Positively Perfect in Every Way

Positive Technologies whiffed the first patent infringement complaint by filing in the name of the wrong entity, Positive-California, when it should have been filed in the name of Positive-Nevada.  Positive calls a mulligan and refiles in the correct entity's name, then the two companies merge into Positive Technologies, Inc.

The standing problems aren't over yet though.  It turns out that the inventor had assigned a 25% ownership interest to a former attorney as part of a fee agreement long before the present suit.  After the present suit was filed and this wild ownership interest uncovered during discovery, the lawyer assigned his ownership interest to Positive, effective on the date of the original assignment to him (N.B., the court refers to the lawyer's assignment to Positive as the "reassignment," although it also says the first assignment was from the inventor, not Positive, to the lawyer.  I'll stick with using the court's denomination of it as a "reassignment.").  The reassignment did not provide for an assignment of the right to sue for past infringement and the attorney also retained the right to enforce his original fee agreement.

A patent infringement suit must be brought by all the owners of a patent, so the defendants raised four arguments for lack of prudential standing:

First, the defendants argued that the lack of standing at the time suit was filed cannot be cured by a nunc pro tunc assignment.  Wrong; other cases where standing was denied were where plaintiffs had no ownership interest at all at the time of suit, i.e., they lacked constitutional standing.  Here, Positive had 75% ownership at the time of suit and lacked only prudential standing, so the standing problem could be cured post-filing. 

Second, the defendants argued that the reassignment didn't grant the right to sue for past damages, which just happened to be when the infringement occurred.  Nunc pro tunc to the rescue again; there was no need to grant the right to sue for past damages since there was no gap in time between the assignment and reassignment.

Third, the reassignment was only to "Positive," not specifying whether it was Positive-Nevada or Positive-California, so it was ambiguous.  No matter; while the grant was effective on a date when both entities existed, it was executed on a date when there was only one "Positive," so no ambiguity as to the assignee.  

Finally, the assignment was illusory because the lawyer did not give up his right to attorney's fees pursuant to the fee agreement, and the fee agreement included the original assignment.  Wrong for the last time; the lawyer did not retain any legal rights to the patents after the reassignment, but only at most an equitable interest.  The equitable interest does not divest Positive of the legal title needed for standing.

I have a hard time swallowing all of this; it seems to be a have-your-cake-and-eat-it-too situation.  The plaintiff succeeded in cutting out links in a chain of title that should be relevant.  The "reassignment" wasn't an assignment-back, it was an assignment to an entirely different party.  Doesn't that affect the past damages question - why does the court talk about a "gap" when it's just two separate transactions between three parties that happened to be effective on the same day?  Why should we assume that the lawyer assigned all that he possessed when he didn't say so, particularly when he clearly was trying to retain some kind of interest in the whole matter?  But if the nunc pro tunc reassignment was to be construed as effectively a nullity (the "no gap" theory), then shouldn't it matter that there were two potential assignees on the effective date?

This "reassignment" was created during the litigation, and, given the plaintiff's success, it was masterfully done.   I like to think about why certain decisions were made and whether there's a reason they made sense, even though in a decision they may just come out looking like sloppy work.  Plus here you have the wild card lawyer who has something good that he knows you need.

You can just imagine that it would be tricky negotiating against a non-patent lawyer who still wants his fees and plans on using his claim to past patent infringement damages somehow to collect them, so you just don't mention them in the agreement and hope for the best (how often have you negotiated an agreement where you just leave it unstated in the belief, or hope, that your position is the better interpretation?).  Is there a side agreement with the lawyer?  And stating that the assignee was just "Positive," rather than exactly who, so you could play it any way you needed to, turned out to be a great call.

Nunc pro tunc was the key, suit proceeds.  Eastern District of Texas, you know.

Positive Technologies v. LG Display Co., No. 2:07 CV 67, 2008 U.S. Dist. LEXIS 73087 (E.D. Tex. Sep. 24, 2008).

© 2008 Pamela Chestek

No Trademark Law at All

I think one of the reasons I like trademarks so much is their split personality. On one hand, they are the most ephemeral of intangible property, only a symbol representing a collective ethos. On the other hand, they can be the subject of the most ordinary transactions, buying and selling, securing a loan. A non-precedential decision from the 6th Circuit, Guaranty Residental Lending, Inc. v. Homestead Mortgage Co. on appeal from a Michigan federal court, demonstrates how a trademark-as-asset can be used to defeat trademark-as-symbol.  Indeed, the case begins, "This appeal of a dismissed trademark suit involves no trademark law."

As typical in these cases, the facts are complicated. Below is a roadmap:

The players:

Plaintiff is Guaranty Residential Lending (GRL) 

Defendant and Counterplaintiff is Homestead Mortgage Co. (HMC)

Non-party former owner of the federally registered trademark HOMESTEAD MORTGAGE is Bob Fitzner, Inc. (BFI)

Counterplaintiff is Bob Fitzner (Fitzner).

The timeline:

1987: GRL first used the marks Homestead USA and Homestead Mortgage in Michigan

1993: BFI filed a federal application for the mark "Homestead Mortgage"

1996: BFI registration issues

May, 1998: A predecessor to GRL (confusingly named "The Homestead Mortgage Company") files an ITU application for the mark Homestead USA, which was refused in 2001 after ex parte appeal on the basis of likelihood of confusion with the BFI mark

March, 2001: Texas forfeited BFI's corporate privileges for failure to pay franchise fee so BFI cannot sue or defend in Texas state or federal courts. Under Texas law, the beneficial ownership of assets passes to the shareholders (here, Fitzner) who can sue or defend in Texas courts on behalf of the company to protect the shareholder's interest

August, 2001: Fitzner files for Chapter 7 bankruptcy personally but did not list the beneficial ownership of trademark as an asset

January, 2002: Fitzner bankruptcy closes (beneficial ownership of trademark remains with the estate because it was unscheduled)

March, 2002: Texas forfeit's BFI's company charter

February, 2003:  Fitzner bankruptcy reopens

Before August or September, 2004: HMC begins using "Homestead Mortgage" in MI

December, 2004: GRL sued HMC for trademark infringement

January 17, 2005: BFI assigns the Homestead Mortgage mark to Fitzner for $1.00, effective March, 2002

January 26, 2005: Fitzner licenses the mark to HMC for $10,000

July, 2005: HMC answers the complaint and Fitzner and HMC file a counterclaim for likelihood of confusion

January, 2006:  Fitzner bankruptcy closes for the second time

GRL claimed that Fitzner lacked the capacity to sue on the trademark counterclaim, since he was not the owner of the Homestead Mortgage mark. The issue in the case was whether, under Texas state corporate and tax law and federal bankruptcy law, the assignment of the mark to Fitzner in 2005 was valid. To summarize a lengthy decision as succinctly as possible, yes.

The court of appeals held that when the corporate charter was forfeited, the beneficial and legal ownerships of the trademark were rejoined. The forfeiture of BFI's charter began a wind-down for a period of three years during which time the corporation has the right to sue or defend in any court and Fitzner no longer had any interest in the mark. Thus, while Fitzner held beneficial title to the mark as of March, 2001, before he filed for bankruptcy, the interest was extinguished when BFI dissolved in March, 2002. At that point in time, BFI held full title to the mark and Fitzner, as the only officer of BFI, had the authority to liquidate and thus assign the mark to himself personally. Any objections to the assignment would have to have been raised by a creditor, not GRL.

The court gave two other reasons why the unscheduled asset was not problematic: first, the forfeiture of corporate privileges affected on the right to sue in Texas courts, not Michigan, so BFI always retained those rights. Second, the court raised an argument sua sponte in dicta, that the bankruptcy code exempts from estate property any power that the debtor may exercise solely for the benefit of an entity other than the debtor. Since under Texas law the beneficial ownership to sue and defend is exercised on behalf of the corporation, not the shareholder personally, Fitzner was not required to schedule it.

Even though Fitzner was (again) in bankruptcy at the time BFI assigned the mark to him, it was an after-acquired asset and not subject to the bankruptcy. Any objections to the assignment would have to be raised in the bankruptcy case (not this trademark case) by a creditor or trustee; further, the time for any objection had past.

So, well-played by HMC, by successfully engineering a counterattack against plaintiff GRL through licensing of a mark. But is HMC really licensing BFI's "mark"?  There is no suggestion that HMC had any relationship with BFI before taking a retroactive license; in other words, BFI and HMC were just two unrelated companies that happened to use the same words, but the words had entirely different sets of meaning (i.e., goodwill) for their respective customers. And speaking of "naked licensing," how could BFI exercise control over HMC's use of the "mark" when there was no relationship whatsoever between the two companies before January 26, 2006? Can one exercise control retroactively?

This is a case where the realities were defeated by the formalities, at least so far.  It will be interesting to watch.

Documents from the district court proceedings here, here and here, in and of themselves an interesting read.

Guaranty Residential Lending, Inc. v. Homestead Mortgage Co., Nos. 07-1773/07-1815, 2008 U.S. App. LEXIS 19025 (6th Cir. Sep. 4, 2008).

© 2008 Pamela Chestek

I Vote for "Belief"

It's off topic, but I am from Massachusetts, it is the Red Sox, and it's almost the playoffs.  The TTABlog reports that a mark was refused registration on the basis that, inter alia, it disparaged the Red Sox.  If disparagement must be of "persons, living or dead, institutions, beliefs, or national symbols," which is the Red Sox?

© 2008 Pamela Chestek

Licensing, Naked or Clothed

In the Moose Tracks ice cream post, I mentioned that it didn't look like a naked license situation because Denali was doing everything necessary to defeat a claim of naked licensing. Nevertheless, there was some suggestion in the survey that consumers didn't associate the "Moose Tracks" flavor with any particular source, i.e., that they thought "Moose Tracks" is a common flavor name, not a trademark.

The problem is that the doctrine of "naked licensing" has taken on a life of its own. Federal law is a creature of statute, yet some cases don't even bother reciting the statutory basis for the naked licensing doctrine. For example, one textbook case explains it this way: "it is well established that where a trademark owner engages in naked licensing, without any control over the quality of goods produced by the licensee, such a practice is inherently deceptive and constitutes abandonment of any rights to the trademark by the licensor." Barcamerica Intern. USA Trust v. Tyfield Importers, Inc., 289 F.3d 589, 598 (9th Cir. 2002). And this, "A trademark owner, for example, can abandon all trademark rights through uncontrolled or 'naked' licensing. 'If a trademark owner allows licensees to depart from its quality standards, the public will be misled, and the trademark will cease to have utility as an informational device.'" TMT North America, Inc. v. Magic Touch GmbH, 124 F.3d 876, 885 (7th Cir. 1997).

But it shouldn't be a question of whether there is a naked license per se; "naked" or "license" do not appear anywhere in the Lanham Act. Rather, it should be a question of statutory abandonment, i.e., whether "any course of conduct of the owner, including acts of omission as well as commission, causes the mark to become the generic name for the goods or services on or in connection with which it is used or otherwise to lose its significance as a mark." Lanham Act § 45, 15 U.S.C. § 1127. 

The problem with identifying "naked licensing" as the issue is that we then look only at the formality of the licensing arrangement, rather than the reality of consumer recognition. And the naked licensing doctrine is a very poor proxy for what consumers may or may not be thinking.

In the Moose Tracks situation, the court said "Denali licenses its registered trademarks and provides the formulas, or recipes, to the licensees. Denali has control over which companies receive licenses . . . . Under its licensing agreements, Denali has control over the quality of the product by requiring the licensees to buy ingredients from authorized vendors. Denali has control over the way its 'Moose Tracks' trademark is presented, by requiring that licensees obtain its approval for advertising and carton design." Thus, apparently no naked licensing. Denali therefore may well be successful in defending itself against a charge of abandonment through naked licensing, but nevertheless the trademark could be sliding into genericism because whether a license is "naked" really isn't the point.

© 2008 Pamela Chestek

Barking up the Wrong Tree

I wrote an article proposing an analytical framework for resolution of disputes where two parties claim to own the same trademark, using settled law on the manufacturer-distributor relationship as a starting point.  The situation comes up fairly often and in this case the tip off is early - "This dispute stems from the shifting business relationship" between the parties. The court didn't have much trouble deciding who owned the mark, though; no strenuous legal analysis required.

Multi-Vet is the plaintiff, Premier the defendant, and the goods are anti-bark spray collars for dogs.  In 2001 Premier entered into a license agreement with Multi-Vet to become the exclusive distributor of its patented anti-bark spray collars in the United States under the trademark "Gentle Spray."  Multi-Vet registered the mark. The 2001 Agreement provided that if the agreement was terminated, Premier could buy the trademark for $1.00.

The parties next entered into a "First Amended and Restated Licensing Agreement" in 2003, which only affected financial terms of the 2001 agreement.  Then the two companies renegotiated their relationship in 2004 and entered into an "Exclusive Distribution Agreement."  The 2004 agreement said that all "intellectual property" used by Multi-Vet in association with its products was the exclusive property of Multi-Vet, with no mention of the potential assignment described in the 2001 agreement. The agreement also had an integration clause stating that it merged, superseded and cancelled the 2003 agreement.

The 2004 agreement was terminated in 2005 and the parties entered into a non-exclusive distribution agreement, including a license to the Gentle Spray trademark.  The distribution agreement was terminated in 2006 but Multi-Vet continued to provide the collars to Premier until 2007.  After that, Premier began manufacturing its own collars.

Multi-Vet brought a trademark infringement claim against Premier in April, 2008 for Premier's use of "Gentle Spray Bark Citronella Anti-Bark Collar" and "Gentle Leader Spray Sense Anti-Bark Collar" (complaint here).  

The conflict addressed in the opinion, however, is on a motion for preliminary injunction by Premier on an infringement counterclaim.  Premier counterclaimed against Multi-Vet's own use of the Gentle Spray mark and trade dress; you can see the two bear a strong similarity:

  

Premier argued that it owned the Gentle Spray trademark, a predictable loser of an argument.  Premier was largely defeated by the evidentiary presumption that Multi-Vet's registration provided.  The purchase option for the mark under the 2001 agreement didn't change anything; it was superseded by the 2004 agreement and later agreements made no mention of a potential transfer.  Letters from Multi-Vet to Premier assuring Premier that it had the exclusive right to use the mark were consistent with the terms of the licenses, not a reflection of the ownership of the mark.

Premier had also itself registered a "No Shock/No Pain" mark.  Multi-Vet claimed to have developed the No Shock/No Pain mark itself in Canada in the mid-1990's, and pointed out that the 2004 agreement confirmed Multi-Vet's ownership of all marks used in association with the product.   This was enough evidence to rebut the presumption of ownership of the "No Shock/No Pain" mark to a degree sufficient to defeat the grant of a preliminary injunction against Multi-Vet.  For good measure, Multi-Vet's use was not likely to be confused with Premier's use because it was so small (the small circle on the left of the back of the packaging):

A packaging trade dress claim failed for procedural reasons; Premier had not pleaded ownership and infringement.

No likelihood of success on the merits, therefore no need to go on to the "irreparable harm" prong of the test for preliminary injunction.  Preliminary injunction on related state law claims also denied.

A procedural faux pas (or a deliberate choice) - only Premier filed a motion for preliminary injunction, which was denied.  Multi-Vet had not, so the products are currently coexisting in the marketplace.  Premier's replacement trademark, Spray Sense, is on the web as "formerly Gentle Spray," so it looks like Premier is phasing out its use.  Query whether the harm to Multi-Vet is even greater, though, if consumers believe that "Gentle Spray" is being phased out when Multi-Vet itself sells it.

Multi-Vet Ltd. v. Premier Pet Prods., Inc., 08 Civ. 3251 (LMM), 2008 U.S. Dist. LEXIS 67466 (S.D.N.Y. Aug. 22, 2008).

© 2008 Pamela Chestek

And the Survey Says --

I had a great response to my survey; my thanks to everyone who participated and to those who drove traffic to the survey, particularly Marty at The Trademark Blog and John at The TTABlog. Here's a link to the original post and here's a pdf version of the blank survey questions, in case you missed it.

The mark being tested was MOOSE TRACKS. 210 people answered the survey questions, 7 from outside the U.S. Here are the results for the U.S. in a snapshot:

	Term	Percentage responding "brand"
1.	Cherry Garcia (brand)	96.0%
2.	Blue Moon (generic)	80.0%
3.	Founder's Favorite (brand)	70.0%
4.	Moose Tracks (test)	64.2%
5.	Death by Chocolate (generic)	61.5%
6.	Jamoca (brand)	57.3%
7.	Rocky Road (generic)	20.5%
8.	Cake Batter (brand)	15.5%
9.	Vanilla (generic)	1.5%
10.	Coffee (generic)	1.0%

The Story

Sometimes you read a case thinking it’s a slam dunk and it comes out just the opposite of what you expected. I recently read Blue Bell Creameries, L.P. v. Denali Co., LLC, about the ice cream flavor "Moose Tracks." My immediate reaction was "but that's a generic flavor!" and I was surprised that genericism wasn't raised as a defense. Moose Tracks ice cream is, to me, vanilla ice cream with a chocolate ribbon and peanut butter-flavored inclusions; the name comes from what a moose track looks like (you might not want to go there). This photo from my camera phone is what I think of (“Moose Tracks” immediately below “Lobster Tracks”):

The decision was only a standard run through the likelihood of confusion factors and the four part test for preliminary injunction, ending with an injunction against DJ plaintiff Blue Bell’s use of MOOO TRACKS. Not only was there no mention of genericism, but the court said "there is no meaningful dispute that Denali’s registered trademark 'Moose Tracks' is a strong mark."

So I asked my hairdresser:

(Me): Have you ever had moose tracks ice cream?

(Him): Look at me, what do you think?

(Me): What's in it?

(Him): Oh, chocolate chips, lots of stuff, depends on which one. Have you ever had Snickers?

(Me): No.

(Him): It's good.

(Me): Who makes moose tracks?

(Him): Lots of companies, you know, it's like all those kinds of ice cream.

I live in New England, where moose are a common theme in advertising and naming, so maybe my reaction was a regional thing. I did some more research. I found many web sites for ice cream shops that list Moose Tracks as an available ice cream flavor with no suggestion it might be anything other than a flavor; I found Moose Tracks ice cream made with "famous Moose Tracks fudge." I found many ice cream shop web sites that use the ® symbol with no suggestion that anyone but the store owned the trademark, but I also found a fair number that mentioned Denali.

Denali owns a number of trademark registrations for variations of MOOSE TRACKS for ice cream and fudge, which it licenses to a number of ice cream manufacturers. From Denali's web site, by my count "Moose Tracks" ice cream is sold under at least 60 different brand names. Here are some licensees:

You can see the Denali moose logo on each container (superimposed over the photo of the ice cream), although there is nothing similar about the font for the words "Moose Tracks" or the trade dress of the cartons. Nevertheless by all indications this was not a "naked license" situation; the decision said "Denali licenses its registered trademarks and provides the formulas, or recipes, to the licensees. Denali has control over which companies receive licenses . . . . Under its licensing agreements, Denali has control over the quality of the product by requiring the licensees to buy ingredients from authorized vendors. Denali has control over the way its 'Moose Tracks' trademark is presented, by requiring that licensees obtain its approval for advertising and carton design."

I had some reason to have assumed that Moose Tracks was just a flavor and not a brand, but was I the only one? So, with a HT to my trademark professor, I decided to do a Teflon survey.

First the disclaimer - it's not a particularly reliable survey. For starters, the surveyed population is made up largely of trademark professionals, hardly a representative population for a genericism survey (albeit a population with a good appetite for ice cream). I asked questions that may not be kosher for an evidentiary survey, but that I hoped might provide some insight into people's responses. I did not qualify respondents and anyone could have stuffed the ballot box. I’m making assumptions based on some fairly low numbers of responses where the margin of error is far larger than the percentage swing. My interest is in the conversation, not finding the “right” answer.

For my survey I wanted a range of choices in the distinctiveness spectrum, so I looked for ice cream flavors, both trademarks and generic, that had lesser or greater descriptiveness (the references I used in deciding whether a flavor is a trademark or generic are hyperlinked):

Test word: Moose Tracks	Ingredient description only	Reference to ingredient with another word	No reference to ingredient
Generic	Vanilla Coffee	Death by Chocolate (Wikipedia,Google)	Rocky Road Blue Moon
Trademark	Cake Batter (® Cold Stone Creamery)	Cherry Garcia (® Ben & Jerry’s) Jamoca (® Baskin Robbins)	Founder’s Favorite (® Cold Stone Creamery)

I would put them in roughly this order along the Abercrombie spectrum, highest to lowest (you may disagree):

1. Blue Moon

2. Rocky Road

3. Moose Tracks

4. Founder's Favorite

5. Cherry Garcia

6. Jamoca

7. Death by Chocolate

8. Cake Batter

9. Coffee

10. Vanilla

Now compare the flavors in order of brand recognition based on the survey results with their spots on the spectrum:

	Term	Brand name or generic term	Rank in distinctiveness spectrum
1.	Cherry Garcia	brand	5
2.	Blue Moon	generic	1
3.	Founder's Favorite	brand	4
4.	Moose Tracks	test	3
5.	Death by Chocolate	generic	7
6.	Jamoca	brand	6
7.	Rocky Road	generic	2
8.	Cake Batter	brand	8
9.	Vanilla	generic	10
10.	Coffee	generic	9

You can see that the little-known, regional generic flavor "Blue Moon" rated very high in brand recognition, demonstrating that we do indeed attribute a lot of trademark significance to the arbitrariness of an identifier. But commercial strength pushed Cherry Garcia to the top of the list, even though it has some descriptive character to it.

But I also asked whether people had encountered the flavors. The ice cream flavors from least to most often encountered were:

	Term	Percentage responding “never seen”
1.	Blue Moon	85.1%
2.	Founder’s Favorite	85.1%
3.	Jamoca	53.6%
4.	Cake Batter	41.5%
5.	Death by Chocolate	37.8%
6.	Moose Tracks	34.9%
7.	Cherry Garcia	4.6%
	Coffee (tie)	4.6%
9.	Rocky Road	2.6%
10.	Vanilla	.5%

I thought that if someone had never actually seen a flavor, his or her response would be more about the inherent distinctiveness of the mark because the person was, after all, just guessing based on the words alone. A comparison of the responses of those who had seen a particular flavor to those that hadn't might give me a different insight into the genericism of the mark.

Comparing the difference in people's responses based on this distinction showed some interesting results. First, the percentage of "don't know" responses dropped; if people had seen it they had formed an understanding. Taking "Blue Moon" again, if people had actually seen the flavor their understanding that it was a generic flavor name increased fairly dramatically:

Blue Moon (generic)	Brand	Common	Don't Know
All responses	80.0%	9.5%	10.5%
Had seen it (29 responses)	69.0% (-11.0%)	31.0% (+20.5%)	0.0%

The same pattern was true in the opposite direction for the brand Founder’s Favorite; for the most part it was the brand recognition that increased for those who had seen it:

Founder's Favorite (brand)	Brand	Common	Don't Know
All responses	70.0%	13.5%	16.5%
Had seen it (29 responses)	79.3% (+9.3%)	13.8% (+0.3%)	6.9%

Cake Batter showed the expected pattern for a brand; brand recognition increased and identification as “common” decreased among those who had seen it (although brand recognition is still very low):

Cake Batter (brand)	Brand	Common	Don't Know
All responses	15.5%	77.0%	7.5%
Had seen it (113 responses)	17.7% (+2.2%)	76.1% (-0.9%)	6.2%

Generic Death by Chocolate was as predicted; generally the recognition of it as a generic flavor increased:

Death by Chocolate (generic)	Brand	Common	Don't Know
All responses	61.5%	29.0%	9.5%
Had seen it (120 responses)	62.5% (+1.0%)	33.3% (+4.3%)	4.2%

But two flavors didn’t behave as one might expect; first, the brand Jamoca appears to be losing recognition amongst those who had seen it:

Jamoca (brand)	Brand	Common	Don't Know
All responses	57.3%	26.1%	16.6%
Had seen it (88 responses)	56.8% (-0.5%)	39.8% (+13.7%)	3.4%

Second, for Moose Tracks both identifications as brand and common increased:

Moose Tracks (test)	Brand	Common	Don't Know
All responses	64.2%	22.9%	12.9%
Had seen it (127 responses)	66.9% (+2.7%)	27.6% (+4.7%)	5.5%

The next survey question might provide some clues why. I also asked people to name the company that makes each flavor. It was Denali’s practice of licensing a flavor name to different manufacturers that I found curious. I had no problem with the concept of a flavor as a brand indicator, but only if it was always associated with the same house brand, like Ben & Jerry's (although of some interest, “Cherry Garcia” is actually a licensed mark, originally owned by Ben & Jerry’s but assigned in 1997 to the estate of Jerry Garcia). Can people recognize that, even though the ice cream itself was made by different companies, the flavor was from a single source?

According to the Denali web site, using Ohio as an example, these brands of ice cream are sold in a "Moose Tracks" flavor at the listed stores:

Ruggles, sold at Meijer, Giant Eagle, Discount Drug Mart, and Acme;

Pierre's, sold at Giant Eagle, Marc's, Acme, Tops, Heinan's, and Dave's;

Reiter, sold at Giant Eagle and Discount Drug Mart;

Giant Eagle Creamery Classic, sold at Giant Eagle;

Tofts, sold at Discount Drug Mart; and

Meijer, sold at Meijer.

So in a Meijer store you can buy two different brands of ice cream in Moose Tracks flavor, in a Giant Eagle store you can buy it in four brands, and so on. What ice cream brands did the survey show people associated with Moose Tracks ice cream?

Ben & Jerry’s is the epitome of successful marketing and commercial strength. Of the 154 people who both identified Cherry Garcia as a brand and provided an answer on the "what company makes it" question, four people didn't know, one person said Baskin Robbins, one said “t&j” and 148 said Ben & Jerry's. In fact, people's association of Ben & Jerry's with ice cream names is so strong that for every flavor, vanilla included, someone named Ben & Jerry's as the manufacturer.

Contrast the result in Cherry Garcia with the generic but more distinctive Blue Moon. Of the 50 people who both identified Blue Moon as a brand and provided an answer on the "what company makes it" question, 44 couldn't name a company, one said Hudsonville, one said Maggie Moo, three identified it as a beer (Marc, I think all yours), and one said the ever-present Ben & Jerry's. So while the responses still tended to “brand” (69%), they actually couldn’t come up with a maker.

"Moose Tracks" had the second highest number of responses to the “who” question after Cherry Garcia, with 58 answering both “brand” and responding on the “who” question. But unlike the universal association between "Cherry Garcia" and "Ben & Jerry's," 37 responses named 21 different companies as the manufacturer of the Moose Tracks flavor. “Moose Tracks” or “Denali” were listed six times, 9 different licensees (Alaskan Classics, Cascade, Gifford's, Hershey’s, Kemp's, Meijer, Pierre's, Spartan, and Velvet) were listed 9 times, and 22 times people listed 11 companies that are not licensees and who do not make a Moose Tracks flavors (Baskin Robbins, Ben & Jerry's, Blue Bell (oops, the DJ plaintiff), Blue Bunny, Breyer's, Dryer’s, Eddie’s, Edy’s, Emack & Bolio's, Friendly's and Perry's). Twenty people couldn’t name a brand, and one said “all.”

One more statistic, I had asked about packaging. My prediction was that people who more often ate ice cream from an ice cream shop (like the top photo) would be less inclined to identify “Moose Tracks” as a brand. The results were the opposite:

Moose Tracks (test)	Brand	Common	Don't Know
Packaged (88 responses)	63.6%	29.6%	6.8%
Unpackaged (28 responses)	78.6%	21.4%	0.0%

We know that people don't have to be able to name the company for a mark to have trademark significance, but is it relevant that when they can name names, they name so many different companies, including so many wrong companies? If a flavor has brand significance when I see it in association with one brand of ice cream, what happens when I go to the store down the street, or look in the next case over, and see another brand with the same flavor? With different trade dress? Is this why those who ate packaged ice cream had a lower “brand” response? The licensing relationship between Denali and its licensees seems to be escaping consumers, why? My ice cream shop photo above is of a licensee’s brand, but “Maine Lobster Tracks” is a registered trademark of the licensee Gifford’s, not Denali - how does that affect the brand significance? Should it matter in a trademark lawsuit whether the consumers are identifying a common company as the manufacturer or many different ones, or is it simply enough that consumers think it’s a brand? Is the identification of many manufacturers evidence that the flavor is going down the genericism road but hasn’t reached the legal identity of “generic” yet?

Even though it was a question specifically about flavor, people overwhelmingly named an ice cream manufacturer in response. Can an ice cream flavor be a trademark and yet associated with many manufacturers? If one is going to use flavors as trademarks, do all flavors associated with the manufacturer have to be distinctive? Are there some kinds of marks, like a flavor, a slogan or distinctive packaging, that can function as a trademark when sold with a primary indicia, i.e., manufacturer brands, but can’t live alone? If so, where does that fit into our standard confusion analysis?

Epilogue

Thanks to the seven of you outside of the U.S. who responded. The only flavors the majority of respondents had seen were vanilla, coffee, Death by Chocolate and Cherry Garcia, and the flavors fell pretty much in the order of inherent distinctiveness (Founder’s Favorite at 89.7%, Blue Moon, Jamoca and Cherry Garcia tied at 71.4%, Death by Chocolate and Rocky Road tied at 57.1%, Moose Tracks and Cake Batter tied at 42.9%, and vanilla and coffee tied at 0.0%).

A few random comments on the survey results. The majority of the responses to "how often do you eat ice cream" was "monthly,” yet about 3/4 of the respondents buy their ice cream in packaging - so hats off to those of you who can manage to buy ice cream at the grocery store and not eat it daily.

Poor Cake Batter ice cream. A teaching moment - proof that you can't turn a generic name into a trademark, even if you get it registered. And you gotta love trademark lawyers; I had several people make comment along the lines of "I know Cold Stone makes Cake Batter, but I still think it's generic" (and they responded "common" on the survey).

I'd love to hear from the person who commented "A very misleading survey." How come?

Thanks to all who helped on this exercise; comments enthusiastically invited. Let it rip.