King Solomon wrote, ‘Iron sharpens iron, and a man sharpens the countenance of his friend’ (Proverbs 27:17). In Talmud Taanit 7a, Rabbi Hama bar Hanina explains that just as iron sharpens iron, two scholars sharpen each other’s understanding through active debate and discussion in the law. This is the spirit in which I receive, and here respond to, Mark Helm’s critique of my work in his article, “Classic Mediators’ Proposals Are Alive and Well” (Mediation Magazine, July 13, 2026). My purpose in this article is not to defeat his critique, but to treat it as an opportunity to sharpen both our understandings and, I hope, the reader’s as well.
I consider Mark Helm a friend. We first crossed paths in the 1970s in Southern California high school debate; we overlapped at Harvard Law School; and we have both been part of the Los Angeles legal community for decades since. Because of this long familiarity, I will refer to him as Mark throughout, and I invite him to refer to me as Jeff.
Mark takes issue with a thesis I have advanced since the American Bar Association issued Formal Opinion 518 on October 15, 2025 [1]: that it is unethical for lawyers to receive, and therefore should be unethical for mediators to make, what I call ‘unvetted’ Mediators’ Proposals — those made without the lawyers consenting in advance to the timing of the proposal, and indicating their willingness to recommend its terms to their clients. My hope is that this reply leads readers to conclude that, with respect to Mediators’ Proposals, our mediation rooms are not yet what they need to be — a reliably welcoming, hospitable, safe environment for the ethical lawyer.
And the fix is easy.
Before any mediator issues a Mediator’s Proposal, the mediator should secure the lawyers’ consent to its timing, and their willingness to recommend its terms to their clients. This willingness should be the result of the lawyers’ own hard work and independent judgment as to the best deal they can get their clients that day. That judgment is exercised through a negotiation of the terms both sides’ lawyers would be willing to recommend to their clients. After the lawyers negotiate those terms, the mediator then formalizes that result in the form of a Mediator’s Proposal. Legal ethics do not permit lawyers to outsource to a mediator the independent judgment to select a number from among the many possible outcomes both sides would accept. Those obligations come from ABA Model Rules of Professional Conduct 1.1, Competent Representation, and 2.1, Independent Professional Judgment.
Let me be clear up front: I am not opposed to Mediators’ Proposals. I use them. I find them valuable. I will continue to use them. Ethically.
In this article, I will not respond to Mark’s every point. I do not want the reader to think, though, that in debater’s terms, I am conceding subsidiary arguments. I am not. And this is not a debate. I just want to get to the heart of it.
The Classic, or Unvetted, Approach
Mark asserts that the following process for issuing Mediators’ Proposals allows lawyers to fulfill their ethical obligations to their clients:
[T]he mediator selects a relatively narrow range of numbers and asks each side privately whether it would seriously consider settling at one or more numbers within that range. If both sides say yes, the mediator then proposes a single number to both sides within the range.[2]
It does not.
The selection of the relatively narrow range of numbers identifies a ZOPA (Zone of Possible Agreement), defined as “the set of all possible outcomes in a negotiation that would be acceptable to all parties.”[3]
And here’s the crux of it: The lawyers’ recommendation of yea or nay on the mediator’s selected number in that range does not allow lawyers to fulfill their ethical duties under Rules 1.1 and 2.1. The lawyers’ duty of competent representation is to get the settlement terms as far as possible toward their side of the ZOPA. In other words, to negotiate the best deal possible for the client which is still minimally acceptable to the other side. The settlement can provide the client value in any number of forms: money, goodwill, finality, any mix of benefits. No matter how the client measures value, the goal is always to maximize it.
This is the starting point of negotiation theory. Here’s how Malhotra and Bazerman express it:
For many people, learning to negotiate more effectively means one thing above all else: “How can I get a better deal for myself?” Or, put another way, “How can I claim the lion’s share of the value in any negotiation?” While Negotiation Genius takes a much broader view of negotiation, we, too, start with this basic goal: getting the best possible deal for yourself.[4]
That is a partisan advocate’s job. The mediator’s job is to be neutral. Those are two different jobs, and the mediator’s selection of an unvetted number someplace in the ZOPA, followed only by the lawyers’ yea or nay on that number, deprives the lawyers of the opportunity to do their job – to fight for their clients and negotiate the best deal possible, as close as they can get to their end of the ZOPA.
In what appears to be his working definition of neutrality, Mark describes the Mediator’s Proposal as “what the mediator thinks is most likely to be agreed to by both sides.”[5] Somewhere in the ZOPA. What is the chance that number will be the same as the number which fulfills the lawyers’ ethical obligation, the number as close to the client’s end of the ZOPA as possible while still being in the ZOPA for the other side?
If the mediator assigns the number subject only to the lawyers’ yea or nay, the lawyers’ ethical duties go unfulfilled. Every client eventually asks the lawyer the acid-test question — a question to which we will return again and again: “Is this the best we can do today?” The ability to answer the acid-test question adequately is the sine qua non of doing the job and fulfilling the duties, and when mediators assign the numbers, lawyers are boxed into a corner of ethical violation.
That is the distinction that makes a difference. A number that is assigned versus a number that is negotiated. Only the second can fulfill lawyers’ ethical duties.
What Marshall Grossman Taught Me
A story proves the point. The ability to push back against a mediator’s number is essential to effective, ethical advocacy.
I spent the first 15 years of my career practicing litigation with Marshall Grossman, a legend throughout California, and indeed the country, in his time.
While the details have been changed enough to protect confidentiality, the story, and the range of numbers, are real.
Nearly 40 years ago, we were in a hard-fought piece of litigation, extraordinarily high stakes for its time. After many hours of a tough mediation, the mediator came into our room and told us the defense was offering $20 million. The mediator was emphatic and a little exasperated. He urged us to take it — a great deal, he said, one he never thought they’d offer.
While the $20 million was not formally a Mediator’s Proposal, we thought the mediator intended to convey it as a take-it-or-leave-it position. But if Marshall had agreed to recommend $20 million, he would have had a problem. He would not have had a good answer to the client’s acid-test question. He would have violated his ethical obligations.
Rule 1.1 requires the thoroughness and preparation reasonably necessary for the representation. A lawyer who hands the client’s matter to someone who is not the client’s lawyer, and then stops working, has not met that standard. Rule 2.1 requires independent professional judgment.
Let’s first consider Rule 1.1. There would have been undue delegation of negotiation responsibilities to the mediator. Who, after all, assigned that $20 million number? Not Marshall. The mediator picked it under some standard or other of neutrality. Did the mediator want Marshall to push back, to challenge his work, to test the limits of whether more was possible? Believe me, the answer is “no.” I was there. The mediator wanted his own negotiation work to substitute for Marshall’s, placing Marshall in the same position as the client, a passive receiver of information. That degree of deference would have put Marshall in violation of the duty of competent representation.[6]
Next, consider Rule 2.1. There would have been an abandonment of independent judgment to the mediator. In this situation, who was exercising independent judgment as to where the case should settle? The mediator. The mediator wanted Marshall to present $20 million to our client as a binary choice, deal or no deal, with Marshall’s judgment on that binary choice to be dependent on where the mediator pegged the number in the first place. Independent judgment is not a yes or no on a number or terms somebody else picked. It is a plenary evaluation of the ZOPA and how far you can get within it. Marshall had to decide, on his own work, whether $20 million was the most the defense would pay. He could not make that evaluation solely from a number the mediator thought would be acceptable to both sides. He could make it only if he tested the number.
Fortunately for our client, 30-something Jeff was not in charge. Marshall, not quite 50, was at the helm, and he was a far better lawyer.
I squirmed in my seat, sweating bullets that Marshall would blow the deal, our client would lose the $20 million in hand, and we’d all have to take the long, risky, expensive path to trial.
Marshall stared the mediator dead in the eye, paused for what seemed like forever, and responded, “No, we need $21.”
When the mediator left the room, I must have looked shell-shocked by what I thought was Marshall’s brazenness. He told me to calm down, that if they would pay $20, they would pay $21, and I should just be patient. Then he asked me whether I appreciated the significant difference that extra million would make to our client, both financially and emotionally.
The mediator returned after what seemed like hours, but was probably five minutes, and said, ‘OK, Marshall, you’ve got your $21.’ Only then could Marshall answer the client’s acid-test question based on his own hard work and independent judgment, not in reliance on the hearsay statement of the mediator as to how high the defense would go.
Suppose the mediator had made $20 million a formal, unvetted Mediator’s Proposal. Suppose the defendant had said “yes,” and Marshall had said, “No, we need $21, and I really mean it.” The mediator would then have needed to go back to the defendant for more. That is a nearly impossible sell. When our hypothetical proposal first issued, the mediator asked both sides to make reciprocal moves in the shared interest of a deal. A subsequent request for more would have been different. It would have asked the defendant to make a unilateral move to indulge what it would have perceived as Marshall’s intransigence. The defendant might have become so agitated that it walked out, jeopardizing everyone’s chance of a deal even though a ZOPA existed.
The Client’s Acid-Test Question
Just as Marshall was able to answer the client’s acid-test question, lawyers should consider their ability to answer it in every case.
Every client who receives a Mediator’s Proposal is disappointed in its terms. Mediators’ Proposals arise, after all, only after the sides have reached a final impasse: the plaintiff’s bottom line and the defendant’s top dollar do not overlap or even meet. The plaintiff must take less, and the defendant must give more than originally intended. Each must recalibrate based on what it learned that day about the facts, the law, and its counterparty’s resolve.
But no client wants to recalibrate any more than necessary. If they can’t get the settlement they originally wanted, they at least want the consolation prize, secure knowledge that they won’t have to take less or give more than they absolutely must to settle. The lawyer’s job is to make sure the client is secure in that knowledge. The mediator’s job is not.
To defend the position that lawyers can still answer the acid-test question when confronted with an unvetted proposal, Mark responds that the parties alone still make the “critical decision,” “if and when they accept the proposal after receiving the advice of their counsel.”[7] But the client’s decision is a different question from the lawyer’s. The client says yea or nay to a number inside the ZOPA. Every number inside the ZOPA will produce a settlement. A yea proves that the number is acceptable. It does not prove that the number is the best the client can do that day. That proof is the lawyer’s job, and only the lawyer’s own work provides it.
Mark’s response leaves the lawyer at risk of client conversations like this one, which no lawyer should have to endure:
Client: The mediator proposes $X, huh? Is that the best we can do today?
Lawyer: I think so. Our mediator is very smart; he has inside information from speaking confidentially with the other side. Settling for $X is better than rolling the dice at trial, and that’s the mediator’s number, so I recommend it.
Client: Is the mediator’s job the same as yours, to get me the best deal possible?
Lawyer: Well, no… The mediator is neutral.
Client: So let me ask you. Your criterion for recommendation is “better than rolling the dice at trial,” right? That’s what you just said.
Lawyer: Yes.
Client: There’s some flexibility there, right? I mean, if the mediator had proposed $X-plus-one-dollar, or $X-plus-one percent, you would still have recommended that as better than rolling the dice at trial, right?
Lawyer: I suppose.
Client: And my sense is that there would be more than one dollar or one percent of flexibility on our side, right?
Lawyer: Yes.
Client: So, isn’t it fair to suppose that there’d be some flexibility on their side, too?
Lawyer: I guess. I’m not sure. I don’t really know.
Client: You know, I didn’t hire you to get me a pretty-good result. I hired you to get me the best result possible to which they would still agree. Tell me, are you just taking the word of that neutral mediator, whose job is not to get the best deal for me? Did you even ask for a better deal? Did you negotiate this on my behalf?
Rule 1.1 requires lawyers to answer “no” to the first question and “yes” to the second two. With an unvetted proposal, lawyers cannot honestly give those answers because they just haven’t done the work.
No lawyer wants to be on the receiving end of questions like these. That’s why Mark’s process leaves lawyers in ethical violation, and why mediators must therefore declare it off-limits.
A disclaimer, a statement that the proposal is only the number the mediator thinks is most likely to be agreed to by both sides, does not rescue the lawyers. Whether the mediator accompanies the proposal with such a disclaimer or not, the number is still the work of the mediator, not the work of the lawyers. The disclaimer does nothing to remedy the lawyers’ violations of Rules 1.1 and 2.1 when they are subjected to the unvetted process.
The Ethical Alternative: The Home Run Approach
What should mediators do instead? We should generate Mediators’ Proposals through what I call “The Home Run Approach.” Let’s use it to take a trip around the bases.
At first base, the mediator confirms that all sides think it’s time for a Mediator’s Proposal. On this, Mark and I agree.
At second base, we vet the terms.
The lawyers are in full control of the final Mediator’s Proposal number. The mediator generally starts this phase by asking each lawyer what they think the Mediator’s Proposal should be. When I do this, I sometimes get pushback: ‘Hey, Jeff, this is supposed to be your number, not mine!’ My answer is that a Mediator’s Proposal is like a question on cross-examination — you don’t ask unless you already know the answer, or you don’t care. I care, so I want to know.
Soon enough, somebody suggests a starting number — sometimes a lawyer does it, sometimes I do. When I do it, I make clear that it’s where we start, not where we finish: ‘Have we reached the “life is too short” point? Do you just want to split the difference? What do you think?’ A starting number is an invitation to negotiate, not a recommendation to settle. Real room to push back has to stay open. From there, the lawyers negotiate with each other, sometimes through shuttle diplomacy, sometimes through direct communication, testing the number, pushing the limits, fighting to do better, until they converge on a number both lawyers are willing to recommend to their clients.
Critically, there is always something to negotiate. I once thought vetting might be a waste of time because it always seemed there was one and only one number or set of terms where the case could settle. Experience has proven me wrong. Many times, vetting produces consensus numbers or terms I didn’t expect and could not have calculated just by splitting a difference or extrapolating from a pattern. And even if there is one and only one possible settlement, the beauty of vetting is that it replaces the mediator’s guesswork and projections with the actual negotiation that proves the point. That direct negotiation between the lawyers, guided by the mediator, is the work that vindicates the lawyers’ obligations under Rules 1.1 and 2.1.
The mediator’s guesswork and projections leave a lawyer vulnerable to a counterparty who uses caucus confidentiality to bluff, cajole, or spin the mediator into proposing a higher or lower number. Vetting eliminates that vulnerability. Testing for a bluff is the lawyer’s job, part of getting the client the best deal possible that day. If the lawyer calls the bluff, good for the lawyer. If the lawyer falls prey to a bluff, the responsibility for that should be borne by the lawyer, not by a mediator who never had a duty to promote the client’s best interest in the first place.
At the end of the vetting process, either the lawyers have agreed on terms they would each recommend to their client, or they have not. If the vetting process doesn’t result in a consensus number both lawyers are willing to recommend, there is no ZOPA, no deal to be had, and no Mediator’s Proposal to make. That sometimes happens, and if the mediator and the lawyers are all doing their jobs, that’s OK. C’est la guerre.
At third base, assuming the lawyers have come to a consensus on vetted terms, those terms are formalized as the Mediator’s Proposal and taken to the clients.
Then, when a client asks the acid-test question — is this the best we can do today — the lawyers can answer yes, honestly, because they fought for the number themselves.
Finally, at home plate, we get a deal. No vetted Mediator’s Proposal I have ever made has been rejected. A double-cross can happen; a lawyer agrees to recommend, then cannot deliver the client. That problem belongs to the lawyer, not the process. Cases will settle, all ethical duties will be fulfilled. Only the vetted process achieves this result.
The Vetted Proposal Solves the Mediator’s Ethical Problems, Too
Lawyers invoke the mediator’s authority to overcome client resistance. Let’s not pretend they don’t. Lawyers across the country sell reluctant clients on disappointing Mediators’ Proposals with one version or another of this mantra: “This is what the mediator says we should do.” And it works. It works for a reason. Any number in the ZOPA leaves both sides better off than no deal, so any such number will settle the case. But settling is not the test of ethical conduct for lawyers. What makes a number right is the lawyer’s secure belief that it is the best the client can do that day as proven through negotiations, not as assumed or projected by the mediator.
Without vetting, though, this causes ethical problems for mediators, too. Our selection of where the case could settle is passed off as our statement of “what we think they should do,” based on our gravitas, and we know it. This sounds like a best-interest recommendation. If it is, that plunks us straight into a violation of Opinion 518, which prohibits mediators from making recommendations of that type.
The counterargument is that the Mediator’s Proposal is free of best-interest evaluations and is merely a tool to get the case over based on mediator guesses about how far each side will go. This does not solve the problem. Rather, it puts the mediator squarely in violation of Standard I(B) of the Model Standards of Conduct for Mediators (2005), which provides that a mediator “shall not undermine party self-determination by any party for reasons such as higher settlement rates, egos, increased fees, or outside pressures from court personnel, program administrators, provider organizations, the media or others.” The counterargument confirms that the mediator has two bad choices. If the proposal carries a best-interest judgment, Opinion 518 forbids it. If it does not carry that judgment, nothing legitimate justifies the mediator’s proposing it; the forbidden reasons of Standard I(B) are all that’s left.
After vetting, the lawyers’ judgment on the best deal possible is an independent one, and the mediator’s formalization of the number as a Mediator’s Proposal is dependent on the result the lawyers’ negotiation achieved. The mediator can say so, plainly, and truthfully: “This is the lawyers’ number, not mine. I formalize it. I do not have to take a position on whether it is the best deal. I vouch for the process, and for the lawyers who negotiated within it.” Here, unlike the unvetted case, the disclaimer works. It is true, and it leaves a legitimate reason for the mediator to propose the number: it is the parties’ own result, reached through their lawyers. The lawyers, as the clients’ fiduciaries, determined that this is the best the parties can do that day, so the number deprives the parties of no self-determination. Standard I(B) is not offended. If a lawyer tells a reluctant client otherwise, the lawyer misstates what the mediator said.
This does not minimize the lawyer’s role. It honors that role’s full dignity.
Where the Conversation Goes From Here
Where has this conversation brought us? Mark is right that Opinion 518 does not mention Mediators’ Proposals. But that is not the test. This article has shown that we can meaningfully apply the principles of Opinion 518 to the case of whether the unvetted, or classic, Mediator’s Proposal is unethical for lawyers to receive and therefore should be unethical for mediators to make. Where do we go next?
We need to have a broader conversation. It involves whether mediators’ practices are consistent with lawyers’ ethical obligations to their clients more generally. That is the challenge the ABA has set forth in Opinion 518. As mediators, we want mediation to be a welcoming, hospitable forum for lawyers, where they can participate without fear that mediators will box them into ethical violations. If we fail that challenge, mediation becomes a forum in which the ethical lawyer cannot take part.
These ethical considerations run smack into a marketplace reality. Many times, lawyers hire us precisely because they want us to make a best-interest recommendation, or at least to confirm, on the merits, the recommendation the lawyer is making. It can give the client comfort. It can help the lawyer get tough-love advice through to a client. It can give a decision-maker who must report back to a board, a boss, a spouse, or a claims supervisor some cover: “The mediator made me do it.”
So, the tension we must address is this: how do we satisfy the demands of the marketplace without violating Opinion 518? Opinion 518 permits a mediator to “offer an opinion as to how a tribunal is likely to rule on an issue.”[8] It tells the lawyer-mediator not to “state that the lawyer-mediator is acting to achieve a party’s best interest or that a proposed settlement is in a party’s best interest,” and to avoid communicating in a manner that might be taken as rendering legal advice.[9]
What about everything in between? There are limitless possibilities. Does “this settlement does not shock the conscience” imply a best-interest judgment? What about “It is within a range of what I would call reasonable settlements”? “I would do it if it were me, but remember, I am not you”? “Your lawyer has done a great job bringing the settlement to this point”? Context makes a difference, too. Mediators who are perceived as higher status need to tread particularly lightly.
One tentative guidepost: we might ask whether we are supporting the lawyer’s recommendation or supplanting it. Supporting is all right. Supplanting is not. When we support the lawyer’s reasons to take a settlement, we can in essence concur with that recommendation, consistent with the working presumption that lawyers are doing their best, are well prepared, and are putting their clients’ interests ahead of their own.
After vetting, we do not have to take a position on whether a proposed settlement is the best deal possible. We take a position on the process: “Your lawyer has had the opportunity to fight hard for your best interests and has done so. I vouch for the process, and for the lawyers who fought for you within it.” The number is the lawyers’ own work, so standing behind the process supports the lawyers.
So, I close with questions and invite Mark and every colleague to help answer them. How can we satisfy the demands of the marketplace without violating Opinion 518? How does the age-old distinction between legal information and legal advice apply in the mediation of commercial cases, where sophisticated clients are represented by sophisticated lawyers? Beyond a vetted proposal, how do we support, but not supplant, the lawyers with whom we work, on whose goodwill we depend to sustain our practices?
While all these questions and more remain open, the answer to one question is clear: Both lawyer and mediator ethics require mediators to vet Mediators’ Proposals before we issue them.
Mark has engaged this argument seriously, more than once, in print, with real citations and real care. That is how an argument this important to our field gets tested: in the open, by people willing to do the work. I thank him for helping me clarify my own thinking and express it in this article.
Let the conversation continue.
Notes
[1] ABA Standing Comm. on Ethics & Prof’l Responsibility, Formal Op. 518 (Oct. 15, 2025).
[2] Mark B. Helm, Classic Mediators’ Proposals Are Alive and Well, Mediation Magazine (July 13, 2026), https://mediationmagazine.adr.org/classic-mediators-proposals-are-alive-and-well/.
[3] Deepak Malhotra & Max H. Bazerman, Negotiation Genius 310 (Bantam Books 2007).
[4] Malhotra & Bazerman, supra note 3, at 16.
[5] Helm, supra note 2.
[6] Cf. Moore v. State Bar, 62 Cal.2d 74, 77 (1964) (three-month suspension of lawyer who entrusted client’s matter to a suspended assistant and thus “mistakenly place[d] himself in the position of his client rather than in that of his client’s lawyer”).
[7] Helm, supra note 2.
[8] Formal Op. 518, supra note 1, at 3.
[9] Id.