California v. Cisco – “Caste” in the Valley

Author: Damo.

Editorial Note: Damo discusses the recent “caste” events in California. There are many serious implications to this. Serious issues for manthan . What is going to be the IK response ?

Are caste identity and casteism the defining feature of Indian American ethnic and religious interaction? Should American companies mandate caste sensitivity training courses? Those who scoff at the absurdity of such questions should perhaps pay attention to the California Department of Fair Employment and Housing (DFEH), a state agency charged with enforcing civil rights laws.

On June 30, CNN along with multiple news sites across the United States and India reported the DFEH was suing Cisco Systems for alleged caste discrimination on behalf of a John Doe ‘Dalit Indian.’ The agency claimed Cisco as a company with an over represented ‘South Asian Indian’ workforce, neglected to protect the rights of an employee “typically darker than the upper-caste Hindus forming the majority of PIOs in the United States” following a “strict Hindu religious and social hierarchy”.

The allegations themselves center on an ‘upon information and belief Brahmin individual’ who knew of Doe’s caste status through attending an unspecified IIT school at the same time as Doe. Eventually Doe alleges, he was not only outed as Dalit to his colleagues, but also lost out on promotions and raises due to the discriminatory attitudes of his upper-caste management and an unresponsive corporate HR department.

The State of California furthermore suggested that given 67% of Dalits in America encounter caste discrimination, this is an issue bigger than Doe, warranting companies employing substantial amounts of Indians to do something about it.[1] [2] And since according to their own complaint the caste system is part and parcel of Hinduism, this problem will only get worse.

“Although Cisco has employed a predominantly South Asian Indian workforce for decades, Cisco was—and continues to be—wholly unprepared to prevent, remedy, or deter the unlawful conduct against Doe or similarly situated lower caste workers. Cisco failed to take any steps whatsoever to prevent “. . . inequalities associated with [c]aste status, ritual purity, and social exclusion [from] becom[ing] embedded . . .” into its workplace, which is a documented problem for “. . . American mainstream institutions that have significant South Asian immigrant populations.”

So, where did the State of California learn so much about caste in America? Did they or another government agency in the United States or India conduct a survey? Did they consult an academic institution?

Guess again.

Such sweeping generalizations are a result of a convenience sample of responses collected through a Typeform survey by a group with the unassuming name of Equality Labs.

Founded by transmedia storyteller and past MIT Center for New Media Studies fellowship recipient Thenmozhi Soundararajan (aka Dalit Diva), the self-described “Ambedkarite power-building organization” appears to have gone from a modest project funded by a $2,000 grant in December 2016 to the de facto authority on caste in America. Apart from an unspecified angel investor, other backers include New Media Ventures, the San Francisco Foundation, and the North Star Fund.

Their mission emphasizes centering South Asians in America originating from marginalized communities; specifically women, queer, trans, non-binary, Dalits, or minority (non-Hindu) religious groups. Not included in the otherwise exhaustive list are lesbians, gays, bisexuals, or people identifying as ‘third gender’ a grouping with both traditional and legal currency in India, Nepal, and even Pakistan. Also notable is the group uses Tamil Eelam in place of Sri Lanka. 

Their blockbuster 52 page report quickly circulated throughout mainstream and diaspora media included a slew of shocking insights as seen in this helpful infographic:  

The fifteen minute survey was endorsed by a coalition of anti-caste discrimination groups[3] and distributed through direct contacts/community groups, generating 1,200 usable responses out of a total of 1,534.[4] These responses constituted the basis of not only their findings on the caste background and discriminatory attitudes of the Indian American population, but also the set of helpful recommendations and reference materials they prepared for schools, governments, and employers.

Figure 1: Perhaps if Cisco had hired an organization like Equality Labs to proactively address casteism, this whole situation could have been avoided!

Also notable were just under thirty percent of responses came from people identifying as something other than heterosexual – including 9.8% bisexual, 6.7% homosexual, 3.5% pansexual, and 2.3% queer.

Wait a minute.

Are Indian-Americans identifying as LGBTQ really double the Indian-Americans identifying as Muslim? Are there really more Indians identifying as pansexual than Jains? Have almost a third American Hindus been stuck in the closet the whole time? On what basis is this truly a representative sample capable of describing a complex issue like caste in a population of over 4 million people? Moreover considering the particular challenges LGBTQ Indian Americans face, to what extent are the issues reported intersectional?

Seriously, is a single convenience sample the definitive source on caste in America for American policy makers, media outlets, or for that matter anyone genuinely interested in advocating for South Asians from marginalized LGBTQ and caste communities?

Would Pew, Gallup, or other experiences researchers would make such broad claims about caste affiliation, income, and attitudes of the Hindu American population as a whole based on such evidence?

Indeed as a paper discussing the pitfalls of convenience sampling noted:

Of all the sampling strategies, convenience sampling is the easiest, least time-intensive, and least expensive to implement, perhaps accounting for its popularity in developmental research. Regarding its disadvantages, results that derive from convenience sampling have known generalizability only to the sample studied. Thus, any research question addressed by this strategy is limited to the sample itself. The same limitation holds true for estimates of differences between sociodemographic subgroups.

Unlike Equality Labs and the State of California, the National Institute of Economic and Social Research (UK) when evaluating caste discrimination in Britain took a different approach in their 121 page report written in 2010.[5] Despite clearly stating that “caste exists in Britain: this is not in dispute,” they stressed that uncovering the extent of caste discrimination in the UK would require a representative survey involving complexities, resources, and timelines beyond the scope of the study – something particularly challenging given the lack of pre-existing British population data segmented by caste.[6]

So why is the State of California uncritically parroting a deeply flawed study in a major lawsuit? Regardless of how the Doe case is resolved, the broader implications remain extremely concerning. To avoid legal liability, will Silicon Valley feel compelled to track employee caste affiliations (like how race or disability are already tracked), or institute mandatory caste-sensitivity training? Who would offer that training? What may have once sounded like absurd, even inflammatory questions suddenly aren’t so far-fetched.

Moreover, such a flawed study could easily be used to legitimize so-called ‘savarna denialism,’, while undermining open and honest discussions South Asian Americans (of all religious, caste, and national origins) need to have regarding sensitive topics – including with non-Hindus. Also obscured are efforts of Indians as part of an Asian American community take to address shared challenges such as colorism, LGBTQ inclusion, or relations with the African American community. It should go without saying that the blemishes of ancient and modern India do not invalidate legitimate concerns over how governments, employers, and educators treat what remains an ethno-religious minority community in the United States.

Currently, Equality Labs owns the narrative in mainstream media and progressive diaspora youth circles – further enabled by the American people’s collective ignorance. They want to ensure caste and Hinduism are intrinsically linked in the minds of regulators and even the American born children of Hindu immigrants – regardless of whether they maintain caste affiliations or not. If such voices remain unchallenged, even Devdutt Pattnaik may find himself one day getting canceled for pink-washing Brahmanical patriarchy. At that point, it might be too late to act.

---

---

References: –

[1] The complaint used content from Human Rights Watch, The Encyclopedia Britannica, Public Radio International, along with Equality Labs. Upon further review, the PRI story itself is largely based on content provided by Equality Labs – and their study forms the primary basis for the “facts” presented regarding the caste composition of Indian Americans (and as a result the primary source discussed here).

[2] Based on Equality Labs’ and the plaintiff’s account, the state’s complaint noted that “unlike Doe, most Indian immigrants in the United States are from upper castes. For example, in 2003, only 1.5 percent of Indian immigrants in the United States were Dalits or members of lower castes. More than 90 percent were from high or dominant castes. Similarly, upon information and belief, the same is true of the Indian employees in Cisco’s workforce in San Jose, California. Doe was expected to accept a caste hierarchy within the workplace where Doe held the lowest status within the team and, as a result, received less pay, fewer opportunities, and other inferior terms and conditions of employment….”

[3] Organizations included the Dalit American Foundation, Ambedkar International Mission (AIM), Ambedkar Association of North America (AANA), the Dalit American Women’s Association, South Asian Americans Leading Together, Associations for India’s Development, Alliance for South Asians Taking Action, and the Indian American Muslim Council.

[4] Methodology is discussed in Appendix II, beginning on page 37.

[5] Also notable is the decision of the authors to focus on jati – saying “…the cases of caste discrimination identified in the literature and in the qualitative case studies related to jati. Given the inclusion of Hindus in the study, the lack of reference to varna, but reference to jati, suggests that perceived caste discrimination (outside biraderi) is based on ‘occupational castes’ rather than ‘religious castes’ i.e. it is a cultural rather than religious construct.”

[6] The report included estimates of the British low-caste population by outside organizations, including caste-focused activist groups ranging from 50,000 to 200,000+.