ILW.COM - the immigration portal Immigration Daily

< Back to current issue of Immigration Daily < Back to current issue of Immigrant's Weekly

Advance Conflict Waivers Are Unethical In Immigration Practice – Debunking Mehta's "Golden Mean"

by Bruce A. Hake

Preface (June 27, 2007)

This article was prepared for the June 2007 annual conference of the American Immigration Lawyers Association (AILA) in Orlando, Florida, where the author appeared on a panel on dual representation in immigration practice. However, AILA refused to publish the article in its conference book. Instead the article was published in the June 1, 2007 issue of Bender's Immigration Bulletin (BIB). At the conference AILA distributed to all attendees free of charge a book entitled "Navigating the Fundamentals of Immigration Law, 2007-08 Edition (AILA G. Akers, ed. 2007) that contains a third version of Cyrus Mehta's Golden Mean article. The BIB will soon be publishing further articles on this topic.


An advance conflict waiver, in the context of this article, is an effort to extract from an employee client, at the start of a dual representation, an agreement that if any apparent conflict arises between the client and his employer, the lawyer will abandon the employee and take the employer's side. This method of caring for foreign business immigration clients is the key idea of "Finding the 'Golden Mean' in Dual Representation," published in two versions in 2005 by lawyer Cyrus Mehta.[1] The Mehta article represents the views of a faction of the U.S. business immigration bar.

The Mehta article alarms this author both professionally and personally. Professionally, it may harm thousands of people who seek immigration status in the United States as well as lawyers who follow its advice. Personally, it undermines the central issue of this author's career in legal scholarship. Thus, at times the language of this critique is passionate, but it is founded on law. Some positions are plainly unethical and it is proper to say so. In particular, the idea that one can generally exact advance conflict waivers from individual clients is per se unethical and likely to lead to malpractice doom. Furthermore, the Mehta article is self-refuting based on its own sources.

As to the history, dual representation first caught this author's attention in 1989 during a discussion on business immigration law, where it was said that lawyers who help businesses to obtain benefits for their alien employees often regard the petitioner as the sole client, thereby adopting what this author went on to term "the Simple Solution." This author was startled. How could a lawyer do legal work for an alien beneficiary without owing lawyer-client duties? That moment has turned into a life campaign. At times, as with this article, it is like fighting an insurgency.

Startled that a lawyer would disclaim duties to persons who seemed obviously to be clients, this author researched the issue deeply. That work culminated in a 1991 law review article, which has been influential.[2] That article shows many reasons why a lawyer who helps an immigration petitioner obtain benefits for a beneficiary has a lawyer-client relationship with the beneficiary. The only exception is if the beneficiary is represented by another lawyer. It does not matter whether the lawyer believes he is acting as the person's lawyer. It does not matter who pays. What matters is what the lawyer does. If he acts as the person's lawyer, he is responsible for the whole range of a lawyer's duties to a client. The law does not discriminate among classes of clients; a person either is or is not a client, and to each client a lawyer owes the full measure of professional duties. In 2004, the American Immigration Lawyers Association (AILA) distributed free of charge to its thousands of members a book on legal ethics.[3] Fourth of 13 articles is an article by this author on dual representation principles.[4] That article concludes: "Loyalty to clients is the foundation of legal ethics. The Simple Solution, a strategy for evading that principle, is clearly unethical." That is the law. Evidently there is no contrary legal authority (except for one anomalous 2003 New York bar opinion, which will require a separate article to distinguish[5]). But some lawyers disagree.

The next year, the Mehta article was published. It was promoted by AILA.[6] A 2006 article by an AILA member states: "There are two general approaches to dealing with dual representation: one suggests that both clients remain on equal footing with you so that you deal with both clients until and unless a conflict of interests arises. This is the approach that was thoroughly discussed in a wonderful article by Bruce Hake. [citing my 2004 article] Another approach was offered by Cyrus Mehta [citing the original Mehta article]."[7] While grateful for the kind words, this author is unhappy to be seen as the proponent of just another alternative to the Mehta article, both approaches being treated as equally worthy.[8] A better view is this recent statement:

[D]ual representation is the norm for immigration practitioners. It is the opinion of the authors that in all cases involving a petitioner and beneficiary (whether business or family cases), the attorney has a lawyer-client relationship with both parties. The sole exception would be those rare occurrences where both parties are each represented by independent counsel. More common in business cases, attorneys sometimes ask one party to acknowledge in writing that the attorney-client relationship doesn't exist or, in the event of a conflict between the parties, the attorney will cease representation of both and only represent one of the parties (typically, the one paying the fees). The viability of these agreements is questionable at best...

Attorneys who ignore the dual representation rules do so at their disciplinary and financial peril. A well-known immigration law firm that advised a corporate client about the mechanics and immigration implications of laying-off an adjustment of status co-client recently settled a malpractice claim brought against it by the foreign national after a jury awarded the plaintiff $365,000. Saraswati v. Wildes, No. GIC742835 (Sup. Ct. San Diego County [CA], Feb. 15, 2001). In Saraswati, the sponsoring employer terminated the employee while the adjustment of status application was pending, resulting in the revocation of the pending application and causing the employee to fall out of status. Of the $365,000 award, $300,000 was for emotional distress.[9]
The Mehta Article's False Premise: It's Not A Golden Mean

The Mehta article is based on a false premise. For years, this author's work has contrasted two approaches: (1) the dual representation rules, which are founded in the law, and (2) the so-called "Simple Solution" (a term coined by this author), which has no foundation in the law. This paradigm has been broadly accepted, so it is difficult for a "Simple Solution" proponent to openly advocate. Instead, the Mehta article postulates a "golden mean" between those two positions. It speaks of both established dual representation principles and the "Simple Solution" as if they are extremes, while the "Golden Mean" is promoted as more rational and moderate. This is a clever and attractive approach, especially in view of the common belief that there is no such thing as objective truth. But in fact, despite disclaimers, the Mehta article is actually a brief for the "Simple Solution."

The updated Mehta article states: "The writer was inspired by Aristotle's Golden Mean. See Will Durant, The Story of Philosophy: The Lives and Opinions of the Great[est] Philosophers (1926)."[10] Aristotle might take issue with this use of his concept. Aristotle's golden mean refers to a virtuous middle way between two extremes. For example, courage is the middle way between the extremes of cowardice and rashness.[11] For the Mehta article's premise to be accurate, established dual representation principles would have to be considered as extreme and as vices. But that is not true. They are the law, and they are noble, because they are founded on the duty of loyalty.[12] In addition, for the Mehta article's premise to be accurate, its proposals would have to (1) occupy a middle ground between two extremes, and (2) be virtuous. But both criteria are not true.

The article's first of two suggested techniques, limiting the scope of representation, is not new and it does not occupy a middle ground; instead, it has always been a part of established dual representation principles. Its second of two suggested techniques, advance conflict waivers, is not a middle ground; instead, it is a radical extreme. Moreover, this second of two techniques is not virtuous; it is unethical, reckless, and destructive. In sum, the Mehta article falsely claims that established dual representation principles are extreme, while promoting radical practices in the name of moderation.

The Mehta Article: A Walkthrough

In its introduction, the updated Mehta article asserts:
The "Golden Mean" is a middle ground approach between the "Simple Solution" and full-throttled [sic] Dual Representation, and provides ways in which an attorney can allocate how each party will be represented in advance, and even limit the representation of one party over the other so as to avoid any conflict. In the event of a conflict, the Golden Mean approach might still enable the attorney to continue to represent one party and not the other.[13]
That is beguiling but false. First, the article does not present a unified "middle ground" approach. What it really does is to advocate the "Simple Solution" by promoting the practice of exacting advance conflict waivers from individual foreigners, who are ostensibly treated as clients, albeit secondary clients, but who in substance are treated as non-clients. Second, regarding the claim that ways are provided "in which an attorney can allocate how each party will be represented in advance," there is nothing new about the idea of appropriately limiting the scope of representation.[14] Third, it does not make sense to say that ways are provided to "even limit the representation of one party over the other so as to avoid any conflict." There is no such thing as "the representation of one party over the other." Fourth, there is no such thing as a special golden mean approach that "might still enable the attorney to continue to represent one party and not the other." The conflicts rules have always provided that after a dual representation is disrupted, the lawyer may continue to represent one client in a different matter, so long as such representation does not compromise confidences and secrets of the other co-client; but the lawyer may not continue to represent one co-client in the disrupted matter.[15]

The next pages of the article discuss the "Simple Solution" and three anomalous decisions and are generally well reasoned.[16] The article then unveils the "Golden Mean" approach" (emphasis added):
While the Simple Solution risks failure if the non-client seeks to establish an attorney-client relationship, it is also difficult for a lawyer to undertake full representation of both parties without considering the potential problems in the event of a conflict and the limits to representation so as to avoid a conflict. After all, the very notion of dual representation results in an inherent limitation of representation as it implicates the attorney's cardinal duties of loyalty and confidentiality towards each client. The Golden Mean approach also envisages dual representation, but it requires the lawyer to thoughtfully limit representation of one of the parties, if necessary, or obtain advance waivers to future conflicts. Under the Golden Mean, an attorney may be able to successfully terminate the dual representation and opt for the Simple Solution too.[17]
In other words, the article expressly admits that the socalled "Golden Mean" is a way for a lawyer to "opt for the Simple Solution" while pretending not to do so. What does that paragraph really say? It says that a lawyer can get the benefits of the "Simple Solution," while pretending to observe dual representation principles, by adopting two techniques: (1) limiting the representation of employee clients, and (2) forcing on them an advance waiver of conflicts. As mentioned above, limiting the scope of representation is appropriate, if done properly. However, the article's second technique of advance conflict waivers is unethical in virtually all immigration cases. Therefore, the Mehta article boils down to one radical proposition, the idea that lawyers can get the benefits of the "Simple Solution," while pretending to follow the dual representation rules, by exacting advance conflict waivers.

The Mehta article's discussion of the technique of limiting the scope of representation is generally correct.[18] But it contains serious errors. For example, it states:
Furthermore, if the foreign national's employment was terminated, the disclosure could also include the fact that the attorney would cease to represent him or her but could take adequate measures to continue to represent the employer in this regard, including the withdrawal of the H-1B petition or other employer-based petitions, as well as represent the employer in unrelated matters.[19]
The bolded portion is reckless. Just such actions led to a $365,000 judgment against an immigration law firm in the Saraswati v. Wildes case discussed above. The Mehta article hangs a note on the bolded portion, but cites no authority. There is no such authority. Instead, the ethics rules provide that when a dual representation is terminated due to an irreconcilable conflict, the lawyer must withdraw from representing both coclients in that matter, but the lawyer may continue to represent one co-client in other matters so long as it does not compromise confidences and secrets of the other co-client.[20] If representation of the employee coclient terminates due to prior agreement (such as upon success in obtaining H-1B status), rather than due to a conflict, the lawyer is still not free to act on behalf of the employer against the alien former client.[21]

Rule 1.9 of the Rules of Professional Conduct governs duties to former clients. The Maryland rule states: "A lawyer who has formerly represented a person in a matter shall not thereafter represent another person in the same or a substantially related matter in which that person's interests are materially adverse to the interests of the former client unless the former client gives informed consent, confirmed in writing." The Mehta article is absurd to claim that an employee could give genuinely informed consent that would permit his former lawyer to stab him in the back by removing his immigration status, removing his employment, and subjecting him and his family to risks of poverty, imprisonment, and deportation. It is very unlikely that such a purported advance consent would ever be enforced. Moreover, it is likely that the attempt to exact such an advance consent would be taken as a sign of bad faith. As this author wrote in 1994:
A lawyer cannot simply pick and choose who is or is not a client for the lawyer's personal advantage. This accords with the important, general principle that in construing the terms of client-lawyer contracts, courts and bar authorities universally resolve all ambiguities in favor of the client and against the lawyer, who has drafted the contract and is in a better position to understand it. An analogous, fundamental principle is that interpretation of legal ethics rules and retainer agreements must be tailored to a client's actual circumstances.[22]
Further, whether or not such a waiver is enforceable, it is doubtful whether the technique makes sound business sense. How could a person acquiesce, when his lawyer is threatening in advance to take the employer's side all the way up to ruinous action, without deeply distrusting the lawyer? How could it be prudent to embark on a general policy of sowing such distrust among all alien clients? Why would a lawyer be so determined to exert superior control that he would insist in advance on his right to a small, compensable business activity such as withdrawing a petition? The entire project reflects a lack of understanding of what it means to be loyal to an individual client and of what it means when the ethics rules insist that a lawyer-client relationship is a personal relationship of trust and accountability.

There Is No Such Thing As A "Primary Client" In Immigration Practice

The rest of the Mehta article concerns advance conflict waivers. Let us first consider the Mehta article's concept of the "primary client." The article rests on a false premise, the notion that a lawyer may properly distinguish between a "primary client" and secondary clients, in other words, that he may systematically elevate loyalty to corporations over loyalty to individual employees. That is functionally equivalent to the "Simple Solution." In fact, no such distinction has ever been recognized in immigration law practice, nor indeed in any other area of law, except for narrow exceptions described below.

One might argue that this article is unfair for criticizing positions (in the original Mehta article) that ostensibly have been abandoned (in the updated Mehta article). Such criticism would be misplaced. First, the original Mehta article is in the literature. It cannot be retracted. Second, the original Mehta article is more likely to be read and cited by lawyers, judges, and others. Third, the concept of a "primary client" continues to be the foundation of the updated Mehta article (in the idea of advance conflict waivers). Fourth, while purporting to abandon the "primary client" concept, the updated Mehta article adds a footnote that seems to signal a continuing belief that the idea may be defensible in immigration practice, as explained below. Therefore, the worst ideas of the first Mehta article (especially the idea of a "primary client") remain alive and continue to require rebuttal.

The original Mehta article has a section called "The Primary Client Syndrome."[23] It starts out boldly (emphasis added): "Even if an attorney is able to assume dual representation, he or she is almost always approached by one of the parties first, and remains in greater contact with one individual over the other. It is impossible for an attorney to be equally loyal to both clients." That is the false idea at the heart of the Mehta article. The passage continues: "The 'primary' client may, in addition to paying the attorney's fees, also disclose information that it expects the attorney to keep confidential from the other party. Such a client will have a greater expectation of loyalty from the attorney despite the fact that the attorney is representing both parties." There is no legal authority for that. It is reasonable for joint clients to agree that certain information of the one will be kept confidential from the other. There is no authority for the idea that a lawyer owes a greater duty to one co-client because that client makes larger claims. The article continues in this vein for several pages, assuming the propriety of setting up one client as a "primary client." But it cites no authority, because there is none.

The updated Mehta article purports to abandon the concept of the "primary client:"
Even if an attorney is able to safely assume dual representation, he or she is almost always approached (and often retained) by one of the parties first, and often remains in greater contact with one client over the other. The prior article termed this the "primary client syndrome," but the term "primary client" will no longer be used as it can suggest unintended prioritization and has the potential to cause confusion.[24]
In fact, the "primary client" concept was not truly abandoned; the concept, if not the term, remains the foundation of the project. Moreover, arguably the term was not abandoned because it might suggest "unintended prioritization" nor because the term is confusing; instead, it is more likely that the term was abandoned because, after objections were raised, some research was done, and it was discovered that there is no supporting legal authority.

That purported abandonment of the "primary client" concept ends with this note: "The term 'primary client' was used in Allegaert v. Perot, 565 F.2d 246 (2d Cir. 1977) (holding that there was no conflict between former client and current or "primary" client where former client had no expectation that information would be kept secret from the primary client)." That is the entire note, which seems to suggest that the "primary client" concept is appropriate, but that the article is abandoning the term for high-minded reasons. Let us put that suggestion to rest. Here is a learned discussion of the Allegaert decision from a 2001 federal decision (emphasis added):
[T]o apply Allegaert to the facts of this case would give insufficient weight to [the lawyer's] duty of loyalty to [his co-client]. The Allegaert exception must necessarily be narrow [n.2: AAllegaert itself describes its holding as one based on 'peculiar facts.' Allegaert, 565 F.2d at 248.] because it requires a finding that one client is secondary to another, and thus, less worthy of an attorney's loyalty. In Allegaert, the secondary client was a large Wall Street brokerage firm that traditionally sought legal advice from its own law firm. The primary client was another brokerage firm controlled by H. Ross Perot that had entered into a joint venture with the secondary client. The primary client and the secondary client had separate counsel during the joint venture negotiations. ... The secondary client in Allegaert was secondary because the primary client's law firm only represented it with respect to a few matters following the joint venture agreement. The secondary client continued to seek legal advice from its own law firm... .[25]
The court went on to reject the application of the Allegaert exception and it disqualified the lawyer whose disqualification was sought. The quotation makes it clear that the anomalous Allegaert decision has no relevance to business immigration practice.

An April 2007 search on Lexisฎ for cases that use the term "primary client" returned 38 cases. Most did not involve law clients. Most of the ones that involved law clients cited the Allegaert decision. None of the cases involved immigration law. None provide any support for the notion that an immigration lawyer may discriminate between primary and secondary clients.

One discussion found in this search is the following (emphasis added):
The lawyer's right to transfer and utilize information for the good of both his clients does not mean that the lawyer has the right to turn against one or the other, for he owes equal loyalty to both his clients. T.C. Theatre Corp., 113 F. Supp. at 268 ("A lawyer's duty of absolute loyalty to his client's interests does not end with his retainer."); First Wisconsin Mortgage Trust v. First Wisconsin Corp., 422 F. Supp. 493, 496 (E.D. Wis. 1976) ("individual fidelity owed a former client" allows no right to prefer one client over another when litigation erupts between them), aff'd 584 F.2d 201 (7th Circ. 1978) (en banc); E.F. Hutton & Co., 305 F. Supp. at 394-396 (same: no right to prefer long-time relationship with corporate employer over one-time relationship with officer-employee. The lawyer owes equal and "independent professional judgment" to both).[26]
Thus, the actual case law cuts against the Mehta article.

The original Mehta article advances other extreme positions that were removed from the updated version. For example, it states (notes omitted): "The moment this [H-1B] employee wishes to inquire about 'portability,' or wishes to explore permanent residency options, however, the attorney will be faced with a conflict of interest. The attorney will no longer be able to represent the foreign national employee."[27] That is very wrong, and it reveals an unreasonable bias in favor of the employer.

First, there is a difference between a potential conflict and an irreconcilable conflict. What is described is only the merest inkling of a conflict. It is true that if an H-1B employee asks the lawyer about portability, there is a potential conflict., but that is a long way from irreconcilable! The wise lawyer knows to deflect the conversation away from conflict, for example, by reminding the employee that the lawyer might be required to withdraw if the employee expresses a fixed intention to act against the employer's interests. Something like this is required both by common sense and by the lawyer's specific duty to try to resolve a potential conflict. It would be rash to announce the lawyer's withdrawal upon first hearing the term "portability."

Second, in a business dual representation, the lawyer does not function like an overseer of slaves; the lawyer is an advisor and counselor. The lawyer could not properly assist the employee to leave the employer for another job, without the employer's consent. Arguably the lawyer could not properly advise the employee about the best options for portability and for permanent residence through another job. But the lawyer can neutrally and objectively answer the employer's questions about the law without taking sides. A wise lawyer would write it into the retainer agreement or engagement letter so that both employer and employee recognize that at times the lawyer might answer questions from the employee regarding his legal options. A corporation that would reject that is arguably not worthy of a lawyer's time, and any lawyer should insist on preserving this degree of professional independence.

That section goes even further: "Worse still, the attorney must reveal this information to the employer as both clients in a dual representation situation are owed the same loyalty and there can be no secrets against each other, unless especially agreed to at the outset of the representation."[28] That is an exaggeration. It is true that the lawyer may have an arguable duty to report to the employer if the employee expresses a fixed intention to act disloyally toward the employer, but each such situation is fact-specific, and drastic action can usually be averted. The rest of this passage says that the mere hearing of the words "portability" or "permanent residency" would require the lawyer to withdraw from representing the employee, report to the employer, instruct the employer that it must fire the employee, withdraw the H-1B petition, and to represent the employer against the employee in an investigation! This passage was removed from the updated Mehta article, but it reveals the article's foundation.

Advance Conflict Waivers

The next section of the Mehta article is entitled "Ethical Basis for Advance Waivers." It starts as follows (emphasis added, note omitted):
On May 11, 2005, the American Bar Association [ABA] issued Formal Opinion 05- 436 stating:
The Model Rules contemplate that a lawyer in appropriate circumstances may obtain the effective informed consent of a client to future conflicts of interest. General and open-ended consent is more likely to be effective when given by a client that is an experienced user of legal services, particularly if, for example, the client is independently represented by other counsel in giving consent and the consent is limited to future conflicts unrelated to the subject of the representation. Rule 1.7, as amended in February 2002, permits a lawyer to obtain effective informed consent to a wider range of future conflicts than would have been possible under the Model Rules prior to their amendment. Formal Opinion 93-372 (Waiver of Future Conflicts of Interest) (footnote omitted) therefore is withdrawn.
ABA opinions deserve respect, but they are nonbinding and advisory in nature. The above 2005 ABA opinion is the Mehta article's primary authority. The citation is self-refuting, because by its own express terms it does not support the conclusions for which it is cited.

First, it says that advance conflict waivers are only available "in appropriate circumstances." That suggests that the technique is inappropriate as an across-theboard approach for all matters of business immigration law.

Second, it says that such a waiver is most likely to be effective when given by an experienced user of legal services. Therefore, such waivers must not be generally proper for business immigration, because most foreigners applying for U.S. immigration status are not familiar with the U.S. legal system and most are not experienced users of legal services.

Third, the opinion says that such waivers are most appropriate where the client is represented by another lawyer. Therefore, for another reason this technique cannot be generally appropriate in business immigration practice. Indeed, the Mehta article has an entire section entitled "Each Party Being Represented by a Separate Attorney is Impractical."[29]

Fourth, the opinion clarifies that such waivers are best if the "consent is limited to further conflicts unrelated to the subject of the representation." Therefore, for yet another reason, this technique cannot generally be appropriate in business immigration practice, because the purpose of the advance conflict waivers advocated by the Mehta article is to permit a lawyer to turn on his employee co-client in favor of the employer, with everything at all times related to the subject of the representation, namely, the employer-employee relationship. The above is the Mehta article 's best authority for its central argument, and it is selfrefuting. The article makes no attempt to explain how this authority actually applies in the business immigration context in light of the above four restrictions.

The Mehta article continues:
More recently, on February 17, 2006, the New York City Bar's Committee on Professional and Judicial Ethics opined that a law firm may ethically request a client to waive future conflicts, including that the client consent to allow the law firm to bring adverse litigation on behalf of another current client, if appropriate disclosure is made and the client is in a position to understand the relevant implications, advantages and risks so that a client can make an informed decision whether to consent or not. [note: "NY City Bar Op. 2006-1 (2006); See also Visa U.S.A. Inc. v. First Data Corporation, 241 F. Supp. 1100 (N.D. Cal. 2003) (holding that law firm not automatically disqualified from representing two adverse clients as use of prospective waiver to future conflict was proper when client was a knowledgeable and sophisticated user of legal services and knowingly consented to future conflict).] Furthermore, in order for an advance waiver to be effective, a disinterested lawyer must be able to determine that the lawyer can competently represent the interests of all affected clients. [note: "Id."] The opinion also endorses the concept of "blanket" or "open-ended" advance waivers with respect to sophisticated clients under narrowly qualified circumstances. [note: "Id."]
Again, given the language of this authority itself, advance conflict waivers cannot be generally appropriate in business immigration practice. Perhaps at times a law firm might encounter a sophisticated foreign executive who is familiar with the U.S. legal system, has frequently been a user of legal services, is fully informed, and has independent legal counsel. An advance waiver of conflicts might (or might not) be upheld against such a person if his employer and lawyer combined against him. But most business immigration clients obviously would not meet these requirements.

The Mehta article continues (note omitted):
It is noted that the New York City Bar opinions have been issued in the context of large corporate clients seeking different law firms in different jurisdictions and different areas of the law. However, the basic underpinning of these opinions can also be useful to the practice of immigration law, which is that a client's choice of counsel is a fundamental right that has been recognized by the New York Court of Appeals in Levine v. Levine, 56 N.Y.2d 42 (1982). In Levine, the court approved a single lawyer representing both the husband and wife to a marital separation agreement, and held that the potentially adverse parties had a right to retain the same lawyer provided "there has been full disclosure between the parties …."
This paragraph has two ideas. First, it notes that permitted advance conflict waivers occur where both co-clients are large corporate clients. Second, repeating text from the bar opinion, the paragraph cites a New York case that held that dual representation may be appropriate even for a marital separation agreement. That citation simply supports established rules. It is unclear what the Levine case has to do with advance conflict waivers. In the bar opinion itself, however, the meaning is clear: in complex situations involving large law firms, large corporations, and multiple matters and areas of law, an overly broad interpretation of the disqualification rules sometimes can prevent a client from using the lawyer of its choice. This has limited relevance, if any, to business immigration practice.

A lawyer might be tempted to scan the headnotes of an authority like this New York City Bar opinion and think, "Yes! Advance conflict waivers are permitted!" But to comply with the law, one must dig to the details. The opinion states towards its end (emphasis added):
In Formal Opinion 2001-2, we articulated a number of factors that a lawyer should consider in determining whether the lawyer can represent multiple clients with differing interests in unrelated matters or in the same matter: ... (d) the sophistication of the client and the client's ability to understand the reasonably foreseeable risks of the conflict; and (e) if the firm is still representing the waiving client when the conflict arises, whether the lawyer's relationship with the clients is such that the lawyer is likely to favor one client over another. These same factors also help determine whether an advance waiver passes muster when that waiver includes substantially related matters in a transactional setting.

We conclude here that a law firm may ethically request an advance waiver that includes substantially related matters if the following conditions are met: (a) the client is sophisticated; (b) the waiver is not applied to opposite sides of the same litigation and opposite sides in a starkly disputed transactional matter; © the law firm is able to ensure that the confidences and secrets of one client are not shared with, or used for the advantage of, another client; (d) the conflict is consentable under the tests of DR 5-105(c); and (e) special consideration is given to the other factors described in Formal Opinion 2001-2.[30]
Earlier in the opinion, it explains that under DR 5- 105(c), an advance waiver is valid only if (1) a "disinterested lawyer would believe that the lawyer can competently represent the interest of each and if each consent to the representation after full disclosure of the implications of the simultaneous representation" and (2) the co-clients give fully informed consent.[31]

At the end of the opinion, it states: "As used in this opinion, a sophisticated client is one that readily appreciates the implications of conflicts and waivers. This would include, but not be limited to, clients that regularly engage outside counsel for legal services, or that have access to independent or inside counsel for advice on conflicts."[32] Moreover, it states (emphasis added):
The sophistication of the client also bears in other ways on the scope of the permissible waiver. For example, there are a few cases suggesting that a client cannot consent to have his or her own lawyer bring claims against the client charging fraud. See, e.g., Rosen v. Rosen, N.Y.L.J. Jan. 31, 2003 (Sup. Ct. Suffolk Cty. 2003) ("this Court simply cannot conceive of a knowing waiver by a client of such significant interests," when one client accused the other of submitting a false court filing). These cases are best understood as reflecting skepticism about whether the client understood and consented to the waiver. They have little relevance to waivers by sophisticated clients, particularly when the client is a large institution and the claims of misconduct involve personnel not involved in the representation of that client.[33]
From the above details of the Mehta article's primary authority, it is clear that most business immigration matters could not meet the requirements for an advance conflict waiver. This is especially true, following cases like Rosen, because the interests of an individual alien client are so significant.

The next part of the Mehta article discusses an anomalous N.Y. State Bar opinion that concluded that it would be possible to structure a representation so that the lawyer would represent an alien wife who was the beneficiary of an I-130 petition but not the U.S. citizen husband petitioner.[34] Earlier, the Mehta article stated correctly that this opinion would not give a lawyer a good basis for not regarding the husband petitioner as a client, but it gave an incorrect reason for this accurate conclusion.[35] In any event, this opinion does not support the propriety of advance conflict waivers.

Next the Mehta article suggests language that a lawyer could use in a notice to both parties in a business dual representation. The first part is generally apt, but this end portion is fraught with danger:
I will follow ABC's direction with respect to the immigration strategies it wishes to pursue in relation to your employment. If at any point in time a conflict arises that would not make it possible [sic] for me to represent you jointly with ABC Corp., you would need to seek independent counsel. However, you also agree that in the event of such a conflict I may continue to represent ABC Corp. In the event of your termination, I will no longer be able to represent you and may take appropriate measures, such as withdrawal of any petitions or applications that ABC Corp. may have filed on your behalf.[36]
It is unwise to state broadly that the lawyer will follow the employer's direction with respect to immigration strategies.[37] This portion should be tailored to the employer-employee relationship, and it would be wise to encourage some kind of employee participation. In addition, such an agreement should mention that in the event of an apparent conflict of interest, the lawyer must try to resolve the conflict.

More importantly, an individual employee client cannot meaningfully consent in advance to the lawyer's taking any "appropriate measures" on behalf of the employer against the employee, including withdrawal of petitions, which can carry catastrophic consequences for the alien. A wiser solution would be something like this:
If I must withdraw from representing you due to an irreconcilable conflict between you and your employer, I will be prohibited from representing either of you further in connection with this representation, although I will be permitted to continue to represent the employer in other matters. If I must withdraw, I will continue to preserve your confidentiality, except as you and your employer have otherwise specifically agreed. Finally, I want you to know that if your employment is terminated, your employer's Human Resources Department may take action against you, including withdrawal of immigration petitions or applications. Therefore, if your employment is terminated, or if I am otherwise forced to withdraw from representing you, you should consider hiring a lawyer to inform you and protect your rights.
Also, any such notice should specify exactly when the representation starts and when it ends.

The Mehta article next notes that some conflicts are not consentable. This is correct. The Mehta article next comments on the form of an advance conflict waiver. As argued here, that is rarely if ever appropriate in immigration cases. The Mehta article next discusses situations where it is impossible to make disclosures necessary to obtain the necessary consent. This is correct. The conclusion reiterates the claim that an advance conflict waiver would permit the lawyer to continue to represent the employer adverse to the employee if the dual representation blows up. As explained above, that is incorrect.

Some Additional Authorities

In addition to the authorities discussed above, many legal authorities recognize that foreigners must often be considered to be especially vulnerable and deserving of special protection under the law. For example, in a 2004 New York case, immigrant women claimed they were "particularly vulnerable to extortionate pressure [from their employer] with the potential to adversely affect their employment and even their liberty."[38] The judge awarded back pay, notwithstanding the holding in Hoffman Plastic Compounds, Inc. v. NLRB, 535 U.S. 137 (2002), that an award of back pay to undocumented aliens who were not authorized to work was foreclosed by federal immigration policy; the judge found that the plaintiff's immigration status was not clear in the record, even though the plaintiff herself had alleged that most of the defendants' employees were undocumented immigrants.

In a 1987 Sixth Circuit case, the court held, in an involuntary servitude prosecution against a farmer, that immigrants are one of the recognized "extremely vulnerable classes."[39]

In a 2002 New York case, the court granted the defendant's motion to suppress evidence of marijuana seized from his truck on the ground that he did not give a meaningful consent to the search.[40] The court reasoned as follows (citations omitted):
In conducting the "objective" analysis of whether police officers reasonably felt a suspect had consented, courts are instructed to pay attention to "the possibly vulnerable subjective state of the person who consents." ... Some factors to be considered in this regard are the age, education, intelligence, and physical and mental condition of the suspect, ... and whether the suspect is a "newcomer to the law." ... The Second Circuit has also indicated that a suspect's grasp of the English language, together with other subjective factors, may be relevant in determining consent. …[41]

In conclusion, the court observed: "Gagnon was a foreigner in the United States. He is most certainly a 'newcomer to the law,' and has little to no knowledge of our system of justice …. He certainly did not have any knowledge that he had a right to refuse consent …."[42] And the court suppressed the evidence on the ground that the consent to the search was not voluntary.

Finally, in a 2005 case the Fifth Circuit upheld a sentencing enhancement in a conviction of a woman for a scheme to defraud undocumented aliens by pretending to be agents of the government and by purporting to help them with immigration forms.[43] The court held that the aliens were indeed "unusually vulnerable victims": "Garza's victims' poverty, language problems, and fears of deportation did make them vulnerable members of society …."[44]

Sophisticated business immigration clients may well be outside the categories described in these cases. But authorities like this demonstrate further that it is questionable whether the Mehta article's key idea of advance conflict waivers is likely to be upheld.[45]


Imagine that you are a foreigner who has been offered a chance to work for a company in the United States. Your prospective employer tells you that a lawyer will take care of this for you. The lawyer sends you questions to answer, forms to fill out, and documents to read. One document is a complex legal writing that seems to say that the lawyer will always take the employer's side in any dispute, all the way up to abandoning you, advocating your termination, and actively taking steps to destroy your immigration status, thereby exposing your family to starvation and the risk of imprisonment and deportation. Or at least you think it means something like that. But you're excited about the job opportunity. And you cannot afford to hire your own lawyer. What are you going to do?

The advance conflict waivers advocated by the Mehta article are a classic take-it-or-leave-it adhesion contract. The difference in bargaining power between the company and the lawyer on one side, and the individual foreigner on the other side, is completely disproportionate. It would be ridiculous to assert that an employee's "consent" to such a waiver is voluntary or informed. It is virtually certain that such a waiver would not be upheld in a malpractice action. Courts are unlikely to enforce such aggressive overreaching. Indeed, the fact that the lawyer tried to exact such a waiver might well be regarded as an aggravating factor and a sign of bad faith. It is certain that using such blanket waiver forms will engender suspicion and disloyalty by the employee toward the employer and the lawyer. It is certain that using such waivers will increase, rather than decrease, a lawyer's exposure to liability. But all those problems can be avoided by following established dual representation principles.

The authorities above show that the advance conflict waivers advocated by the Mehta article are unethical in immigration practice if exacted from an individual who is not represented by an independent lawyer. It would be reckless to use them. Any lawyer who really wants to follow the golden mean will avoid them. Once advance conflict waivers are removed from the Mehta article, all that is left is established dual representation principles, seasoned with an unreasonable bias in favor of corporate clients. A prudent lawyer will not be fooled.


1 Immigration and Nationality Law Handbook 29 (AILA 2005-06 ed.) ("the original Mehta article"). The article was revised as Cyrus Mehta, Finding The "Golden Mean" In Dual Representation -- Updated, 06-8 Immigr. Briefings (Aug. 2006) ("the updated Mehta article").

2 Bruce A. Hake, Dual Representation in Immigration Practice: The Simple Solution Is the Wrong Solution, 5 Geo. Immigr. L.J. 581-639 (Fall 1991) ("Hake 1991 article").

3 AILA, Ethics in a Brave New World: Professional Responsibility, Personal Accountability, and Risk Management for Immigration Practitioners (John L. Pinnix et al., eds. 2004).

4 Bruce A. Hake, Dual Representation in Immigration Practice, in id. at 28 ("Hake 2004 article").

5 The updated Mehta article at 2-5, and also at 7-8, discusses three anomalous decisions, including that 2003 opinion (N.Y. State Bar Op. 761), that superficially may seem to support to the Simple Solution. The article shows why they do not really do so. However, see infra n.35 and accompanying text, showing that the Mehta article's reasons to discredit the 2003 opinion are incorrect, so the opinion remains problematic.

6 E.g., in November 2005, it was posted in the Just Posted section of the AILA Infonet, as if it were a new development, apparently the only member article that has been prominently posted like that. It appeared in connection with AILA's November 8, 2005 teleconference on "Ethical Issues For The Immigration Practitioner," in which the main topic was a contrast between the Mehta article and this author's work. The Mehta article is no longer on the Infonet; the Hake 2004 article does appear there in the Ethics section as AILA Doc. No. 05060724.

7 Kristina Rost, Ethical Dilemmas: A Never Ending Quandary, Immigr. L. Today 8 (AILA January/February 2006).

8 It's like cable news, where truth is not recognized, all positions are considered relative, and all positions are assumed to have alternatives of equal weight. In reality, some positions are objectively true while their opposites are objectively wrong. Fair legal scholarship does not assume that everything is relative and everyone has an equally valid opinion. Instead, it must show authority that actually supports a position. This author's articles have done so; the Mehta article has not. E.g., the U.S. Department of State recognizes the fact that journalism (which is twin to scholarship) must be founded on truth, not on a confusion of alternatives (emphasis added):
The concept of objectivity in journalism developed almost a century ago, as a reaction to the sensational, opinion-driven reporting that was common in most newspapers of the day. The term "objectivity" wasoriginally used to describe a journalistic approach or method; journalists would seek to present the news in an objective way, without reflecting any personal or corporate bias.

… Fairness is not the same thing as balance, however. Balance suggests that there are only two sides to any story, which is rarely the case, and that each side should be given equal weight. Journalists who seek that kind of artificial balance in their stories actually may produce coverage that is fundamentally inaccurate. ... A story giving equal time or space to the views of both groups would be misleading.
U.S. Dep't of State, Handbook of Independent Journalism (July 2006), located at /journalism/whatis.htm [last accessed May 17, 2007].

9 Paul L. Zulkie & Charles H. Kuck, Ethics and the Immigration Practitioner: Can One Straddle the Fence and Survive? Immigr. L. Today 44 (AILA November/December 2006).

10 Updated Mehta article, note 2.

11 "Yet there is a road to [virtue], a guide to excellence * * *: it is the middle way, the golden mean. The qualities of character can be arranged in triads, in each of which the first and last qualities will be extremes and vices, and the middle quality a virtue or an excellence. So between cowardice and rashness is courage ... ." Will Durant, The Story of Philosophy 75-76 (Pocket Books 1973, first published in 1926).

12 See Frisch, Gee, Grauer, Hake, Patrick, Ripley, Rosen & Whelan, Ethical Issues In Immigration Practice: A Roundtable Discussion, 90-8 Immigr. Briefings (Aug. 1990).

13 Updated Mehta article at 2. "Full-throttled" is not defined.

14 See, e.g., Hake 1991 article at 606-08; Hake 2004 article at 34.

15 Hake 1991 article at 620-24.

16 Updated Mehta article at 2-5; but see infra n.35 and accompanying text.

17 Updated Mehta article at 5-6. Incidentally, it is misleading to assert that "the very notion of dual representation results in an inherent limitation of representation as it implicates the attorney's cardinal duties of loyalty and confidentiality toward each client." In a typical dual representation, the lawyer has duties of loyalty and confidentiality towards each co-client, but this does not mean that the representation of either is inherently limited; it means that there is a potential for conflict, which if it arises must be dealt with according to well-established rules.

18 Id. at 6.

19 Id. at 6 (note omitted; emphasis added).

20 See Hake 1991 article at 620-26.

21 See Bruce A. Hake, Loyalty to a Former Client, 13 Immigr. L. Rep. 205-210 (Fragomen, Del Rey & Bernsen law firm Sept. 15, 1994).

22 Id. at 211.

23 Original Mehta article at 30-32.

24 Id. at 3 & n.8.

25 Exterior Systems, Inc., d/b/a/ Fabwell, Inc. v. Noble Composites, Inc. et al., 175 F. Supp. 2d 1112, 1120-21 (N.D. Ind. 2001).

26 Felix v. Balkin, 49 F. Supp. 2d 260, 270 (S.D.N.Y. 1999).

27 Original Mehta article at 31. This is a major theme of the original Mehta article. In the introduction at 29, it claims that when lawyers "assume the full responsibilities of dual representation, [they are then] compelled to withdraw from representing both parties in the event of even the slightest conflict (emphasis added)." This language about the "slightest" conflict is repeated in the updated Mehta article at 2. This is a red herring that unfairly discredits established dual representation principles, which actually hold that withdrawal is necessary only after an irreconcilable conflict has resisted all efforts to resolve it.

28 Original Mehta article at 31.

29 Updated Mehta article at 2-3.

30 New York City Bar Op. 2006-1 at 4 (2006).

31 Id. at 3.

32 Id. at 6.

33 Id.

34 Updated Mehta article at 7-8 (discussing N.Y. State Bar Op. 761 (2003)).

35 Id. at 4 (arguing that: "Notwithstanding this opinion, it would be difficult for the attorney to avoid considering the citizen husband as the client because U.S. Citizenship and Immigration Services (USCIS) requires the petitioner to sign a Notice of Appearance (Form G-28) in petition proceedings."; and citing 8 C.F.R. ' 292.4(a)). This is incorrect. That C.F.R. section only requires a Form G-28 if a lawyer makes an appearance for a petitioner. Thus, it does not require the petitioner to sign a G-28, because there is no law that says a G-28 must be filed with a petition; it only requires the petitioner to sign the G-28 if a G-28 is filed by a lawyer to notice his appearance. It is conceivable that a lawyer might represent a beneficiary, but not the petitioner, and file (or have the petitioner file) the petition without a G- 28. The USCIS would accept that. This would cause practical problems, such as no notice to the lawyer or beneficiary, but it is lawful. This 2003 New York State Bar opinion truly is incompatible with established dual representation principles. It deserves a detailed response in a future article. The opinion arose in the context of lawyers who represents indigent clients and need a way to protect the rights of abused wives. It is a good example of the maxim that hard cases make bad law.

36 Updated Mehta article at 8-9.

37 Consider two rules that underlie all dual representation situations (quoting the Maryland Rules of Professional Conduct): (1) "A lawyer shall not accept compensation for representing a client from one other than the client unless there is no interference with the lawyer's independence of professional judgment or with the client-lawyer relationship" (Rule 1.8(f)(2)); (2) "A lawyer may limit the scope of the representation if the limitation is reasonable under the circumstances and the client gives informed consent." Rule 1.2(c)). There are limits to how far the scope may be limited, and this may go too far.

38 Molina v. J.F.K. Tailor Corp., 01 Civ. 4016 (RLC) (KNF), 2004 U.S. Dist. LEXIS 7872 (S.D.N.Y. 2004).

39 U.S. v. Kozminski, 821 F.2d 1186, 1193 (6th Cir. 1987).

40 U.S. v. Gagnon, 230 F. Supp. 2d 260 (N.D.N.Y. 2002).

41 Id. at 268.

42 Id. at 272.

43 U.S. v. Garza, 429 F.3d 165 (5th Cir. 2005).

44 Id. at 174.

45 See also Rule 1.14(a) of the Maryland Rules of Professional Conduct, "Client with Diminished Capacity": "When a client's capacity to make adequately informed decisions in connection with a representation is diminished whether because of minority, mental impairment or for some other reason, the lawyer shall, as far as reasonably possible, maintain a normal client-lawyer relationship with the client." Rule 1.14(b) provides that in such circumstances a lawyer may need to take special action to protect the client. Colorado Regulatory Counsel (bar counsel) has stated to AILA members at a local conference that foreign national clients appear to fall under this rule, meaning that immigration lawyers have an enhanced responsibility toward their foreign clients compared to other lawyers, similar to the enhanced responsibilities a lawyer has to a mentally disabled client. Email from AILA member James Bonn (Apr. 13, 2007).