Too square

We all understood by now that under Wal-Mart, trade dress that is deemed to be “product packaging” — unlike “product configuration” — “may be inherently distinctive and therefore, registrable, without proof of acquired distinctiveness.” Now John Welch writes about a TTAB decision in a case called In re Icelandic Provisions, Inc. concerning a container for some kind of Viking food (no, not this) called “skyre.” The question presented by the distinctively-represented applicant, appealing a refusal to register product packaging on the Principle Register, was, how square is too square to be inherently distinctive?

In the TTAB, the test for what qualifies as inherently distinctive comes from Seabrook Foods. Under Seabrook, inherently distinctiveness in packaging is determined by examining (1) whether the design is a common basic shape or design; (2) whether the design is unique or unusual in the field in which it is used; (3) whether the design is a mere refinement of a commonly-adopted and well-known form of ornamentation for the particular class of goods viewed by the public as a dress or ornamentation for the goods; or (4) whether the design is capable of creating a commercial impression distinct from any accompanying matter.

So now for the skyre container. From the opinion:

The Examining Attorney . . . presents evidence showing that overall square and rectangular-shaped containers are commonly used in the yogurt and dairy industryand that tapered sides and rounded corners are also common features of containers utilized in those industries. [T]he Examining Attorney argues that Applicant’s parallelogram-shaped containers are merely a refinement of the square- and rectangular-shaped containers commonly used in the dairy industry.

Applicant contests the relevance of the Examining Attorney’s evidence showing a parallelogram to be a common basic shape because, according to Applicant, none of the evidence shows a three-dimensional figure having a horizontal cross-section in the shape of a non-rectangular parallelogram, let alone one having tapered sides and rounded corners like Applicant’s proposed mark. However, the evidence introduced by the Examining Attorney directly relates to three-dimensional representations of the two-dimensional figures shown.

As shown in the record, manufacturers of similar products use containers of various shapes, several of them having tapered side walls and rounded corners, to hold their goods, which also prominently display their word marks, logos and decorative images relating to their contents, some samples of which are shown.
The evidence submitted by the Examining Attorney shows that many yogurt and similar dairy products are sold in containers having square, rectangular and oblong shapes, i.e., three-dimensional configurations, having tapered sides and rounded corners. Applicant’s configuration, substantially consisting of the shape of a nonrectangular parallelogram in horizontal cross-section, is a recognized geometric shape and a mere common refinement of such configurations. As such, purchasers and prospective customers for Applicant’s goods would be unlikely to regard its three dimensional configuration as identifying and distinguishing Applicant’s skyr and indicating its source.

Thus doing you-know-what to the circle.  John remains with one question, however, focusing on this passage from the opinion:

The wording and design elements applied to Applicant’s container contribute to the overall trade dress, and we see no basis for a finding that the shape of Applicant’s container creates a separable commercial impression. . . .

Prospective customers viewing yogurt and related dairy product containers including those shown in the record and those used for Applicant’s goods, are likely drawn to the products by viewing the prominent word marks, logos and the decorative colors used on the packaging i.e., indicia other than the relatively common geometric shapes of the product containers.

Square in a good way

John asks:

If, in considering a container shape, the Board looks to the labeling and other indicia on the container, can any container be inherently distinctive? I.e., won’t the consumer nearly always be “drawn to the products by viewing the prominent word mark, logos and the decorative colors?” Is a survey a must in these cases?

Good question, and it jumps out at you when you consider the fourth Seabrook factor: whether the design is capable of creating a commercial impression distinct from any accompanying matter.  How did the TTAB not turn that standard entirely on its head in making a reckoning with reference to “accompanying matter,” i.e., labels and logos?

But I’ll raise him one.  Let’s back up to the how-unusual-is-your-square question.  Again, from the Board:

Turning to the second factor, i.e., whether the subject matter sought to be registered is unique or unusual in the field in which it is used, Applicant contends that nothing in the evidence demonstrates that the applied-for mark is not “unique or unusual” in its field. Admittedly, there is no evidence showing use of the same exact configuration as that used by Applicant. . . .

[But under the third factor, whether the design is a “mere refinement, t]he evidence submitted by the Examining Attorney shows that many yogurt and similar dairy products are sold in containers having square, rectangular and oblong shapes, i.e., three-dimensional configurations, having tapered sides and rounded corners. Applicant’s configuration, substantially consisting of the shape of a nonrectangular parallelogram in horizontal cross-section, is a recognized geometric shape and a mere common refinement of such configurations.

This raises the question of how the second and third Seabrook factors are supposed to interact.  Again:  Number (2) is whether the design is unique or unusual in the field in which it is used, and (3) is whether the design is a mere refinement of a commonly-adopted and well-known form of ornamentation for the particular class of goods viewed by the public as a dress or ornamentation for the goods.  So here the Board admits, yeah, true, no one has ever done this before — but it’s a mere refinement.

A distinct honor

So what is the point of Seabrook factor number (2)?  If a unique design is always subject to being deemed a mere refinement, uniqueness is irrelevant.  If, on the other hand, something is found not to be a mere refinement, I don’t need uniqueness; uniqueness is axiomatic to “not merely a refinement.”

So let’s read two of the case squibs cited by the Board.  One is J. Kinderman & Sons Inc., 46 USPQ2d 1253, 1255 (TTAB 1998) [can’t find a link], in which the TTAB comes out and says “while the designs [of packaging for lights] applicant seeks to register may be unique in the sense that we have no evidence that anyone else is using designs which are identical to them, they are nonetheless not inherently distinctive.” This is almost the same exact formulation as in our skyre case: We’ve never exactly seen it before, but it’s not that unusual.

Digging a bit deeper in time, we have In re Right-On Co. Ltd., in which the TTAB took this inherently non-distinctive distinction by the horns, holding, “While the [pocket stitching] design may in fact be unique, i.e., it is the only such design being used, it also must possess an ‘original, distinctive and peculiar appearance,’” citing In re McIlhenny Co. Well, if it doesn’t have an original, distinctive and peculiar appearance, how can you say it’s the only such design being used?

Gryberg

I’m not the first person to wonder. In “The Judicial Role in Trademark Law,” 52 B.C. L. Rev. 1283, 1316-1317, Michael Grynberg writes, quoting Seabrook in the set-off passage below:

Secondary meaning is a familiar trademark standard, one generating a substantial body of case law over decades.  This test could be imported directly to evaluation of product design trade dress without significant modification. That is, the question whether consumers have come to associate a product design with a single source is the same question asked of descriptive trademarks and trade dress in the traditional secondary meaning inquiry.

In contrast, trademark law has not developed a similarly predictable means for determining whether trade dress is inherently distinctive. Some courts try to impose the spectrum of distinctiveness used for classifying word marks. They try to apply adjectives developed for words to shapes, and consider whether a particular design is fanciful or descriptive. Others try to tailor a test for determining whether trade dress is inherently distinctive and ask whether

it was a “common” basic shape or design, whether it was unique or unusual in a particular field, whether it was a mere refinement of a commonly-adopted and well-known form of ornamentation for a particular class of goods viewed by the public as a dress or ornamentation for the goods, or whether it was capable of creating a commercial impression distinct from the accompanying words.

To a large extent this “test” does little more than restate the question. How do we know if a design is “capable of creating a commercial impression distinct from the accompanying words?” By knowing whether it is distinctive. And we know that . . . how?

When we see it?

Yes, it’s pretty tricky to prosecute these marks, or to protect them in litigation. Sure, I can posit an understanding of the difference between “unique, unprecedented” and “inherently distinctive,” but it depends a lot on adjectives such as “very” or “really” or adjectival phrases such as “a whole bunch.”

Not such a square deal, really — for clients, their diligent counsel, or trademark law.

 

Ron Coleman

I write this blog.

One thought on “Too square

  1. It’s a test about artwork. Trying to come up with a concrete legal test for that is probably impossible. Trade dress, design patents, obviousness, and . . . what is considered “indecent” are all “know it when you see it” type tests. Not surprised!

Leave a Reply

Your email address will not be published. Required fields are marked *