Dr. Rappeport claims that because the confusion levels changed
when modifications other than removal of the name were made that
such change cannot be the result of the name. Therefore, he
continues, any change in the confusion level approximates
confusion that results from factors other than the name. This
seems to do indirectly that which could be done directly by
removing the NutraSweet name from the box.
Appropriate controls would effectively eliminate the name as a
factor causing confusion. Here the name cannot be eliminated as
a factor causing confusion because the name was never eliminated
from the allegedly infringing box. It would have been
instructive to see how participants would have responded to a
variation of the NutraSweet box that replaced the well-known
brand name. Since this was not done we simply do not know.
Likewise as to the survey's use of the color blue. As the
court noted in its 1999 decision the color blue serves a
functional purpose in the aspartame-based sweetener market and
is not a protectable element of plaintiffs trade dress. Color is
generally not protectable and drawing a line between a shade and
a color is an exercise in hairsplitting in which this court will
not engage. The NutraSweet box was the only box in the second
room of a similar shade to the NatraTaste box. The controls do
not net out this legally irrelevant similarity.
In the most critical respects the revised survey is no better
than the previous ones this court discounted as seriously
flawed. The essential comparison made by the respondents is
nearly identical. As in the prior surveys the respondents saw
the NatraTaste box in one room and in the second room they saw a
variation of the original NutraSweet box along with the Four
Brands. Only one of the Four Brands boxes has a predominantly
blue background. This box is a very dark blue several shades
darker than the light blue hues of the NatraTaste and NutraSweet
boxes. None of the Four Brands has a name sounding remotely like
There was another obvious flaw in the new controls. The
variations of the NutraSweet box, both light blue and the darker
blue, with the coffee cup removed are inappropriate and
thoroughly unrealistic. When the coffee cup was removed it was
not replaced with any other graphic and every other detail on
the box was unchanged. In place of the coffee cup was a
conspicuous absence in the middle of the front panel. The
sweetener packet, tilted on angle so as to lean against the cup,
was left tilting in a seemingly random fashion. The results of
the survey indicate that these two non-coffee cup variations
caused the lowest levels of confusion. This is hardly
surprising. These boxes look unauthentic and contrived.
Comparing the confusion levels caused by these two so-called
controls to the other NutraSweet box variations cannot result in
any legally meaningful data.
Dr. Jacob Jacoby objected to Dr. Rappeport's use of "controls
that could never be compelled to be used in the marketplace and
are therefore unrealistic stimuli for present purposes."
Plaintiff responded to this criticism by pointing out that "a
sugar pill is the standard medical control." Plaintiffs reliance
on the placebo analogy is misplaced. Medical experiments test
the substance of the medication and its affects. In a medical
trial the placebo is designed to be visually identical to the
tested medication so that the participants cannot distinguish it
from the other medication. Participants are not supposed to know
they are taking something different.
By contrast, the value of a trade dress is its ability to
visually distinguish the product from others in the same market.
Visual modifications that cannot be enforced in
the market do not provide meaningful evidence of actual
confusion. The test is not whether the products can or do create
confusion in the abstract but whether that confusion affects
"the purchasing and selling of the goods or services in
question." Lang v. Retirement Living Publishing Co.,
949 F.2d 576, 582-3 (2d Cir. 1991). "Trade [dress] infringement protects
only against mistaken purchasing decisions and not against
confusion generally." Id. at 583 (quoting Restatement (Third)
of Unfair Competition § 20 (Tent. draft No. 2, 1990) reporter's
note at 179).
Any evidence of legally cognizable confusion provided by the
revised survey is purely indirect. Plaintiff cannot attribute
any of the confusion to any element of their trade dress nor can
it eliminate any irrelevant factors as sources of confusion. The
analysis of the survey results does not focus on the actual
number of confused responses but on changes and differences in
the number of those responses. It is unclear what these
differences tell the court. Plaintiff can only point to the
study for the proposition that certain modifications in the
NutraSweet box cause more confusion than others. That is a vague
and abstract conclusion and does not raise a genuine issue as to
any material actual confusion.
Dr. Rappeport testified that analysis of the "Why do you say
that?" question asked of respondents to clarify the reasons for
their confusion has value "depending on what you get out of the
control[s] and how clear the data is." The court agrees. Because
the new controls did not eliminate the name and color as factors
causing confusion and the data was not clear in its conclusions
the court examined the reasons that respondents gave for their
confusion. The court obtained the responses as recorded in Dr.
Leo Kaplan's declaration. Plaintiff did not dispute the accuracy
of the responses as reported by Dr. Kaplan so the court will
assume they are accurate.
The following table organizes the confused responses and lists
the reason(s) each respondent gave for their confusion.
Why do you say that [NutraSweet is made by the same company
that made NatraTaste?]
Quotes of reasons given, organized by category