The Attorney-Client Privilege and the Corporate Client: Where Do We Go after Upjohn?

Source: Michigan Law Review , Jan., 1983, Vol. 81, No. 3 (Jan., 1983), pp. 665-692
Published by: The Michigan Law Review Association

Stable URL: [Link]

REFERENCES
Linked references are available on JSTOR for this article:
[Link]
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@[Link].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
[Link]

The Michigan Law Review Association is collaborating with JSTOR to digitize, preserve and
extend access to Michigan Law Review

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
The Attorney-Client Privilege and the Corporate Client:
Where Do We Go After Upjohn?

The purpose of the attorney-client privilege is to encourage full

and frank communication between clients and their attorneys.' The
increased communication which presumably results from the privi-
lege2 enables attorneys to render better legal advice, ostensibly in-
creasing compliance with the law and improving the administration
of justice.3 Once litigation arises, however, the privilege removes

1. Upjohn Co. v. United States, 449 U.S. 383, 389 (1981); Trammel v. United States, 445
U.S. 40, 51 (1980); Fisher v. United States, 425 U.S. 391, 403 (1976); Hunt v. Blackburn, 128
U.S. 464, 470 (1888).
The attorney-client privilege is the oldest of the evidentiary privileges, dating back to at
least the sixteenth century. 8 J. WIGMORE, EVIDENCE ? 2290, at 542-43 (McNaughton rev.
1961) [hereinafter cited as J. WIGMORE]. see also Gergacz, Attorney-Corporate Client Privilege,
37 Bus. LAW. 461, 473 (1982) (suggesting that the privilege originated with the Romans); see
generally Hazard, An Historical Perspective on the Attorney-Client Privilege, 66 CALIF. L. REV.
1061, 1070 (1978) (suggesting that the history of privilege provides no authority for construing
the privilege broadly); Gardner, A Re-evaluation of the Attorney-Client Privilege, 8 VILL. L.
REV. 279, 286-304 (1963). Radin, The Privilege of Confidential Communication Between Lawyer
and Client, 16 CALIF. L. REV. 487, 489 (1928).
The justification for the privilege has changed over time. Originally, the privilege was
designed to preserve "the oath and honor of the attorney." The privilege belonged to the
attorney, not the client. Only the attorney could waive the privilege because it was his honor
that was at stake. In addition, the privilege terminated at the conclusion of the particular
client's litigation. See J. WIGMORE, supra at 543-45.
The "oath of honor" jusitification was discarded in the late 1700s. Today, the privilege
belongs to the client. It exists independently of any litigation, and can only be waived by the
client. J. WIGMORE, supra at 544-45. Moreover, if the client has not waived the privilege, the
attorney must assert the privilege upon any attempt to require him to testify about confidential
conversations or to produce confidential documents. This duty is a "matter of profesional re-
sponsibility in preventing the policy of the law from being violated." Schwimmer v. United
States, 232 F.2d 855, 863 (8th Cir. 1956); see also MODEL CODE OF PROFESSIONAL RESPONSI-
BILITY Canon 4 ("A Lawyer Should Preserve the Confidences and Secrets of a Client").
2. While it is presumed that the privilege encourages communications between clients and
attorneys, that presumption has not been verified empirically. See, e.g., Kobak, The Uneven
Application of the Attorney-Client Privilege to Corporations in the Federal Courts, 6 GA. L. REV.
339, 339 (1972); Note, The Attorney-Client Privilege and the Corporation in Shareholder Litiga-
tion, 50 S. CAL. L. REV. 303, 306 (1977) ("[N]o empirical data is available which will accu-
rately demonstrate whether the privilege actually encourages communications between
attorneys and their clients.") (footnote omitted). The only empirical study conducted to date
produced inconclusive results. See Note, Functional Overlap Between the Lawyer and Other
Professionals. Its Implicationsfor the Privileged Communications Doctrine, 71 YALE L.J. 1226,
1262 (1962) [hereinafter cited as Yale Note] (55 of 108 laymen questioned would be less likely
to provide information to counsel if the privilege were abolished, 37 would not be less likely,
and 16 did not know).
Nevertheless, Note, Attorney-Client Privilege For Corporate Clients. The Control Group
Test, 84 HARV. L. REV. 424, 427-29 (1970) [hereinafter cited as Harvard Note] argues that the
privilege may not increase communications between corporate clients and their attorneys.
While this proposition may in fact be true, courts will probably not challenge this long held
assumption. See note 5 infra.
3. Upjohn Co. v. United States, 449 U.S. 383, 389 (1981). The Upjohn Court stated that
the privilege "promote[s] broader public interests in the observance of law and administration

665

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
666 Michigan Law Review [Vol. 81:665

these communications from the discovery process, thereby impeding

the search for the truth.4 Because the privilege has clear costs and
largely speculative benefits, courts and commentators argue for the
narrowest possible construction of the privilege that is "consistent
with the logic of its principle."5
In applying the privilege to individual clients, the courts have
generally struck a balance between these competing interests by ap-
plying Professor Wigmore's eight part formulation of the privilege.6

of justice. The privilege recognizes that sound legal advice or advocacy serves public ends and
that such advice or advocacy depends upon the lawyer being fully informed by the client."
449 U.S. at 389. See MCCORMICK'S HANDBOOK OF THE LAW OF EVIDENCE ? 87, at 175 (E.
Cleary 2d ed. 1972) [hereinafter cited as MCCORMICK]; Miller, The Challenges to the Attorney-
Client Privilege, 49 VA. L. REV. 262, 268-70 (1963) (attorneys can give more forthright advice,
thereby reducing the possibility of misleading clients). Of course, both clients and attorneys
benefit from the privilege. Clients benefit from being able to disclose information to attorneys
without fear of having the information subject to discovery; attorneys benefit from being fully
informed of their client's affairs. See Gergacz, supra note 1, at 464-68.
4. J. WIGMORE, supra note 1, ? 2291, at 554. (The privilege is "an obstacle to the investiga-
tion of the truth."). The privilege also contravenes the modern theory of discovery that "par-
ties . . . [should] obtain the fullest possible knowledge of the issues and facts before trial."
Hickman v. Taylor, 329 U.S. 495, 500 (1947); United States v. Bryan, 339 U.S. 323, 331 (1950)
("[T]he public . .. has a right to [know] every man's evidence."). See generally 8 C. WRIGHT
& A. MILLER, FEDERAL PRACTICE AND PROCEDURE ? 2001 (1970).
5. J. WIGMORE, supra note 1, ? 2291, at 554 (footnote omitted). A strong argument can be
made that the privilege should not exist. According to Dean Wigmore, one fundamental con-
dition for establishing a privilege is that "[t]he injury that would inure to the relation by the
disclosure of the communications must be greater than the benefit thereby gained for the cor-
rect disposal of litigation." J. WIGMORE, supra note 1, ? 2285, at 527 (emphasis in original).
The Model Code of Evidence accepts the common perception that this condition is satisfied,
MODEL CODE OF EVIDENCE Rule 210 Comment a (1942), but many commentators argue that
the issue is unresolved. J. WIGMORE, supra note 1, ? 2285, at 528; see, e.g., Weissenberger,
Toward Precision in the Application of the Attorney-Client Privilegefor Corporations, 65 IOWA L.
REV. 899, 899 (1980) ("The most that is usually said on behalf of the privilege is that, without
it, the rendition of legal services would be virtually impossible.") (footnote omitted); Kobak,
supra note 2, at 339; Harvard Note, supra note 2, at 425 "[T]here is some reason to doubt
whether the privilege even achieves the benefits claimed of it.") (footnote omitted); Note, Evi-
dence - Privileged Communications - The Attorney-Client Privilege in the Corporate Setting:
A SuggestedApproach, 69 MICH. L. REV. 360, 364 (1970) [hereinafter cited as Michigan Note]
("[I]t is difficult - if not impossible - to determine whether the privilege does in fact promote
freedom of consultation and encourage full disclosure of material information .. ."); Yale
Note, supra note 2, at 1236-46 (1962) (after discussing costs and benefits of the privilege, the
Note concludes that bar and bench would vigorously oppose the elimination or alteration of
the privilege and therefore change "is unlikely.").
Nevertheless, almost no one seriously advocates abolishing the privilege. See Chirac v.
Reinicker, 24 U.S. (11 Wheat.) 280, 294 (1826) (privilege is "indispensable for the purposes of
private justice"); compare MCCORMICK, supra note 3, at 176 (society would be "outraged" at
any attempt to abolish the privilege) and Hazard, supra note 1, at 1062 ("[T]he issue concern-
ing the attorney-client privilege is not whether it should exist, but precisely what its terms
should be."); with Gardner, A Personal Privilegefor Communications of Corporate Clients
Paradox or Public Policy?, 40 U. DET. L.J. 299, 309, 354 (1963) ("[N]o well-reasoned argument
jusitifies the extension of the privilege to corporations.") and Radin, supra note 1, (questioning
the policy reasons behind the privilege).
6. See, e.g., Fisher v. United States, 425 U.S. 391, 403 (1976); United States v. Osborn, 561

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 667

Applying the privilege to corporations7 has proven more problematic

because corporations can request legal advice or make decisions
based on that advice only through their agents.8 In addition, unlike
most individuals, corporations-particularly large ones-continu-
ously consult attorneys,9 which creates the intolerable possiblity of
F.2d 1334 (9th Cir. 1977); United States v. Judson, 322 F.2d 460 (9th Cir. 1963); Colton v.
United States, 306 F.2d 633, 639 (2d Cir. 1962), cert. denied, 371 U.S. 951 (1963).
Under Dean Wigmore's standard, a communication is privileged if the following condi-
tions are satisfied.
(1) Where legal advice of any kind is sought (2) from a professional legal advisor in his
capacity as such, (3) the communication relating to that purpose, (4) made in confidence
(5) by the client, (6) are at his instance permanently protected (7) from disclosure by
himself or by the legal advisor, (8) except if the protection be waived.
J. WIGMORE, supra note 1, ? 2292, at 554.
7. Originally, courts assumed that the privilege applied to communications from corporate
clients. See, e.g., Schwimmer v. United States 232 F.2d 855, 863-64 (8th Cir. 1956); Cole v.
Hughes Tool Co., 215 F.2d 924, 930-31 (10th Cir. 1954); Belanger v. Alton Box Bd. Co., 180
F.2d 87, 93-94 (7th Cir. 1950). One district court challenged this assumption, holding that the
privilege did not apply to corporate communications. Radiant Burners, Inc. v. American Gas
Assn., 207 F. Supp. 771 (N.D. Ill. 1962), affd on rehearing, 209 F. Supp. 321 (1962). However,
the decision was reversed on appeal. Radiant Burners, Inc. v. American Gas Assn. 320 F.2d
314 (7th Cir.), cert. denied, 375 U.S. 929 (1963).
In Upjohn, the Supreme Court "assumed that the privilege applies when the client is a
corporation." Upjohn Co. v. United States, 449 U.S. 383, 390 (1981) (citing United States v.
Louisville & Nashville R.R., 236 U.S. 318, 336 (1915), where the Court declared: "The desira-
bility of protecting confidential communications between attorney and client as a matter of
public policy is too well known ... to need extended comment now."); Bell v. Maryland, 378
U.S. 226, 263 (1964) (appendix 1 to opinion of Douglas, J.) (concurring opinion) ("A corpora-
tion, like any other client, is entitled to the attorney-client privilege."); Simon, The Attorney-
Client Privilege As Appliedto Corporations, 65 YALE L.J. 953, 955 (1956) ("[I]f our laws tend to
be good, good legal advice should benefit society as well as General Motors."). But see Gard-
ner, supra note 5, at 354.
8. Dartmouth College v. Woodward, 17 U.S. (4 Wheat.) 518, 636 (1819) (Marshall, C.J.)
("A corporation is an artificial entity, invisible, intangible, and existing only in contemplation
of law."); Diversified Indus., Inc. v. Meredith, 572 F.2d 596, 608 (8th Cir. 1977) (en banc) ("[A]
corporation can speak or hear only through its ... agents .. ."); 2 J. WEINSTEIN & M. BER-
GER, WEINSTEIN'S EVIDENCE ? 503(b)[04], at 503-41 to 44 (1981) [hereinafter cited as WEIN-
STEIN]; Rosenfeld, The Transformation of the Attorney-Client Privilege. In Search of an
Ideological Reconciliation of Individualism, The Adversary System, and the Corporate Client's
SEC Disclosure Obligations, 33 HASTINGS L.J. 495, 516 (1982); Simon, supra note 7, at 956 ("A
corporation can only speak through its representatives."); Weissenberger, supra note 5, at 905;
Michigan Note, supra note 5, at 374 (corporations have "decision makers" and "givers of infor-
mation") (emphasis in original).
9. See Upjohn v. United States, 449 U.S. 383, 392 (1981) ("In light of the vast and compli-
cated array of regulatory legislation confronting the modern corporation, corporations, unlike
individuals, 'constantly go to lawyers to find out how to obey the law.' ") (quoting Burnham,
The Attorney-Client Privilege in the Corporate Arena, 24 Bus. LAW. 901, 913 (1969)). See, e.g.,
Gergacz, supra note 1, at 478 ("corporations have a regular need for legal advice"); Weis-
senberger, supra note 5, at 902 ("[c]orporations tend to have more continuing interaction with
lawyers than do natural persons."); Note, The Attorney-Client Privilege - Identifying the Cor-
porate Client, 48 FORDHAM L. REV. 1281, 1307 (1980) ("integral role of attorneys in the daily
function of corporations") (footnote omitted). Corporate clients also rely on attorneys for
business advice, which enlarges the potential zone of silence. See, e.g., Simon, supra note 7, at
969 ("lawyers . . . participate in business decisions"); Weissenberger, supra note 5, at 902
(need "incisive test for separating legal from nonlegal communications, [or else] . . . the zone
of silence . . . may become abusively large.") (footnote omitted). See generally Yale Note,
supra note 2. The court's difficulty in separating legal advice from business advice is vividly
illustrated in cases where an attorney assisted with the preparation of a tax return. Compare

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
668 Michigan Law Review [Vol. 81:665

insulating from discovery all information provided attorneys unless

the privilege is narrowly applied.10 Given the conceptual difficulty
of developing a standard that provides corporations with the same
protection as that afforded individuals, while preventing corporate
attorneys from becoming repositories of undiscoverable information,
it is not surprising that courts have formulated several different stan-
dards for applying the attorney-client privilege to corporate
communications. l
In Upjohn v. United States, 12 the Supreme Court rejected one of

Canaday v. United States, 354 F.2d 849, 857 (8th Cir. 1966) (lawyer who prepared tax returns
acted as a mere scrivener; therefore communications to him not privileged) with Colton v.
United States, 306 F.2d 633, 639 (2d Cir. 1962), cert. denied 371 U.S. 951 (1963) (communica-
tions during preparation of tax returns privileged).
10. Simon, supra note 7, at 955-56 originally phrased this potential abuse as the "zone of
silence":
Where corporations are involved, with their large numbers of agents, masses of docu-
ments, and frequent dealings with lawyers, the zone of silence grows large. Few judges -
or legislators either, for that matter - would long tolerate any common law privilege that
allowed corporations to insulate all their activities by discussing them with legal advisers.
Simon's "zone of silence" is not as ominous as it would first appear. See notes 90-98 and
accompanying text.
11. Prior to the Upjohn decision, federal courts used numerous different standards to deter-
mine whether the privilege applies to a particular corporate communication. This Note dis-
cusses four of those tests: the control group test, the subject matter test, the modified subject
matter test, and the Ampicillin test. The first three tests are discussed in detail in this Note.
See notes 12-18, 27-41, 42-74 infra and accompanying text. For a discussion of Ampicillin, see
note 60 infra.
12. 449 U.S. 383 (1981). In Upjohn, the drug manufacturer's counsel conducted an inter-
nal investigation concerning payments to foreign government officials by one of its subsidiar-
ies. Managers of the foreign operations completed questionnaires and several employees were
interviewed. Upon completion of the investigation, Upjohn disclosed several questionable
payments to the Securities and Exchange Commission under the Commission's Voluntary Dis-
closure Program. The Internal Revenue Service then began a criminal investigation into the
tax consequences of the payments and demanded to see the questionnaires and interview
notes. Upjohn claimed they were protected by the attorney-client privilege and by the work-
product doctrine. See note 28 infra for a discussion of the work-product doctrine.
Investigations of corporate behavior such as illegal payments, commercial bribery, and
other questionable payments have become quite common since the duty to disclose such be-
havior was imposed by the passage of the Foreign Corrupt Practices Act of 1977, Pub. L. No.
95-213, ?? 101-104, 91 Stat. 1494 (1980) (amending 15 U.S.C. ? 78q(b), 78ff(a) (1976)). See
generally Block & Barton, Internal Corporate Investigations. Maintaining the Confidentiality of
a Corporate Clients Communications with Investigative Counsel, 35 Bus. LAW. 5, 6 (1979).
Prior to passage of the Foreign Corrupt Practices Act, the Commission introduced the Volun-
tary Disclosure Program in 1976 to allow corporations to disclose illegal or questionable pay-
ment voluntarily and avoid the SEC's enforcement mechanism. See generally [1975-1976
Transfer Binder] FED. SEC. L. REP. (CCH) 1 80,600.
Courts are divided on whether such disclosures themselves constitute a waiver of the privi-
lege and the issue was not resolved in Upjohn. Some courts have held that such disclosure
waives the attorney-client privilege, see, e.g., In re Weiss, 596 F.2d 1185, 1186 (4th Cir. 1979)
(grand jury proceedings); SEC v. Dresser Indus. Inc., 453 F. Supp. 573, 576 (D.D.C. 1978),
affd, 628 F.2d 1368 (D.C. Cir.), cert. denied, 449 U.S. 993 (1980) (SEC proceedings), while
others have held that, in the interest of fostering disclosure, such disclosures do not waive the
privilege as to other parties seeking the same documents. See, e.g., Diversified Indus., Inc. v.
Meredith, 572 F.2d 596, 611 (8th Cir. 1977) (en banc); In re Grand Jury Subpoena Dated July
13, 1979, 478 F. Supp. 368, 373 (E.D. Wis. 1979). The Court of Appeals in Upjohn adopted a
middle position: "The . .. voluntary disclosures . .. amount to a waiver of the privilege only

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 669

the more popular standards developed by the lower courts - the

control group test. Under that test, the privilege only extends to
communications made by those employees "who play a substantial
role in deciding and directing the corporation's response to the legal
advice given."'3 The test arguably offered a reasonable solution to
the unique problems posed by corporate clients because it en-
couraged corporate officials possessing decision-making power to
communicate with counsel, but prevented corporations from creating
a broad "zone of silence" by funnelling communications from lower-
level employees to counsel.'4

with respect to the facts actually disclosed." United States v. Upjohn Co., 600 F.2d 1223, 1227
n.12 (6th Cir. 1979). The Supreme Court did not address the waiver issue in Upjohn, leaving
one to wonder whether the Sixth Circuit's approach is still viable. See Comment, Upjohn Co.
v. United States: A Functional Expansion of the Attorney -Corporate Client Privilege, 67 IOWA
L. REV. 161, 176 n.117 (1981) [hereinafter cited as Iowa Comment] (suggesting that the Sixth
Circuit's approach "may well [have] surviv[ed] intact"); Compare Nath, Upjohn: A New Pre-
scriptionfor the Attorney - Client Privilege and Work Product Defenses in Administrative Inves-
tigations, 30 BUFFALO L. REV. 1, 58-65 (arguing that disclosure to the SEC should constitute a
waiver); with Comment, Corporate Self-Investigations Under the Foreign Corrupt Practices Act,
47 U. CHI. L. REV. 803, 816-23 (1980) (arguing that the courts should either adopt the subject
matter test or expand the self-investigative privilege for disclosure made pursuant to the For-
eign Corrupt Practices Act).
13. United States v. Upjohn Co., 600 F.2d 1223, 1226 (6th Cir. 1979) (footnote omitted),
revd, 449 U.S. 383 (1981).
The control group test was first announced in Philadelphia v. Westinghouse Elec. Corp.,
210 F. Supp. 483 (E.D. Pa.), mandamus andprohibition denied sub nom. General Elec. Co. v.
Kirkpatrick, 312 F.2d 742 (3d Cir. 1962), cert. denied, 372 U.S. 943 (1963). Judge Kirkpatrick,
in an often quoted paragraph, declared:
Keeping in mind that the question is, is it the corporation which is seeking the lawyer's
advice when the asserted privileged communication is made?, the most satisfactory solu-
tion, I think, is that if the employee making the communication, of whatever rank he may
be, is in a position to control or even to take a substantial part in a decision about any
action which the corporation may take upon the advice of the attorney, or if he is an
authorized member of a body or group which has that authority, then, in effect, he is (or
personifies) the corporation when he makes his disclosure to the lawyer and the privilege
would apply.
210 F. Supp. at 485.
Prior to the Upjohn decision, the control group test was followed by the Third, Sixth, and
Tenth Circuits, as well as various District Courts. See, e.g., United States v. Amerada Hess
Corp., 619 F.2d 980 (3d Cir. 1980); United States v. Upjohn Co., 600 F.2d 1223, 1224-25 (6th
Cir. 1979), revd, 449 U.S. 383 (1981); In re Grand Jury Investigation, 599 F.2d 1224 (3d Cir.
1979); Natta v. Hogan, 392 F.2d 686, 692 (10th Cir. 1968); Jarvis, Inc. v. American Tel. & Tel.
Co., 84 F.R.D. 286, 291 (D. Colo. 1979); United States v. Lipshy, 492 F. Supp. 35, 42-43 (N.D.
Tex. 1979); Virginia Elec. & Power Co. v. Sun Shipbuilding & Dry Dock Co., 68 F.R.D. 397,
400 (E.D. Va. 1975); Burlington Indus. v. Exxon Corp., 65 F.R.D. 26, 35-36 (D. Md. 1974);
United States v. International Bus. Mach. Co., 66 F.R.D. 206, 209-10 (S.D.N.Y. 1974); Con-
goleum Indus., Inc. v. GAF Corp., 49 F.R.D. 82, 84 (E.D. Pa. 1969), affd, 478 F.2d 1398 (3d
Cir. 1973).
14. See United States v. Upjohn Co., 600 F.2d at 1227; Accord Nath, supra note 12, at 52-
53; Harvard Note, supra note 2, at 429-32. Cf. Philadelphia v. Westinghouse Elec. Corp., 210
F.2d at 485 (suggesting that control group test was necessary to avoid a conflict with Hickman
v. Taylor, 329 U.S. 495 (1947)). Several commentators disagree with the Westinghouse ration-
ale. See, e.g., Weissenberger, supra note 5, at 909 n.47; Weinschel, Corporate Employee Inter-
views and the Attorney-Client Privilege, 12 B.C. IND. & COM. L. REV. 873, 875 (1971) ("It does
not follow from the Hickman decision that the attorney-client privilege does not extend to
communications by [noncontrol group] employees to corporate attorneys."). See note 28 infra.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
670 Michigan Law Review [Vol. 81:665

Notwithstanding these benefits, the Upjohn court rejected the

control group test, reasoning that its narrow scope frustrated the pur-
poses of the privilege in three respects. First, the test inhibited the
rendering of sound legal advice by denying the privilege to commu-
nications from middle- and lower-level employees15 even though

In Upjohn, a number of the communications from company employees to counsel would

not have been privileged because the employees were not members of the control group. How-
ever, the Sixth Circuit, while virtually conceding that senior corporate officials are control
group members, declined to determine which other employees were members of the control
group, and remanded the case to the district court. 600 F.2d at 1227-28.
15. 449 U.S. at 391-92; see, e.g., Diversified Indus., Inc. v. Meredith, 572 F.2d 596, 608-09
(8th Cir. 1977) (en banc) (quoting Weinschel, supra note 14, at 876); Michigan Note, supra
note 5, at 374.
This view, however, is not without its skeptics. See, e.g., In re Grand Jury Investigation
599 F.2d 1224, 1236 (3d Cir. 1979); Nath, supra note 12, at 41-44. These critics basically argue
that fear of reprisal from superiors or loyalty to the corporation motivates lower-level employ-
ees to communicate with counsel. See In re Grand Jury Investigation, 599 F.2d at 1236
("[Nion-control-group employees have little reason not to relate information to corporate
counsel, especially where a superior has instructed them to do so."); Nath, supra note 12, at 43
("The dominant impetus is, and remains the employee's loyalty to this corporation."). These
critics reason that the employee makes the communication at his own risk because the corpora-
tion, acting through its control group, can waive the privilege, thereby exposing the lower-level
employee to prosecution. See Diversified Indus., Inc. v. Meredith, 572 F.2d at 611 n.5; cf
Note, Permian Corporation v. United States and the Attorney-Client Privilegefor Corporations.
Unjustified Severity on the Issue of Waiver, 77 Nw. U.L. REV. 223 (1982) (discussing corporate
waiver in the context of disclosures to regulatory agencies). Because the privilege belongs to
the corporation, the employee cannot claim the privilege if he is prosecuted. In re Grand Jury
Proceedings, Detroit, Mich., August 1977, 434 F. Supp. 648, 650 (E.D. Mich. 1977), affd, 570
F.2d 562 (6th Cir. 1978); In re Grand Jury Subpoena Duces Tecum, 391 F. Supp. 1029, 1034
(S.D.N.Y. 1975). Thus, these authorities conclude that the only possible justification for the
employee's communication is his duty to the corporation. Accordingly, the privilege should
not extend to such communications because they will be made in the absence of the privilege.
While this argument demands further analysis by the courts and commentators, several
responses may be offered in defense of a test that extends the privilege to communications of
lower-level employees. First, the argument for denying the privilege to lower-level employees
simply replaces the assumption that the privilege encourages communication with another as-
sumption - that it is really the employee's duty to the corporation that is the impetus for
disclosure, without offering any evidence to support such an assumption. Second, even though
in theory the control group can waive the privilege, this threat of waiver may be more ap-
parant than real. If the lower-level employee is reasonably assured that the communication
will be privileged and that the privilege will not be waived, he will be more willing to disclose
information to counsel even when that information may be incriminating.
More fundamentally, the employee's perception of what loyalty to the corporation or the
fear of reprisal dictates in a particular situation may depend on the availability of the privilege.
If the corporation cannot assert the privilege regarding communications from lower-level em-
ployees, the discovery of which may dramatically increase the risk that the corporation will
incur civil or criminal liability, the corporation will have a very strong incentive to discourage
such self-initiated reports. Would, for example, the management of Upjohn have initiated a
major internal investigation with the knowledge that its fruits would be discovered by the
government? The vigor with which the corporation fought discovery suggests that it would not.
And if a corporation would not institute an investigation because of the apprehension that the
record developed internally might become subject to discovery, that same apprehension proba-
bly suffices to foster a corporate environment hostile to employee reports of corporate wrong-
doing. The importance of such an environment should not be under-emphasized:
Organizational processes, then, create an internal moral and intellectual world in which
the individual identifies with the organization and the organization's goals. The survival
of one becomes linked to the survival of the other, and a normative environment evolves
that, given difficulty in attaining organizational goals, encourages illegal behavior to at-

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 671

they often possess the information most needed by counsel to pro-

vide sound legal advice.16 Second, the test frustrated attorneys' ef-
forts at ensuring compliance with the law by denying the privilege to
communications from attorneys to lower level employees who often
are responsible for implementing corporate policy.17 Third, the un-
certainty caused by varying definitions of "control group" inhibited
communications between corporate officials and attorneys.18

tain those goals. But some finer distinctions must be made. Not all agents of an organiza-
tion will act illegally in the organization's behalf. The nature of the response - lawful or
unlawful behavior - will be shaped by structural factors both internal and external to the
organization. While organizations may experience structural tensions to violate, varia-
tions in subunit membership, position in the information system, and in rewards and pun-
ishments may undermine the organization's ability to unify the goals and actions of its
members with its own goals and actions, producing either deviance or conformity to legal
norms.

Vaughan, Toward Understanding Unlawful Organizational Behavior, 80 MICH. L. REV. 1377,

1391 (1982) (footnotes omitted). Conversely, if the privilege attaches to such self-initiated re-
ports of possible misconduct, the corporation has every incentive to encourage these inquiries.
Such an environment, moreover, would indicate to employees that loyalty to the corporation
dictates a willingness to report potential illegality, rather than to suppress it for fear of ultimate
discovery by adversaries in litigation. Perhaps this contribution to internal policing does less
to encourage compliance with the law than the expanded enforcement of liability that would
follow the unconfined discovery of internal requests for legal advice. But as Upjohn itself reaf-
firmed, the existence of the privilege in any form depends on the judgment that informed
counsel can do more to secure obedience to law than can unlimited discovery. Given that
corporate decision-makers confront some risk of liability even when their own investigations
are not subject to discovery, the extension of this reasoning to the corporate context does not
seem implausible.
16. Nath and other critics have also suggested that the privilege does not encourage com-
munications between any corporate employees and attorneys. Nath argues that "potential ex-
posures to securities and tax law violations, derivative suits, charges of currency violations, and
other independent business reasons [provide] the impetus" for investigation of questionable
behavior. Nath, supra note 12, at 44. Thus, the communications had to take place, quite apart
from the privilege. See Harvard Note, supra note 2, at 428-29 (in some situations, business
reasons motivate communication to counsel); Note, The Attorney-Client Privilege: Fixed Rules,
Balancing and Constitutional Entitlement, 91 HARV. L. REV. 464, 474-75 (1977) ("continuing
desire of corporation to obtain legal advice" vitiates need for certainty in applying the privi-
lege.).
Again, the fundamental flow of these arguments is that they replace one unverifiable as-
sumption with another. Moreover, the logical extension of these arguments is that the privi-
lege should not extend to individuals who conduct business in a sole proprietorship. Given
prevailing authority, such a position is untenable.
Moreover, the existence of independent incentives to investigate potential wrongdoing does
not diminish the additional incentive provided by the privilege, especially when mediated
through the corporate reaction to internal disclosures. Where the fear of liability inspires the
internal investigation, the corporation may be especially hostile to learning the worst in discov-
erable form. Since the fear of litigation will arise independently of information subsequently
obtained, the corporation will not lose its incentive to investigate because confirmation of this
fear is not discoverable; even without internal disclosure, whatever prompted the corporate
apprehension of suit or prosecution will suffice to encourage compliance with the law. The
privilege thus may reinforce other incentives for self-regulation, while encouraging internal
disclosure which otherwise might not occur.
17. 449 U.S. at 392 (quoting Duplan Corp. v. Deering Milliken, Inc., 397 F. Supp. 1146,
1164 (D.S.C. 1974)) ("After the lawyer forms his or her opinion, it is of no immediate benefit
to the Chairman of the Board or the President. It must be given to the corporate personnel
who will apply it.").
18. 449 U.S. at 393 ("An uncertain privilege, or one which purports to be certain but re-

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
672 Michigan Law Review [Vol. 81:665

Unfortunately, the Upjohn Court expressly declined to comment

on the other formulations of the privilege developed by the lower
courts.19 Thus, except for the factors listed in the opinion,20 the

suits in widely varying applications by the courts, is little better than no privilege at all.").
Unpredictability was one of the reasons Upjohn filed for certiorari. See Petitioner's Brief for
Certiorari at 10, Upjohn Co. v. United States, 449 U.S. 383 (1981)
The courts had in fact employed a varying definition of "control group." Compare Con-
goleum Indus. v. GAF Corp., 49 F.R.D. 82, 83-85 (E.D. Pa. 1969) (Director of Research is not
a control group member) with Hogan v. Zletz, 43 F.R.D. 308, 315-16 (N.D. Okla. 1967), affd in
part sub nom. Natta v. Hogan, 392 F.2d 686 (10th Cir. 1968) (Group leader in Research and
Development Department is a control group member). See also Duplan Corp. v. Deering
Milliken, Inc., 397 F. Supp. 1146, 1164 (D.S.C. 1974) (terms such as "manager," "officer,"
"representative" are insufficient to establish membership in control group); Philadelphia v.
Westinghouse Elec. Corp., 210 F. Supp. 483, 485 (E.D. Pa.), mandamus andprohibition denied
sub nom. General Elec. Co. v. Kirkpatrick, 312 F.2d 742 (3d Cir. 1962), cert. denied, 372 U.S.
943 (1963) (manager, executive, officer, "are not clearly defined terms."); WEINSTEIN, supra
note 3, ? 503(b)[04] at 503-46 (control group variable), and 503-48 ("superior" a variable
term); Amicus Curiae Brief of ABA at 5, Upjohn Co. v. United States, 449 U.S. 383 (1981)
("[E]ach time there is a request for advice by a corporate client, [counsel] must first inquire
into the position, status, and responsibility of the corporate employee making the
request . . .").
19. 449 U.S. at 396 ("[W]e decide only the case before us, and do not undertake to draft a
set of rules which should govern challenges to investigatory subpoenas.").
According to the court, a case-by-case approach is consistent with the spirit if not the letter
of Federal Rule of Evidence 501. 449 U.S. at 396-97. Chief Justice Burger strongly disagreed
with this part of the opinion, arguing that the need for certainty warranted the articulation of a
standard that would "afford guidance to corporations, counsel. . . and federal courts.") 449
U.S. at 402.
While the majority's interpretation of the legislative history of rule 501 is not clearly erro-
neous, it appears equally plausible that Congress was looking to the courts to develop a uni-
form rule. First, given that one of the primary reasons for developing the Federal Rules of
Evidence was to eliminate "uncertainties in the law of evidence and variations among the
circuits," Federal Rules of Evidence.' Hearings on H.R. 5463 Before the Senate Comm. on the
Judiciary, 93d Cong., 2d Sess. 4 (1974) (statement of Rep. Hungate), it would behoove the
Court to adopt a standard that would be followed by all circuits.
Second, the timing of the promulgation of rule 501 suggests that Congress wanted the
courts to formulate a standard. Originally, the common law privileges were to be handled by
promulgating thirteen distinct rules. One of those rules codified the control group test. How-
ever, after the Supreme Court affirmed the subject matter test, see note 27 infra, the Advisory
Committee deleted the section on corporations. The House and Senate then disagreed on the
treatment of privileges generally, and abandoned 12 rules in favor of rule 501, a general propo-
sition that the law of privileges "shall be governed by the principles of the common law as they
may be interpreted by the courts of the United States in the light of reason and experience."
See WEINSTEIN, supra note 8, 1 503[01], ? 503(b)[04], at 503-47. A reasonable interpretation of
this sequence of events is that Congress wanted the courts to develop a test for applying the
privilege to corporations. See 93d Cong., 2d Sess., 120 Cong. Rec. 40891 (1974) (remarks of
Rep. Hungate) (The phrase "governed by the principles of the common law" was "intended to
provide the courts with the flexibility to develop rules or privilege on a case-by-case basis.")
(emphasis added).
Finally, one could argue that establishing a standard to which the facts of a particular case
are applied does constitute an application of the privilege on a case-by-case basis. While en-
couraging judicial flexibility and experiment in a new and difficult area of law may foster
salutary ferment in the lower courts, the scope of the corporate attorney-client privilege is
unlikely to benefit from further development absent additional Supreme Court guidance. The
issue is at least as old as the Court's recognition of the privilege in a corporate context. See
United States v. Louisville & N. R.R., 236 U.S. 318, 336 (1915) (protecting confidential com-
munications between corporations and attorneys is desirable). The control group test was first
articulated in Philadelphia v. Westinghouse Elec. Corp., 210 F. Supp. 483, 485 (E.D. Pa. 1962),
mandamus andprohibition deniedsub nom. General Elec. Co. v. Kirkpatrick, 312 F.2d 742 (3d

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 673

Supreme Court placed no restrictions on the federal courts' discre-

tion either to choose among the alternative tests already developed
or to formulate a totally new standard.21 Part I of this Note exam-
ines two of the more popular standards, the Seventh Circuit's "sub-
ject matter test"22 and the Eighth Circuit's "modified subject matter
test"23 and concludes that neither approach is entirely consistent

Cir. 1962), cert. denied, 372 U.S. 943 (1963). If twenty years of competition among the possi-
ble tests has not clarified the issues and the options, it seems most unlikely that further search-
ings by the lower courts will reveal a definitive answer. Indeed, the most recent "innovation"
among the lower courts is the Illinois Supreme Court's decision to adopt the control group test.
See Consolidation Coal Co. v. Bucyrus-Erie Co., 89 Ill. 2d 103, 118-20, 432 N.E.2d 250, 257-58
(1982). Given the costs of uncertainty, and the necessity of eventually adopting a standard (if
only by rejecting erroneous formulations as they come, one by one, before the Court), the
Chief Justice's concurrence advances a persuasive argument for significant clarification by the
Supreme Court at the next opportunity.
20. The Court listed several significant factors: (1) The communications were made by
Upjohn employees at the direction of corporate superiors, (2) so that Upjohn could recieve
legal advice from counsel; (3) the communications concerned matters within the scope of the
employee's duties (4) which were not available from upper-level directors; (5) the employees
were told the purpose of the communications; and (6) the communications were considered
confidential when made and were not disseminated outside the corporation. 449 U.S. at 394.
The Court's failure to explain the significance of these factors has engendered confusion
among the courts and the commentators. See, e.g., Baxter Travenol Laboratories, Inc. v. Le-
may, 89 F.R.D. 410, 413-14 (S.D. Ohio 1981) (applying the six-factor test of Upjohn); In re
LTV Sec. Litigation, 89 F.R.D. 595, 602 (N.D. Tex. 1981) ("In Upjohn, the Court made clear
that the .. privilege applies to communications made by corporate employees . .. regardless
of echelon .. ."); SEC v. Gulf & Western Indus., Inc. 518 F. Supp. 675, 681 (D.D.C. 1981)
("Upjohn v. United States did not articulate . .. specific guidelines regarding corporation and
the privilege as some had anticipated."); Gergacz, supra note 1, at 511 ("The Court in Upjohn
was not satisfied with the control group test, nor did it embrace the subject matter test. In-
stead, it ... applied ... an analysis of various factors .. ."); Nath, supra note 12, at 54
(Court adopted subject matter test); Sexton, A Post-Upjohn Consideration of the Corporate
Attorney-Client Privilege, 57 N.Y.U. L. REV. 443, 462 (1982) ("[T]he Court did not embrace the
subject matter test or any of its modifications."); Iowa Comment, supra note 12, at 174-75
(courts should use four of the Upjohn court's six factors); Comment, Corporate Attorney-Client
Privilege.- The Confusion Remains After Upjohn, 17 NEW ENG. L. REV. 925, 943-44 (1982)
(court adopted six-factor test, but "it is unclear whether each element" must be satisfied.).
21. Even if a single test is adopted, total uniformity will not be achieved. In diversity suits,
state rules of evidence apply. FED. R. EVID. 501 ("in civil actions and proceedings, with respect
to an element of a claim or defense as to which State law supplies the rule of decision, the
privilege . .. shall be determined in accordance with State law."). Ordinarily, federal law of
privileges applies in federal questions cases, although theoretically state privilege law may
control. See WEINSTEIN, supra note 8, ? 501[02], at 501-20 to 21; cf. Kobak, supra note 2, at
341-52 (arguing that federal courts should not develop a federal common law for the privilege,
but rather should adopt the state law).
In addition, state courts are not required to apply the Federal Rules of Evidence. See
Stern, Attorney-Client Privelege: Supreme Court Repudiates the Control Group Test, 67 A.B.A.
J. 1142 (1981). Most state courts have looked to Upjohn for guidance, see, e.g., Marriott Corp.
v. American Academy of Psychotherapists, Inc., 157 Ga. App. 497 (1981), Leer v. Chicago,
Milwaukee, S. P. & Pac. Ry., 308 N.W.2d 305 (Minn. 1981), cert. denied, 455 U.S. 939 (1982);
Macey v. Rollins Envtl. Servs., Inc., 179 N.J. Super. 535 (1981). The Illinois Supreme Court
has, however, repudiated Upjohn and applied the control group test. See Consolidation Coal
Co. v. Bucyrus-Erie Co. 89 Ill. 2d 103, 116-20, 432 N.E.2d 250, 256-58 (1982). Thus, even if
one standard were adopted for federal courts, total uniformity would still not be achieved.
22. See notes 27-41 infra and accompanying text.
23. See notes 42-74 infra and accompanying text.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
674 Michigan Law Review [Vol. 81:665

with the purposes of the privilege. Part II argues that the courts
should adopt the Eighth Circuit's test with two further modifications.
One revision is but a demand for clarification and consistency: the
courts should explicitly adopt Dean Wigmore's legal advice require-
ment for corporate clients.24 The other modification is more radical:
the command requirement should be eliminated. Under this ap-
proach, every employee may stand in the position of the client for
purposes of the privilege.25 The Note argues that these modifications
will better enable attorneys to provide sound legal advice, assist at-
torneys in ensuring that corporations comply with the law, and pre-
vent attorneys from becoming repositories of undiscoverable
information.26

I. PRESENT STANDARDS AND THEIR WEAKNESSES

A. The Subject Matter Test

Ten years prior to Upjohn, the Seventh Circuit, in Harper & Ro
Publishers Inc. v. Decker,27 adopted a formulation of the privile
that is much broader than the control group test: The privilege a
plies to communications from any employee to an attorney if "th
employee makes the communication at the direction of his superio
in the corporation and where the subject matter upon which the a
torney's advice is sought by the corporation and dealt with in th
communication is the performance by the employee of the duties
his employment."28 The court rejected the control group test be

24. The modified subject matter test allows the corporation to satisfy itsprimafacie burd
on the legal advice requirement by showing that the communication was made to an attorne
See notes 48-65 infra and accompanying text.
25. Under the modified subject matter test, lower- and middle-level employees must spea
to counsel at the direction of superiors before their communications are privileged. See note
57-74 infra and accompanying text.
26. See Part II, infra.
27. 423 F.2d 487 (7th Cir. 1970), affd per curiam by an equally divided court, 400 U.S. 348
(1971).
28. 423 F.2d at 491-92.
Because the court found the subject matter of the communication to be within the em-
ployee's scope of employment, it did not decide whether the privilege should attach to em
ployee communications that discuss matters the employee observed as a witness. 423 F.2d a
491. Hickman v. Taylor, 329 U.S. 495 (1946), probably prevents the privilege from applying in
such a situation. In Hickman, several crew members of a tugboat were killed in an accident
The tugboat company's attorney interviewed surviving crew members. The plaintiffs in th
suit against the tugboat company sought discovery of these documents. The defendant's lawyer
claimed the documents were protectd by the attorney-client privilege. The Court held that the
privilege did not apply because the statements were taken from "witnesses." 329 U.S. at 508
However, the Court never explained why the employees' agency relationship did not privilege
the communications. See 8 J. WIGMORE, supra note 1, ? 2317, at 618; Weissenberger, supra
note 5, at 909 n.47 ("It remains a mystery, however, why the court did not consider an agency
relationship between the partners and the crew members, which would have allowed applica-
tion of the privilege to the employees."). Compare Simon, supra note 7, at 959 n.20 (the crew
members had claims of their own pending, so their statements were not made on behalf o

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 675

cause it rejected the premise that only those employees who direct
the corporation's response to legal advice personify the corpora-
tion.29 Instead, the court concluded, without analysis,30 that any em-
ployee "is sufficiently identified with the corporation" whenever the
communication satisfies the conditions of its "subject matter test."31
The subject matter test has two principal virtues: it recognizes
the interdependency of corporate personnel by extending the privi-
lege to communications from lower-level employees,32 and it encour-
ages the flow of information to corporate counsel.33 The test,
however, fails effectively to distinguish communications made to se-
cure legal advice from those communications made to attorneys for
other purposes. Because only communications made to secure legal
advice are privileged,34 the test fails to provide an entirely satisfac-
tory solution to the unique problems posed by applying the attorney-
client privilege to communications from corporate clients.
The Harper & Row opinion is susceptible to two interpretations
regarding the legal advice requirement: either it fails to impose a

their employers) with Weinschel, supra note 14, at 875 (Hickman decision based on the content
of the documents, not the communicant).
Prior to Upjohn, the lower courts disagreed on the interpretation of the "witness" limitation
imposed by Hickman. Compare United States v. United Shoe Mach. Corp., 89 F. Supp. 357-58
(D. Mass. 1950) (communications of all employees within the privilege if they are made pursu-
ant to a request for legal advice) with Philadelphia v. Westinghouse Electric Corp., 210 F.
Supp. 483, 484 (E.D. Pa. 1962) (under Hickman, lower-level employees are witnesses, not cli-
ents, and therefore their communications are not privileged). For criticism of the Westing-
house interpretation, see, e.g., Weissenberger, supra note 5, at 909 n.47 ("[I]t is questionable
whether Hickman actually stands for the proposition that corporate employee witnesses are
never the "client" for purposes of the privilege."); Weinschel, supra note 14, at 875 (Hickman
only applies if a corporate employee describes what he saw as a witness); Michigan Note, supra
note 5, at 371, 376 (same). While the Upjohn court did not embrace the United Shoe interpre-
tation of Hickman, it implicitly rejected the Westinghouse interpretation because it indicated
the privilege would apply to communications of lower-level employees that discuss matters
within the scope of their employment. 449 U.S. at 394. The clear implication of Upjohn is that
the communications of employees who discuss matters they observe as mere witnesses will not
be privileged - a position that the commentators have long advocated. See, e.g., Kobak,
supra note 2, at 365 ("The exception for the 'bystander' employee is a minimum necessity of
the attorney-client privilege ...."); WEINSTEIN, supra note 8, ?503(b)[04], at 503-44 to 45;
Weissenberger, supra note 5, at 912 n.59; Michigan Note, supra note 5, at 371.
29. 423 F.2d at 491.

30. Gergacz, supra note 1, at 492 ("Harper & Row never explained its dissatisfaction wi
the control group test ....").
31. 423 F.2d at 491.

32. See Upjohn Co. v. United States, 449 U.S. at 391 ("[middle level and lower-leve
employees. . . have the relevant information needed by corporate counsel . ."); Diversifi
Indus., Inc. v. Meredith, 572 F.2d 596, 609 (8th Cir. 1978) (en banc) (If middle- and lowe
level employees are not interviewed, it is difficult to determine what happened.); Weinschel
supra note 14, at 876 (the attorney faces a "Hobson's choice" under control group test).
33. See, e.g., Diversified Indus., Inc. v. Meredith, 572 F.2d 596, 609 (8th Cir. 1978) (en
banc); Gergacz, supra note 1, at 492 (more information provided to counsel).
34. See, e.g., WEINSTEIN, supra note 8, ? 503(b)[04], at 503-48; Weissenberger, supra no
5, at 901; 8 J. WIGMORE, supra note 1, ? 2290 at 542.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
676 Michigan Law Review [Vol. 81:665

legal advice requirement or its command requirement35 serves as a

substitute test for the legal advice requirement.36 If the test does not
impose a legal advice requirement, as many authorities argue, a
communication becomes privileged whenever an attorney receives
the particular document.37 As a consequence, corporations can
abuse the privilege and create a zone of silence by directing "all
employees to channel business reports to corporate attorneys,
thereby rendering all such communications privileged."38 If the
command requirement is a substitute for the legal advice require-
ment, the test is totally ineffective because it fails to examine the
purpose of the communication.39 The standard also enables a corpo-
ration to create the same zone of silence abuse because superiors can
routinely approve communications between lower-level employees
and corporate counsel.40 Thus, under both interpretations of Harper
& Row, the subject matter test lacks an effective legal advice require-
ment, which enables the corporation to protect communications that
are not properly within the scope of the privilege.41

B. The Modifed Subject Matter Test

In Diversified Industries, Inc. v. Meredith,42 the Eighth Circuit
adopted the "modified subject matter test," which provides that any

35. The command requirement is the first element of the subject matter test; it requires that
the "employee [make] the communication at the direction of his superiors in the corpora-
tion .. ." 423 F.2d at 491.

36. 423 F.2d at 491-92 (implication from court's failure to mention legal
requirement).
37. See, e.g., Diversified Indus., Inc. v. Meredith, 572 F.2d at 609; WEINSTEIN, sup
8, ? 503(b)[04], at 50347 to 49.
38. Note, Privileged Communications - Inroads On the "Control Group" Test In T
porateArea, 22 SYRACUSE L. REV. 759, 766 (1971) (footnote omitted). See, e.g., Unite
v. El Paso Co., 682 F.2d 530, 541 (5th Cir. 1982) (general claim that corporate tax dep
funnels work through its attorneys insufficient to prove privilege); Nath, supra note 12, at
("If... a communication made directly to the corporation's top management would
privileged, but a communication directed to the corporation's lawyer would, mana
would have an incentive to order that all important communications take place directl
through counsel.") (footnote omitted); Weissenberger, supra note 5, at 912 (Harper & R
creates excessive zone of silence); Harvard Note, supra note 2, at 432-33. (privilege
available).
39. See, e.g., Weissenberger, supra note 5, at 912 (criticizing Harper & Row because privi-
lege is triggered whenever the superior authorizes the communication); cf. WEINSTEIN, supra
note 8, ? 503(b)[04], at 50348 (test should include separate legal advice requirement).
40. See Nath, supra note 12, at 49-50.
41. See text at note 34 supra. Very few courts have wholly relied on Harper & Row be-
cause of its overbroadness. See Gergacz, supra note 1, at 492 (courts have embraced subject
matter approach, but have fashioned tests to narrow the scope of privilege). For two decisions
that adopted Harper & Row, see Sylgab Steel & Wire Corp. v. Imcoco-Gateway Corp., 62
F.R.D. 454 (N.D. Ill. 1974); Hasso v. Retail Credit Co., 58 F.R.D. 425 (E.D. Pa. 1973). Harper
& Row was never cited by the Supreme Court in Upjohn.
42. 572 F.2d 596 (8th Cir. 1977), modified on rehearing en banc 572 F.2d 606 (8th Cir.
1978).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note--Where Do We Go After Upjohn? 677

corporate employee's communication is privileged if the corporation

satisfies the following conditions:43
(1) the communication was made for the purpose of securing legal ad-
vice; (2) the employee making the communication did so at the direc-
tion of his corporate superior; (3) the superior made the request so that
the corporation could secure legal advice; (4) the subject matter of the
communication is within the scope of the employee's corporate duties;
and (5) the communication is not disseminated beyond those persons
who, because of the corporate structure, need to know its contents.44
This test, the Eighth Circuit reasoned, offered a reasonable compro-
mise between the control group test and the subject matter test. Un-
like the control group test, this test properly accounts for the realities
of corporate life because it privileges communications from informa-
tion holders even when they are not high-level decision makers, but
it avoids the "zone of silence" abuse of the subject matter test by
imposing a legal advice requirement.45
The Eighth Circuit's test is a substantial improvement over the
subject matter test because it narrows the availability of the privilege
by imposing a confidentiality requirement46 and an explicit legal ad-
vice requirement.47 The test, however, has two important flaws: (1)
The Diversified court's application of its legal advice requirement un-
dermines the requirement's utility; and (2) the command require-
ment inhibits disclosures by middle- and lower-level employees and
frustrates attorneys' attempts to ensure corporate compliance with
the law. Both shortcomings frustrate the purposes of the privilege.

1. The Legal Advice Requirement

In Diversified, the court allowed the corporation asserting the

privilege to meet itsprimafacie burden on the legal advice require-

43. Traditionally, the party asserting the privilege bears the burden of proof. See, e.g.,
Weil v. Investment/Indicators, Research & Management, Inc., 647 F.2d 18, 25 (9th Cir. 1981);
United States v. Bump, 605 F.2d 548, 551 (9th Cir. 1978); In re Ampicillin Antitrust Litigation,
81 F.R.D. 377, 394 (D.D.C. 1978); Duplan Corp. v. Deering Milliken, Inc., 397 F. Supp. 1146,
1161 (D.S.C. 1974) ("The adversary party, by virtue of the obvious fact that it has not seen the
documents, cannot be expected to bear the burden of establishing a lack of privilege."); Eutec-
tic Corp. v. Metco, Inc., 61 F.R.D. 35, 41 (E.D.N.Y. 1973); Honeywell, Inc. v. Piper Aircraft
Corp., 50 F.R.D. 117, 120 (M.D. Pa. 1970).
44. 572 F.2d at 609. The Diversified court adopted the standard proposed by Judge Wein-
stein. See WEINSTEIN, supra note 8, ? 503(b)[04], at 503-47 to 50.
45. 572 F.2d at 608-09.

46. 572 F.2d at 609. Defining confidentiality in the corporate context is problematic be
cause the corporate structure often requires three or more people to know the contents of a
particular communication. A Michigan Note proposed that a communication is made in con
fidence if the communication was distributed only to those who, because of the corporat
structure, need to know its contents. See Michigan Note, supra note 5, at 377-79. Judge Wein
stein incorporated this definition into his modified subject matter test. See WEINSTEIN, supr
note 8, 1 503(b)[04], at 503-49 to 50.
47. 572 F.2d at 609.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
678 Michigan Law Review [Vol. 81:665

ment by showing that the communication was made to a lawyer.48

The court ruled that unless the party opposing the privilege makes
"a clear showing to the contrary," the privilege attaches to the com-
munication.49 Future courts should reject this allocation of the bur-
den of proof for two reasons. First, the factual premise of the burden
allocation is erroneous. The Diversified court assumed that corpora-
tions normally employ attorneys to handle requests for legal advice
and that rarely will corporate counsel provide business advice.50 But
this premise has little to commend it. Unlike most individual clients,
corporations constantly communicate with attorneys.51 Many of
these communications admittedly involve legal questions generated
by the complexity and size of corporate operations.52 Many corpo-
rate executives, however, rely on attorneys for business advice, be-
cause the attorney knows the client's business as a result of their
constant interaction.53 Given that corporate attorneys provide both
legal and business advice to corporate clients, it makes little sense to
allow corporations to satisfy the legal advice requirement by a mere

48. 572 F.2d at 610. The court cited Dean Wigmore for support of its position:
It is not easy to frame a definite test for distinguishing legalfrom nonlegal advice ....
[T]he most that can be said by way of generalization is that a matter committed to a
professional legal adviser is primafacie so committedfor the sake of the legal advice which
may be more or less desirable for some aspect of the matter, and is therefore within the
privilege unless it clearly appears to be lacking in aspects requiring legal advice.
Obviously, much depends upon the circumstances of individual transactions.
8 J. WIGMORE, supra note 1, ? 2296, at 566-67 (emphasis in original) (footnote omitted).
49. 572 F.2d at 610.

50. The Diversftied court's citation of Wigmore, see note 48 supra, for its position is suspect
at best. Dean Wigmore'sprimafacie analysis was developed for natural persons, not corpora-
tions. See Weissenberger, supra note 5, at 916. Thus, his allocation of the burden of proof
makes sense for individuals, not corporations. In addition, a strong argument can be made
that an individual engaged in a business enterprise is not different from a corporation, and
therefore Wigmore's analysis should be limited to the noncommercial context.
51. See note 9 supra.
52. See generally, Kobak, supra note 2, at 340 (legal advice needed "to separate whims of
. . . client[s] from immediate gratification"); Simon, supra note 7, at 955; Iowa Comment,
supra note 12, at 16 ("[C]orporations may possess the most glaring need for unfettered legal
advice in modern society.").
The privilege is the same for in-house and outside counsel. See, e.g., Burlington Indus., Inc.
v. Exxon Corp., 65 F.R.D. 26, 36-37 (D. Md. 1974); Air-Shield, Inc. v. Air Reduction Co., 46
F.R.D. 96, 97 (N.D. Ill. 1968); Malco Mfg. Co. v. Elco Corp., 45 F.R.D. 24, 26 (D. Minn.
1968); New York Underwriters Ins. Co. v. Union Constr. Co., 285 F. Supp. 868, 869 (D. Kan.
1968); United States v. United Shoe Mach., 89 F. Supp. 357, 360 (D. Mass. 1950).
53. See note 9 supra; Rast, What the Chief Executive Looksfor in his Corporate Law De-
partment, 33 Bus. LAW. 811, 813 (1978) (legal advice should be combined with business ad-
vice). One commentator, however, argues that in-house attorneys are "so keenly aware of the
dimensions of their roles that they would ordinarily abjure business admonitions better left to
management or other business personnel." The author readily admits his thesis is "entirely
conjectural," and needs empirical testing. Kobak, supra note 2, at 353. Conjecture is hardly a
compelling basis for allowing large corporations to gain the benefits of the presumption of
"legal advice" whenever a communication is sent to counsel. But cf Program, A Businessman's
View of Lawyers, 33 Bus. LAW. 817, 819 (1978) ("[U]nless it's a special relationship, the chief
executive normally doesn't like to receive business advice mixed in with legal advice.") (Com-
ments of Donald N. Frey, Chairman of the Board, Bell & Howell).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note--Where Do We Go After Upjohn? 679

showing that the communication was made to an attorney.54

Second, placing the "real" burden on the party opposing the
privilege may nullify the legal advice requirement. A corporation
can always satisfy itsprimafacie burden because, unless the commu-
nication was sent to a lawyer, the issue of the attorney-client privi-
lege would not arise. The burden of proof then shifts to the
opposing party, who must establish that the communication was not
made to secure legal advice. That party labors under two handicaps.
First, without access to the communication, it will be difficult to ar-
gue intelligently that the corporation did not seek legal advice with
that particular communication.55 Second, even if the party in-
troduces circumstantial evidence that tends to establish that the cor-
poration did not seek legal advice in the communication, the
evidence may not suffice to satisfy the "clear showing to the con-
trary" standard imposed by the Diversified court.56 Thus, the Diversi-
fied court's allocation of the burden of proof on the legal advice
requirement creates the possiblity that all communications to attor-
neys will be privileged, in effect eliminating the legal advice
requirement.

2. The Command Requirement

Under the Eighth Circuit's test, the privilege applies to communi-
cations from lower- and middle-level employees to counsel only if
the employee's superior authorizes such communication and the su-
perior is seeking legal advice on behalf of the corporation.57 The
command requirement frustrates the purposes of the privilege in two
respects. First, the varying definition of "superior" undermines the
predictability essential to ensuring the full benefits of the privilege.

54. See, e.g., Diversified Indus., Inc. v. Meredith, 572 F.2d 596, 612-14 (8th Cir. 1978) (en
bane) (Henley, J., concurring in part, dissenting in part) ("A communication is not privileged
simply because one of the parties to it is a lawyer.") (citations omitted). Although Judge Hen-
ley criticized the majority for its allocation of the burden of proof, he based his dissent on his
finding, after in camera inspection of the pertinent documents, that the attorneys were not
employed to provide legal services. 572 F.2d at 614.
55. Diversified Indus., Inc. v. Meredith, 572 F.2d at 614 (Henley, J., concurring in part,
dissenting in part) ("[T]he party seeking disclosure does not know in advance and has no way
of knowing why the matter in question was turned over to the lawyer or why the communica-
tions were developed or what they amounted to or contained.").
56. See Diversified Indus., Inc. v. Meredith, 572 F.2d at 614 (Henley, J., concurring in part,
dissenting in part) ("[A]part from in camera proceedings . . . there is no way for the party
seeking disclosure to [overcome] the prima facie case of privilege .. ."); Weissenberger,
supra note 5, at 917.
The Diversified court's analysis also fails to provide a test to distinguish communications
made to secure legal advice from communications made for other purposes. Thus, even in in
camera proceedings, judges are left without guidance. See Weissenberger, supra note 5, at
917, 919 (division of opinion on the legal advice issue demonstrates Diversified's failure to
develop a meaningful legal advice requirement).
57. See text at note 44 supra. For purposes of this Note, these two conditions will be
referred to as the command requirement.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
680 Michigan Law Review [Vol. 81:665

Second, in some situations, the requirement inhibits the flow of in-

formation from middle and lower level employees and frustrates at-
torneys' attempts to ensure corporate compliance with the law.

a. Uncertainty. The command requirement creates uncertainty

in applying the privilege because the definition of "superior" de-
pends on the circumstances of each case.58 The term can never ac-
quire a uniform meaning because the size of corporations and the
division of responsibility within them vary greatly.59 Thus, defining
"superior" results in the same uncertainty as was created by the defi-
nition of "control group." The Upjohn Court condemned that un-
certainty and those comments are equally applicable here.60
Judge Weinstein, the originator of the modified subject matter
test adopted in Diversified, argues that defining uncertainty in func-
tional terms can avoid uncertainty: A superior is one "who is in a
position to initiate a request for legal advice and whose actions indi-
cate that his purpose in directing disclosure to the attorney was in
contemplation of the rendition of legal services."61 But Judge Wein-
stein's solution begs the question because it does not define when a
person is in the position, and therefore has the authority, to request
legal advice. Moreover, Judge Weinstein admits the uncertainty of
his own test by recognizing that in some circumstances a communi-
cation made by a lower-level employee on his own initiative will be
privileged.62 He fails, however, to explain why the privilege should
apply in some circumstances but not in others.63 Thus, Judge Wein-

58. Cf. WEINSTEIN, supra note 8, 1 503(b)[04] at 503-48 ("The question of who is a supe-
rior. . . is a difficult one.").
59. Rast, supra note 53, at 812 ("[B]usinesses differ. They differ as to their products and
services. They differ in structure, in size, and in scope of operations.").
60. See note 18 supra and accompanying text. This criticism applies to the Ampicillin test.
Under that test, a communication is privileged if four requirements are satisfied:
(1) The particular employee or representative of the corporation must have made a com-
munication of information which was reasonably believed to be necessary to the decision-
makingprocess concerning a problem on which legal advice was sought; (2) the communi-
cation must have been made for the purpose of securing legal advice; (3) the subject mat-
ter of the communication to or from an employee must have been related to the
performance by the employee of the duties of his employment; and (4) the communication
must have been a confidential one.
In re Ampicillin Antitrust Litigation, 81 F.R.D. 377, 385 (D.D.C. 1978) (emphasis in original).
While the Ampicillin court must be applauded for eliminating the command requirement, see
notes 86-105 infra and accompanying text, the first element of the test creates uncertainty be-
cause it does not indicate who determines whether the communication is reasonably necessary
to the decision-making process concerning a legal problem and when that belief must be held.
See Gergacz, supra note 1, at 495-96; cf. Fordham Note, supra note 9, at 1289 (test offers "only
cosmetic differences, in an already murky field"); New England Note, supra note 20, at 938
(test does not provide predictability, but it does provide realistic treatment of corporate life).
61. See WEINSTEIN, supra note 8, ? 503(b)[04], at 503-48 to 49.
62. See WEINSTEIN, supra note 8, ? 503(b)[04], at 503-48 to 49.
63. See WEINSTEIN, supra note 8, ? 503(b)[04], at 503-48 to 49.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 681

stein's analysis does little to resolve the uncertainty resulting from

the command requirement.
The purpose of the privilege is to encourage full and frank com-
munication between clients and their attorneys.64 The command re-
quirement undermines this objective by introducing uncertainty
about whether the privilege will apply because of the employee's po-
sition in the corporation.

b. Intererring With the Flow of Information to Counsel. The

command requirement vests superiors with the authority either to
prohibit communications from lower-level employees or to alter
their contents.65 Consequently, the command requirement distorts
the flow of information to counsel, frustrating attorneys' efforts to
obtain accurate information and to ensure corporate compliance
with the law.
Initially, it is important to recognize that superiors are not better
able to recognize potential legal problems simply by virtue of their
position in the corporation. Generally, superiors are not lawyers.
The increased interaction with attorneys that accompanies higher
rank in the corporation does not eliminate this educational barrier,66
although it may mitigate it. In addition, superiors, as well as coun-
sel, must to some extent rely on subordinates either to alert them to
potential legal problems or to provide information for a previously
recognized legal problem, because these employees often possess
knowledge unavailable to superiors or to counsel.67
Under the command requirement, a lower-level employee who
identifies a potential legal problem must either seek approval from
his superior to discuss the matter with counsel or forward the infor-

64. See note 1 supra and accompanying text.

65. See text at note 44 supra.
66. Cf. Program, supra note 53, at 831 (1978) ("[L]aw is very complex, bewildering, and
frustrating for most of us who have not been in the legal profession.") (comment of Donald P.
Kelley, President, Esmark, Inc.)
67. Cf. Upjohn Co. v. United States, 449 U.S. 383, 394 (1981) (information not available to
upper-echelon); Rast, supra note 53 at 814:
I look to the Legal Department for advice which will prevent legal problems from arising.
Its ability to do this must come from two sources. First, an intimate knowledge of the
buisness with particular sensitivity to all changes in policy and operating procedures. And
second intimate knowledge of all changes in the law which might possibly have an impact
on the business. I have a responsibility with respect to the first point, to see that the Legal
Department is continuously represented in the highest councils of management. The sec-
ond point, knowledge of changes in the law, is the sole responsibility of the lawyer
himself.
These comments indicate that corporate executives expect attorneys to keep abreast of
changes in law, while they are primarily responsible for providing the attorneys with the facts
attorneys need to formulate legal opinions. However, even representation at the highest levels
of management may not ensure the attorney receives complete and accurate information. Cf.
Upjohn Co. v. United States, 449 U.S. at 386 (independent accounting firm apprised in-house
counsel of questionable payments to officials of foreign governments).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
682 Michigan Law Review [Vol. 81:665

mation to the superior who, in turn, forwards it to counsel along with

a request for legal advice.68 While this requirement is reasonable for
corporate governance,69 it only interferes with the information flow
to counsel.
This interference can take several forms. First, every intervening
layer of administrative authority separating information from coun-
sel adds to the probability of omission or distortion.70 The superior
may erroneously, albeit innocently, conclude that no legal problem
exists, or that additional legal advice would only duplicate previous
consultations.71 The superior may delete some information in the
mistaken belief that it does not relate to the particular problem.
Thus, even honest and conscientious management may implement
the command requirement so as to frustrate the purpose of encour-
aging an open and complete flow of information to counsel.
The second possibility of intervention involves questionable con-
duct on the part of the superior. Where the lower-level employee
advances information potentially implicating the superior in inap-
propriate or perhaps even criminal conduct, the command require-
ment permits the superior to conceal his own wrongdoing through
distortion or inaction.72
Finally, the need to go to the superior before consulting the cor-
porate counsel exerts a chilling effect on lower-level employees.
Even if their information is innocuous, or although damaging would
receive scrupulous investigation from the superior, the perceived ex-
posure to retaliation from one's immediate superior can only inhibit

68. See text at note 44 supra.

69. See note 67 supra. It can hardly be gainsaid that corporate executives would condone
the repeated violations of corporate lines of responsibility that would occur if lower-lever em-
ployees talk directly to counsel. This concern, however, is totally irrelevant in determining
whether the privilege attaches to such a communication.
70. See Coffee, Beyond the Shut-Eyed Sentry. Toward a Theoretical View of Corporate Mis-
conduct and an Effective Legal Response, 63 VA. L. REV. 1099, 1138 (1977) ("Even absent the
distorting impact of preexisting attitudes on information flow, experimental evidence suggests
that serial relay of information results in significant information loss. Information theorists
have formulated the rule that each additional relay in a communications system halves the
message while doubling the 'noise.' ") (footnotes omitted); Haft, The Effect of Insider Trading
Rules on the Internal Effciency of the Large Corporation, 80 MICH. L. REV. 1051, 1053 (1982)
("In large corporations, this information may pass through as many as fifteen hierachical levels
from bottom to top. Organizational and communications scholars have confirmed the obvi-
ous: merely because the information is transmitted so many times, the message becomes dis-
torted.") (footnotes omitted).
71. See Haft, Business Decisions by the New Board: Behavioral Science and Corporate Law,
80 MICH. L. REV. 1, 55 (1982) (Behavioral scientists "have shown, for example, that most
individuals misinterpret highly relevant evidence in making intuitive judgments."). The phe-
nomenon of cognitive dissonance increases the probability of this result. See Coffee, supra
note 70, at 1137-38. See generally L. FESTINGER, A THEORY OF COGNITIVE DISSONANCE
(1957).
72. Cases of corporate misconduct that have come to light confirm this possibility. See
Coffee, supra note 70, at 1131 n.109 ("informal internal organizations" actively concealed im-
proper payments by Lockheed and a secret political fund at Firestone).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 683

subordinates' requests for legal advice.73 The potential futility of

such a request adds to the likely unwillingness of subordinates to
request such advice through their superiors. Given that a vulnerable
subordinate must weigh these risks at each level of the corporate bu-
reaucracy, the command requirement clearly contravenes the privi-
lege's purpose of fostering full disclosure to corporate counsel.74

II. A PROPOSED SOLUTION

The previous Part criticized both the subject matter test and
modified subject matter test for frustrating the purposes of the at
ney-client privilege. This Part argues that the courts should ad
the modified subject matter test, with two further changes. First,
legal advice requirement should be reformulated to better dist
guish communications made to secure legal advice from comm
cations made for other purposes. Second, the command requirem
should be eliminated, because it restricts the flow of informati
from middle- and lower-level employees to corporate counsel, w
out producing any benefits that are consistent with the logic of
privilege.

A. Modification of the Legal Advice Requirement

Because the modified subject matter test fails to provide a totally
effective legal advice requirement, communications made to secure
business advice may be privileged.75 Consequently, the test removes
the particular communication from the discovery process without
producing any social benefit in the form of increased communica-
tions to secure legal advice.76 The frequent interaction between cor-
porate clients and their attorneys77 exacerbates this distortion of the
privilege's competing interests.
To avoid this problem, and to provide courts with a standard for
making the legal advice determination,78 the legal advice require-
ment should focus on the advice that is being sought by the corpora-

73. The proposition that self-interest influences information transfers within organizations
is intuitively plausible and widely accepted. See, e.g., R. CYERT & J. MARCH, A BEHAVIORAL
THEORY OF THE FIRM 81-82, 109-10 (1963).
74. See text at notes 1-3 supra.
75. See text at notes 48-56 supra.
76. See notes 1-5 supra and accompanying text.
77. See note 9 supra. Cf. Program, supra note 53, at 844 ("Because of the increasing com-
plexity of the law, we now have almost continual contact with our attorneys instead of spo-
radic and infrequent consultations.") (comment of Jerome Van Gorkum, President of Trans
Union Corporation).
78. See note 56 supra. One court has stated that the contents of documents must be ex-
amined to determine whether the corporation was seeking legal advice. In re Ampicillin Anti-
trust Litigation, 81 F.R.D. 377, 394 (D.D.C. 1978). However, simply stating this truism
without providing a test is of little utility.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
 684 Michigan Law Review [Vol. 81:665

tion.79 Dean Wigmore's standard test achieves these objectives.80

Accordingly, a communication should satisfy the legal advice re-
quirement if the advice sought by a corporate employee, pursuant to
which the communication is made, is advice that can be given only
by a qualified legal practitioner.8' To limit the privilege to its nar-
rowest application, the courts should strengthen this requirement by
requiring the corporation to prove that the purpose of the communi-
cation was to secure legal advice, before the burden of proof shifts to
the party opposing the privilege.82
This procedure has two advantages over the Diversified court's
formulation of the legal advice requirement. First, instead of focus-
ing on the party to whom the communication is made, the proposed
solution requires scrutiny of each communication thereby avoiding
the presumption of legal advice resulting from the Diversfied court's
burden allocation.83 Second, the proposed solution provides a stan-

79. See In re Ampicillin Antitrust Litigation, 81 F.R.D. 377, 394 (D.D.C. 1978) ("whether
it was legal advice that was being sought") (emphasis in original); cf. New England Note,
supra note 20, at 943-44 ("The focus of the privilege is no longer solely on who within the
corporate hierarchy communicates with legal counsel, but rather what is communicated and
how the information correlates with the particular employee's position.") (emphasis in
original).
80. See 8 J. WIGMORE, supra note 1, ? 2292 at 554 ("[l]egal advice of any kind is sought
. . . from a professional legal advisor in his capacity as such .. .").
81. The proposed test is but a slight modification of the first two elements of Dean Wig-
more's test. See note 6 supra. Judge Henley, in his partial dissent in Diversified implicitly
adopted this standard. See Diversified Indus., Inc. v. Meredith, 572 F.2d at 596, 614-15 (8th
Cir. 1978) (en banc) (Henley, J., concurring in part, dissenting in part) (reviewing documents
in camera and concluding that advice could have been given by any person possessing com-
mon sense and business prudence).
One commentator suggests a "but for" analysis for determining when a corporation seeks
legal advice: "If a communication would not have existed but for the pursuit of legal advice,
the communication is privileged." Weissenberger, supra note 5, at 918 (emphasis in original).
Such an approach fails for three reasons. First, it is circular: declaring that a communication is
privileged when it is made solely in the "pursuit of legal advice" is not helpful when "pursuit
of legal advice" is not defined.
Second, the "but for" test is too restrictive. By requiring that the document only be sent to
lawyers, the test incorporates the confidentiality requirement into an analysis of whether legal
advice was sought by the corporation. However, because corporate executives depend on one
another for information to perform their respective duties, it seems not only unrealistic, but
also unjust to deprive the corporation of the privilege simply because some corporate execu-
tives receive copies of the particular communication. See Michigan Note, supra note 5, at 377-
79.
Third, the proposal argues, based on the facts of Diversified, that, as a general proposition,
if the corporation has a fiduciary or other duty to undertake an investigation, irrespective of
any "legal advice" considerations, communications are not made "but for" securing legal ad-
vice, and, therefore, should not be privileged. Id at 920. Corporations, however, conduct
investigations precisely to determine whether or not disclosure to shareholders, the IRS, or the
SEC, is mandated. Therefore, the decision to investigate is not made independent of a request
for legal advice but precisely because the corporation has legal obligations which it must fulfill.
Therefore to argue that the communications were not made to secure legal advice is fallacious.
82. E.g., In re Ampicillin Antitrust Litigation, 81 F.R.D. 377, 394 (D.D.C. 1978); WEIN-
STEIN, supra note 8, ? 503(b)[04], at 503-47 to 48. The party seeking to obtain the privilege has
traditionally borne the burden of proof. See note 43 supra.
83. See notes 50-54 supra and accompanying text.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 685

dard that courts can apply in in camera proceedings.84

This solution has two admitted shortcomings. Clients may at-
tempt to circumvent the requirement by framing requests for busi-
ness advice in language that indicates the corporation was seeking
legal advice.85 In addition, scrutiny of each communication pro-
duces uncertainty, while achieving the full benefits of the privilege
requires certainty.86
While these shortcomings diminish the utility of the proposed so-
lution, they inhere in any judicial application of a general standard
to a particular case. The alternative - aper se rule that all commu-
nications to attorneys are either privileged or not privileged - elimi-
nates the objectional features of the proposed formulation of the
legal advice requirement, but the costs of aper se rule clearly out-
weigh the benefits that would accrue.87 Instead, one must rely on the
judiciary to apply this standard to each case to determine whether or
not the claim of legal advice is a sham. Once the courts agree on the
proper standard for the privilege, they will develop a consistent body
of case law that will enable attorneys to predict with reasonable cer-
tainty whether a particular communication will be privileged.
Moreover, the uncertainty created by the approach advocated
here would not apply to communications clearly requesting legal ad-
vice. These communications, including those relating to potential
civil or criminal liability for corporate misconduct, are those where
the free flow of information to counsel can do the most to ensure
effective legal advice and corporate compliance with the law. Any
test will produce some uncertainty in borderline cases; the test de-
fended by this Note relegates that uncertainty to communications of
at least an arguably business nature. Since chilling these communi-
cations imposes only private costs (costs which increase the likeli-
hood that they will take place notwithstanding the uncertain
availability of the privilege), the consequences of uncertainty under

84. See note 56 supra. An in camera review also offers one possible solution to the prob-
lem of communications made partly for business purposes and partly to secure legal advice.
The judge can order discovery of only those portions of the communication not related to
securing legal advice. The courts rely on such selective disclosure in many other situations.
For example, in resolving an administrative agency's request for grand jury transcripts, the
court must limit disclosure to the material for which the agency has "particularized need." In
such cases, the court can grant access to some parts of the transcript but not others. See In re
Disclosure of Testimony Before the Grand Jury, 580 F.2d 281, 287 (8th Cir. 1978); Weis-
senberger, supra note 5, at 924 ("In reality, most cases can be resolved by virtue of in camera
severance of privileged from nonprivileged matters.").
85. Cf Nath, supra note 12, at 55-58.
86. See note 18 supra and accompanying text.
87. If all communications to attorneys are privileged, communications with attorneys are
encouraged, but attorneys become repositories of undiscoverable information. If all communi-
cations to attorneys are not privileged, the number of communications to attorneys will de-
crease, and attorneys will face a substantial obstacle in their quest to render sound legal advice
and to ensure corporate compliance with the law. See notes 1-5 supra and accompanying text.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
686 Michigan Law Review [Vol. 81:665

the proposed approach may well prove less significant than

anticipated.
The solution proposed by this Note is more consistent with the
purposes of the privilege than the Diversified court's approach be-
cause it requires corporations to establish affirmatively that legal ad-
vice was being sought at the time of the communication. By
requiring such a showing, the proposed requirement eliminates the
overinclusiveness objection to the Diversified court's approach, and
achieves an optimal balance between encouraging clients to commu-
nicate with counsel to secure legal advice and preventing corporate
attorneys from becoming repositories of undiscoverable information.
Thus, the proposed formulation of the legal advice requirement is
preferable to that imposed by the Diversified court.

B. Elimination of the Command Requirement

The command requirement frustrates the policies of the attorney-
client privilege in two respects. First, it engenders uncertainty in ap-
plying the privilege because the definition of "superior" varies with
the facts of each case.88 Second, it not only creates the possibility
that counsel will not receive completely accurate information from
the corporation, but also frustrates attorneys' efforts at ensuring cor-
porate compliance with the law by empowering superiors, at all
levels of the corporation, to suppress subordinate initiated investiga-
tions where the propriety of the superior's acts may be called into
question.89 In the light of these considerations, it seems logical to
eliminate the requirement.90
However, eliminating the requirement would result in deeming
all employees the "client" for purposes of the privilege. Thus, every
employee could request legal advice. Moreover, all communications
made pursuant to such a request that discussed matters within the
scope of the employee's corporate duties would be privileged.91

88. See text at notes 57-64 supra.

89. See text at notes 68-74 supra.
90. As one commentator has stated in criticizing the control group test:
An employee who lacks power may not be wanting in zeal. If such an earnest soul per-
ceives a problem while acting within the scope of his employment and subsequently
secures legal advice, the communication ought to be privileged, even if the outcome is
determined by those to whom he reports. The potential for social benefit inherent in the
communication does not vary drastically with the status of its narrator.
Kobak, supra note 2, at 368; accord Iowa Comment, supra note 12, at 175 ("[I]f the corporation
receives legal advice as a result of an employee's unsolicited, confidential communication, the
purpose of the privilege is still served.") (footnotes omitted). Even Judge Weinstein, who pro-
posed the modified subject matter test, admits that the command requirement need not be
satisfied in certain situations; unfortunately, he does not provide an adequate explanation of
the extenuating circumstances that warrant abandoning the requirement. See WEINSTEIN,
supra note 8, 1 503(b)[04], at 50347 to 49.
91. Several courts have stated that the privilege applies to communications of all employ-
ees pursuant to a request for legal advice. However, none of the cases involved the issue of

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 687

Two objections can be made to such an expansive definition of

the "client." The first is that the zone of silence will expand. The
second is that treating all employees as representatives of the client
may distort the corporation's decision-making structure. This sec-
tion analyzes these arguments, finds them wanting, and concludes
that the command requirement should be eliminated.

1. Enlarging the Zone of Silence

One possible justification92 for the command requirement is that

it may be needed to narrow the scope of the privilege.93 While this
goal is laudable, the use of the command requirement to achieve it is
misplaced, because the requirement limits neither of the factors that
creates the "zone of silence."
The zone of silence is a function of the number of communica-
tions to counsel and the ability to distinguish communications made
to secure legal advice from communications made for other pur-
poses.94 The first factor can not be controlled by changing the re-
quirements of the privilege because business enterprises, whether
they are conducted as sole proprietorships or as corporations, will
increase requests for legal advice as the size and complexity of oper-

whether a lower-level employee must first be authorized to speak to counsel before his commu-
nication is privileged. See, e.g., In re LTV Sec. Litigation, 89 F.R.D. 595, 599-603 (N.D. Tex.
1981) (The "Upjohn . . . court made clear that the privilege applies to communications made
by corporate employees concerning matters pertinent to their job tasks, regardless of echelon,
if sought by the corporation's attorney in order to ... render legal advice .. ."); SEC v. Gulf
& Western Indus., Inc. 518 F. Supp. 675, 681 n.9 (D.D.C. 1981) (leaving open the issue of who
must request legal advice); In re Westinghouse Elec. Corp. Uranium Contracts Litigation, 76
F.R.D. 47, 56-57 (W.D. Pa. 1977); United States v. Aluminum Co. of America, 193 F. Supp.
251, 252 (N.D.N.Y. 1960) (president requested that subordinate provide counsel with informa-
tion); Zenith Radio Corp. v. Radio Corp. of America, 121 F. Supp. 797, 795 (D. Del. 1954);
United States v. United Shoe Mach. Corp., 89 F. Supp. 357, 359 (D. Mass. 1950).
92. The Diversified court did not offer an explanation for the command requirement.
However, Judge Weinstein, whose modified subject matter test was adopted by the Diversified
court, criticized United States v. United Shoe Mach. Corp., 89 F. Supp. 357, 359 (D. Mass.
1950), for its holding that all communications "by an officer or employee" of the corporation
would be privileged, on the ground that "such a broad approach is incompatible with other
features of our procedural system [because it] runs competely counter to the modem trends of
broad discovery .... ." WEINSTEIN, supra note 8, ?503(b)[04], at 503-44 to 45. Stated con-
versely, Judge Weinstein's criticism becomes a zone of silence argument against eliminating
the command requirement. But see note 90 supra for the inconsistency in Judge Weinstein's
position.
93. See Nath, supra note 12, at 48 ("no subject matter of a nature significant to the corpo-
ration would not be privileged under [the subject matter test]"); Simon, supra note 7, at 955
(suggesting, without support, that extending the privilege to communications of all employees
creates a privilege broader than that afforded individuals); cf. Gardner, supra note 5, at 342
(privilege should only apply to small corporations because such corporations are alter-ego of a
few individuals).
94. See, e.g., Simon, supra note 7, at 955-56 (variables that affect the size of the zone of
silence include the "number of agents, masses of documents, and frequent dealings with
lawyers").

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
688 Michigan Law Review [Vol. 81:665

ations increase.95 The second factor is not properly addressed by the

command requirement, because whether the superior authorizes the
communication to counsel does not determine whether the legal ad-
vice is being sought by the corporation.96 Instead, that issue is more
accurately resolved through an effective legal advice requirement.97
Thus, abandoning the command requirement will not enlarge the
zone of silence created by the privilege.98
One could argue that limiting the number of employees who can
request legal advice also limits the scope of the privilege,99 but there
is no logical reason for imposing such a limitation. If the employee
requests legal advice for the benefit of the corporation, it makes little
sense to withhold the privilege simply because the particular em-
ployee is not a superior. Regardless of the employee's position, the
corporation will receive the same benefit from the legal advice.
Moreover, the privilege's primary purpose - encouraging communi-
cations between clients and attorneys - is fulfilled irrespective of

95. See Rast, supra note 53, at 812 (emphasis added):

I think it is essential that all of us remember that businesses differ. They differ as to their
products and services. They differ in structure, in size, and in scope of operations. And
for our purpose here today, corporations quite obviously dier in their exposure to the vari-
ous aspects of the law.
96. See Diversified Indus. v. Meredith, 572 F.2d 596, 609 (8th Cir. 1978) (en banc) (the
second element of the test requires that "the employee making the communication did so at
the direction of his corporate superior.").
97. See text at notes 75-83 supra for discussion of the legal advice requirement.
98. One could also argue that the zone of silence is not nearly as ominous as it would first
appear. The privilege only applies to communications, not the underlying facts. Therefore, the
party seeking to obtain the information need only depose the communicant. See Upjohn Co.
v. United States, 449 U.S. at 395-96 (citing Philadelphia v. Westinghouse Elec. Corp., 205 F.2d
830, 831 (E.D. Pa. 1962)). Second, independently preexisting documents, even if given to at-
torneys, will not be privileged unless they were privileged in the clients' hands. Fisher v.
United States, 425 U.S. 391, 404-05 (1976); J. WIGMORE, supra note 1, ? 2307, at 591-94;
Miller, The Corporate Attorney Client Privilege and the Work Product Doctrine. Protectionfrom
Compelled Disclosure in Criminal Investigation of a Corporation, 12 U.S.F. L. REV. 569, 587
(1978). Third, the privilege does not apply to planning fraudulent or criminal activities. See,
e.g., In re Westinghouse Elec. Uranium Contracts Litigation, 76 F.R.D. 47, 57 (W.D. Pa.
1977); Hazard, supra note 1, at 1063.
Opponents counter that applying the privilege to all corporate communications exacerbates
the privilege's deleterious effect on the discovery process. See note 4 supra and accompanying
text. The cost of obtaining information increases. See, e.g., Upjohn Co. v. United States, 449
U.S. at 386-87 (employees around the world). Employees' memories fade, their testimony falls
into line with the official corporate position, and opposing counsel become powerless to verify
the testimony. See, e.g., Nath, supra note 12, at 48 n.128, 51; Hazard, supra note 1, at 1062 (the
privilege is "a device for cover-ups").
The reponse to this argument lies in the policy judgment implicit in the privilege's exis-
tence: whenever the purposes of the privilege are fulfilled, society must bear the costs the
privilege imposes on the discovery process. As the Upjohn court concluded, extending the
privilege to communications of lower-level employees fulfills those purposes. Upjohn Co. v.
United States, 449 U.S. at 391-93. This argument for limiting the privilege is inconsistent with
that judgment, and should therefore be rejected.
99. Cf. Nath, supra note 12, at 37-54 (arguing for control group test); Harvard Note, supra
note 2 (same).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 689

the communicant's position in the corporation.1?? Thus, the zone of

silence argument should be rejected because it advocates limitation
for its own sake, amounting in principle to an argument against the
existence of the privilege.

2. Corporate Decision-Making
A second argument against the elimination of the command re-
quirement is that permitting all employees to request legal advice
and to submit communications pursuant to such requests will disturb
the corporate decision-making structure. Superiors would not learn
of the information that subordinates provide counsel. Legal fees
would arguably increase. Perhaps worst of all, subordinates might
attempt to undermine a superior's authority by bringing frivolous
claims that call into question the superior's abilities or allegiance to
the company.
More importantly, this analysis is theoretically appealing because
it achieves a symmetry between individuals and corporations in ap-
plying the privilege.?10 Superiors are deemed to personify the inter-
ests of the corporation because they function like individuals who
employ agents: each possesses the authority to request legal advice
and each controls the behavior of subordinates or agents.102 Thus, it
seems logical to require a superior to request legal advice and then
authorize communications from lower-level employees to counsel.
This argument is ultimately unpersuasive for three reasons. First,
the fundamental issue is not whether the corporation's decision-mak-
ing structure will be disturbed; rather it is whether protecting re-
quests for legal advice made by lower-level employees is consistent
with the purposes of the privilege.103 All the purposes of the privi-
lege are fulfilled when a lower-level employee requests legal advice.
Counsel receives information on which to base legal advice. That

100. See Kobak, supra note 2, at 368, 371; Iowa Comment, supra note 12, at 175; cf. WEIN-
stein, supra note 8, at ? 503(b)[04], at 503-48 to -49 (when lower-level employee initiates re-
quest for legal advice, corporation "must show that the communication, from its inception, was
intended to be made in order to obtain legal services . .. ").
101. The courts and commentators for the most part have analyzed the scope of the privi-
lege for corporations by drawing an analogy to individuals who employ agents. Both the sub-
ject matter test and modified subject matter test implicitly accept the premise that the scope of
the privilege should be determined by analogy to individuals who employ agents. See, e.g.,
Harper & Row Publications, Inc. v. Decker, 423 F.2d 487 (7th Cir. 1970), affd. per curiam by an
equally divided court, 400 U.S. 348 (1971) (privilege only applies to communication of lower-
level employees if superiors have requested such communications); Diversified Indus. v. Mere-
dith, 572 F.2d 596, 602 (8th Cir. 1978) (en bane) (same). Several commentators have expressly
adopted this analytical approach. See, e.g., Simon, supra note 7, at 963-66; Gardner, supra
note 5, at 384-88.
102. An individual's privilege extends to communications of his agents. 8 J. WIGMORE
supra note 1, ? 2317, at 618.
103. Cf. Kobak, supra note 2, at 371 ("The appropriate question to ask in every case is not
who contacted an attorney, but why?").

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
690 Michigan Law Review [Vol. 81:665

this information may not have otherwise come to the attorney's at-
tention enables him to serve the client more effectively. Moreover, if
the information concerns questionable or patently illegal behavior of
a superior, which in all likelihood will be the only cause for lower-
level employees to communicate directly to counsel, corporate attor-
neys will be able to further corporate compliance with the law by
notifying the appropriate corporate officials. Upjohn suggests that
the scope of the privilege depends upon its own underlying logic,
notwithstanding adverse implications for other policies. Given that
the proposed test fulfills the purposes of the privilege, the chance
that it may also disturb the management structure of the corporation
may be simply irrelevant.
Second, an attempt to equate superiors at all levels of the corpo-
ration with individuals who employ agents is misguided because, un-
like individual clients, the interests of the corporation and those of a
particular superior do not necessarily coincide. When an individual
requests legal advice and requires his agent to provide information
pursuant to that request, that course of action is, by definition, in the
client's best interest. In contrast, a superior, who may not even rank
among those in the upper echelon of the corporation, may frustrate a
subordinate's investigation of a particular legal matter to avoid scru-
tiny of his own behavior. In such a situation, it is the subordinate
who represents the interests of the client-corporation. Because all
employees at all levels of the corporation may at one time or another
represent the corporation's interest,104 the analogy to an individual
client who employs agents is erroneous.
Third, the supposed interference with corporate operations that
would result from eliminating the command requirement is more ap-
parent than real. Faced with threat of demotion or dismissal, mid-
dle- or lower-level employees will not initiate a request for legal
advice unless they believe that working within the corporate struc-
ture would prove fruitless. In this situation, however, the
subordinate's ability to communicate with counsel assumes the most
importance. Superiors can otherwise place their own interest above
that of the corporation. Thus, abolishing the command requirement
will lead, in practice, to requests for legal advice by lower-level em-
ployees when such a request is in the best interest of the corporation.
One could argue, however, that a corporation's decision-making
structure should not be ignored because, even if subordinates may
request legal advice, supervisors can simply refuse to act on counsel's

104. See Kobak, supra note, at 368; WEINSTEIN, supra note 8, ? 503(b)[04] at 503-48 to -49
("[A] communication which a rather low level employee made on his own initiative may be
privileged if the court can find that the employee consciously divulged his information to the
attorney for the purpose of obtaining legal advice which would benefit the corporation.");
Iowa Comment, supra note 12, at 175.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
January 1983] Note-Where Do We Go After Upjohn? 691

advice.105 Assuming that following the advice furthers the corpora-

tion's interest, upper-level executives such as the Chief Executive Of-
ficer or members of the Board of Directors, could place their
personal interest above that of the corporation and refuse to follow
counsel's advice. No lower-level employee has the authority to chal-
lenge their behavior, so the subordinate's authorization to request
legal advice would have no effect.
The fallacy of this argument is that the net effect is not the same.
The command requirement increases exponentially the probability
that the purposes of privilege will be frustrated because superiors at
all levels are empowered to interfere with or totally prohibit the flow
of information to counsel. These supervisors, who may be relatively
low-ranking corporate officials, may not themselves possess the
power to contact counsel, but they do have the power to prohibit the
flow of information to their superiors in the corporate hierarchy.
Thus, these supervisors may effectively prevent disclosure of poten-
tially illegal behavior.
In contrast, the elimination of the command requirement allows
only the highest ranking officials to place their personal interests
above that of the corporation. Not only does this of itself dramati-
cally lessen the possibility of supressing legally relevant information,
but these officials may respond to other influences that inhibit them
from acting contrary to the best interest of the corporation. Their
fiduciary duty to the corporation is one such influence.106 Another is
the coercive ability of corporate counsel, who for ethical reasons
must represent the interests of the corporation.'07 Together these in-
fluences diminish the risk of high ranking officials subverting the
corporation's best interest. Therefore, while eliminating the com-
mand requirement may not totally effectuate the privilege's purpose
of ensuring corporate compliance with the law, it will, together with
other constraints on the behavior of upper-level officials, signifi-
cantly increase the probability of achieving this objective.
Neither argument for retaining the command requirement is ulti-
mately persuasive. Even though the zone of silence arguably will
increase, no reason supports limiting the privilege on the basis of the
communicant's position in the corporation. In addition, eliminating

105. A variation of this argument is often cited in support of the control group test. See,
e.g., Nath, supra note 12, at 39-40 ("How much of a shield is necessary - at a bare minimum
- to accomplish the job of effectively guiding corporate legal decisionmaking") (footnote
omitted).
106. See, e.g., CORPORATE DIRECTOR'S GUIDEBOOK, 33 Bus. LAW. 1595 (1978); Kaplan,
Fiduciary Responsibility in the Management of the Corporation, 31 Bus. LAW. 883, 886-88
(1976); Knauss, Corporate Governance - A Moving Target, 79 MICH. L. REV. 478, 487-500
(1981). Cf. Meinhard v. Salmon, 249 N.Y. 458, 463-64, 164 N.E. 545, 546 (1928).
107. MODEL CODE OF PROFESSIONAL RESPONSIBILITY EC 5-18 (1978). See generally
Burke, The Duty of Confidentiality and Disclosing Corporate Misconduct, 36 Bus. LAW. 239,
258-60 (1981).

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]
692 Michigan Law Review [Vol. 81:665

the requirement furthers the purposes of the privilege, making any

possible interference with the allocation of decision-making respon-
sibility of little consequence.

CONCLUSION

Although the Supreme Court rejected the control group test in

Upjohn, it declined to provide a standard for applying the attorney-
client privilege to corporations. In its wake remains continued un-
certainty for corporations and confusion for courts. The standards
employed by the lower courts contain elements that frustrate the
purposes of the privilege.
This Note argues that the courts should adopt the modified sub-
ject matter test with two further refinements. The first such modifi-
cation, which narrows the scope of the privilege, is the explicit
adoption of Dean Wigmore's standard for the legal advice require-
ment. The second, which broadens the privilege's scope, is the elimi-
nation of the command requirement. Together these modifications
shift the emphasis of the judicial inquiry from the communicant's
status in the corporation to the purpose of the communication. This
approach, it is submitted, produces results more consistent with the
privilege's purposes.

This content downloaded from

[Link] on Tue, 17 Mar 2026 [Link] UTC
All use subject to [Link]