ATTORNEY-CLIENT PRIVILEGE

ATTORNEY-CLIENT PRIVILEGE
CONTINUING CONFUSION ABOUT
ATTORNEY COMMUNICATIONS,
DRAFTS, PRE-EXISTING DOCUMENTS,
AND THE SOURCE OF THE FACTS COMMUNICATED
Adapted from 48 Am. U.L. Rev. 967 (1999)
by Paul R. Rice
Introduction
The attorney-client privilege has become one of the most complex and, therefore, litigated
privileges. This has been due, in significant part, to the difficulties created by the concept of
confidentiality. From the creation of the privilege, through its preservation, to the factual proof
of the legitimacy of privilege claims, the requirement of confidentiality has created time-consum-
ing and costly responsibilities for both litigants and judges. In addition, confidentiality has been
interpreted as imposing a superfluous secrecy requirement that has led to conflicting decisions
and practices.
As judges and lawyers have forgotten fundamental principles, confusing and other
difficulties have mounted. Most prevalent among these has been the failure to remember the dis-
tinction between communications and information and the principle that the privilege protects
only communications. This has led to (1) confusion about how the privilege is applicable to the
attorney’s communication to the client; (2) unnecessary restrictions on the source of information
communicated by the client to the attorney; (3) misunderstandings about pre-existing documents
communicated by the client to the attorney; and (4) erroneous decisions regarding drafts of docu-
ments prepared for dissemination to third parties.
Whose Communications are Protected by the Privilege?
To understand which communications qualify for the protection of the attorney-client
privilege, it is necessary to first understand the rationale for the privilege. It is believed that if the
client is assured that what she says to a lawyer cannot later come back to harm her, she will be
more open and candid in her communications with the attorney--that is, willing to communicate
things that she otherwise would suppress. As a consequence, the attorney will be better
informed, and therefore, able to give more accurate advice. Therefore, the client’s ability to
conform her conduct to the requirements of the law will be improved. The basic privilege
protects only what the client communicates to the attorney.

---

1. Basic Protection
client attorney
÷ ÷ ÷

To adequately protect the client’s communications to the attorney, the responsive com-
munications from the attorney to the client are also protected by the privilege to the extent that
the responses reveal the content of what the client previously communicated. This is a derivative
protection that is dependent upon the continued privileged character of the previous revealed
communications. Therefore, to establish the privilege for the responsive communication of the
attorney, the client must prove that both sides of the conversation were privileged.
2. Derivative Protection
÷ ÷ ÷
client
attorney

² ² ²



Because of inevitable disclosures in attorney communications, many courts have begun to
define the attorney-client privilege loosely as protecting communications “between” the attorney
and client. This was spurred, in part, by the language in Proposed Federal Rule of Evidence 503
that used the same preposition. The problem created by this characterization of the privilege is
that it extends the privilege’s protection far beyond what is necessary to further its limited goal--
-2-

---

encouraging open and candid communications from the client. There is no apparent reason why
clients would be less candid in communications with attorneys if attorney communications
revealing information from other sources are discoverable.
As illustrated in the following graphic, the use of the preposition “between” obscures the
scope of the protection. It could mean no more than its common law progenitor illustrated at
3.a.--communications to the attorney and responsive communications that reveal the content of
what the client originally said. On the other hand, it might include any other communication
from the attorney, regardless of whether it reveals only the law that the attorney communicated to
the client (independent of its application to the client’s circumstances), illustrated at 3.b., or facts
acquired from third parties, illustrated at 3.c..
3. “Between”
÷
client ÷ ÷
attorney
a. ² ² b.

Law
² ² ²
8






c.

Facts
This recharacterization of the scope of the privilege protection resulted in a radical
expansion of the protection. Without either acknowledging this expansion or appreciating its
consequences, courts have widely adopted it. Although many courts have intended the expansion
to include only the responsive advice of the attorney, 3.b., most have given no indication of any
limitation. A few courts have even gone so far as to explicitly extend the privilege’s coverage to
all communications from the attorney, even those relaying only information from third parties,
3.c..
This expansion of the privilege has been appealing to judges for two reasons. First, it is
roughly accurate. Most responsive communications will reveal or depend on, to some extent, the
content of prior communications of the client. Second, and more importantly, this interpretation
-3-

---

of the privilege simplifies the resolution process. Eliminating the derivative theory removes the
need to prove two valid claims in order to successfully assert one. All the proponent has to
demonstrate under this expanded definition is that the lawyer communicated legal advice or
assistance to the client in confidence. The hidden dangers in this practice surfaced in the Ninth
Circuit's decision in United States v. Bauer, 132 F.3d 504, 508 (9th Cir.1997).
Bauer involved the prosecution of a bankruptcy petitioner for fraudulently failing to
report all of his assets in his bankruptcy petition. In his defense Bauer claimed that he innocently
withheld information about certain assets believing that they did not have to be reported.
Consequently, the principal focus of the trial was on Bauer's knowledge and intent. To establish
a knowing misrepresentation, the government called the defendant's former bankruptcy attorney
who, over the objections of Bauer, was required to testify that he had informed the defendant that
as a bankruptcy petitioner Bauer had a legal obligation to report all of his property in the petition,
and that any false statement would constitute perjury. On the appeal of his conviction, Bauer
claimed that the trial judge's order violated his attorney-client privilege. The Ninth Circuit
agreed.
The trial judge required the disclosure by the bankruptcy attorney on the belief that when
an attorney advises the client about the rules of the court, he is not acting as an attorney, but as an
officer of the court conveying public information. This decision was influenced by a well-estab-
lished body of law holding that when an attorney notifies a client of the dates on which the client
has been ordered by the court to appear for sentencing, there can be no reasonable expectation on
the part of the client that such communications are confidential. Consequently, the attorney can
be required to reveal that the client was advised of the appearance date when he is later tried for
failing to appear. The same result obtains in ascertaining when a deficiency notice from the IRS
was received by the client so that the time from which a petition for review should have been
filed can be established.
Without explaining why informing a client of where and when the law (represented in an
order by the presiding judge or a notice of deficiency from the IRS) requires the client to person-
ally appear is not legal assistance protected by the privilege, the court held that informing the
client of where and when the law (represented in an act of Congress) requires that a client’s
assets “appear” is legal advice that is privileged. This decision was a product of the
recharacterization of the attorney-client privilege as protecting communications “between” the
attorney and client--affording a direct protection to all responsive attorney communications.
Under the classical derivative theory for responsive attorney communications, this attor-
ney's communication about the abstract requirements of the law would not have been privileged
because it did not apply or interpret those principles in light of the client’s unique circumstances,
and thus did not reveal prior privileged communications of the client. And this would be true
regardless of whether the attorney's communications were characterized as “legal advice” or the
“transmission of public information.”
-4-

---

The Privilege Protects Communications, Not Information
One of the more difficult principles of the attorney-client privilege has been the distinc-
tion between the privilege’s protecting communications and not information. Confusing the
two, courts often have denied the privilege claim because the information within
communications has not been confidential, even if the communications themselves were
confidential.
The privilege protects what the client communicated to the attorney in confidence for the
purpose of obtaining legal advice or assistance. It protects the fact that the information was com-
municated (graphically illustrated below in the box at level 1), not the information itself, aside
from the fact of the communication. Therefore the client’s knowledge of Facts A, B, & C is not
privileged and they otherwise can be disclosed by him to others or inquired about by third parties.
If the client voluntarily testifies to the same facts communicated to the attorney, or
informally discloses them to third parties, but does not disclose that he previously related the
same information to his attorney, the privilege protection is unaffected. This is illustrated at level
2 in the graphic. Similarly, as illustrated in level 3, the client can be required to disclose Facts
A, B & C, because they are within his knowledge, but he cannot be required to disclose that he
communicated Facts A, B, & C to his attorney.
-5-

---

client PROTECTED COMMUNICATION attorney
1.


û Facts A, B, C û
úClient can voluntarily disclose
the same facts to a third party
without effecting the privilege
third party
2. û
û
Facts A, B, C
úClient can be required to disclose the same
facts to an opposing party during discovery
without effecting the privilege

Opposing party
3. û
û
Facts A, B, C
The subsequent disclosure of Facts A, B, & C waives the privilege protection
for the contents of the privilege box only if the client also volunteers that
those same facts were communicated to the attorney--viz., discloses that they
are the content of the box.
-6-

---

The privilege focuses on the communication. It is unconcerned about, and does
not affect, the information within the communication, as it exists outside the box. It does not
bestow an independent protection on that information and the privilege is not affected by the
information’s nature and status. The information does not have to be confidential for the
communication in which is incorporated to be confidential, and therefore, privilege.
The Source and Nature of the Information Communicated to the Attorney is
Irrelevant to Privilege Protection Applicable to the Client’s Communications
A persistent misperception is that a client’s communication of nonconfidential informa-
tion precludes the communication from being confidential, and therefore, the attorney-client
privilege from being applicable.
For example, an IRS field agent may communicate with IRS attorneys in Washington,
D.C., on a novel issue that he is confronting. That communication should be privileged if its pur-
pose is to obtain legal advice, regardless of the nature or source of the facts and information
included within the communication. Even if the agent’s communication predominantly consists
of tax information acquired from a taxpayer, the communication should be protected by the
government’s attorney-client privilege. However, the D.C. Circuit Court of Appeals in Tax
Analyst v. IRS, 117 F.3d 607 (D.C. Cir. 1997), recently held that because of the nonconfidential nature
of the information communicated, the government cannot have a reasonable expectation that
either the agent’s communication or the attorney’s response is confidential, and therefore,
privileged.
This decision was wrong. It erroneously focused on the information (the nature and
source of which is irrelevant to the privilege protection), rather than on the client’s
communication and the attorney’s response that reveals the communication’s content.
The court in Tax Analyst concluded that the communications from the attorneys in what is
called the Field Service Advice Memoranda (hereafter FSA), were not based on “‘confidential in-
formation obtained from the client,’” but rather on communications from outsiders, i.e., individu-
al taxpayers, that did not contain “‘any confidential information concerning the Agency.’”
Second, the court held that since the opinions of counsel were being used as a basis for agency
policy, those opinions were, in substance, law created by the IRS that should be applied with
consistency to all taxpayers, and therefore, should be subject to discovery by taxpayers.
The first line of reasoning--requiring the client to communicate confidential information
in order to assert the privilege--reflects a fundamental, albeit widely held, misunderstanding
about the privilege. The nature of the information contained in the communications from the
client to the attorney is irrelevant to the communications’ privileged status. Regardless of where
the client acquired the information, or of the information’s confidential or public nature, the
content of what the client communicated to the attorney is privileged. That is, if the client
acquires the information in his communication from the public record, the daily newspaper, or
direct communications from third parties (such as taxpayers), the content of what the client
communicated is privileged, and therefore, should not be discoverable from either the attorney or
the client. As illustrated in the following graphic, what is privileged is what was communicated-
-7-

---

-what is in the privilege box. The status of the information outside the privilege box is irrelevant
to the privilege status of the communications that employ the information.
SOURCE OR TYPE OF INFORMATION
IRRELEVANT

C
From third parties ú

C
From public record ú PRIVILEGE PROTECTION
client
attorney

û û û



C
Nonconfidential facts ü
C
Technical facts ü
ú While these facts may be discoverable
from the client, that does not effect the
privilege status of the communications
into which they are incorporated
In Tax Analysts v. IRS, 117 F.3d 607 (D.C. Cir 1997) the IRS field agent acquired nonconfi-
dential facts from a third party and sought advice based on those communicated facts. The agent
communicated these facts to the Office of Chief Counsel seeking legal advice on how best to
handle a particular issue, and the agent did so with the expectation of confidentiality. If the
responsive communications from the Office of Chief Counsel explored the legal position that
-8-

---

should be taken by the field agents, focusing on the unique facts of each inquiry, the responsive
communication should be afforded the derivative protection of the privilege. The Court
acknowledged that the FSAs gave guidance to field agents “with reference to the situation of a
specific taxpayer” and included “a statement of issues, a conclusions section, a statement of facts,
and a legal analysis section.” These memoranda “‘state any limitations or conditions to which a
conclusion may be subject, ” and are exploratory and descriptive in nature “‘so that the strengths
and weaknesses of a case are presented and developed candidly, directing attention to the author-
ities against the conclusions arrived at as well as those which support them.’”
The Court in Tax Analyst appeared to have followed the classical derivative definition of
the privilege for responsive legal advice, but inappropriately required that it reveal or “‘rest on
confidential information obtained from the client,’” rather than on confidential communications
of the client.
The Court’s second line of reasoning in Tax Analysts for requiring the disclosure of the
FSA’s was the unique nature of the legal opinions of IRS counsel. Because the IRS is interpret-
ing the law and applying it to individual taxpayers, the court characterized Chief Counsel’s
opinions as “making law” that should be accessible to taxpayers.
No private attorney has the power to formulate the law to be applied to others. Matters
are different in the governmental context, when the counsel rendering the legal opinion in
effect is making law. Here the Office of Chief Counsel is one of the principal tax law-
givers within the Executive Branch. Nearly all the interpretations of the tax laws the IRS
applies in assessing and collecting taxes emanate from the Office of Chief Counsel. . . .
As we have discussed previously, FSAs issued by the Chief Counsel create a body of pri-
vate law, applied routinely as the government’s legal position in its dealings with tax-
payers. It is this quality . . . that [makes them] significant.
While the Court had previously acknowledged that the FSAs are not binding on IRS field
personnel, it still characterized documents as “making law” simply because they are held in “high
regard” and “generally followed.” This, of course, is also true in the private context. Legal opin-
ions of counsel, particularly outside counsel, are highly respected and generally followed by
clients. This, however, does not change the privileged character of those opinions. It is the
actions that are taken by the client, whether IRS or private entity, that converts the advice into a
“body of law” controlling future actions. Although the client can be compelled to explain the
rationale for its actions, which may be identical to the legal advice previously obtained, the con-
tent of that advice should remain privileged until the client chooses to waive it by stating that the
client is specifically relying on it. If the logic of the opinion in Tax Analysts were generally
followed it would, at the very minimum, jeopardize the privileged status of the legal advice
obtained by all governmental regulatory agencies.
The Scope of Waiver--Drafts of Documents Prepared for Release to Third Parties
The most frequently litigated of all attorney-client privilege issues is waiver. It involves
difficult judgments about the preservation of confidentiality, and the scope of the resulting waiv-
-9-

---

er once it is found. Decisions concerning the existence and scope of waiver are based on the
standard of fairness. The question that must be answered is “[h]as the disclosure sufficiently
revealed the contents of the protected communication, so that to recognize the privilege, and
thereby continue to permit the concealment of the undisclosed portions, (or, in the case of an
entire communicaton having been disclosed, other related communications on the same subject),
would be permitting the client to have his cake and eat it?” Courts often express this in terms of
using the privilege as “both a sword and a shield.”
Differences of opinion are inevitable with these types of judgments. The judgments are
fact- dependent; each judge perceives the relevant facts differently; and even if their perceptions
were consistent, there is no way for judges to assign consistent values to each fact relative to its
importance in the “fairness” calculation.
There is a limited area of waiver, however, where courts can reasonably be criticized for
misinterpreting the scope of waiver. This is where documents like government reports,
prospectuses, or patent applications are prepared for dissemination to third parties.
The privilege usually protects confidential communications that the client intends to con-
tinue to hold in confidence. The client communicates with the attorney with the expectation that
the attorney will advise him on appropriate courses of action, and will assist in the taking of that
action. Even if the recommended actions are taken, the confidential communications of both the
attorney and client remain confidential and unavailable to third parties because the actions do not
necessarily reveal the content of prior communications. The situation is a bit different when the
client consults the attorney for the purpose of drafting an instrument that ultimately will be made
public. The construction and dissemination of the instrument inevitably reveals a great deal of
substance about the content of the prior communications between the attorney and client.
If a lawyer is consulted for the purpose of drafting a communication or instrument that
eventually will be disclosed to third parties (a prospectus, for example), are its prior drafts pro-
tected by the attorney-client privilege? Stated differently, if those drafts were initially considered
confidential, would the client’s continuing expectation be reasonable after the dissemination of
the final approved version?
In the following graphic the attorney has prepared three prior drafts of the instrument that
will be disseminated. When the Fourth Draft is approved for release, and expectations of confi-
dentiality with respect to that draft are relinquished, should the scope of the resulting privilege
waiver extend to the three previous drafts? A number of courts have said yes. They have held
that when a document that will be disseminated to third parties is being prepared by the attorney,
there can be no reasonable expectation of confidentiality in the communications leading to the
creation of the final product because it has been forfeited. These decisions alternatively hold
that the client never had a reasonable expectation of confidentiality. Both conclusions are
unrealistic and unfair.
-10-

---