Case Name: Starbucks Corporation v. Sardarbuksh Coffee & Co. 

Citation: CS (Comm.) 1007/2018 Delhi High Court 

Court: Delhi High Court

Coram: Manmohan, J. 

Introduction 

Starbucks Corporation v. Sardarbuksh Coffee & Co. is an interesting case at the cross-over of world domination and commercial copy-cat. It asks the question now in trademark law that few could put as clearly as it is thus: When does cultural adaptation become a “deceptive similarity”? As the December 2017 Delhi High Court settlement decree illustrates, the answer is not the actual meaning of a mark, but in commercial logic or structure of the common impression. Seattle-based Starbucks Corporation, a global brand with more than 30,000 stores in over 80 countries, first gained a presence in India through a joint venture company with Tata in 2012, and had a more streamlined retail presence by 2015. It is widely recognized as a trade dress (twin-tailed Norse siren surrounded by concentric circles with the brand name, in the graphic design with the green and black color scheme) and has acquired a well-known mark status under Section 11(6) of the Trade Marks Act, 1999 in accordance with decades of transnational use and vast secondary meaning. Sardarbuksh Coffee & Co. came up with its brand on the same background and introduced a mark with a mix of elements from the Norse siren and a turbaned Sikh commander (a sardaar figure) that copied almost perfectly the concentric circular geometry, the bilateral typography, the spatial hierarchy of the central image, and the predominate green and black palette. The phonetic approximation SARDARBUKSH as a structural echo of STARBUCKS added to the visual layering. In 2017, Starbucks sent cease and desist letters. In absence of compliance, it initiated a suit under CS (Comm.). Application for an injunction, rendition of accounts and damages 1007/2018. Eventually, the parties settled, with the defendant changing its name to “Sardarji-Bakhsh” and modifying the colour scheme resulting in a “consent decree” as the court termed it.  

Architectural Mimicry and Total Commercial Impression  

Trade Dress The analytical core of this case is the doctrine of trade dress and any statement that limits it to a “sound-alike” issue is a disservice to the doctrine. Under Section 2(zb) of the Trade Marks Act, 1999 (which covers “shape of goods, their packaging and combination of colours”) and as developed by the courts in Zippo Manufacturing v Zippo Dot Com and by the domestic courts in Colgate Palmolive v Anchor Health, trade dress is protection for the totality of a brand’s visual appearance as perceived by the marketplace. The current law is the Total Commercial Impression (TCI) test, which asks whether a commercial impression of the defendant’s mark, as seen by the average customer acting with ordinary care, causes a likelihood of confusion with the plaintiff’s mark. The test is designed to be anti-dissectionist, meaning that it forbids courts from dissecting a mark into its elements and then assessing each of these elements separately. The question under TCI analysis is not whether the turbaned sardaar is visually similar to the Norse siren, but whether the gestalt of the Sardarbuksh branding how the figure is used, how it is presented and positioned, how it is applied and the structure in which that figure is used, is similar to the recognizable architecture of the Starbucks trade dress in the first instance. That architecture can be broken down into 3 distinct layers. The first is the concentric circular geometry: the same geometric principle is used in both marks, where the nested circular frames provide the primary organizational principle, making them a medallion format. This is not a universal coffee-store design this is a design decision that Starbucks has made that’s made it unique around the world. In addition, the bilateral typographic placement, where both marks are positioned along the inner circumference of the outer circle, and the central image is placed in the same relative spatial area. Third, the green-and-black chromatic dominance: This combination of field color (forest green) and detailing color (black) has more than just been a preferred color; it has become a signifier of source, and a field color that Starbucks have exclusively used for their coffee in India, with sustained use globally. The substitution of the siren with sardaar by the defendant was not a legally sufficient change to remove the protection granted for trade dress under Section 29(2) as the protection is granted on the structural architecture rather than on the literal iconographic content of the central image. Culturally different, but functioning not that much differently than the siren, the turbaned commander was positioned at the centre of the medallion, surrounded and enveloped in an identical typographic arc and the brand name in the same typographic embrace. The replacement of the icon was a superficial change which preserved the deeper iconological text. The doctrine of initial interest confusion adds to this analysis. For an urban Indian consumer, who is not a stranger to seeing brands every day from the posters on the streets, the navigation apps on two-wheelers and even as thumbnail images in food delivery apps, the threshold of recognition rests on peripheral rather than granular details like logos. If a consumer enters a Sardarbuksh outlet from fifty metres or looks at its icon on a Swiggy or Zomato app at lower resolution, it would be able to capture the layout of the medallion on the screen and the green colour palette before any disambiguation regarding its content or figures occurs. The Ninth Circuit’s statement in Brookfield Communications v. West Coast Entertainment that “initial interest confusion” is “actionable even if eventual purchase confusion is remedied,” is particularly relevant to this “scanning” activity. Although the term has so far not been adopted by Indian courts, the mischief, namely diverting the attention of consumers by simulating the brand in the early stages is clearly within the scope of the passing off doctrine, as stated in Perry v. Truefitt (1842) and in India, in Laxmikant V. Patel v. Chetanbhai Shah (2002) 3 SCC 65. 

Precedent and Doctrinal Legacy 

This blog examines the need to modernize the Evidentiary Burden with the introduction of Digital Footprints as per Bharatiya Sakshya Adhiniyam, 2023. The judgment is a consent judgment which precludes further evidentiary investigation. But, in the context of an adversarial trial, where this dispute is likely to proceed in the absence of the Indian Evidence Act, 1872, and in its new form, the Bharatiya Sakshya Adhiniyam (BSA), 2023, which came into existence from 1 July 2024, the issue of empirically proving actual consumer confusion and imperfect memory becomes all the more relevant. The traditional way to establish consumer confusion via witness depositions, consumer survey evidence and expert testimony is still viable but is clearly unhelpful for platform-mediated commerce. The BSA’s expansion of the definition of “electronic record” under Section 2(1)(t) (expanding the definition of “electronic record” to cover almost everything) and its new Section 63 (electronic record admissibility, successor to Section 65B IEA) greatly enhances a trade dress plaintiff’s evidentiary arsenal. The following three types of digital evidence would be highly probative. First, there is documentary evidence of misdelivered order data from food delivery platforms such as Zomato, Swiggy and others; when consumers order food from the “Starbucks” listing on Zomato/Swiggy but ended up ordering from Sardarbuksh or vice versa. Platform operators are responsible for storing the logs of all search queries, click-through paths and corrections of orders on the server side. These logs, certified by the authorized signatory of the platform entity under the BSA section 63 certificate mechanism, would be considered primary electronic evidence, and therefore often not subject to the duty to produce the originator’s device. Second, the social media mis-identification records: Instagram posts, Facebook check-ins and Google reviews by consumers that included the Starbucks hashtags, #Starbucks, #StarBucksIndia and the consumer’s photo caption with the wrong brand name are admissible as electronic documents under BSA Section 2(1)(t). Their retrieval by API export or by download from the platform as certificate certified data meets the certificate requirement. Such posts demonstrate the standard that Corn Products Refining Co. v. Shangrila Food Products AIR 1960 SC 142 calls the “imperfect recollection” of the consumer; the average consumer who does not have a “legally learned eye” to make a side-by-side comparison of marks. Third, WhatsApp and messaging application logs the targeting of friends or family to “the Starbucks near Connaught Place” by consumer communication to a friend or family member, when the relevant store turned out to be the Sardarbuksh, would be admissible under BSA Section 63 read with Section 57(1)(j) (electronic messages). The confusion which had been present in Arjun Panditrao Khotkar v. Kailash Kushanrao Gorantyal (2020) 7 SCC 1, under Section 65B IEA, regarding the identity of the “person in charge”, has been clarified, which will make consumer generated WhatsApp evidence more available to the plaintiffs practically. This body of evidence would all help a plaintiff to move past impressionistic survey evidence, into a more data-driven proof that the rate of actual confusion is X percent, rather than being algorithmic. This shift, from deponent testimony to platform log certification, is the one most important modernization in IP evidence that is offered by the BSA framework.  

The Settlement as Pragmatic Judicial Balancing 

The settlement arrived under the court’s supervision, and is enshrined in a consent decree by Manmohan J., which is a compromise of judicial recognition of competing interests. The defendant did not have to stop doing business altogether but only use the name “Sardarji-Bakhsh” to indicate that it maintained the cultural essence of the Punjabi sardaar without incorporating its structural phonetic and visual parasitism into the model, and change the color scheme from Starbucks’ exclusive green-and-black aesthetic. The court’s further explanation that the defendant will be able to file a claim against third parties who use the term BAKHSH is of great importance to the law as it gives some credibility to the concept that when such a term is rebranded by the defendant, they acquire their own proprietary interest in the term, which will encourage compliance and prevent free-riding by later copyists. A fine application of the consent-decree system to create positive externalities for the IP system. The court in Aya Singh Tirlok Singh v. Munshi Ram Atma Ram, MANU/DE/0014/1968, has pointed out that the courts have a role to play in moulding commercial rational choices through an implicit directive to pay back half of the plaintiff’s court fees, a procedural footnote but a reminder that the court is actively involved in shaping commercially rational outcomes. For example, it is in the interest of precedent, in that this case brings together several important propositions. First, it reiterates that foreign marks that are widely known in India receive strong protection against the copying of the trade dress in India, under the interplay of the provisions of passing off and the well-known mark provisions provided by Section 11(6)-(10) of the Trade Marks Act. Secondly, it indicates that the Total Commercial Impression test will be used in a sophisticated manner because it will enable it to break through iconographic substitution. Third, it shows that Indian courts, both in settlement proceedings, will craft a nuanced result, allowing for coexistence, but not suppression, when the underlying business is a valid one, but the brand is parasitic. The decision is a blueprint for brand protection strategy in the Indian market: As brand owners, well-known international marks can and should exercise their rights under the doctrine of “trade dress” in India on the basis of “structural similarity”, while the BSA’s “digital evidence” architecture now provides brands with the evidentiary tools to establish, with empirical rigor, the consumer confusion required for the proof of passing off and infringement.