UNITED STATES COURT OF APPEALS
FOR THE FIRST CIRCUIT
No. 94-1498
FOSTER-MILLER, INC.,
Plaintiff, Appellant,
v.
BABCOCK & WILCOX CANADA,
Defendant, Appellee.
APPEAL FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF MASSACHUSETTS
[Hon. Richard G. Stearns, U.S. District Judge]
Before
Selya, Circuit Judge,
Aldrich, Senior Circuit Judge,
and Boudin, Circuit Judge.
James J. Foster, with whom Michael A. Diener and Wolf,
Greenfield & Sacks, P.C. were on brief, for appellant.
Peter L. Resnik, with whom Cherie L. Krigsman and McDermott,
Will & Emery were on brief, for appellee.
February 9, 1995
SELYA, Circuit Judge. In Boit v. Gar-Tec Prods., Inc.,
SELYA, Circuit Judge.
967 F.2d 671 (1st Cir. 1992), we urged district courts to take a
flexible approach in handling motions to dismiss for lack of in
personam jurisdiction, and, concomitantly, to tailor procedures
for use in those purlieus. Turning from the general to the
particular, we recommended that district courts employ varying
levels of scrutiny in connection with such motions, adapting the
level of scrutiny to the exigencies of the individual case. See
id. at 674-78. Among other possibilities, we suggested using a
special intermediate standard when "factual issues are common to
both the jurisdictional question and the claim on the merits . .
. ." Id. at 677.
The case before us today an appeal by Foster-Miller,
Inc. (FMI) from an order dismissing its commercial tort action
against Babcock & Wilcox Canada (BWC) illustrates vividly that
Boit's flexible approach demands circumspection in its
application. In this case, the district court applied Boit's
intermediate standard too rashly when, eager to test whether a
legally sufficient showing of jurisdiction had been made, it
neither gave the parties adequate notice that it intended to use
this special standard nor ensured that FMI had a fair opportunity
to gather and present the evidence necessary for such a showing.
While we are not without sympathy for the district judge he
inherited this case midstream, and Boit, in retrospect, should
have emphasized the need to forewarn litigants of a trial court's
intention to go beyond the prima facie standard typically
2
associated with motions to dismiss under Fed. R. Civ. P. 12(b)(2)
we cannot permit the dismissal order to stand.
I. THE FACTS
I. THE FACTS
We sketch the operative facts, drawing liberally from
the lower court's opinion. See Foster-Miller, Inc. v. Babcock &
Wilcox Can., 848 F. Supp. 271 (D. Mass. 1994).
The parties to this appeal are quondam competitors:
FMI is a Massachusetts corporation engaged in furnishing sludge
and particle removal services for nuclear steam generators; BWC
is a Canadian firm that, among other things, services such
generators. At its core, the litigation concerns a virtual
meltdown of the parties' relationship, which in turn detonated a
lawsuit. The tale follows.
As early as 1988, FMI and BWC entertained the prospect
of a joint venture to furnish sludge removal services to Ontario
Hydro, a Canadian utility. Although the joint venture idea
stalled and the principals went their separate ways, Canada
remained an alluring target. But the road to prosperity had a
large pothole. FMI's then-extant technology, known by the
acronym "CECIL," featured flexible lances that directed powerful
bursts of water at pockets of sludge found within the hard-to-
reach crannies of nuclear steam boilers. While this system had
distinct competitive advantages over BWC's rival rigid lance
system, neither system performed satisfactorily in the cleansing
of Canadian boilers (known in the trade as Candu boilers).
Determined to detour around the "can't do Candu"
3
pothole and penetrate the Canadian market, BWC set out to design
a lance of unprecedented flexibility. In 1989, while BWC's
development efforts were underway, Ontario Hydro (acting on
behalf of a consortium of Canadian utilities) retained FMI to
study the feasibility of adapting FMI's flexible lance technology
for use in Candu boilers. As part of this endeavor, FMI
contracted with a well-known supplier, U.S. Composites (CompCo),
to create a new type of hose.
In March 1990, Robert A.S. Lee, an FMI employee who had
been instrumental in perfecting CECIL, attended an industry
conference in Tennessee. Daniel St. Louis, a BWC engineer
involved in that company's push to fashion a flexible lance,
attended the same session. During a previous encounter, the men
had casually discussed high pressure hoses. On this occasion,
their conversation became more detailed and focused on the
possibility of reinforcing high pressure hoses with certain
fibers. The discussion proved prophetic: a few weeks
thereafter, CompCo delivered the special hose that FMI had asked
it to design. The hose was thought in certain quarters to
represent a technological breakthrough. One of its more
revolutionary features was a double-layered Kevlar sheath that
supplied desired reinforcement.
On May 11, 1990, an Ontario Hydro representative, James
Malaugh, traveled to FMI's plant in Waltham, Massachusetts, to
assess FMI's progress. Seeking expertise and insight, Malaugh
invited St. Louis to join him. Nonplussed, FMI allowed St. Louis
4
to attend only after BWC signed a confidentiality agreement. The
agreement, duly executed by a ranking official of BWC and
transmitted via facsimile machine from Canada, acknowledged that
FMI "anticipate[d] disclosing . . . certain information of a
novel, proprietary, or confidential nature," and memorialized
BWC's promise "not to use [the] information for any purpose
unless specifically authorized in writing by FMI." The agreement
also stipulated that FMI would be entitled to relief for any
breach.
William Leary, the FMI engineer in charge of the
Ontario Hydro project, hosted the Waltham session. The
participants debated various aspects of flexible lance
technology, including the preferred characteristics of the hose
and possible methods of reinforcement. At one point Leary,
responding to a direct question by St. Louis, identified CompCo
as FMI's supplier. Not long after the Waltham meeting, St. Louis
contacted CompCo and inquired about the possibility of that
company fabricating a similar hose for BWC. St. Louis'
suggestion that FMI would not object proved overly sanguine;
after consulting with FMI, CompCo rebuffed BWC's overtures.1
Undaunted, BWC forged ahead in its research effort. It
eventually succeeded in manufacturing its own flexible lance,
suitable for Candu boilers. Thereafter, Ontario Hydro awarded
BWC a lucrative contract.
1Nonetheless, BWC managed to obtain a sample of the special
hose. The parties dispute whether this occurred before or after
the May 11 meeting.
5
II. THE LITIGATION
II. THE LITIGATION
On November 12, 1993, FMI, claiming to have gotten
hosed, commenced suit against BWC in the United States District
Court for the District of Massachusetts. Invoking diversity
jurisdiction, 28 U.S.C. 1332 (1988), FMI charged breach of the
confidentiality agreement, misappropriation of trade secrets, and
unfair competition. BWC moved to dismiss for lack of in personam
jurisdiction or, in the alternative, on the basis of forum non
conveniens.
Judge Keeton drew the case. Concerns about the
parties' trade secrets slowed discovery to a crawl. At a
conference held on December 16, 1993, Judge Keeton scheduled a
hearing on the motion for January 4, 1994, restricted discovery
for the time being to matters "bear[ing] upon the jurisdictional
issue," and advised counsel that, absent an agreement dissolving
the discovery deadlock, he would use the traditional prima facie
standard, not the special intermediate standard, in evaluating
the motion to dismiss.
At that point, fate intervened. In a routine shuffling
of cases ancillary to the appointment of several new judicial
officers, this case was plucked from Judge Keeton and reassigned
to Judge Stearns. The January 4 hearing never materialized.
Instead, Judge Stearns heard the motion to dismiss on February 1
and 2, 1994. Though the discovery dispute had not been resolved,
Judge Stearns, to FMI's obvious chagrin, undertook not only to
probe the existence of the basic facts on which jurisdiction
6
might be premised but also to adjudicate certain ultimate facts
(e.g., whether the participants actually disclosed any
confidential information at the Waltham meeting).2 And he
applied the special intermediate level of scrutiny rather than
the more easily satisfied prima facie standard.
The district court granted the motion to dismiss.
After stressing the importance of the Waltham meeting to the
jurisdictional issue BWC, after all, had no other significant
contacts with the forum Judge Stearns articulated two bases for
refusing to exercise jurisdiction. First, he found it unlikely
either that proprietary information had been disclosed at the
meeting or that such information came into BWC's possession as a
result of the meeting; therefore, FMI's cause of action did not
arise from BWC's participation in the meeting as required by the
Massachusetts long-arm statute. See Foster-Miller, 848 F. Supp.
at 276-77. Second, and alternatively, the judge concluded that,
even if FMI's claims did arise from BWC's participation in the
Waltham meeting, it would be unreasonable for a Massachusetts-
based court to exercise jurisdiction over BWC, in part because
the court might not be able to grant effective injunctive relief.
See id. at 277.
FMI moved for reconsideration, specifically withdrawing
2Because the scope of this inquiry caught FMI off guard, it
tendered a series of post-hearing offers of proof in an effort to
make up lost ground. We do not comment on the timing of FMI's
proffers. Further proceedings in the district court are
obligatory, see infra Parts IV(C) & V, at which time new
evidentiary submissions can be assembled.
7
its prayer for an injunction. Judge Stearns denied the motion
without comment. This appeal ensued.
III. SPECIFIC PERSONAL JURISDICTION
III. SPECIFIC PERSONAL JURISDICTION
Prior to reexamining the Boit framework, we rehearse
certain general principles of law relating to specific in
personam jurisdiction.
Personal jurisdiction implicates the power of a court
over a defendant. In a federal court, both its source and its
outer limits are defined exclusively by the Constitution. See
Insurance Corp. of Ireland v. Compagnie des Bauxites de Guinee,
456 U.S. 694, 702 (1982).
There are two different avenues by which a court may
arrive at personal jurisdiction. One frequently traveled route
leads to general jurisdiction. "General jurisdiction exists when
the litigation is not directly founded on the defendant's forum-
based contacts, but the defendant has nevertheless engaged in
continuous and systematic activity, unrelated to the suit, in the
forum state." United Elec. Workers v. 163 Pleasant St. Corp.,
960 F.2d 1080, 1088 (1st Cir. 1992) (Pleasant St. I). Here,
BWC's forum-related contacts are far too scanty to justify the
invocation of general jurisdiction. See Foster-Miller, 848 F.
Supp. at 273 (marshalling certain undisputed facts).
When general jurisdiction is lacking, the lens of
judicial inquiry narrows to focus on specific jurisdiction. As
the label implies, this focus requires weighing the legal
8
sufficiency of a specific set of interactions as a basis for
personal jurisdiction. See Helicopteros Nacionales de Colombia,
S.A. v. Hall, 466 U.S. 408, 414-15 & nn.8-9 (1984) (recognizing
"general" and "specific" jurisdiction and distinguishing between
them); Pleasant St. I, 960 F.2d at 1088 (similar); Donatelli v.
National Hockey League, 893 F.2d 459, 462-63 (1st Cir. 1990)
(similar). In that exercise, the applicable constitutional
limits assume critical importance. We explain briefly.
The existence of specific personal jurisdiction depends
upon the plaintiff's ability to satisfy two cornerstone
conditions: "first, that the forum in which the federal district
court sits has a long-arm statute that purports to grant
jurisdiction over the defendant; and second, that the exercise of
jurisdiction pursuant to that statute comports with the
strictures of the Constitution." Pritzker v. Yari, F.3d ,
(1st Cir. 1994) [No. 93-2374, slip op. at 9]; see also
Ticketmaster-New York, Inc. v. Alioto, 26 F.3d 201, 204 (1st Cir.
1994); Hahn v. Vermont Law Sch., 698 F.2d 48, 51 (1st Cir.
1983).
Although we deem the first of the cornerstone
conditions to be self-explanatory, the second condition requires
amplification. This condition implicates three distinct
components, namely, relatedness, purposeful availment (sometimes
called "minimum contacts"), and reasonableness:3
3This trilogy forms an interesting contrast with the
jurisprudence of the branch of the Massachusetts long-arm statute
that applies in many business disputes. See Mass. Gen. L. ch.
9
First, the claim underlying the litigation
must directly arise out of, or relate to, the
defendant's forum-state activities. Second,
the defendant's in-state contacts must
represent a purposeful availment of the
privilege of conducting activities in the
forum state, thereby invoking the benefits
and protections of that state's laws and
making the defendant's involuntary presence
before the state's courts foreseeable.
Third, the exercise of jurisdiction must, in
light of the Gestalt factors, be reasonable.
Pleasant St. I, 960 F.2d at 1089; accord Pritzker, F.3d at
[slip op. at 10-11]; Ticketmaster, 26 F.3d at 206.
IV. APPLYING THE JURISDICTIONAL RULES
IV. APPLYING THE JURISDICTIONAL RULES
A long-arm statute is plainly available for FMI's use.
See Mass. Gen. L. ch. 223A, 3(a) (1992). Section 3(a), quoted
supra note 3, is not modest in its reach. Its language is
expansive, and its words are to be generously applied in order to
determine whether a given defendant fairly can be said to have
participated in the forum's economic life. See Pleasant St. I,
960 F.2d at 1087 (collecting cases). Since section 3(a) applies
here, we turn directly to the second of the two cornerstone
223A, 3(a) (1992) (providing in relevant part for the exercise
of "personal jurisdiction over a person, who acts directly or by
an agent, as to a cause of action in law or equity arising from
the person's . . . transacting any business" in Massachusetts).
The Massachusetts Supreme Judicial Court has held that, to
animate the statute, the facts must show both that the defendant
transacted business in Massachusetts, and that the plaintiff's
claim arises out of the transaction of that business. See Tatro
v. Manor Care, Inc., 625 N.E.2d 549, 551 (Mass. 1994); Good Hope
Indus., Inc. v. Ryder Scott Co., 389 N.E.2d 76 (Mass. 1979).
From our coign of vantage, these two requirements appear to
correspond, in reverse order, to the first two of the three
constitutional components, i.e., the "transacting any business"
requirement corresponds to "minimum contacts," while the "arising
from" requirement corresponds to relatedness.
10
conditions that constitute the foundation for a finding of
specific in personam jurisdiction.
As we have said, the condition comprises three
components. The first minimum contacts is not legitimately
in issue. The seminal jurisdictional fact that BWC voluntarily
dispatched a representative to Massachusetts for commercial
advantage pursuant to a written contract with a Massachusetts
firm cannot be gainsaid. In taking this action, BWC
purposefully conducted activities in the forum state, thereby
making a suit foreseeable. See id. at 1089. Hence, we endorse
the district court's conclusion that BWC transacted business in
Massachusetts to such an extent, and in such a manner, as to
satisfy the minimum contacts requirement. See Foster-Miller, 848
F. Supp. at 276; compare Pritzker, F.3d at [slip op. at
14] (finding jurisdiction in part because the nonresident
defendant, by contract, had "knowingly acquir[ed] an economically
beneficial interest" in a forum-based commercial venture).
Setting the matter of minimum contacts to rest, we come
face to face with the next component: relatedness. In this
case, evaluating that requirement reduces to whether FMI's claim
arises from BWC's minimum contacts. To place this issue into
proper perspective, we first limn the options that are available
to a district court in handling a motion to dismiss for want of
jurisdiction over the person. We then refine that framework and
scrutinize the decision below in light of our handiwork.
A. Establishing and Testing Personal Jurisdiction.
A. Establishing and Testing Personal Jurisdiction.
11
It is apodictic that the plaintiff, who bears the
burden of proving the existence of in personam jurisdiction, must
carry the devoir of persuasion on the elements of relatedness and
minimum contacts. See Ticketmaster, 26 F.3d at 207 n.9; Martel
v. Stafford, 992 F.2d 1244, 1247 n.5 (1st Cir. 1993); Donatelli,
893 F.2d at 468. But this is merely one step along the path; to
allocate the burden is neither to define the evidentiary showing
necessary to meet it nor to explain whether that showing varies
from context to context.
We addressed these important issues in Boit. There, we
tried to formulate a procedural matrix that would serve to endow
the decisional process with appropriate degrees of economy and
manageability. That endeavor produced a trio of standards, each
corresponding to a level of analysis, that might usefully be
employed when a trial court comes to grips with a motion to
dismiss for want of personal jurisdiction.
The most conventional of these methods permits the
district court "to consider only whether the plaintiff has
proffered evidence that, if credited, is enough to support
findings of all facts essential to personal jurisdiction." Boit,
967 F.2d at 675. To make a prima facie showing of this calibre,
the plaintiff ordinarily cannot rest upon the pleadings, but is
obliged to adduce evidence of specific facts. See id. Withal,
the district court acts not as a factfinder, but as a data
collector. That is to say, the court, in a manner reminiscent of
its role when a motion for summary judgment is on the table, see
12
Fed. R. Civ. P. 56(c), must accept the plaintiff's (properly
documented) evidentiary proffers as true for the purpose of
determining the adequacy of the prima facie jurisdictional
showing. Despite the lack of differential factfinding, this
device is a useful means of screening out cases in which personal
jurisdiction is obviously lacking, and those in which the
jurisdictional challenge is patently bogus. However, the
approach offers little assistance in closer, harder-to-call
cases, particularly those that feature conflicting versions of
the facts. See, e.g., General Contracting & Trading Co. v.
Interpole, Inc., 899 F.2d 109 (1st Cir. 1990).
A second option open to the court is to embark on a
factfinding mission in the traditional way, taking evidence and
measuring the plaintiff's jurisdictional showing against a
preponderance-of-the-evidence standard. In Boit, we stated that
this standard may appropriately be invoked when a court
determine[s] that in the circumstances of a
particular case it is unfair to force an out-
of-state defendant to incur the expense and
burden of a trial on the merits in the local
forum without first requiring more of the
plaintiff than a prima facie showing of facts
essential to in personam jurisdiction. A
court may so determine, for example, when the
proffered evidence is conflicting and the
record is rife with contradictions, or when a
plaintiff's affidavits are "patently
incredible . . . ."
Boit, 967 F.2d at 676 (offering examples). Virtually by
definition, the preponderance standard necessitates a full-blown
evidentiary hearing at which the court will adjudicate the
13
jurisdictional issue definitively before the case reaches
trial.4 In that mode, the court will "consider[] all relevant
evidence proffered by the parties and mak[e] all factual findings
essential to disposition of the motion." Id. But this method
must be used discreetly. For one thing, pretrial evidentiary
hearings are relatively cumbersome creatures, and, if used
routinely, can squander judicial resources. For another thing,
since this method contemplates a binding adjudication, the
court's factual determinations ordinarily will have preclusive
effect, and, thus, at least in situations in which the facts
pertinent to jurisdiction and the facts pertinent to the merits
are identical, or nearly so, profligate use of the preponderance
method can all too easily verge on a deprivation of the right to
trial by jury.
In Boit, we recognized these difficulties. We also
recognized that the prima facie and preponderance-of-the-evidence
standards are merely two of several possible models, and that
trial courts need not confine themselves to choosing between
these two levels of evidentiary scrutiny. See id. at 677. In
the special circumstance in which the assertion of jurisdiction
is bound up with the claim on the merits, the possibility of
preclusion renders use of the preponderance standard troubling,
4Such hearings frequently are convened under the aegis of
Fed. R. Civ. P. 12(d), which provides in pertinent part that
certain defenses, including the defense of lack of in personam
jurisdiction, "shall be heard and determined before trial on
application of any party," unless the court orders a deferral
until time of trial.
14
while the possibility of permitting a dubious case to proceed
beyond the pleading stage, and even to trial, though the court
eventually will be found to lack jurisdiction, renders use of the
prima facie standard undesirable.
The Boit panel anticipated that, when this special
circumstance arose, trial courts might steer a middle course by
engaging in some differential factfinding, limited to probable
outcomes as opposed to definitive findings of fact, thereby
skirting potential preclusionary problems while at the same time
enhancing the courts' ability to weed out unfounded claims of
jurisdiction. Utilizing this intermediate standard, a district
court, "even though allowing an evidentiary hearing and weighing
evidence to make findings . . . may merely find whether the
plaintiff has shown a likelihood of the existence of each fact
necessary to support personal jurisdiction." Id. This showing
constitutes an assurance that the circumstances justify imposing
on a foreign defendant the burdens of trial in a strange forum,
but leaves to the time of trial a binding resolution of the
factual disputes common to both the jurisdictional issue and the
merits of the claim. See id. at 678.
Unlike the prima facie standard, and like the
preponderance standard, this third method, which we sometimes
call the "likelihood standard,"
involves factfinding rather than merely
making a ruling of law regarding sufficiency
of the evidence to present a fact question.
Like the first and unlike the second method,
however, the third method avoids potentially
troubling issues of "issue preclusion" or
15
"law of the case" (at least when the court
denies the motion) because a determination by
such an intermediate standard . . . does not
purport to be a finding by the same standard
on the same issue as will be decided at
trial.
Id.
We acknowledge that having an array of standards at the
ready may be thought too much of a good thing. However, even
though an intermediate standard will not be used with great
frequency, the need for one is manifest. We can postulate a
variety of "common facts" scenarios in which the facts necessary
to sustain personal jurisdiction are intimately bound up with
facts necessary to establish the merits of the underlying claim.
See, e.g., Ann Althouse, The Use of Conspiracy Theory to
Establish In Personam Jurisdiction: A Due Process Analysis, 52
Fordham L. Rev. 234, 247-51 (1983) (noting, though not adequately
resolving, the problem created in situations where proving the
facts "upon which jurisdiction depends is viewed as inextricably
tied to the substantive merits of the case"). It is precisely
because of the incidence of these situations situations in
which the issue of jurisdiction is factually enmeshed with the
merits of the suit that we recognized in Boit the need for an
intermediate standard of proof and, correspondingly, an
intermediate standard of judicial analysis.
B. Standards of Review.
B. Standards of Review.
We are reluctant to end our discussion of the methods
available to district courts for testing jurisdictional waters
without mentioning appellate review. As a practical matter, the
16
standard of review will depend in the first instance on whether
the court of appeals is reviewing the district court's choice of
an analytic method or its application of such a method. As for
the court's initial choice from among the three standards we have
discussed prima facie, likelihood, or preponderance appellate
review is de novo. This accords with the general principle that
a trial court's determinations as to the legal rules that govern
a party's proof, including those that dictate what quantum of
proof the law requires, are subject to plenary review. See,
e.g., Putnam Resources v. Pateman, 958 F.2d 448, 471 (1st Cir.
1992); see also Soto v. United States, 11 F.3d 15, 17 (1st Cir.
1993) (holding that "if a district court applie[s] an erroneous
legal standard to the facts," de novo review obtains).
As for the district court's subsequent application of
the method that it chooses, the standard of review will vary from
method to method. If the district court employs the prima facie
standard, then appellate review is de novo. See United Elec.
Workers v. 163 Pleasant St. Corp., 987 F.2d 39, 44 (1st Cir.
1993) (Pleasant St. II); Boit, 967 F.2d at 675; see also Garita
Hotel Ltd. Partnership v. Ponce Fed. Bank, 958 F.2d 15, 17 (1st
Cir. 1992) (explaining that appellate courts traditionally review
rulings on motions to dismiss de novo, "applying the same
criteria that obtained in the court below"). If the district
court departs from the conventional method of adjudicating
motions to dismiss and relies upon the preponderance-of-the-
evidence standard to determine the existence vel non of personal
17
jurisdiction, then appellate review is for clear error. See
CutCo Indus., Inc. v. Naughton, 806 F.2d 361, 364-65 (2d Cir.
1986); see also Fed. R. Civ. P. 52(a).
If the district court employs the intermediate
standard, then appellate review is for abuse of discretion. Cf.
Boit, 967 F.2d at 678 (suggesting a deferential standard of
appellate review). Two considerations point to the applicability
of this deferential mode of review in this situation. First, the
nature of a likelihood analysis is such that it falls naturally
within the realm of discretionary decisionmaking. Second, from a
practical standpoint, a likelihood analysis simply does not seem
amenable to either of the other standards. Unlike the classic
motion to dismiss, in which the plaintiff's assertions are
accepted as true, a likelihood analysis requires the judge to
pass upon the accuracy and integrity of the plaintiff's
assertions. Yet, in contrast to a preponderance-of-the-evidence
analysis, these determinations are not true findings of fact, for
they lack definiteness to some degree, and they also lack the
preclusive quality that would otherwise normally attach.
Consequently, we believe that abuse of discretion is the proper
standard of review. In practical terms, this means that we will
set aside the challenged ruling only if we descry "a meaningful
error in judgment." Anderson v. Cryovac, Inc., 862 F.2d 910, 923
(1st Cir. 1988); accord Rosario-Torres v. Hernandez-Colon, 889
F.2d 314, 323 (1st Cir. 1989) (en banc).
Of course, whatever method is chosen and however it may
18
be applied, appellate review of the trial court's legal
conclusions about whether its findings do or do not support
the exercise of in personam jurisdiction is always nondeferential
and plenary. See Boit, 967 F.2d at 678.
C. Applying the Likelihood Standard.
C. Applying the Likelihood Standard.
After convening an evidentiary hearing and bringing the
likelihood standard to bear, Judge Stearns found it unlikely
either that FMI disclosed legally protected information at the
Waltham meeting or that BWC obtained confidences as a consequence
of the meeting. See Foster-Miller, 848 F. Supp. at 276-77. FMI
assigns error. Its appeal raises potentially difficult questions
about the application of Boit's likelihood standard in certain
types of cases.
We hasten to note that the paradigm case involving
the use of Boit's intermediate standard as a basis for exercising
jurisdiction, i.e., as a basis for denying a motion to dismiss
hardly seems problematic. In such a scenario, the plaintiff is
permitted to proceed in its forum of choice, yet the defendant
has the consolation of having been afforded a detailed
demonstration, beyond a mere prima facie showing, of why the
court deems it fair to exercise jurisdiction, at least
provisionally. It is only when Boit's intermediate standard is
used as a basis for declining the exercise of jurisdiction, i.e.,
as a basis for granting a motion to dismiss, that the prospect of
mischief looms. One can easily imagine cases in which the
likelihood standard might be applied to adjudicate facts that are
19
only marginally related to jurisdiction, or are very closely
related to the merits of the plaintiff's substantive claims, thus
prematurely extinguishing a plaintiff's ability to present its
case in a full and fair manner.5
The short of it is that, whatever its merits in the
abstract, Boit's intermediate standard requires caution in its
application, especially when it appears that a dismissal may
result. Indeed, although Boit does, in dictum, 967 F.2d at 677-
78, propose to authorize such dismissals, it is noteworthy that,
apart from the opinion of the court below, there is no other
reported case, Boit included, that has sanctioned a dismissal
pursuant to a district court's use of the likelihood standard.
In general, this is as it should be. To the limited
extent that dismissals under Boit's intermediate standard are
justified at all, they will happen only rarely. Even then, the
exact bounds of permissible application may not always be
evident. Nonetheless, we believe it is better to tolerate the
inconvenience of mild doctrinal uncertainty rather than to forgo
altogether the utility of an intermediate standard and method of
analysis. See generally Stephen L. Carter, Constitutional
Adjudication and the Indeterminate Text: A Preliminary Defense
5Conceivably, such an adjudication may also serve to thrust
the judge into a role that, depending upon the circumstances,
more appropriately belongs to the jury. See, e.g., Jacob v. City
of New York, 315 U.S. 752, 756 (1942) (noting basic principle
that merely because a "case is close and a jury might find either
way . . . is no reason for a court to usurp the function of the
jury"); Nunes v. Farrell Lines, Inc., 227 F.2d 619, 621-22 (1st
Cir. 1955) (applying principle of Jacob and vacating directed
verdict).
20
of an Imperfect Muddle, 94 Yale L.J. 821 (1985) (recognizing the
impossibility of removing all uncertainty from legal doctrine);
Oliver Wendell Holmes, Jr., The Path of the Law, 10 Harv. L. Rev.
457, 465 (1897) (warning that, in respect to judicial decisions,
"certainty generally is illusion, and repose is not the destiny
of man"). The bottom line, clearly, is that judges employing
Boit's intermediate standard should proceed with great care.
In any event, these potential difficulties are
peripheral to the instant appeal. In this instance, the flaw is
less Judge Stearns' initial decision to switch the signals
previously given by Judge Keeton and instead apply the
intermediate standard, but more his failure to apprise FMI
squarely of this change of plan a failure that was aggravated
by FMI's inability to engage in appropriate discovery and then to
present the totality of its evidence within the context of a
likelihood analysis.
When judges elect on their own initiative to use
innovative methods in an effort to accelerate the progress of a
case, they must take pains to ensure that parties are given
satisfactory notice, reasonable access to discovery, and a
meaningful opportunity to present evidence. See, e.g., Stella v.
Town of Tewksbury, 4 F.3d 53, 55-56 (1st Cir. 1993) (stating
these principles in the context of sua sponte summary judgment);
Jardines Bacata, Ltd. v. Diaz-Marquez, 878 F.2d 1555, 1560-61
(1st Cir. 1989) (similar). While the likelihood standard has
value, the latent risks associated with its use are not
21
insignificant, and they should be ameliorated to the extent
practicable.
Here, the lower court did not afford FMI the process
that was due. To be sure, Judge Stearns advised counsel at a
status conference on January 20, 1994, that he was pondering the
use of the likelihood standard, but he neither eased the existing
restriction on discovery nor superseded Judge Keeton's prior
directives. The prima facie standard remained the default
setting, and Judge Stearns' intentions remained open to
conjecture until the day of the hearing. Indeed, while the court
transmitted mixed signals to some degree, it closed the January
20 conference by specifically announcing that the question of
misappropriation would not be subject to anything more rigorous
than scrutiny under a prima facie standard. On this point, Judge
Stearns' intention could not have been more explicit. He told
the lawyers: "I am going to, for purposes of this hearing, . . .
basically accept whatever [FMI] allege[s] to be true in terms of
the misappropriation."
Following this pronouncement, the court never gave the
litigants suitable forewarning of a change of heart, or of the
extent to which it would apply the likelihood standard. To
understand the gravity of this omission, it is important to
understand the restriction imposed on discovery by Judge Keeton,
and how that restriction arose. On December 15, 1993, FMI moved
"to examine the documents and other materials maintained by BWC
which would be relevant to statements in the affidavits of Mr.
22
St. Louis and others concerning contacts with, statements made
by, and other information received from Foster-Miller . . . ."
BWC objected. The next day Judge Keeton, ruling ore tenus,
restricted FMI's discovery to matters "bear[ing] upon the
jurisdictional issue." All other discovery, he ruled, was "out
of bounds" for the time being. We think that this limitation,
coupled with the judge's simultaneous indication that he would
evaluate the motion to dismiss under the prima facie standard,
effectively prevented FMI from engaging in merits-related
discovery. And when Judge Stearns then shifted abruptly from the
forecasted prima facie standard to the more intrusive likelihood
standard, the preexisting restriction which remained intact on
Judge Stearns' watch hamstrung FMI.6
Since this imperfect communication obviously
prejudiced FMI's ability fairly to meet the rigors that an
across-the-board use of the likelihood standard imposed in the
circumstances of this case, we must set aside the court's
conclusion that FMI's suit did not "arise from" BWC's activities
in the forum state. To that extent, then, the dismissal order
6This is because the two standards involve markedly
different quanta of proof. So long as a prima facie standard
obtained, FMI had neither a right nor a reason, in the course of
"jurisdictional discovery," to ferret out all the supporting
evidence regarding the confidential nature of what had been
discussed in Waltham. By the same token, it had neither a right
nor a reason to document fully the allegedly improper uses of
such confidences by BWC. But once the court shifted to a
likelihood standard, the scope, tenor and degree of the
prospective inquiry changed, and FMI was caught, like a fawn in
the headlights of a speeding van, without the discovery it needed
to prove its point.
23
succumbs.
V. ASSESSING REASONABLENESS
V. ASSESSING REASONABLENESS
Our odyssey is not yet at an end. In addition to
holding that FMI's claim did not arise from BWC's in-forum
contacts, the district court held, alternatively, that it would
be unreasonable to exercise jurisdiction over BWC. See Foster-
Miller, 848 F. Supp. at 277; see generally Asahi Metal Indus. Co.
v. Superior Court, 480 U.S. 102, 113-16 (1987) (undertaking
reasonableness inquiry); Burger King Corp. v. Rudzewicz, 471 U.S.
462, 476-78, 482-85 (1985) (similar). Since the rapid-fire shift
of standards probably tainted this conclusion as well, we could
simply vacate the alternative holding. We choose instead to
dissect it for three reasons: the district court's rationale is
troubling, it has been reported in a published opinion, and the
underlying issue may arise on remand.
The hallmark of reasonableness in the context of
personal jurisdiction is "fair play and substantial justice."
International Shoe Co. v. State of Washington, 326 U.S. 310, 320
(1945). We have tended to channel the quest for that imperative
through a template that highlights five factors. The factors
include:
(1) the defendant's burden of appearing, (2)
the forum state's interest in adjudicating
the dispute, (3) the plaintiff's interest in
obtaining convenient and effective relief,
(4) the judicial system's interest in
obtaining the most effective resolution of
the controversy, and (5) the common interests
of all sovereigns in promoting substantive
social policies.
24
Pleasant St. I, 960 F.2d at 1088. We have called the points that
compose this template "the gestalt factors" because, in any given
case, they may neither be amenable to mechanical application nor
be capable of producing an open-and-shut result. Their primary
function is simply to illuminate the equitable dimensions of a
specific situation, thereby "put[ting] into sharper perspective
the reasonableness and fundamental fairness of exercising
jurisdiction" in that situation. Pritzker, F.3d at [slip
op. at 19].
In the case at bar, the trial court found that the
first, fourth, and fifth factors did not favor one outcome over
the other, but that the remaining two factors discouraged the
exercise of jurisdiction. See Foster-Miller, 848 F. Supp. at
277. The court then invoked a sixth factor the ability of a
Canadian court to apply Massachusetts law competently and
impartially and concluded that even if "some harm had flowed to
Foster-Miller from the Waltham meeting," the suit should be
dismissed based on "considerations of due process." Id.
The district court's analysis is flawed. First and
foremost, the court's added consideration the absence of any
reason to believe that a Canadian court would display bias or
prove incapable of applying Massachusetts law has no place in
the minimum contacts calculus. Though the five gestalt factors
should not necessarily be deemed collectively exhaustive, see,
e.g., FDIC v. British-American Ins. Co., 828 F.2d 1439, 1442 (9th
Cir. 1987) (recognizing additional factors), the element seized
25
upon by the court below is of no relevance here. Moreover, it is
already committed to the doctrine of forum non conveniens. See
Gulf Oil Corp. v. Gilbert, 330 U.S. 501, 509 (1947); see also
Burger King, 471 U.S. at 477 & n.20 (specifically distinguishing
between the primary role of the enumerated gestalt factors and
the secondary role of considerations relevant to forum non
conveniens). The doctrines of personal jurisdiction and forum
non conveniens share certain similarities, but they embody
distinct concepts and should not casually be conflated. Compare
Allan R. Stein, Forum Non Conveniens and the Redundancy of Court-
Access Doctrine, 133 U. Pa. L. Rev. 781, 788-89 (1985)
(distinguishing the doctrines) with Margaret G. Stewart, Forum
Non Conveniens: A Doctrine in Search of a Role, 74 Cal. L. Rev.
1259 (1986) (arguing that certain factors currently considered
under forum non conveniens doctrine should be subsumed under
personal jurisdiction analysis). Consequently, the district
court's self-proclaimed sixth factor adds nothing useful to the
jurisdictional mix.7
The court also adopted several other questionable
7To drive this conclusion home, we note two related points.
First, the very case on which the district court relied in
assigning weight to the added factor, Howe v. Goldcorp Invs.,
Ltd., 946 F.2d 944 (1st Cir. 1991), cert. denied, 112 S. Ct. 1172
(1992), is a forum non conveniens case, not a personal
jurisdiction case. Second, we are unable to discern a link
between the judge's hosannas to the Canadian court system and his
conclusion that a federal district court sitting in Massachusetts
lacks jurisdiction. Assuming that neither of two courts poses an
undue risk of biased or incompetent adjudication, there is
nothing to be counted against either of them in working the
jurisdictional calculus.
26
positions, likely influenced by its mistaken blending of the
theories of personal jurisdiction and forum non conveniens, in
the course of ascertaining that the second and third gestalt
factors militated against the exercise of jurisdiction. For
example, the court deviated from the thrust of the second factor
by centering much of its discussion on "the implications of this
litigation for a Canadian industry upon whom [sic] an entire
population depends for electric power" and on the extent to which
Canada's interests "dwarf" those of Massachusetts. Foster-
Miller, 848 F. Supp. at 277. This emphasis distorts the
directive that a court pondering the second factor must mull "the
forum state's interest in adjudicating the dispute," Pleasant St.
I, 960 F.2d at 1088. The purpose of the inquiry is not to
compare the forum's interest to that of some other jurisdiction,
but to determine the extent to which the forum has an interest.
See, e.g., Burger King, 471 U.S. at 483 & n.26 (flatly rejecting
the notion that a non-forum state's "acknowledged interest might
possibly render jurisdiction in [the forum] unconstitutional" and
observing that "minimum-contacts analysis presupposes that two or
more States may be interested in the outcome of a dispute").
The district court's analysis is equally awry in its
treatment of the third gestalt factor (which requires an
assessment of "the plaintiff's interest in obtaining convenient
and effective relief," Pleasant St. I, 960 F.2d at 1088).
Although finding that "it is more convenient for the plaintiff to
litigate this matter in domestic comfort," Foster-Miller, 848 F.
27
Supp. at 277, the court offset this finding by invoking, inter
alia, a presumed inability "to award plaintiff the full measure
of relief that it seeks" because of doubts concerning both the
propriety and the efficacy of enjoining a foreign national whose
presence in Massachusetts had been fleeting. Id. This concern
is beside any relevant point where a plaintiff's inability to
obtain certain kinds of relief is wholly a product of her own
choice of forum. At any rate, the plaintiff here explicitly
informed the court of its willingness to forgo injunctive relief
if necessary to salvage jurisdiction.
We have another problem with the district court's
assessment of the third gestalt factor. The court downplayed
FMI's convenience because the company "does business in Canada
and is represented by presumably capable Canadian attorneys."
Id. Putting this spin on the matter emphasizing that the
plaintiff could just as easily litigate in a Canadian court
effectively nullifies the plaintiff's choice to litigate its suit
not in Canada but in Massachusetts. Though such judicial second-
guessing may be proper in deciding transfer-of-venue motions or
when the plaintiff's supposed convenience "seems to be . . . a
makeweight," contrived purely for strategic advantage, courts
considering jurisdictional issues generally should "accord
plaintiff's choice of forum a degree of deference in respect to
the issue of its own convenience . . . ." Ticketmaster, 26 F.3d
at 211. So it is here.
We will not comment on the lower court's assessment of
28
the first, fourth, and fifth gestalt factors. It is evident from
what we have written to this point that the order of dismissal
cannot plausibly rest on the existing assessment of
reasonableness.
VI. CONCLUSION
VI. CONCLUSION
We have come full circle, back to our beginnings. The
Boit framework is an expression of pragmatism an authoritative
recognition, informed both by experience and by the demands
placed on the federal bench, that it is desirable for trial
judges, when feasible, to give meaningful, yet not unduly
burdensome, scrutiny to the question of jurisdiction at the early
stages of particular types of cases. The pragmatic nature of the
framework requires courts to proceed with caution, mindful of the
risks of overapplication and of the need to give parties ample
notice and opportunity to demonstrate that jurisdiction is, or is
not, proper. In the case at hand, the district court failed to
provide these latter necessities to FMI. The court then
compounded its error by weighing extraneous elements in
attempting to strike a balance on reasonableness. Thus, we find
merit in FMI's appeal.
We need go no further. We vacate the order of
dismissal. On remand, the district court should alert the
parties in advance to the level of scrutiny that it will apply to
the pending motion and the factual questions to which the
standard will pertain. The court should also allow such further
discovery, if any, as may be desirable in light of its
29
intentions. At the appropriate juncture, the court may accept
submissions in such form as it deems proper and make its
determination on relatedness. If the court deems the basic
jurisdictional tests satisfied, it should then undertake a
reasonableness analysis that comports with our precedents.8 We
take no view on the ultimate resolution of the issues to be
addressed on remand, or on the proper weighing of the gestalt
factors. Our concern at this stage is primarily with the court's
methodology.
Vacated and remanded. Costs in favor of appellant.
Vacated and remanded. Costs in favor of appellant.
8We note that BWC's Rule 12(b) motion raised the matter of
forum non conveniens as an independent basis for dismissal. The
district court declined to reach that issue. See Foster-Miller,
848 F. Supp. at 277 n.4. On remand, this issue may be raised
again.
30