator nor judge will have seen the witnesses being deposed. They will know that neither judge nor evaluator will have studied all the potentially probative documents. And, in most instances, they will understand that neither an evaluator nor a settlement judge can know how the trial judge will construe all of the relevant law or how she will rule on important motions in limine.

Appreciating that their evaluator and their settlement judge are necessarily handicapped by all these informational uncertainties, lawyers and at least well-advised litigants understand that what their neutral host is trying to do is guess how other minds (the trial judge’s and the jurors’) will respond to evidence, argument, and personalities that are only partially known — sometimes only very partially. Lawyers and at least sophisticated clients will understand that it often is difficult to predict the character and outcome even of “pure” legal reasoning by someone else’s legal mind. And there is an exponential increase in uncertainty when the target of prediction shifts to how a group of unidentified non-legal reasoners (jurors) will respond to witnesses, lawyers, and a collection of evidence whose contents cannot be confidently predicted. So the “assessing” that an evaluator or a settlement judge does necessarily involves a substantial degree of “best professional guessing”' — an intellectually soft and malleable undertaking that many participants in litigation will understand enjoys little or none of the reassuring discipline of mathematics or good science.

The elusiveness of the targets of inquiry, and the uncertainties that necessarily accompany the assessment process, can make clients and lawyers worry about the vulnerability to influence of a settlement judge’s mind. We might well expect the intensity of that worry to increase when we remind ourselves about the qualifications our rules require our evaluators to meet and when we compare what is involved in the ENE process to what is involved (typically) in a settlement conference.

Every lawyer who serves as an evaluator in our program must have been a member of the bar for at least fifteen years and must have substantial expertise in the particular subject matter of the case in which he or she serves. Our rules presume that our evaluators will have litigated many cases of the kind they are asked to evaluate. So both litigants and settlement judges perceive our evaluators to be experts in the substantive law that will govern disposition of the case — as well as in the real world circumstances that can affect the availability and quality of pertinent evidence.

On the other hand, our settlement judges are, by necessity, generalists. Especially after they have been on the bench for a substantial period (and thus away from practice), our settlement judges often will have less substantive law expertise and relevant litigation experience than the evaluator had. Clients and lawyers who understand this reality may well fear that the evaluator’s assessments would enjoy considerable sway in the settlement judge’s mind.

The level of our concern about the intensity of litigants’ concerns deepens when we compare the structure and content of an ENE session to the processes that parties encounter in judicially hosted settlement conferences. There are several differences that are significant for present purposes. One is that evaluators often have more time to devote to the ENE process than judges have to devote to a settlement conference. Often evaluators have more freedom than judges to take the time necessary to be more systematic and to probe a little more deeply. Moreover, unlike settlement conferences, an ENE session is, by rule-design, a highly structured process with a predictable, uniform shape. It is dominated by systematically organized presentations of evidence and law — all of which must occur in the presence of all parties. So the process has a structure and a focus that mimics, albeit imperfectly, the structure and focus of a trial on the merits. The orderliness and focus of the ENE session are designed to increase the reliability of the parties’ and the neutral’s assessment of the merits of the claims and defenses, as well as to enable all participants to understand clearly what the bases are for the evaluator’s views.

The structure and focus of judicially hosted settlement conferences, in sharp contrast, are much less consistent from case to case and judge to judge. In many of their forms, settlement conferences feature less systematic and less open exploration of evidence than would occur in an ENE session. In many settlement conferences, the parties and the judge engage in most of their consequential communication about the merits of the case in private caucuses- — so neither party can be confident that he or she knows all the information that informs the judge’s assessment of the merits. And in many settlement conferences the host of the process is required to devote a substantial percentage of her time (usually in private caucuses) to matters that have no direct bearing on how the case might come out if tried to judgment. For example, settlement judges sometimes need to help parties or lawyers work through emotions or interpersonal issues, or to explore the implications (outside the litigation) of various possible courses of action or proposed terms of a settlement agreement. Settlement judges often need to spend time trying to identify alternative components of settlement packages. They almost always need to devote considerable time to the negotiation process itself — a process that more than occasionally seems to play out in surprising independence of the relevant evidence and law. So a settlement judge often spends less time on evidence and law than an evaluator spends, spends that time less systematically, and spends a substantial portion of that time in private caucuses— outside the vision of opposing parties.

What are the implications of these differences between an ENE session and a settlement conference for our concern about how litigants might react if our rules permitted a party to disclose to a settlement judge the substance of an assessment of the case that had been made earlier in the pretrial period by one of our evaluators? A party whose case suffered a negative assessment in an ENE might well worry, as plaintiffs counsel did in the case at bar, that a settlement judge who was given a copy of the evaluator’s assessment might be unduly influenced by it. Seeing that the settlement judge has less time to consider the merits and is not as systematic in that consideration as the evaluator was, a litigant might well fear that the judge’s understanding of the matter is superficial and is vulnerable to influence.

Moreover, if the settlement conference includes private caucusing, which most of ours do, litigants will understand that they cannot know everything that has gone into the settlement judge’s mind that might influence his or her assessment of the claims and defenses. Litigants might even worry that the assessments the judge offers have been skewed by some unidentified factor (learned in private caucus with the other side) that has little or nothing to do with the merits.

In short, knowing that the settlement judge is likely to feel distracted by other duties and pressed for time, that she may be inclined to cut quickly to the chase, and that she is likely to have something less than a firm grasp of either the subtleties of the pertinent substantive law or the details of the evidence, it would not be surprising for litigants to fear that their settlement judge might give considerable weight to an assessment of the merits that was formed by one of our well-qualified early neutral evaluators — even if that assessment was developed relatively early in the pretrial period.

Thus, the prospect that the settlement judge would learn an evaluator’s assessment could cause three kinds of harms. First, it could lead some parties to actively resist participating in the ENE program at all — even when ENE would otherwise promise to deliver more value to the litigants than any other ADR process. Second, it could drive parties to be less forthcoming, more tactical and calculating, and more rigid when they participate in an ENE session. That kind of behavior would reduce the value of the ENE process to all the litigants and to the court. Third, the prospect that the settlement judge would learn the evaluator’s assessment could undermine litigants’ confidence in the independence and integrity of the settlement judge’s views about the merits of the case — thus reducing the effectiveness of the settlement conference process and its value to the parties and to the court.

The last mentioned of these kinds of harms warrants some elaboration. One significant benefit that settlement conferences can offer litigants is a fresh look at their dispute, a second opinion about how the litigation might play out if it were pursued through judgment. For example, a party whose claims were assessed negatively by an evaluator earlier in the pretrial period might continue, in fact, to face a very steep uphill climb in the case — but that party might still have considerable difficulty acknowledging this reality, even after his own lawyer has tried in private to persuade him that the likelihood that he will prevail is quite small. A settlement judge whose views are perceived as truly independent is much more likely to be of real help to such a party than a settlement judge whom the party fears has been influenced by the opinions of the evaluator.

More generally, a second opinion, whether positive or negative, can deliver whatever value it really deserves only to the extent that it is perceived as independent. If it is, the parties can proceed to examine its underpinnings and assess for themselves how much persuasive power to give it. But if the “second opinion” is not perceived as independent, the parties have much less reason to take it seriously and to bother moving to the second stage of processing it, i.e., examining the quality of the bases for it. So by prohibiting settlement judges from learning the opinions of evaluators, the court can eliminate one potentially disabling source of concern about whether the settlement judge’s views about the merits of the case are truly independent.

We should acknowledge one additional consideration that has played a role in shaping the rules about the confidentiality of ENE communications. This consideration is rooted in assumptions and fears that some lawyers have expressed about communication within the court, communication behind closed doors between judges. Despite what the rules purport to promise and to prohibit, some lawyers and clients simply will not believe that a settlement judge would never disclose to an assigned judge what occurred during a settlement conference or what impressions the settlement judge formed of the merits of the case or the parties’ positions. This cynicism about the flow of information and impressions behind the court’s closed doors might be especially acute when the case dispositive power rests in the hands of a district judge but the host of the settlement conference is a magistrate judge. The power of the former over the latter might simply be assumed — and that assumption might inspire fear that if the district judge asks, the magistrate judge will tell. In fact, the magistrate judges of this court are quite independent and the district judges of this court not only respect, but also cultivate, that independence. When we write rules, however, we cannot ignore the effect on our procedures and programs that litigants’ assumptions and fears can have — even when we know those-, assumptions and fears are misplaced. Assumptions and fears are real parts of the real world in which our rules must work to promote their purposes.

Anticipating the kinds of threats we have described in the preceding paragraphs to the value of ENE sessions and to the value of settlement conferences, the drafters intended the ENE Local Rules to protect the confidentiality of ENE communications to the maximum extent possible. That intent must inform our interpretation and application of these Rules.

IV.

Responses to Defense Counsel’s Arguments for Permitting One Party (Without the Concurrence of the Others) to Disclose the Evaluation to the Settlement Judge

Before concluding this opinion it is important to address considerations that defense counsel argued (thoughtfully) should support an interpretation of the local rules that would permit a party, acting without the required stipulation of the others and the neutral, to disclose an evaluator’s views to a settlement judge.

The principal argument advanced by defense counsel in the case at bar begins with the contention that the phrase “assigned judge” as used in the ADR Local Rules refers only to the judge who is charged with responsibility for exercising plenary jurisdictional power over the litigation — the one judge with power to enter dispositive rulings on substantive issues, the one judge who would preside at trial if the case advanced to that stage. Proceeding from this starting point, the argument continues by asserting that because ADR Local Rule7-5 expressly requires “the set tlement judge” to treat settlement communications as “confidential information,” and forbids the “the settlement judge” from disclosing confidential information to “the assigned judge,” the barrier “that should always be in place between the ‘assigned judge’ and any substantive activities engaged in by the parties under the ADR rules” is neither breached nor jeopardized when a party discloses ENE communications only to a “settlement judge.”

There are two infirmities in this rule-based argument. The first, and the most important for present purposes, is that the language of ADR Local Rule 7-5 fails to establish as solid a “barrier” between the settlement judge and the assigned judge as defense counsel suggests. As we pointed out in a preceding section, a potentially gaping hole in that barrier is created by the phrase “[ejxcept as provided by a case-specific order” that qualifies the prohibition on disclosing settlement communications to the assigned judge. Uncircum-scribed, that language leaves litigants and settlement judges clueless about how likely it might be that such a “case-specific order” would be issued after any given settlement conference. And that uncertainty threatens a fundamental predicate for an effective ENE session: the promise in the rules that the “assigned judge” will never learn (from anyone, including a settlement judge) what the evaluator thinks of the parties’ positions or what the parties shared with the evaluator during the ENE process.

The second infirmity in the rule based-argument by defense counsel arises from ambiguity in the phrase “assigned judge.” That phrase, unexplained, very likely would be understood by most counsel to refer only to the one judge who may exercise plenary power over the merits of the case, the judge who would rule on potentially dispositive motions and who would preside at trial. Cf. Civil Local Rule 3-3(a). Defense counsel, understandably but incorrectly, assumes that this is the full extent of the intended reach of the phrase “assigned judge” in the ADR Local Rules. While the judge with potentially case-dis-positive power clearly was the focus of the Rule’s primary concern, he or she is not the only judicial officer to whom the drafters of the ADR Local Rules intended this phrase to apply.

Instead, the intent of these Local Rules is to distinguish between, on the one hand, a “settlement judge,” who can exercise no power over any of the parties’ substantive or litigation rights, and, on the other hand, any judge who may exercise the judicial institution’s power over any such rights. Thus, the intent of the rule is to prohibit disclosure of ENE communications (absent stipulation) not only to the judge who would preside at trial, but also to any judicial officer (magistrate judge, bankruptcy judge, or district judge) to whom responsibility has been assigned to address discovery disputes or to fix any other terms of the pretrial process. So the line that the Rule draws, for any given case, is between any judge who can exercise power over litigation rights or substantive rights and a judge who cannot.

In addition to his rule-based arguments, defense counsel contended that it was necessary to disclose the evaluator’s assessment in order to comply with the directives issued by the settlement judge in advance of the conference. Defense counsel pointed out that the settlement judge’s pre-conference Order required each party to include in its written settlement conference statement a summary of the proceedings to date, a candid evaluation of the party’s likelihood of prevailing, and a statement of the party’s position on settlement, including a history of past settlement discussions.

Defense counsel believed that in order to respond fully and accurately to these mandates it was necessary to disclose to the settlement judge not only that an ENE session had been held, but also the substantive conclusions or opinions reached by the evaluator. The evaluator’s views, which had been favorable to the defendant, helped inform the defendant’s analysis of the case and had become part of the basis for the defendant’s settlement position. Reasonably, defense counsel inferred that the settlement judge expected him to explain why his client was taking the position he was taking on settlement— to set forth the bases for that position. Because the evaluator’s opinions constituted a significant element of that explanation, defense counsel believed he should share those opinions with the settlement judge. Defense counsel also apparently believed, again not unreasonably, that because the defendant was advocating the same analysis that the evaluator had presented, disclosing the evaluation would help demonstrate to the settlement judge that the defendant’s analysis was reliable, that the positions he was taking were reasonable, and that he was proceeding in good faith.

Defense counsel could point to one additional provision of the settlement judge’s order that arguably required the parties to disclose the evaluator’s assessment. That order demanded that the parties be prepared to discuss any impediments to settlement that they perceived. In the circumstances of this case, defense counsel might well anticipate that the evaluator’s assessment could become an impediment to settlemen1&wkey;that it was likely, in fact, to make it much more difficult to persuade the defendant to make a more generous settlement offer.

During oral argument in these proceedings, defense counsel also pointed out that if the defendant was a public entity it might be called upon to explain to its constituents why it had taken the positions on settlement it had taken. If the evaluator’s assessment of the case played a significant role in informing that position, the public entity would not be able to explain fully and defend its handling of the settlement process if it could not disclose the evaluator’s views to the voters. Defense counsel also intimated that the lawyers for public entities might not even be able to explain fully the bases for their recommendations to the decision-making body, e.g., a city council, if sunshine statutes would enable the press or members of the public to gain access to the record of the decision-makers’ proceedings.

The specter raised by this last suggestion has no roots in the court’s ADR Local Rules and likely has few roots, if any, in the real world. The rules that govern ENE in this court require the informed participation of the parties themselves and permit disclosure of ENE communications to persons “involved in the litigation.” The decision-making body of a public entity would be deemed to be a “party” that is “involved in the litigation” and, therefore, would be fully entitled to learn from its lawyers or other representatives what the evaluator opined. Moreover, at least in many jurisdictions, the law permits decision-making bodies for public entities to discuss some matters (e.g., related to personnel) in some circumstances in “executive” sessions — beyond the reach of sunshine laws. More to the immediate point, however, no public entity need worry that it would be deemed to have violated this Court’s Local Rules if, as the result of initiatives taken by others and in conformance with legal mandates, someone who was not “involved in the litigation” was able to compel disclosure of the public entity’s deliberations and thus to learn about something that happened during an ENE proceeding.

Defense counsel’s argument that is based on the proposition that public entities need to be able to explain fully to their constituents the bases for their settlement positions attacks a much broader policy than is in issue in these proceedings: it attacks the fundamental notion that to maximize the value of ENE to litigants, and to protect their fairness interests, it is essential that ENE communications not be made public. The other points made by defense counsel, however, are pertinent to the narrower issue that is before the Court in the present proceedings: whether litigants should be permitted to disclose evaluator’s assessments to settlement judges. While counsel’s other points are “on issue-target,” however, the interests they seek to advance do not outweigh the competing considerations that animated the drafters of ADR Local Rule 5-12.

Not being permitted to disclose an evaluator’s assessments to a settlement judge would not disable a litigant from adopting the reasoning that supported those assessments and presenting that reasoning, without identifying its source, to the settlement judge. So the Court’s Rule would not prevent a party from using the substance of what the party learned through the ENE process in judicially hosted settlement negotiations. The purposes of ENE include helping all parties understand the relevant law more accurately, expanding their information base, refining their legal and evidentiary analyses, and sharpening the joinder of issues. These kinds of contributions to the quality of justice are not imperiled by the Rule that defendant challenges here. Instead, what that Rule prohibits is identifying the Court’s evaluator as the human source of the arguments or predictions that underlie a party’s settlement position. Those arguments and predictions will have to stand ■ on their own legs; their strength will be assessed on their inherent merits, not on the identity of their author.

Our construction of ADR L.R. 5-12 does prevent a party from fully explaining why or how his client arrived at the position it is taking in settlement if a factor in that process was the outcome of the ENE session. Our Rule also could prevent a party from being able to describe fully the history of settlement negotiations and all the factors that played some role in the settlement dynamic. For example, if a defendant made a settlement offer in one amount before an ENE session, then, because of a favorable assessment of its position by the Court’s neutral, lowered its offer after the ENE session, our Rule would prohibit that party from explaining, at least in full, the reasons for the downward movement of its settlement position. Thus, our Rule in some measure compromises the ability of a party to show the settlement judge that its behavior in the settlement dynamic is in good faith and has respect-worthy reinforcement.

We should not assume, however, that any such compromise is severe. After all, the settlement judge will know that an ENE took place (that fact will be part of the public record). If there was a significant change in a party’s settlement position after the ENE session, we can expect it to occur to the settlement judge that the views that the evaluator expressed during that session might at least help explain the change. Moreover, as we have already pointed out, there is nothing in the ENE Rules that prevents a party from sharing with the settlement judge, in tight analytical detail, all the reasoning that supports that party’s current settlement positio— even if substantial elements of that reasoning emanated initially from the evaluator. Our settlement judges are likely to be much more interested in the mature bases for the parties’ positions at the time of the conference than in re-traversing the paths over which the parties traveled in the past.

Our conclusion is this: while not being able to disclose the role the evaluator’s assessment has played in developing a party’s settlement position could make it somewhat more difficult for a party to persuade a settlement judge that it is being reasonable and that its analysis of the case is sound, and while in some cases this prohibition will disable a party from describing all of the interchanges that have occurred during the history of past negotiations, the losses so caused are far less significant than the harms that would be threatened by a rule that permitted one party, over the objection of his opponent, to disclose the evaluator’s assessment (or other ENE communications) to the settlement judge. For the reasons described at length earlier in this opinion, permission to make such disclosures would threaten the viability and utility of ENE at its core— and could compromise the value to litigants of inputs from a settlement judge. In short, the interests that might be advanced by changing our Rule to permit such disclosures are far too modest to justify the harms that such a change would threaten. It follows that the Court must decline to re-write, by strained interpretation, ADR Local Rule 5-12.

IT IS SO ORDERED.

. Under the ADR Local Rules for the Northern District of California, the "ADR Magistrate Judge ... shall hear and determine all complaints alleging violations of these local rules.” ADR L.R. 2-2. One purpose of committing this responsibility to one designated magistrate judge is to preserve the confidentiality of ADR communications to the fullest extent possible — and to assure litigants that no such communications will be disclosed to any judge who could exercise power over the parties’ rights in the underlying case. Thus, the ADR Local Rules provide that complaints alleging violations of the ADR rules must be submitted directly to the chambers of the ADR Magistrate Judge, may not be filed, and may not be disclosed to the judge (or judges) who exercise the court's power over the litigation itself. Proceedings to determine whether a violation of the ADR rules has occurred are conducted on the record (to protect the process rights of the persons involved) but under seal. ADR L.R. 2-4. These proceedings remain confidential and no judge who may exercise power over the litigants’ rights in the underlying action may have access to these sealed records. To be sure that these confidentiality provisions were honored, the ADR Magistrate Judge secured permission to make this ‘abstracted’ opinion public from counsel for the litigants in the case in which this issue arose. The court also shared this opinion with those lawyers before making it publicly available. . ADR Local Rule 5 — 12(b) permits disclosure of otherwise confidential ENE information when "all parties and the evaluator” enter a stipulation to make the disclosure in question. . The ADR Magistrate Judge has formed no opinion about the merits of the parties’ post-tions in the underlying litigation. . ADR Local Rule 5-12(a)(l). . ADR Local Rule 5-12(a)(3). . This language is quoted from the prohibition on disclosure that is set forth in ADR Local Rule 5 — 12(a)(1). . In the case at bar, defense counsel emphasized the length of time the evaluator committed to the ENE session (almost an entire day) in explaining why he (counsel) thought it was important to share the evaluator’s assessments with the settlement judge. . By well-considered intention, the court’s ADR Local Rules do not impose an independent requirement that parties participate in ADR proceedings or settlement conferences "in good faith.’’